From b.j.smith at ieee.org Wed Feb 1 10:14:55 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] Now Firefox 1.5 thanx to PrefBar 3.3 Beta 3 ... Message-ID: <20060201151455.64032.qmail@web34111.mail.mud.yahoo.com> FYI, I've finally moved away from Firefox 1.0.7 since PrefBar 3.3 Beta 3 largely works with Firefox 1.5. It's largely just a few cosmetic issues with the new version's beta that are left. http://prefbar.mozdev.org/ -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Wed Feb 1 11:34:21 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] Problems w/ KVM switch .... Message-ID: <43E0E30D.3010309@HiWAAY.net> .... I recently (this weekend) installed a Linkskey 4-port KVM switch so that I could monitor my 3 PCs from 1 monitor/kbd/mouse. The PCs are: 2.4 GHz P4, SuSE linux 9.2, all stock, running in run-level 3; 933 MHz PIII, SuSE 8.2, run-level 5; & a WIN2K Gateway. My problem is that whenever I switch *from* the Gateway *to* the 933, the mouse goes (& apparently stays) crazy. If I can eventually get to a console or shell window, I reboot. This is a hassle :-). Has anyone else had similar problems ? If so, how did you solve them ? TIA -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From b.j.smith at ieee.org Wed Feb 1 14:55:55 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] [Job] LAN Administrator (at my employer) Message-ID: <20060201195555.74391.qmail@web34111.mail.mud.yahoo.com> This is _not_ a 3rd party deal. This would be direct with my employer. Please contact me off-list if you are interested. I have no idea what the pay rate is (I'm not doing IT at this company). We're also still looking for C developers too. JOB DESCRIPTION XXX in Lake Mary, FL has an immediate permanent opening available for a LAN ADMINISTRATOR. Candidates should meet the following criteria before submitting information: Support a number of corporate infrastructure components related to their specific functional area of responsibility (Windows Servers). Install, configure, maintain, and monitor system hardware and software. Document system configuration and installation and administration procedures. Troubleshoot complex problems and resolve trouble tickets as assigned. Evaluate and modify established best practices for infrastructure maintenance. Perform system administration activities across multiple platforms including daily system capacity, performance monitoring, patch and fix maintenance, and backup/restore activities. Qualifications: Three or more years of networking experience, 3 or more years experience configuring Cisco routers in LAN and WAN environments. Requires 2 or more years experience with the following: TCP/IP, CSU/DSU s, VPN and Firewall appliances. Working knowledge of Cisco Catalyst switches and VPN technology is a plus. *** Must be able to manage multiple projects, organize priorities, escalate critical events and communicate effectively. *** BS/BA degree or related technical degree. Cisco/Microsoft Certification with related work experience could be a substitute for a college degree. -- Bryan J. Smith b.j.smith@ieee.org (407) 489-7013 (Mobile) P.S. Linux is also a major plus since much of our product line and the IT support infrastructure that you may touch regularly involves Linux. But we're clearly an MS ADS/Office setup IT-wise (which is where most of the "work" comes from, and not us engineers ;-). -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Thu Feb 2 09:00:12 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] ServerWorks IIILE SuperMicro 370DLE mdma2? In-Reply-To: <200601311916.01453.jasonb@edseek.com> Message-ID: <20060202140012.8396.qmail@web34107.mail.mud.yahoo.com> Jason Boxman wrote: > I thought it was supposed to do udma2? > I'm stuck in mdma2. Is that right? Probably not. The OSB4 southbridge is capable of "Ultra33" aka Ultra DMA mode 2 (8MHz DDR @ 16-bit = 33MBps). This not only which not only provides a significant speed boost over Multiword DMA mode 2 (8MHz SDR @ 16-bit = 16MBps), but more importantly, gives you CRC checking. There is _no_ error checking pre-Ultra DMA, which is the #1 reason why I didn't adopt ATA for hard drives (stuck with SCSI and its parity checking) until Ultra DMA became available. There is a newer revision of the OSB4 in the "-SL" editions that offers improved ATA compatibility. In all honesty, the ServerWorks ATA and select other peripherals are rather immature/low-featured. ServerWorks excels at bridging peripheral busses/channels (so much so that all current Intel designs are licensed from them -- especially the ServerWorks designed E7200/7500 series), but they leave much to be desired in peripherals before their Intel cross-licensing agreement. Hence why I use off-chipset controllers for fixed disk storage. Although the OSB4 seems to work fine as a Ultra33 controller for ATAPI optical devices. I use the Matsushita/Panasonic and LG GSA DVD-RAM/R drives on them without issues. Consider a low-cost 3Ware Escalade 7006-2, or try to find an Escalade 6200/6410 (for kernel 2.4 or earlier) or 7200/7000-2/7410/7500-4LP on eBay. > nebula:/tmp# hdparm -i /dev/hda > /dev/hda: > Model=Maxtor 6Y120L0, FwRev=YAR41VW0, SerialNo=Y41NK87E > Config={ Fixed } > RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57 > BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, > MultSect=off > CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=240121728 > IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} > PIO modes: pio0 pio1 pio2 pio3 pio4 > DMA modes: mdma0 mdma1 *mdma2 > UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 udma6 > AdvancedPM=yes: disabled (255) WriteCache=enabled > Drive conforms to: (null): > * signifies the current active mode > nebula:/tmp# hdparm /dev/hda > /dev/hda: > multcount = 0 (off) > IO_support = 0 (default 16-bit) > unmaskirq = 0 (off) > using_dma = 1 (on) > keepsettings = 0 (off) > readonly = 0 (off) > readahead = 256 (on) > geometry = 65535/16/63, sectors = 240121728, start = 0 Try running this command: hdparm -X66 /dev/hda The "-X" sets the mode. 64 Ultra DMA +2 mode = 66. Yes, that looks wrong, but -X68 is Ultra66 (Ultra DMA mode 4 = 16MHz DDR @ 16-bit = 66MHz). The OSB4's ATA bus arbitrator does have a _very_large_ "read-ahead" buffer (128KiB), which is very nice. Most other ATA logic have only a small 16 sector (8KiB) readahead. > Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 > ide: Assuming 33MHz system bus speed for PIO modes; override > with idebus=xx > SvrWks OSB4: IDE controller at PCI slot 0000:00:0f.1 > SvrWks OSB4: chipset revision 0 > SvrWks OSB4: not 100% native mode: will probe irqs later > ide0: BM-DMA at 0xffa0-0xffa7, BIOS settings: hda:DMA, hdb:pio > ide1: BM-DMA at 0xffa8-0xffaf, BIOS settings: hdc:DMA, hdd:pio > Probing IDE interface ide0... > hda: Maxtor 6Y120L0, ATA DISK drive > ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 > Probing IDE interface ide1... > hdc: NEC CD-ROM DRIVE:282, ATAPI CD/DVD-ROM drive > ide1 at 0x170-0x177,0x376 on irq 15 > nebula:/tmp# lspci > 0000:00:00.0 Host bridge: ServerWorks CNB20LE Host Bridge (rev > 06) > 0000:00:00.1 Host bridge: ServerWorks CNB20LE Host Bridge (rev > 06) > 0000:00:01.0 SCSI storage controller: QLogic Corp. ISP1080 SCSI > Host Adapter (rev 01) > 0000:00:02.0 VGA compatible controller: Matrox Graphics, Inc. > MGA 2064W > [Millennium] (rev01) > 0000:00:06.0 Ethernet controller: Intel Corp. 82557/8/9 > [Ethernet Pro 100] > (rev 08) > 0000:00:0f.0 ISA bridge: ServerWorks OSB4 South Bridge (rev 51) > 0000:00:0f.1 IDE interface: ServerWorks OSB4 IDE Controller > 0000:00:0f.2 USB Controller: ServerWorks OSB4/CSB5 OHCI USB > Controller (rev 04) > 0000:01:01.0 RAID bus controller: 3ware Inc 3ware 7000-series > ATA-RAID (rev 01) Do you have a spare port on the 3Ware? If so, then use the disk in JBOD mode on it instead. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Thu Feb 2 09:20:15 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] Cheap PCI GbE NIC that has fair specs for $15 ... Message-ID: <20060202142015.29112.qmail@web34101.mail.mud.yahoo.com> I know a lot of you are buying mainboards with GbE on-board. But just in case you need to add a GbE PCI card (when you don't have PCIe) to your system, but don't want to spend a lot of money, the RTL8169 MAC isn't a bad solution. Now it doesn't do jumbo frames last time I checked, but if you're not spending money, then you're probably not too worried about that. It does have 802.3x flow control (make sure your switch does too!) which is pretty much essential at GbE (especially with only 1500 byte frames). It also has a 64KiB SRAM receive buffer, which allows the card to queue up a good 47 1500 byte frames before becoming full. Lastly, the 32-bit card is 3.3V with 5V tolerances (universal), and can run at 66MHz (which is nice in case you want to put it in a PCI-X slot). Ignore the specs on TCP off-load support, that's just transmission only -- let alone the transmit buffer is only 8KiB (so forget jumbo frames). But for $15, I'm not complaining. An example card is this $15.99 at NewEgg.COM, the NW-500-GB: http://www.newegg.com/Product/Product.asp?Item=N82E16833332506 I know the Linux 2.6 kernel has RTL8169 support for sure, and I'd assume more recently 2.4 kernels have the driver as well. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From glaiacona at aikencountysc.gov Thu Feb 2 10:10:27 2006 From: glaiacona at aikencountysc.gov (George Laiacona) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] Mac OS and firewalls. Message-ID: Looking for some knowledge on Mac OS, specifically the Safari browser, the Transfer program, FTP, and PIX firewalls. Are there any known issues? I've got an FTP rule created in the PIX, it works for IE, but not Safari. I'm getting connection timed out errors trying to log in to an FTP server outside my network. I can log into any FTP server inside my firewall, however, so I know the Mac is configured correctly. Browsing also works as it should through my Novell BorderManager proxy. George. From dmckenna at thelimucompany.com Thu Feb 2 10:20:29 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] Mac OS and firewalls. Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D5C1@tlc001.tlcusa.thelimucompany.com> > Looking for some knowledge on Mac OS, specifically the Safari > browser, the Transfer program, FTP, and PIX firewalls. Do you have Transfer and Safari set to use passive-mode transfers? -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Thu Feb 2 10:20:56 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:36 2006 Subject: [Pc_Support] Mac OS and firewalls. In-Reply-To: Message-ID: <20060202152056.83666.qmail@web34106.mail.mud.yahoo.com> George Laiacona wrote: > Looking for some knowledge on Mac OS, specifically the Safari > browser, the Transfer program, FTP, and PIX firewalls. > Are there any known issues? I've got an FTP rule created in the > PIX, it works for IE, but not Safari. I'm getting connection > timed out errors trying to log in to an FTP server outside my > network. I can log into any FTP server inside my firewall, > however, so I know the Mac is configured correctly. Browsing > also works as it should through my Novell BorderManager proxy. It's most likely an active v. passive detail. Find out what the Safari brower defaults to and check the firewall for what is supports. I have a feeling it's active and several PIX products default to not allowing an active connection. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Thu Feb 2 10:46:56 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] ServerWorks IIILE SuperMicro 370DLE mdma2? In-Reply-To: 20060202140012.8396.qmail@web34107.mail.mud.yahoo.com Message-ID: <28272348c1dff39029ed4a1bc6487ca5@localhost> > In all honesty, the ServerWorks ATA and select other peripherals are > rather immature/low-featured. ServerWorks excels at bridging > peripheral busses/channels (so much so that all current Intel designs > are licensed from them -- especially the ServerWorks designed > E7200/7500 series), but they leave much to be desired in peripherals > before their Intel cross-licensing agreement. > > Hence why I use off-chipset controllers for fixed disk storage. > Although the OSB4 seems to work fine as a Ultra33 controller for > ATAPI optical devices. I use the Matsushita/Panasonic and LG GSA > DVD-RAM/R drives on them without issues. Aren't ServerWorks boards also really designed with the philosophy of "We'll throw some IDE ports on there, but you're really supposed to use the SCSI ports"? At least that is definitely how it seems in my file server (ServerWorks IIIHE). Of course this isn't a problem for me, since that machine is pure SCSI ;-) (unless you count the SATA drives hanging off a 3ware 9500-4LP) Then again, even pure-SCSI servers often still use IDE for their CD/DVD-ROM drives these days. -Derek From glaiacona at aikencountysc.gov Thu Feb 2 11:19:05 2006 From: glaiacona at aikencountysc.gov (George Laiacona) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Mac OS and firewalls. Message-ID: Yes, I thought of that. Sorry I didn't mention it in the original post. George. >>> dmckenna@thelimucompany.com 02/02/06 10:20 AM >>> > Looking for some knowledge on Mac OS, specifically the Safari > browser, the Transfer program, FTP, and PIX firewalls. Do you have Transfer and Safari set to use passive-mode transfers? -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include _______________________________________________ Pc_support mailing list Pc_support@matrixlist.com http://lists.matrixlist.com/mailman/listinfo/pc_support From jasonb at edseek.com Thu Feb 2 22:31:11 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] ServerWorks IIILE SuperMicro 370DLE mdma2? In-Reply-To: <20060202140012.8396.qmail@web34107.mail.mud.yahoo.com> References: <20060202140012.8396.qmail@web34107.mail.mud.yahoo.com> Message-ID: <200602022231.11574.jasonb@edseek.com> On Thursday 02 February 2006 09:00, Bryan J. Smith wrote: > Try running this command: > hdparm -X66 /dev/hda Command succeeds, but the output from `hdparm -i /dev/hda` is the same. It doesn't seem to get turned on. No errors, though. Mystery. > Do you have a spare port on the 3Ware? > If so, then use the disk in JBOD mode on it instead. It's full. :) -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From jasonb at edseek.com Thu Feb 2 22:35:58 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Cheap PCI GbE NIC that has fair specs for $15 ... In-Reply-To: <20060202142015.29112.qmail@web34101.mail.mud.yahoo.com> References: <20060202142015.29112.qmail@web34101.mail.mud.yahoo.com> Message-ID: <200602022235.58068.jasonb@edseek.com> On Thursday 02 February 2006 09:20, Bryan J. Smith wrote: > I know a lot of you are buying mainboards with GbE on-board. But > just in case you need to add a GbE PCI card (when you don't have > PCIe) to your system, but don't want to spend a lot of money, the > RTL8169 MAC isn't a bad solution. Interesting. Everytime I try to go GbE I find it doesn't work out. When I tried last January I found I didn't have the bandwidth on my PCI bus to really transfer any data faster than Fast Ethernet speeds. Actually, the real reason I was looking into GbE was backups from my fileserver to the backup server, but the backup server is an old Katmai? slot P3 and even using 'arcfour' as the encryption with ssh, I get less than 100 Mbps throughput. So, I guess my problem is the encryption overhead associated with running `rsync` over `ssh`. Oh well. > Now it doesn't do jumbo frames last time I checked, but if you're not > spending money, then you're probably not too worried about that. It > does have 802.3x flow control (make sure your switch does too!) which > is pretty much essential at GbE (especially with only 1500 byte > frames). It also has a 64KiB SRAM receive buffer, which allows the > card to queue up a good 47 1500 byte frames before becoming full. > Lastly, the 32-bit card is 3.3V with 5V tolerances (universal), and > can run at 66MHz (which is nice in case you want to put it in a PCI-X > slot). Neat. > Ignore the specs on TCP off-load support, that's just transmission > only -- let alone the transmit buffer is only 8KiB (so forget jumbo > frames). But for $15, I'm not complaining. > > An example card is this $15.99 at NewEgg.COM, the NW-500-GB: > http://www.newegg.com/Product/Product.asp?Item=N82E16833332506 > > I know the Linux 2.6 kernel has RTL8169 support for sure, and I'd > assume more recently 2.4 kernels have the driver as well. Wow, that is insanely cheap. I have a GbE switch. Almost makes we want to buy a few, but I never seem to need to push enough data around to warrant it. Maybe I'll buy two and move stuff between my workstation and fileserver over NFS and see what happens. -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From glaiacona at aikencountysc.gov Fri Feb 3 08:45:53 2006 From: glaiacona at aikencountysc.gov (George Laiacona) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Mac OS and firewalls. Message-ID: I have verified that the Mac is using PASV for FTP transfers. I've even opened up the rule a bit to allow for any IP connection outbound from that machine. George. >>> "Bryan J. Smith" 02/02/06 10:20 AM >>> George Laiacona wrote: > Looking for some knowledge on Mac OS, specifically the Safari > browser, the Transfer program, FTP, and PIX firewalls. > Are there any known issues? I've got an FTP rule created in the > PIX, it works for IE, but not Safari. I'm getting connection > timed out errors trying to log in to an FTP server outside my > network. I can log into any FTP server inside my firewall, > however, so I know the Mac is configured correctly. Browsing > also works as it should through my Novell BorderManager proxy. It's most likely an active v. passive detail. Find out what the Safari brower defaults to and check the firewall for what is supports. I have a feeling it's active and several PIX products default to not allowing an active connection. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** _______________________________________________ Pc_support mailing list Pc_support@matrixlist.com http://lists.matrixlist.com/mailman/listinfo/pc_support From dmckenna at thelimucompany.com Fri Feb 3 10:15:34 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Cheap PCI GbE NIC that has fair specs for $15 ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D627@tlc001.tlcusa.thelimucompany.com> FYI there are a load of <$50 gig-e switches available on NewEgg: http://www.newegg.com/Product/ProductList.asp?Submit=ENE&N=2050400030+70 001468&Subcategory=30&description=&srchInDesc=&minPrice=&maxPrice=&ATTR1 =&ATTR2=&ATTR3=&ATTR4=2050400030+70001468&ATTR5= -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Fri Feb 3 11:51:11 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Cheap PCI GbE NIC that has fair specs for $15 ... In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D627@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060203165111.99904.qmail@web34101.mail.mud.yahoo.com> From: Damien McKenna > FYI there are a load of <$50 gig-e switches available on NewEgg: > http://www.newegg.com/Product/ProductList.asp?Submit=ENE&N=2050400030+70 > 001468&Subcategory=30&description=&srchInDesc=&minPrice=&maxPrice=&ATTR1 > =&ATTR2=&ATTR3=&ATTR4=2050400030+70001468&ATTR5= Note that several of those lower-costing devices do not have 802.3x. Without 802.3x, especially on a SOHO network with cheaper components, it's very easy to get very poor performance From dmckenna at thelimucompany.com Fri Feb 3 12:00:03 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Cheap PCI GbE NIC that has fair specs for $15 ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D630@tlc001.tlcusa.thelimucompany.com> > Note that several of those lower-costing devices do not have > 802.3x. Without 802.3x, especially on a SOHO network with > cheaper components, it's very easy to get very poor performance That's annoying, I did the original search based off that keyword. Incidentally, the first one listed ($12 AR thing from ABS) has it, as does the D-Link for $26 AR, but the NetGear for $27 AR does not. OK, here's a link with the appropriate keyword: http://www.newegg.com/Product/ProductList.asp?Submit=ENE&N=2050400030+70 001468&Subcategory=30&description=&srchInDesc=IEEE802.3x&minPrice=&maxPr ice=&ATTR1=&ATTR2=&ATTR3=&ATTR4= -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Fri Feb 3 15:56:09 2006 From: b.j.smith at ieee.org (=?ISO-8859-1?Q?Bryan=20J=2E=20Smith?=) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] I love Walmart ... Message-ID: Been busy so like a few times in the past, I took my truck to a Walmart Express Lube at lunch. First time I tried out the Seminole Town Center location. After I make a left on to Rinehart I start to smell something funny. I look at my gages and oil pressure as well as engine temp are fine. But by the time I travel another 90 seconds down Rinehart, I still smell it - unlike anything before. I pull over at the 7-11 at Rinehart and Lake Mary and open the hood. The cap is on and everything seems okay for the first second. Then I notice an orange glow and look around the left side of my engine. Nice BASKETBALL SIZE FIREBALL at the base of my engine slowly making its way all over whatever it can BURN!!! After jumping back a few feet, I ran into 7-11 to not-so-casually to exclaim "my engine is on fire! Where are the gallon bottles of water?!" As if the same people from Walmart seem to work at 7-11, they say nothing. I repeat myself, and finally 2 customers are far more helpful than the 7-11 crew. Sigh, what an interesting society of Americans make up our retail and everyday service trades today, eh? Waiting on AAA to call me back with news of a truck and some much needed legal advise. -- Sent from my Treo From b.j.smith at ieee.org Fri Feb 3 15:04:03 2006 From: b.j.smith at ieee.org (=?ISO-8859-1?Q?Bryan=20J=2E=20Smith?=) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] I love Walmart ... Message-ID: Been busy so like a few times in the past, I took my truck to a Walmart Express Lube at lunch. First time I tried out the Seminole Town Center location. After I make a left on to Rinehart I start to smell something funny. I look at my gages and oil pressure as well as engine temp are fine. But by the time I travel another 90 seconds down Rinehart, I still smell it - unlike anything before. I pull over at the 7-11 at Rinehart and Lake Mary and open the hood. The cap is on and everything seems okay for the first second. Then I notice an orange glow and look around the left side of my engine. Nice BASKETBALL SIZE FIREBALL at the base of my engine slowly making its way all over whatever it can BURN!!! After jumping back a few feet, I ran into 7-11 to not-so-casually to exclaim "my engine is on fire! Where are the gallon bottles of water?!" As if the same people from Walmart seem to work at 7-11, they say nothing. I repeat myself, and finally 2 customers are far more helpful than the 7-11 crew. Sigh, what an interesting society of Americans make up our retail and everyday service trades today, eh? Waiting on AAA to call me back with news of a truck and some much needed legal advise. -- Sent from my Treo From b.j.smith at ieee.org Fri Feb 3 16:54:42 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] I love Walmart ... In-Reply-To: Message-ID: <20060203215442.58540.qmail@web34111.mail.mud.yahoo.com> It's now at the dealer. When the fire was smoldering, I found some rags that had fell off from just below where the fire was. So I bagged them for inspection by the dealer. There was clearly some sort of grease all over portions of the engine. The dealer also noted the power steering cap wasn't on all the way and power steering fluid was caked, but that was away from the fire. I just had this truck at the dealer in October for my last oil change. They have almost a full record of anything done on that truck, short of when I go to someone else for a oil change. I typically avoid the "lube" places and try to mix in my oil changes and lube with any tire rotations, typically at my local tire place if not the dealer. Oh well, now I know better! I love Walmart. ;-> Bryan J. Smith wrote: > Been busy so like a few times in the past, I took my truck to a > Walmart Express Lube at lunch. First time I tried out the Seminole > Town Center location. > After I make a left on to Rinehart I start to smell something > funny. I look at my gages and oil pressure as well as engine temp > are fine. > But by the time I travel another 90 seconds down Rinehart, I still > smell it - unlike anything before. I pull over at the 7-11 at > Rinehart and Lake Mary and open the hood. > The cap is on and everything seems okay for the first second. Then > I notice an orange glow and look around the left side of my > engine. > Nice BASKETBALL SIZE FIREBALL at the base of my engine slowly > making its way all over whatever it can BURN!!! > After jumping back a few feet, I ran into 7-11 to not-so-casually > to exclaim "my engine is on fire! Where are the gallon bottles of > water?!" As if the same people from Walmart seem to work at 7-11, > they say nothing. I repeat myself, and finally 2 customers are far > more helpful than the 7-11 crew. > Sigh, what an interesting society of Americans make up our retail > and everyday service trades today, eh? > Waiting on AAA to call me back with news of a truck and some much > needed legal advise. > -- > Sent from my Treo -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Fri Feb 3 18:23:50 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] I love Walmart ... In-Reply-To: <20060203215442.58540.qmail@web34111.mail.mud.yahoo.com> References: <20060203215442.58540.qmail@web34111.mail.mud.yahoo.com> Message-ID: <43E3E606.5040205@HiWAAY.net> Bryan J. Smith wrote: >It's now at the dealer. > >When the fire was smoldering, I found some rags that had fell off >from just below where the fire was. So I bagged them for inspection >by the dealer. > >There was clearly some sort of grease all over portions of the >engine. The dealer also noted the power steering cap wasn't on all >the way and power steering fluid was caked, but that was away from >the fire. > >I just had this truck at the dealer in October for my last oil >change. They have almost a full record of anything done on that >truck, short of when I go to someone else for a oil change. > >I typically avoid the "lube" places and try to mix in my oil changes >and lube with any tire rotations, typically at my local tire place if >not the dealer. Oh well, now I know better! > >I love Walmart. ;-> > > >Bryan J. Smith wrote: > > >>Been busy so like a few times in the past, I took my truck to a >>Walmart Express Lube at lunch. First time I tried out the Seminole >>Town Center location. >>After I make a left on to Rinehart I start to smell something >>funny. I look at my gages and oil pressure as well as engine temp >>are fine. >>But by the time I travel another 90 seconds down Rinehart, I still >>smell it - unlike anything before. I pull over at the 7-11 at >>Rinehart and Lake Mary and open the hood. >>The cap is on and everything seems okay for the first second. Then >>I notice an orange glow and look around the left side of my >>engine. >>Nice BASKETBALL SIZE FIREBALL at the base of my engine slowly >>making its way all over whatever it can BURN!!! >>After jumping back a few feet, I ran into 7-11 to not-so-casually >>to exclaim "my engine is on fire! Where are the gallon bottles of >>water?!" As if the same people from Walmart seem to work at 7-11, >>they say nothing. I repeat myself, and finally 2 customers are far >>more helpful than the 7-11 crew. >>Sigh, what an interesting society of Americans make up our retail >>and everyday service trades today, eh? >>Waiting on AAA to call me back with news of a truck and some much >>needed legal advise. >>-- >>Sent from my Treo >> >> Well, as a degreed ME (Mech.Eng. MSME Ga.Tech, Fall 1984, specializing in prime-movers (Carnot- (gasoline) & Diesel-cycle piston cylinder engines, gas- and steam- turbines), and more recently, solid-fueled rocket motors), I'll provide any informed speculation I can if needed :-). Could you be a bit more specific on where the fire was & what was burning (oily rags, oil/gas/other-fluids spilled on or leaking from/on engine, etc.) ? -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From b.j.smith at ieee.org Sat Feb 4 07:14:58 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] I love Walmart ... In-Reply-To: <43E3E606.5040205@HiWAAY.net> References: <20060203215442.58540.qmail@web34111.mail.mud.yahoo.com> <43E3E606.5040205@HiWAAY.net> Message-ID: <1139055298.4784.12.camel@bert64.oviedo.smithconcepts.com> On Fri, 2006-02-03 at 17:23 -0600, William A. Mahaffey III wrote: > Well, as a degreed ME (Mech.Eng. MSME Ga.Tech, Fall 1984, specializing > in prime-movers (Carnot- (gasoline) & Diesel-cycle piston cylinder > engines, gas- and steam- turbines), and more recently, solid-fueled > rocket motors), I'll provide any informed speculation I can if needed > :-). Could you be a bit more specific on where the fire was & what was > burning (oily rags, oil/gas/other-fluids spilled on or leaking from/on > engine, etc.) ? According to the Mazda dealer, there's no permanent damage to the engine compartment. They couldn't tell much from the rags, but when the rags fell, they fell into a puddle of water at the 7-11 (it's been overcast/raining for the last 36 hours or so). The best guess now is that the rags I picked up off the ground after throwing water on the fire and knocking them off of what looks like the exhaust manifold was the cause and sole "fuel" for the fire. So, in a nutshell, we thing it was rags left on the exhaust manifold. It looked like the starter was engulfed in a fireball, which is right next to the exhaust manifold. After I came out from getting a couple of jugs of water from 7-11, the fire was down a bit, and what I initially thought was some hanging/burn insulation, wire or some other material was the clearly charred rags I picked up. Again, when I threw one gallon underneath, one of those rags fell down. Another fell down after I threw the other gallon at the back of my engine near the start/exhaust manifold. I drove ~15-20 miles home without incident. I'll keep my eye on it. -- Bryan J. Smith mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------ Some things (or athletes) money can't buy. For everything else there's "ManningCard." From ae4ko at amsat.org Mon Feb 6 10:13:32 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive Message-ID: <43E7214C.12847.603C7A@ae4ko.amsat.org> I need parts to try to recover some data on dead harddrives... I have a friend who had a catastrophic power supply failure in his PC. So much so that it killed every PC Board in the machine. He had two harddrives in the box at the time and both will not power up. Naturally, there is data on the drive that he wants recovered. My only option at this point is to find some drives with the same circuit board and try to swap electronics. So what I have is this: WD1200 (Western Digital 120GB) MDL: WD1200JB-75CRA0 DATE: 06OCT2003 DCM: HSEHNA2CH Rev A00 (Rev A is also listed on the PC Board) Quantum Fireball Plus LM (3.5 Series) GTLA LM15A011-01-A 15.0AT (Assy: 20-12130 is listed on the PC Board) If anyone has a drive that I could swap (even temporarily) PC boards with, please let me know. Thanks. --am From b.j.smith at ieee.org Mon Feb 6 11:52:12 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] LCDs drop to $130: 14" Proview PL482s for $179 - $50 = $129 AR Message-ID: <20060206165212.85719.qmail@web34103.mail.mud.yahoo.com> Only 1024x768 with 38ms refresh, no DVI, but if you want an LCD for cheap just for the desk, this 14" LCD (equivalent to a 15", 14" viewable, CRT) is affordable from Amazon.COM: http://dealnews.com/deals/Proview-PL482-s-14-LCD-Monitor-for-130-after-rebate/109286.html -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From jasonb at edseek.com Mon Feb 6 13:21:17 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] LCDs drop to $130: 14" Proview PL482s for $179 - $50 = $129 AR In-Reply-To: <20060206165212.85719.qmail@web34103.mail.mud.yahoo.com> References: <20060206165212.85719.qmail@web34103.mail.mud.yahoo.com> Message-ID: <33055.216.134.200.78.1139250077.squirrel@nebula.internal.foo> Bryan J. Smith said: > Only 1024x768 with 38ms refresh, no DVI, but if you want an LCD for > cheap just for the desk, this 14" LCD (equivalent to a 15", 14" > viewable, CRT) is affordable from Amazon.COM: > http://dealnews.com/deals/Proview-PL482-s-14-LCD-Monitor-for-130-after-rebate/109286.html I want one of those that you can see clealy from strange 45 degree angles and such. What spec do you want for that? Actually my BenQ FP731 at work isn't bad. I wouldn't mind buying two of those, but I'd rather get 19" LCDs if I'm going to spend the money anyway... I keep waiting for that Dell Ultrasharp 19" non-widescreen one to get cheaper. They had it last week for like $300ish again. Maybe it was the 17" though. From dmckenna at thelimucompany.com Mon Feb 6 15:37:56 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Dead laptop drive, recovery companies? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6A0@tlc001.tlcusa.thelimucompany.com> Could anyone recommend somewhere that would do a physical examination of a laptop drive to ascertain the recoverability of its data after said drive decided to go the way of the dodo? The drive has something loose on the inside that is audible when shaken, so at this point we'd like someone who knows what they're doing to simply tell us whether we should proceed with a full recovery or if it's a lost case. Thanks. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From whittake at sbaflorida.com Mon Feb 6 15:50:13 2006 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <43E7214C.12847.603C7A@ae4ko.amsat.org> References: <43E7214C.12847.603C7A@ae4ko.amsat.org> Message-ID: <1139259013.807.22.camel@localhost.localdomain> On Mon, 2006-02-06 at 10:13 -0500, Aaron Morrison wrote: > I need parts to try to recover some data on dead harddrives... > > I have a friend who had a catastrophic power supply failure in his > PC. So much so that it killed every PC Board in the machine. He had > two harddrives in the box at the time and both will not power up. > Naturally, there is data on the drive that he wants recovered. My > only option at this point is to find some drives with the same > circuit board and try to swap electronics. > > So what I have is this: > > WD1200 (Western Digital 120GB) > MDL: WD1200JB-75CRA0 > DATE: 06OCT2003 > DCM: HSEHNA2CH > Rev A00 > (Rev A is also listed on the PC Board) Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. But I am lost as to how this would help his situation? Also, I would need assurance that if he/you blows my 120 GB drive you will get me a new one :) Homer > > Quantum Fireball Plus LM (3.5 Series) > GTLA LM15A011-01-A 15.0AT > (Assy: 20-12130 is listed on the PC Board) > > If anyone has a drive that I could swap (even temporarily) PC boards > with, please let me know. > > Thanks. > > --am > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support From tim at mcdonough.net Mon Feb 6 23:06:31 2006 From: tim at mcdonough.net (Tim McDonough) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Dead laptop drive, recovery companies? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6A0@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6A0@tlc001.tlcusa.thelimucompany.com> Message-ID: <43E81CC7.8040801@mcdonough.net> Damien McKenna wrote: > Could anyone recommend somewhere that would do a physical examination of > a laptop drive to ascertain the recoverability of its data after said > drive decided to go the way of the dodo? The drive has something loose > on the inside that is audible when shaken, so at this point we'd like > someone who knows what they're doing to simply tell us whether we should > proceed with a full recovery or if it's a lost case. Thanks. > They aren't cheap but over the years Ontrack has been 3 for 3 in my personal experiences. -- Tim From work at sprynet.com Mon Feb 6 23:21:04 2006 From: work at sprynet.com (John Hayden) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <1139259013.807.22.camel@localhost.localdomain> Message-ID: The 120 drive I can not help with, but the quantum I have several that you can look at and see if they will help, I will be at the HAMcation fair grounds wed - Sunday 7am till dark.30 You can call me to set up a meet if you want to get them before Friday opening J.T. Hayden KG4BFJ 1419 Oregon Ave. St. Cloud, FL. 34769 Home/Office - 407.891.1835 office/mobile - 407.922.3091 -----Original Message----- From: pc_support-bounces@matrixlist.com [mailto:pc_support-bounces@matrixlist.com] On Behalf Of Homer Whittaker Sent: Monday, February 06, 2006 15:50 To: This is the PC Support list. Subject: Re: [Pc_Support] Help Request: HDD parts to revive dead drive On Mon, 2006-02-06 at 10:13 -0500, Aaron Morrison wrote: > I need parts to try to recover some data on dead harddrives... > > I have a friend who had a catastrophic power supply failure in his PC. > So much so that it killed every PC Board in the machine. He had two > harddrives in the box at the time and both will not power up. > Naturally, there is data on the drive that he wants recovered. My > only option at this point is to find some drives with the same circuit > board and try to swap electronics. > > So what I have is this: > > WD1200 (Western Digital 120GB) > MDL: WD1200JB-75CRA0 > DATE: 06OCT2003 > DCM: HSEHNA2CH > Rev A00 > (Rev A is also listed on the PC Board) Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. But I am lost as to how this would help his situation? Also, I would need assurance that if he/you blows my 120 GB drive you will get me a new one :) Homer > > Quantum Fireball Plus LM (3.5 Series) > GTLA LM15A011-01-A 15.0AT > (Assy: 20-12130 is listed on the PC Board) > > If anyone has a drive that I could swap (even temporarily) PC boards > with, please let me know. > > Thanks. > > --am > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support _______________________________________________ Pc_support mailing list Pc_support@matrixlist.com http://lists.matrixlist.com/mailman/listinfo/pc_support begin 666 smime.p7s M,( &"2J&2(;W#0$'`J" ,( "`0$Q"S )!@4K#@,"&@4`,( &"2J&2(;W#0$' M`0``H((.03""!' P@@-8H ,"`0("$!^TS9CZN4FQ0V*Y59U/UV$P#08)*H9( MAO<-`0$%!0`P@:XQ"S )!@-5! 83`E53,0LP"08#500($P)55#$7,!4&`U4$ M!Q,.4V%L="!,86ME($-I='DQ'C Q.#<$R/-P37'^]9.$0IG7O4&6VNE"#M;*18ZYX>R- ;%O 6E`WR" MRRD0KN&(@;W6GM/^+5;!%<[C)IT5+A#[!H\P!-ZGM&.T_[&WRSQ;KPPVRPD$MOK-35HD,L`L)/:X$;GTP'08#51T.!!8$%)6QM/"4MKW'VM$1 M"2&^P:])_1![, X&`U4=#P$!_P0$`P(!!C /!@-5'1,!`?\$!3 #`0'_,!$& M"6"&2 &&^$(!`00$`P(``3!8!@-5'1\$43!/,$V@2Z!)AD=H='1P.B\O8W)L M+G5S97)TG#MX&PMTIL MT @<("Y/K;#%+MF&'4UUZ9V.D^,$6A]I+@O_FO/YTB,'F1J',&&B(G@3JLBH M'Q-+ZX&Q0%Z$?&[M$I/"4KZ=CXB6X79O8%%TG3[_W H+M/V[U>[_D^#A[_+) MQ,-]?E)$CK09%YO3B#,U&Y@Y$,75+DWUEIS7?;Q,5Z7VB#]!K2-+FD[;9Z]_ M$L#S6[8XXENZ7/8"616=9 N2?HHI#:T`=0Q?U,S&S2?=E23E.$5]8\UJ>SK; MVK751XRW#0/A"QCS(F0H*7HRH!SC^A8[\>WGA:-4X>(\(?RCJ7CKS%3,]!\: M%BG-@;JF;,6.GF#G_6%)0-YT-"71]3""!*(P@@.*H ,"`0("$$2^#(M0`"2T M$=,V)25GR8DP#08)*H9(AO<-`0$%!0`P@:XQ"S )!@-5! 83`E53,0LP"08# M500($P)55#$7,!4&`U4$!Q,.4V%L="!,86ME($-I='DQ'C 3$>,!P&`U4$"A,55&AE(%5315)44E535"!.971W M;W)K,2$P'P8#500+$QAH='1P.B\O=W=W+G5S97)TQQZI)*H%_*8%?*3S,FD,I6[#$\PO9B@"XOE M;ANB1OIXO*)OJUE>I2_/RMIMJB_KK*&S:JJW+F@Z;ZOS:)UL]"J2Z`W>S9 M/:X$;GTP6 8#51T?!%$P3S!-H$N@289':'1T<#HO+V-R;"YU(Q7"L.[NYA56\$/L\Y MWL4;293DZR!,M.:>4"YRV8WUJJ.S2MI6'&"7@-R"HJU*O8HK_PL)M,;7( 1% MY,V `;JZ*V[.JM>2_N2OZ_0F'18J?VPPE3WZ*%Q5P( MIGCF4T^QZ+?3%)Z3IL-DXZQ^<1A61E;C$?,!T&"2J&2(;W#0$)`180=V]R:T!S<')Y;F5T+F-O;3"!GS -!@DJ MADB&]PT!`0$%``.!C0`P@8D"@8$`G +SYVI#I63G`A1ON6P2`L8+4^-W`TD% M#ALENF0C,AHWFA"O^>)8#=OT4J0*N(^I$OAN`9T1#5&B\T?]*+WNVE*?L;RU MOONQ43)K:N\^_E#E)_R+,'DV7F411SGL(X:"GJU%>.-[9H\8# .!@-5'0\!`?\$! ,"!: P# 8#51T3`0'_! (P`# @ M!@-5'24$&3 7!@@K!@$%!0<#! 8+*P8!! &R,0$#!0(P$08)8(9(`8;X0@$! M! 0#`@4@,$8&`U4=( 0_,#TP.P8,*P8!! &R,0$"`0$!,"LP*08(*P8!!04' M`@$6'6AT='!S.B\OC!X,#L&""L&`04% M!S "AB]H='1P.B\O8W)T+F-O;6]D;V-A+F-O;2]!9&14RN'/0#P>'>T) M1>_UW!' 6!#>;A:'%BF9+OY(JJT3GK,K>ENA^*V)Y_+&+A?=M\;G?T:S@(I) M6;DYA]7352#N6VL+-H.P@F,,%DQQ\8&]7E<.^(0`6LIDA=/.AQL\$E+.9=TQ ME5F,KT\K>8<:``>FE7UINBD?J*Z8L0Z>+@IDQ)MAMZ&4A(+F) +KCFIJ1V? M2!C84*)%MB-3/CP?S-5L/8D"S:'<5$M4&[_88#758>:2::NO@39+`16+3XQ\ M\E71PG9:1^H1A P& 8)*H9(AO<-`0D#,0L&"2J& M2(;W#0$'`3 1A;RA(;=0/]E8"KL'G2EB`N\!\-Z>)_8LF],_PHF7_S4#");] References: Message-ID: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> Bryan (and those on the list with a Treo 650), > Sent from my Treo The Lord (and my new company) has so blessed me with such a beast - any suggestions on 'Must Have' software / Hacks / Sites / etc? Thanks in advance, dave From b.j.smith at ieee.org Tue Feb 7 06:45:38 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> References: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> Message-ID: <1139312738.4720.26.camel@bert64.oviedo.smithconcepts.com> On Mon, 2006-02-06 at 23:23 -0600, David Simmons wrote: > Bryan (and those on the list with a Treo 650), > The Lord (and my new company) has so blessed me with such a beast - any > suggestions on 'Must Have' software / Hacks / Sites / etc? Actually, I have a Treo 600, although I'm planning to get a 650 soon (and give my wife my 600). -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From ozz at ozz.is-a-geek.net Tue Feb 7 07:32:50 2006 From: ozz at ozz.is-a-geek.net (ozz@ozz.is-a-geek.net) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <1139312738.4720.26.camel@bert64.oviedo.smithconcepts.com> References: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> <1139312738.4720.26.camel@bert64.oviedo.smithconcepts.com> Message-ID: <46862.69.176.47.130.1139315570.squirrel@www.ozz.is-a-geek.net> > On Mon, 2006-02-06 at 23:23 -0600, David Simmons wrote: >> Bryan (and those on the list with a Treo 650), >> The Lord (and my new company) has so blessed me with such a beast - any >> suggestions on 'Must Have' software / Hacks / Sites / etc? > > Actually, I have a Treo 600, although I'm planning to get a 650 soon > (and give my wife my 600). I toyed with the idea of getting a 650, but heard some real horror stories about reception, etc. from people I knew who had them. In fact, my local cell provider is refusing to even carry the new 700 because the 650 bombed so badly. I opted for the Audiovox 6600 (the 6700 is not available to any providers in my area yet, and won't be for at least another 6 months, and I didn't want to wait that long). Regards, Ozz. From ae4ko at amsat.org Tue Feb 7 08:35:37 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <1139259013.807.22.camel@localhost.localdomain> References: <43E7214C.12847.603C7A@ae4ko.amsat.org> Message-ID: <43E85BD9.30513.52CF337@ae4ko.amsat.org> On 6 Feb 2006 at 15:50, Homer Whittaker wrote: > > Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. > But I am lost as to how this would help his situation? Also, I would > need assurance that if he/you blows my 120 GB drive you will get me a > new one :) Homer I plan on swapping the electronic circuit board on the bottom long enough to copy the data off of the platters. And yes, if your board or drive dies, it will be replaced. Just be sure you have backups of everything! Are you still willing to help? --am From ae4ko at amsat.org Tue Feb 7 08:35:36 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: References: <1139259013.807.22.camel@localhost.localdomain> Message-ID: <43E85BD8.18887.52CF161@ae4ko.amsat.org> I'll be there probably Friday and Saturday (when I'm not testing). I'll try to hook up with you Friday. Thanks! --am On 6 Feb 2006 at 23:21, John Hayden wrote: > The 120 drive I can not help with, but the quantum I have several that > you can look at and see if they will help, I will be at the HAMcation > fair grounds wed - Sunday 7am till dark.30 You can call me to set up > a meet if you want to get them before Friday opening > > > J.T. Hayden > KG4BFJ > 1419 Oregon Ave. > St. Cloud, FL. 34769 > > Home/Office - 407.891.1835 > office/mobile - 407.922.3091 > > -----Original Message----- > From: pc_support-bounces@matrixlist.com > [mailto:pc_support-bounces@matrixlist.com] On Behalf Of Homer > Whittaker Sent: Monday, February 06, 2006 15:50 To: This is the PC > Support list. Subject: Re: [Pc_Support] Help Request: HDD parts to > revive dead drive > > On Mon, 2006-02-06 at 10:13 -0500, Aaron Morrison wrote: > > I need parts to try to recover some data on dead harddrives... > > > > I have a friend who had a catastrophic power supply failure in his > > PC. So much so that it killed every PC Board in the machine. He > > had two harddrives in the box at the time and both will not power > > up. Naturally, there is data on the drive that he wants recovered. > > My only option at this point is to find some drives with the same > > circuit board and try to swap electronics. > > > > So what I have is this: > > > > WD1200 (Western Digital 120GB) > > MDL: WD1200JB-75CRA0 > > DATE: 06OCT2003 > > DCM: HSEHNA2CH > > Rev A00 > > (Rev A is also listed on the PC Board) > > Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. > But I am lost as to how this would help his situation? Also, I would > need assurance that if he/you blows my 120 GB drive you will get me a > new one :) Homer > > > > > Quantum Fireball Plus LM (3.5 Series) > > GTLA LM15A011-01-A 15.0AT > > (Assy: 20-12130 is listed on the PC Board) > > > > If anyone has a drive that I could swap (even temporarily) PC boards > > with, please let me know. > > > > Thanks. > > > > --am > > > > _______________________________________________ > > Pc_support mailing list > > Pc_support@matrixlist.com > > http://lists.matrixlist.com/mailman/listinfo/pc_support > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support > > > begin 666 smime.p7s > M,( &"2J&2(;W#0$'`J" ,( "`0$Q"S )!@4K#@,"&@4`,( & From ae4ko at amsat.org Tue Feb 7 08:45:41 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <46862.69.176.47.130.1139315570.squirrel@www.ozz.is-a-geek.net> References: <1139312738.4720.26.camel@bert64.oviedo.smithconcepts.com> Message-ID: <43E85E35.24209.53629EB@ae4ko.amsat.org> On 7 Feb 2006 at 7:32, ozz@ozz.is-a-geek.net wrote: > > On Mon, 2006-02-06 at 23:23 -0600, David Simmons wrote: > >> Bryan (and those on the list with a Treo 650), > >> The Lord (and my new company) has so blessed me with such a beast - > >> any suggestions on 'Must Have' software / Hacks / Sites / etc? > > > > Actually, I have a Treo 600, although I'm planning to get a 650 soon > > (and give my wife my 600). > > I toyed with the idea of getting a 650, but heard some real horror > stories about reception, etc. from people I knew who had them. In > fact, my local cell provider is refusing to even carry the new 700 > because the 650 bombed so badly. > Well, I'm on Sprint with my 650 (and had a 600 for 2 years) and have had no problems with reception. I've had problems with their network. I've known when someone called me, but the phone never rang -- went straight to voicemail. That's not the phone's fault, but I could easily see a wonk at a store leading someone to believe it was. Having said that, there are some issues out there. Mine randomly resets, my friends does not. The bluetooth implementation is flakey - I have the popular Motorola HS850 headset and I had to learn to work around the quirks. The 700w has WindowsCE based software which I don't personall care for (very clunky for a phone/PDA). I can see why some people like them and I can see why some stores won't carry them. --am > I opted for the Audiovox 6600 (the 6700 is not available to any > providers in my area yet, and won't be for at least another 6 months, > and I didn't want to wait that long). > > Regards, > Ozz. From ae4ko at amsat.org Tue Feb 7 08:52:19 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> References: Message-ID: <43E85FC3.30297.53C3E7E@ae4ko.amsat.org> On 6 Feb 2006 at 23:23, David Simmons wrote: > The Lord (and my new company) has so blessed me with such a beast - > any suggestions on 'Must Have' software / Hacks / Sites / etc? > > Thanks in advance, > > dave KeyCaps600 (650) - lets you type caps and numbers without the shift or option key. pTunes 3 (shareware) - nice player that supports Internet streaming, Ogg Vorbis, WMA, mp3, WAV. FileZ -- good for any Palm, lets you look at the data files and resources. ---- Not necessarily "must have" but handy for me -- they eat network bandwidth, so make sure your plan can cover the data use. TuSSH -- ssh client PalmVNC -- VNC client Chatter Email -- email client that supports POP and IMAP (and "push" IMAP if the server supports it) - shareware pdaNet -- software that makes your palm a modem for Internet access. (shareware) From dave at dgnal.net Tue Feb 7 09:23:34 2006 From: dave at dgnal.net (David Simmons) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <43E85FC3.30297.53C3E7E@ae4ko.amsat.org> References: <43E85FC3.30297.53C3E7E@ae4ko.amsat.org> Message-ID: <1424.71.252.176.10.1139322214.squirrel@qtmail.dgnal.net> > Not necessarily "must have" but handy for me -- they eat network > bandwidth, so make sure your plan can cover the data use. These are all great - Thanks! I've been reading alot where people are finding/activating drivers for a WiFi card to add to the 650 the title VoIP phone...anyone using/doing this? dave From b.j.smith at ieee.org Tue Feb 7 09:21:10 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <46862.69.176.47.130.1139315570.squirrel@www.ozz.is-a-geek.net> Message-ID: <20060207142110.3109.qmail@web34114.mail.mud.yahoo.com> ozz@ozz.is-a-geek.net wrote: > I toyed with the idea of getting a 650, but heard some real horror > stories about reception, etc. from people I knew who had them. In > fact, my local cell provider is refusing to even carry the new 700 > because the 650 bombed so badly. Remember, the 700w is _Pocket_PC_, _not_ PalmOS like the 600/650. I've only had the GSM versions (AT&T/Cingular, T-Mobile) of the Treo 600, not the CDMA versions (Sprint, Verizon). The GSM versions were just fine for me, although once Cingular F'up AT&T Wireless' network as of about February 2005, it went to crap (long story). My Treo 600 I've had since April 2005 on T-Mobile's network has been excellent. Even though T-Mobile has a smaller GSM network than Cingular, there is _no_ roaming charges, and I got _better_ reception than post-February 2005 on Cingular's network. > I opted for the Audiovox 6600 (the 6700 is not available to any > providers in my area yet, and won't be for at least another 6 > months, and I didn't want to wait that long). Is that PalmOS? -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 7 09:24:18 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <43E85BD9.30513.52CF337@ae4ko.amsat.org> Message-ID: <20060207142418.88613.qmail@web34112.mail.mud.yahoo.com> Aaron Morrison wrote: > I plan on swapping the electronic circuit board on the bottom long > enough to copy the data off of the platters. And yes, if your > board or drive dies, it will be replaced. Just be sure you have > backups of everything! Ironically, I only have 80GB and 160GB Western Digital drives. Western Digital is not very good in labelling models in the same series as either Hitachi/IBM, Maxtor or Seagate, so I don't know if they are the same series as your 120GB. BTW, I _never_ buy just *1* drive of the same model anymore, but at least 2. Typically I'm doing at least RAID-1, but I also like to buy 4 and use them in different systems so I avoid the single system issue like you had. My rationale is always that I'll recycle them as server drives on my Escalade 7000/8000 series cards in arrays of 4+ in 2 or so years when they are no longer current for my desktops. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From mflang at bellsouth.net Tue Feb 7 09:32:19 2006 From: mflang at bellsouth.net (mflang@bellsouth.net) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive Message-ID: <20060207143219.MDRB13866.ibm65aec.bellsouth.net@mail.bellsouth.net> > > Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. > I plan on swapping the electronic circuit board on the bottom long > enough to copy the data off of the platters. And yes, if your board A good question before you proceed is, do you have the proper bit to remove the screws on the board? Most Western Digital drives use a six-point star bit with a pretty flat tip to remove a screw with a very flat head. And it's hard to find this bit (I don't have one and I've searched), so it's hard to get these screws off. I use a small lock wrench from Sears, and carefully (lots of time and patience!) and slowly nudge them off. Mind the wrench's jaw doesn't scratch the tracings on the board! And if you have one of those bits, where the heck did you get it? Max. From wam at HiWAAY.net Tue Feb 7 11:43:59 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <20060207143219.MDRB13866.ibm65aec.bellsouth.net@mail.bellsouth.net> References: <20060207143219.MDRB13866.ibm65aec.bellsouth.net@mail.bellsouth.net> Message-ID: <43E8CE4F.3080609@HiWAAY.net> mflang@bellsouth.net wrote: >>>Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. >>> >>> >>I plan on swapping the electronic circuit board on the bottom long >>enough to copy the data off of the platters. And yes, if your board >> >> > >A good question before you proceed is, do you have the proper bit to remove the screws on the board? Most Western Digital drives use a six-point star bit with a pretty flat tip to remove a screw with a very flat head. And it's hard to find this bit (I don't have one and I've searched), so it's hard to get these screws off. I use a small lock wrench from Sears, and carefully (lots of time and patience!) and slowly nudge them off. Mind the wrench's jaw doesn't scratch the tracings on the board! > >And if you have one of those bits, where the heck did you get it? > >Max. > > Sounds like a Torx bit .... Sears & others have them, used in automotive & apparently computers :-). -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.matrixlist.com/pipermail/pc_support/attachments/20060207/de890d31/attachment.html From ozz at ozz.is-a-geek.net Tue Feb 7 12:24:57 2006 From: ozz at ozz.is-a-geek.net (ozz@ozz.is-a-geek.net) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <1424.71.252.176.10.1139322214.squirrel@qtmail.dgnal.net> References: <43E85FC3.30297.53C3E7E@ae4ko.amsat.org> <1424.71.252.176.10.1139322214.squirrel@qtmail.dgnal.net> Message-ID: <56244.69.176.47.130.1139333097.squirrel@www.ozz.is-a-geek.net> On Tue, 7 Feb 2006 06:21:10 -0800 (PST) "Bryan J. Smith" wrote: > > ozz@ozz.is-a-geek.net wrote: > > I toyed with the idea of getting a 650, but heard some real horror > > stories about reception, etc. from people I knew who had them. In > > fact, my local cell provider is refusing to even carry the new 700 > > because the 650 bombed so badly. > > Remember, the 700w is _Pocket_PC_, _not_ PalmOS like the 600/650. > > I've only had the GSM versions (AT&T/Cingular, T-Mobile) of the Treo > 600, not the CDMA versions (Sprint, Verizon). The GSM versions were > just fine for me, although once Cingular F'up AT&T Wireless' network > as of about February 2005, it went to crap (long story). GSM coverage here is very patchy to say the least. The 650s I mentioned and my 6600 are all CDMA. > My Treo 600 I've had since April 2005 on T-Mobile's network has been > excellent. Even though T-Mobile has a smaller GSM network than > Cingular, there is _no_ roaming charges, and I got _better_ reception > than post-February 2005 on Cingular's network. > > > I opted for the Audiovox 6600 (the 6700 is not available to any > > providers in my area yet, and won't be for at least another 6 > > months, and I didn't want to wait that long). > > Is that PalmOS? It's PocketPC unfortunately, however, it can be coaxed to run Linux. See http://www.handhelds.org/moin/moin.cgi/BlueAngel Regards, Ozz. From b.j.smith at ieee.org Tue Feb 7 12:30:47 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <43E8CE4F.3080609@HiWAAY.net> Message-ID: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> "William A. Mahaffey III" wrote: > Sounds like a Torx bit .... Sears & others have them, used in > automotive & apparently computers :-). Yep, many are T-10. A few vendors play games like putting a "key" in the center, so you can't use a standard Torx bit. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 7 12:40:08 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <56244.69.176.47.130.1139333097.squirrel@www.ozz.is-a-geek.net> Message-ID: <20060207174009.81260.qmail@web34114.mail.mud.yahoo.com> ozz@ozz.is-a-geek.net wrote: > GSM coverage here is very patchy to say the least. I was shocked that I got better GSM reception on Cingular and T-Mobile than CDMA, even old analog, from Verizon, when in the West Virginia mountains. > It's PocketPC unfortunately, however, it can be coaxed to run > Linux. > See http://www.handhelds.org/moin/moin.cgi/BlueAngel I'm interested in a supported OS out-of-the-box that is build for PDA functionality, works with stock connectors, etc... That's pretty much PalmOS only for me. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 7 12:44:55 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] ABS 8-Port GbE Switch for $51.99 - $20 Rebate ... Message-ID: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> Just FYI, this ABS 8-port GbE Switch is a decent entry-level switch. Although it only has a measly 144KiB buffer, it does have the all-important 802.3x hardware flow control support. Although Jumbo Frame support is listed, there is _no_ VLAN support, so all ports must be using the same frame rate. There is no indication whether or not it is auto-MDX (straight/crossover). It has a standard 3-prong A/C out, no external AC/DC adapter, which is a very nice touch for the price. http://dealnews.com/deals/ABS-8-Port-Gigabit-Ethernet-Switch-for-32-after-rebate/109438.html -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From dmckenna at thelimucompany.com Tue Feb 7 12:52:38 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:37 2006 Subject: [Pc_Support] ABS 8-Port GbE Switch for $51.99 - $20 Rebate ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6DD@tlc001.tlcusa.thelimucompany.com> > Although it only has a measly 144KiB buffer, it does have the > all-important 802.3x hardware flow control support. Although Jumbo > Frame support is listed, there is _no_ VLAN support, so all ports > must be using the same frame rate. OK, newbie qn here, but what difference does the frame size make? -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Tue Feb 7 12:55:45 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] ABS 5-Port GbE Switch for $31.99 - $20 Rebate (5-port version) In-Reply-To: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> Message-ID: <20060207175545.55624.qmail@web34101.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > Just FYI, this ABS 8-port GbE Switch is a decent entry-level > switch. Although it only has a measly 144KiB buffer, it does > have the all-important 802.3x hardware flow control support. > Although Jumbo Frame support is listed, there is _no_ VLAN support, > so all ports must be using the same frame rate. There is no > indication whether or not it is auto-MDX (straight/crossover). > It has a standard 3-prong A/C out, no external AC/DC adapter, which > is a very nice touch for the price. There is also a 5-port version, same specs: http://dealnews.com/deals/ABS-5-Port-Gigabit-Ethernet-Switch-for-12-after-rebate/109024.html BTW, someone noted on the 5-port review that Jumbo Frames did _not_ work for 2 cards they have previously tested as working with another switch. Doesn't surprise me, sometimes the specification "Jumbo Frame" has been used to only support 4,500 byte frames, but not the common 9,000 byte. It's interesting to note the rebates are _separate_, only 1 product listed on each rebate. This becons the question, can you get both? In any case, a cheap GbE switch with 802.3x support and an internal AC 3-prong is very, very nice for the price. http://images10.newegg.com/uploadfilesfornewegg/rebate/SH/ABS33-332-003Jan10Feb2806sl10.pdf http://images10.newegg.com/uploadfilesfornewegg/rebate/SH/ABS33-332-004Jan10Feb2806sl10.pdf Although I'm still planning on buying a refurbished NetGear GSM73xx series for my home network for a few hundred bucks, I can always use more switches for testing, InstallFests or as a spot/improptu repeater. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 7 13:12:05 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] ABS 8-Port GbE Switch for $51.99 - $20 Rebate ... In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6DD@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060207181205.17732.qmail@web34110.mail.mud.yahoo.com> Damien McKenna wrote: > OK, newbie qn here, but what difference does the frame size make? At 1,500 byte standard frames, an 10Base node must process at least 1,000 full-sized frames in a second using the full data transfer rate (up to 14,880 smaller frames per second if payloads are smaller). That increases to at least 10,000 frames in a second for FE. At GbE, that's almost 100,000 frames in a second! By using 9,000 byte Jumbo Frames, that reduces the number to under 15,000 frames in a second -- more on-par with FE. Overhead of incoming UDP/IP datagrams and, especially, TCP/IP segments can seriously overload the system, resulting in dropped frames for a node with today's commodity GbE hardware and their small buffers. Without 802.3x flow control, this can utterly _destroy_ performance. The overwhelming majority of commodity TCP/IP off-load engines are for _transmission_only_. They are _not_ used for receiption, where it's really needed. To get such, you need a $600+ GbE HBA and some well cached/buffered equipment. Until then, the best recommendations, in order of cost ... - 802.3x - NICs/switches with larger caches/buffers (NICs difficult to find) - Jumbo Frames, all nodes - Jumbo Frames, VLANs with localized traffic (separate routing) - Jumbo Frames, VLANs with layer-3 switching (wire-speed routing) GbE can be and is quite often implemented like crap. You do _not_ want to use SOHO networking equipment even in a small business. More than a few nodes and you can _toast_ any reason for GbE, reducing it to even worse than FE performance if you don't even have all nodes/switches with 802.3x. Lack of flow control with cheap GbE hardware is the new "collision-like nightmare." -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From dmckenna at thelimucompany.com Tue Feb 7 13:26:54 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] ABS 8-Port GbE Switch for $51.99 - $20 Rebate ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E2@tlc001.tlcusa.thelimucompany.com> Thanks for that, Bryan. So would there be any point in spending $100 for a few PCI cards and a small switch to boost the interoperability of a small number of servers? -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Tue Feb 7 14:46:40 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] GbE Design -- WAS: ABS 8-Port GbE Switch In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E2@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060207194640.57715.qmail@web34102.mail.mud.yahoo.com> Damien McKenna wrote: > Thanks for that, Bryan. > So would there be any point in spending $100 for a few PCI cards > and a small switch to boost the interoperability of a small number > of servers? Depends. If you are trying to do a separate, segmented "out-of-band" connection for servers, that's one good idea. Make sure the NICs _and_ switches can handle 9,000 byte jumbo frames. I don't know of too many $20 GbE cards that do Jumbo Frames though. You typically have to spend $50+ on a GbE NIC that has 16+KiB receive and transmit SRAM caches. Then add in $100 for a decent GbE switch that truly does Jumbo Frames and has 1+MiB of buffer. It doesn't have to do 802.1q VLAN, but it _must_ allow 9,000 byte Jumbo Frames (they will typically support up to 16KiB). If you want to start doing in-band (same NIC for everything), I really recommend you look at a good layer-3 switch. But that's more $$$, although under $1,000 these days even for GbE. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From jasonb at edseek.com Tue Feb 7 14:57:12 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs Message-ID: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> Anyone know just how much waste is created from these 'always-on' appliances? I just realized my TV is 'always' on. The DVD player is more obvious, since it has a standby light. What about CRT monitors in 'energy saver' mode? Is there any sizable savings in screwing with this stuff, or is it not worth the effort to calculate up how much power things draw? Also, where I live now has all halogen lights. How do those compare to CFBs? Thanks. From dmckenna at thelimucompany.com Tue Feb 7 15:01:55 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E6@tlc001.tlcusa.thelimucompany.com> > Anyone know just how much waste is created from these 'always-on' > appliances? Check the manuals of the items you're concerned with, I think they're required to state it. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Tue Feb 7 15:11:31 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> Message-ID: <20060207201131.47884.qmail@web34113.mail.mud.yahoo.com> Jason Boxman wrote: > Anyone know just how much waste is created from these 'always-on' > appliances? I just realized my TV is 'always' on. The DVD player > is more obvious, since it has a standby light. What about CRT > monitors in 'energy saver' mode? The problem with televisions, unlike computer monitors, is that they are not very interactive. There's really no way to gage feedback on use, especially when you are using disseparate components that have separate remote controls. DVD players are at least partially interactive, or the movie eventually ends withing a few hours. Computers are completely interactive and power down can occur within minutes and be fairly accurate on usage. > Is there any sizable savings in screwing with this stuff, or > is it not worth the effort to calculate up how much power things > draw? I went _completely_ fluorescent a couple years ago (over a matter ofa few weeks) and noticed not only a considerable savings on my electric bill, but I stopped replacing blubs. E.g., I like to leave the 40W light above my stove on 24x7 on as it provides just enough lumination throughout my house so I don't trip if I get up in the middle of the night. I was replacing incandescent blubs every 6 months or so. I replaced it with a 60W equivalent, 13W actual, fluourescent light about 2 years ago. I haven't replaced it since, and it provides even more light to reach the entire other end of the house, while still being low enough. > Also, where I live now has all halogen lights. How do those > compare to CFBs? Halogen lights are probably the "purest" light, fluorescent lights are the worst. _Both_ are _better_ than incandescent lights in energy efficiency. Use halogen when you want the best lighting. Use fluorescent when you want general lighting. Halogen has greater heat generation and must be cleaned more. Fluorescent makes a number of trade-offs for its increased efficiency. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Tue Feb 7 15:20:20 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> Message-ID: <43E90104.4020403@HiWAAY.net> Bryan J. Smith wrote: >"William A. Mahaffey III" wrote: > > >>Sounds like a Torx bit .... Sears & others have them, used in >>automotive & apparently computers :-). >> >> > >Yep, many are T-10. > >A few vendors play games like putting a "key" in the center, so you >can't use a standard Torx bit. > > > > Roger that. His description said flat in middle, so it sounds like he is in the clear .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.matrixlist.com/pipermail/pc_support/attachments/20060207/d67eb402/attachment.html From glaiacona at aikencountysc.gov Tue Feb 7 15:22:12 2006 From: glaiacona at aikencountysc.gov (George Laiacona) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... Message-ID: Actually, halogen lamps produce light that is more yellow than natural sunlight. Not a problem for humans, but might be so for anyone with aquariums. Algae love light that is more yellow. Some flourescent bulb manufacturers are producing standard bulbs that are more white, or "hotter" than the standard 40 watt bulb. 6500K to 7300K instead of the typical 4500K. Natual sunlight is about 9300K. (Layman's terms, and pretty simple description). I've read a lot about lighting for my aquarium hobby. I am by no means an expert. George. >>> "Bryan J. Smith" 02/07/06 3:11 PM >>> Jason Boxman wrote: > Anyone know just how much waste is created from these 'always-on' > appliances? I just realized my TV is 'always' on. The DVD player > is more obvious, since it has a standby light. What about CRT > monitors in 'energy saver' mode? The problem with televisions, unlike computer monitors, is that they are not very interactive. There's really no way to gage feedback on use, especially when you are using disseparate components that have separate remote controls. DVD players are at least partially interactive, or the movie eventually ends withing a few hours. Computers are completely interactive and power down can occur within minutes and be fairly accurate on usage. > Is there any sizable savings in screwing with this stuff, or > is it not worth the effort to calculate up how much power things > draw? I went _completely_ fluorescent a couple years ago (over a matter ofa few weeks) and noticed not only a considerable savings on my electric bill, but I stopped replacing blubs. E.g., I like to leave the 40W light above my stove on 24x7 on as it provides just enough lumination throughout my house so I don't trip if I get up in the middle of the night. I was replacing incandescent blubs every 6 months or so. I replaced it with a 60W equivalent, 13W actual, fluourescent light about 2 years ago. I haven't replaced it since, and it provides even more light to reach the entire other end of the house, while still being low enough. > Also, where I live now has all halogen lights. How do those > compare to CFBs? Halogen lights are probably the "purest" light, fluorescent lights are the worst. _Both_ are _better_ than incandescent lights in energy efficiency. Use halogen when you want the best lighting. Use fluorescent when you want general lighting. Halogen has greater heat generation and must be cleaned more. Fluorescent makes a number of trade-offs for its increased efficiency. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** _______________________________________________ Pc_support mailing list Pc_support@matrixlist.com http://lists.matrixlist.com/mailman/listinfo/pc_support From wam at HiWAAY.net Tue Feb 7 15:38:37 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> Message-ID: <43E9054D.30101@HiWAAY.net> Jason Boxman wrote: >Anyone know just how much waste is created from these 'always-on' >appliances? I just realized my TV is 'always' on. The DVD player is more >obvious, since it has a standby light. What about CRT monitors in 'energy >saver' mode? > >Is there any sizable savings in screwing with this stuff, or is it not worth >the effort to calculate up how much power things draw? > >Also, where I live now has all halogen lights. How do those compare to CFBs? > >Thanks. > > I seem to recall back to the '70's when we were told that the single biggest power user in the house was/is the fridge, followed closely by central A/C, then other large appliances (washer/dryer, depending on use cycles, chest freezer, beer fridge in the family room, etc.), *then* lighting, even back then w/ virtually *all* incandescent lights. I am (almost) all flourescent (sp?), have been for about 5 years. They are definitely long lasting, but didn't really make much of a dent in my power bill. I recently unplugged 2 of my 3 SGI Octanes (power burps got 'em), thus quieting 2 750 watt (!!!!) PSU's, we'll see how much difference that makes :-). I'm holding out w/ the beer-fridge & chest freezer though .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From octo at logicprobe.org Tue Feb 7 15:37:35 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> Message-ID: > Is there any sizable savings in screwing with this stuff, or is it not worth > the effort to calculate up how much power things draw? While I havn't messed with appliance stand-by power, I have taken increasing interests in computer and HVAC power usage. Also, I'm someone who has a very high electric bill which I've come to accept as a side-effect of my chosen ubergeek lifestyle. Trying to reduce your electric bill often feels like a game of trade-offs, with the ever-present question of how much you care about recurring monthly costs versus immediate one-time costs. The bottom line doesn't change as easily as you'd think. Assuming you have no desire to reduce the level of service/functionality provided by equipment/computers/etc, I've found the following factors: 1) Variation in per-kWh charges (fuel surcharges do change) 2) Variation in climate that affects HVAC usage 3) Power draw of equipment 4) Cost of replacing equipment with lower-power equivalent(s) You can't do anything about #1, and #2 oscilates over the course of the year. You can affect #3, though you'll often find that due to #4 you are lowering monthly recurring costs by a small enough amount that the bottom line won't even out or drop for a very long time. Then again, #4 can also be offset by lower-power equipment causing a reduction in HVAC usage. In the end, its a game that is a lot harder to win than you'd think. These days, I try to reduce my computer-power to what I see as the minimal that won't lose functionality (so no huge power-hungry boxes, if I can do the same job on something a bit smaller). Even still, my rack presently consumes about 1kW continuously, and my monthly bills range from $200-350 depending on time of year. (I've also noticed that managed network devices can use more power than you'd think.) -Derek From octo at logicprobe.org Tue Feb 7 15:39:57 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <43E9054D.30101@HiWAAY.net> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> <43E9054D.30101@HiWAAY.net> Message-ID: > didn't really make much of a dent in my power bill. I recently unplugged 2 of > my 3 SGI Octanes (power burps got 'em), thus quieting 2 750 watt (!!!!) > PSU's, we'll see how much difference that makes :-). I'm holding out w/ the > beer-fridge & chest freezer though .... Having a 750W PSU does *not* mean you're actually using anywhere near that much power. I've got a machine with a 670W PSU that only draws 300W, and plenty of similar examples. The only way to know how much power something uses is to measure it. -Derek From pberry2 at cfl.rr.com Tue Feb 7 16:05:06 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <20060207143219.MDRB13866.ibm65aec.bellsouth.net@mail.bellsouth.net> References: <20060207143219.MDRB13866.ibm65aec.bellsouth.net@mail.bellsouth.net> Message-ID: <200602071605.06710.pberry2@cfl.rr.com> On Tuesday 07 February 2006 09:32, mflang@bellsouth.net wrote: > > > Aaron: I have a 120 GB WD1200 as the second hd on one of my boxes. > > > > I plan on swapping the electronic circuit board on the bottom long > > enough to copy the data off of the platters. And yes, if your board > > A good question before you proceed is, do you have the proper bit to remove > the screws on the board? Most Western Digital drives use a six-point star > bit with a pretty flat tip to remove a screw with a very flat head. And > it's hard to find this bit (I don't have one and I've searched), so it's > hard to get these screws off. I use a small lock wrench from Sears, and > carefully (lots of time and patience!) and slowly nudge them off. Mind the > wrench's jaw doesn't scratch the tracings on the board! > > And if you have one of those bits, where the heck did you get it? > > Max. > I bought two TORX T8 x 60 tools, that are a total 5 inches long, permanent screwdriver type, with a red handle, by email from one of the tool vendors. Made in Germany by van-moly boan Part #362 Imported by Wiha tools, slod by Tecratools, it cost me $4.30 each plus postage. Got it to assemble Intel passive heatsinks on PII processors. http://www.tecratools.com/pages/service/torx_drivers.html Others, different, more expensive: one is here, a bit more expensive, though at $8.99: http://aps-mobile.com/cellular-phone-accessories/cellular-accessory/TORX8.html Here is the 1/4" socket from SK tools, $2.22 http://www.sjdiscounttools.com/skt81798.html -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From dmckenna at thelimucompany.com Tue Feb 7 16:07:58 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E7@tlc001.tlcusa.thelimucompany.com> > (I've also noticed that managed network devices can use more > power than you'd think.) I'd wager that anything that uses a generic 3-prong power cord is probably going to use a good chunk of power, e.g. the ABS GigE switches Bryan mentioned today. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Tue Feb 7 16:15:39 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E7@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060207211539.71923.qmail@web34113.mail.mud.yahoo.com> Damien McKenna wrote: > I'd wager that anything that uses a generic 3-prong power cord is > probably going to use a good chunk of power, e.g. the ABS GigE > switches Bryan mentioned today. Huh? Just because they put the A/C to D/C conversion inside instead of outside doesn't mean anything. It all depends on how it's done. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 7 16:17:52 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs Message-ID: <20060207211752.17179.qmail@web34105.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > Huh? Just because they put the A/C to D/C conversion inside > instead of outside doesn't mean anything. It all depends on how > it's done. In fact, most FE/GbE switches are Broadcom and RealTek reference designs. They typically use less than 10W overall. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From jasonb at edseek.com Tue Feb 7 16:26:19 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060207201131.47884.qmail@web34113.mail.mud.yahoo.com> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> <20060207201131.47884.qmail@web34113.mail.mud.yahoo.com> Message-ID: <46092.216.134.200.78.1139347579.squirrel@nebula.internal.foo> Bryan J. Smith said: > Jason Boxman wrote: >> Anyone know just how much waste is created from these 'always-on' >> appliances? I just realized my TV is 'always' on. The DVD player >> is more obvious, since it has a standby light. What about CRT >> monitors in 'energy saver' mode? > > The problem with televisions, unlike computer monitors, is that they > are not very interactive. There's really no way to gage feedback on > use, especially when you are using disseparate components that have > separate remote controls. Well, I do turn it off when I'm not watching it and there is anyone else around so there's no risk of it being on unattended. I rarely use it more than an hour a night anyway. >> Is there any sizable savings in screwing with this stuff, or >> is it not worth the effort to calculate up how much power things >> draw? > > I went _completely_ fluorescent a couple years ago (over a matter ofa > few weeks) and noticed not only a considerable savings on my > electric bill, but I stopped replacing blubs. > > E.g., I like to leave the 40W light above my stove on 24x7 on as it > provides just enough lumination throughout my house so I don't trip > if I get up in the middle of the night. I was replacing incandescent > blubs every 6 months or so. I replaced it with a 60W equivalent, 13W > actual, fluourescent light about 2 years ago. I haven't replaced it > since, and it provides even more light to reach the entire other end > of the house, while still being low enough. Yeah, I tried that at my last place, but I rarely have _any_ lights on, so the savings wasn't immediately detectable. I'm one of those light police that turn off every single light that isn't being used. The nice thing about CFB that I've heard is there isn't much start up cost, so it's pretty 'free' to turn them on and off liberally. Not having to replace bulbs all the time was nice, too, but also not my primary goal. (Although how much are those incandescent bulbs per unit anyway. I haven't replaced one in so long, I forgot about that cost.) >> Also, where I live now has all halogen lights. How do those >> compare to CFBs? > > Halogen lights are probably the "purest" light, fluorescent lights > are the worst. _Both_ are _better_ than incandescent lights in > energy efficiency. > > Use halogen when you want the best lighting. > Use fluorescent when you want general lighting. > > Halogen has greater heat generation and must be cleaned more. > Fluorescent makes a number of trade-offs for its increased > efficiency. Where I live now, everything is pretty wired for cute mini-halogen lights. All the fixtures I have seen that are part of the structure of the building, if you will, take little halogen bulbs. Not sure how I'd even get a CFB in one of those little sockets. The kitchen has one of those overhead 5V or 12V setups with six little halogen lights installed. As you can see, few, if any places can accept traditional bulbs or CFBs. I'm also a cold whore, so I nearly never use the heater in the winter in Florida. It routinely hovers around 60-70F and it doesn't bother me. I sleep with my fan on maximum speed at night, even in the dead of winter. Now, during the summer my CHA gets a real workout. I usually set the 'stat at around 75-77, at least at the old place. I don't know how well air circulates at my new place. It has a pretty modern CHA unit that I'm told is pretty efficient, so I might get off easy. One thing's for sure. My five boxes sure put off a ton of BTUs. You can seriously feel the difference in temperature, which the A/C has to eventually cool down to a reasonable level for me. I guess I could replace everything with lower power stuff, but as Derek mentioned, that's a game of numbers and a potential loser. When I do buy new stuff, I'll certainly keep an eye on power efficiency, but I won't upgrade just to chase potential savings. On that note, how bad does it hurt to have PSUs that have lower efficiency ratings. What's that mean, anyway? For example the ones I have now have something like 70% or 75% efficiency. From pberry2 at cfl.rr.com Tue Feb 7 16:45:12 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E6@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E6@tlc001.tlcusa.thelimucompany.com> Message-ID: <200602071645.12330.pberry2@cfl.rr.com> On Tuesday 07 February 2006 15:01, Damien McKenna wrote: > > Anyone know just how much waste is created from these 'always-on' > > appliances? > > Check the manuals of the items you're concerned with, I think they're > required to state it. My 17" LCD monitor uses 40 watts when on, 20 watts on standby, 0 watts when we turn it off. We notice very little warm up time... have two My 17"CRT monitor uses 400 watts when on, and 40 watts in standby mode, 0 watts when off. Have 5. $89.95 each, vs. $300 for the LCDs. Upgrading will take a while on my budget. Fortunately, I have two KVM switches, so, many monitors are spares. My computers that have "350" to "400" watt psu's actually burn that, but, I have to run a dozen UPS's, which also use about 400 watts, each. So, my 4 KW of stuff that is on is a necessary cost of my computer endeavor/hobby. Now then, let's talk parasitic devices. Those wall-warts that charge cell phones, run toys, charge batteries, power switch-hubs, all suck 5 to 60 watts, all the time, just by being plugged in! Add them up. Then, put them all on switched outlets, and turn off when not in actual use. Fans? Fans cool people, not rooms. A ceiling fan can run 300-400 watts per hour. If you leave the room turn it off! Check out your Water Heater. Is the temp on both heating elements set to 120 degrees? In the past ten years or so, most water heaters are set to about 120 degrees, from the factory. I installed a $40 timer clock, (in homes that I own!) and set my heater for two heating cycles each day, as the water stays hot for up to about 8 hours, and the heating cycles should be on for about 2 to 3 hours. REMEMBER! Only Crazy Patrick plays with Electircity! If you don't know what you are doing, then you need to hire a licensed electrician to install the clock, as it switches 250 Volts AC!! That is LETHAL voltage, if you get 'bit'! And, be safe and disable circuits, before working! The Dishwasher has a HOT water heater internally. It really uses some KW. We don't run any special loads that do it. Normal cycle, and, we have a water softener that uses Osmosis (salt process). Everytime I buy a home, I install some energy saving items, and the clock is one. Another is to change all the light bulbs on porches to the flourescent tube type that burn 10 watts, instead of 60 watts, or more. There are some energy management programs that your computer can use, to run the X10 light and appliance outlets/switches in your home. At $5 to $10 per remote, with a cost of about $40 for the base unit, you can have remote control of many devices in your home. http://www.x10ideas.com/ http://www.x10.com/techtoysab2.htm http://www.x10.com/activehomepro/activehome-pro.html?44cddac65a757a6e43e91398 There are also some Open Source versions, free for download. http://www.kevinboone.com/home-automation.html -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From pberry2 at cfl.rr.com Tue Feb 7 16:57:10 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> Message-ID: <200602071657.10971.pberry2@cfl.rr.com> On Tuesday 07 February 2006 12:30, Bryan J. Smith wrote: > "William A. Mahaffey III" wrote: > > Sounds like a Torx bit .... Sears & others have them, used in > > automotive & apparently computers :-). > > Yep, many are T-10. > > A few vendors play games like putting a "key" in the center, so you > can't use a standard Torx bit. Called a "security" or "Tamper-proof" torx. Most the drives and cpu heatsinks I see are T8. start at $6.63 here: http://www.wihatools.com/362trSF.htm and regular Torx are starting at $4.30 here: http://www.wihatools.com/indexes/indxtorx.htm -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From b.j.smith at ieee.org Tue Feb 7 17:06:49 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <46092.216.134.200.78.1139347579.squirrel@nebula.internal.foo> Message-ID: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> Jason Boxman wrote: > Well, I do turn it off when I'm not watching it and there is > anyone else around so there's no risk of it being on unattended. > I rarely use it more than an hour a night anyway. I'm the same way, although between my wife and I, it's on maybe 3-4 hours/evening total. > Yeah, I tried that at my last place, but I rarely have _any_ > lights on, so the savings wasn't immediately detectable. I'm > one of those light police that turn off every single light that > isn't being used. My wife and I do a bit of work off-computer, so we need 100-200W illumination on a regular basis. Whether it's the 100W in our bedrooms, the 3x60W cluster in our living room or bedroom, etc... Going down to 23W (100W) or 3x13W (3x60W) made a good $10-20/month difference. > The nice thing about CFB that I've heard is there isn't much start > up cost, so it's pretty 'free' to turn them on and off liberally. Most lights today shouldn't have much of a startup cost. > Not having to replace bulbs all the time was nice, too, but > also not my primary goal. (Although how much are those > incandescent bulbs per unit anyway. I haven't replaced > one in so long, I forgot about that cost.) The cost is about equal overall. I just got tired of replacing them regularly. I haven't had to replace an CFB yet, except one I broke. > As you can see, few, if any places can accept traditional bulbs > or CFBs. Well, my house was built in 1987, so there's plenty of traditional sockets. Plus any fixtures I've put in take the traditional sockets as well. > I'm also a cold whore, so I nearly never use the heater in the > winter in Florida. It routinely hovers around 60-70F and it > doesn't bother me. I'm the same, although I try to keep it above 65F so I don't go losing my foundation (typically below 55F). > Now, during the summer my CHA gets a real workout. I usually set > the 'stat at around 75-77, at least at the old place. I don't know > how well air circulates at my new place. I used to set it at 75-77F, but with the hurricanes in 2004 taking off part of my roof, that was a mold magnet. I immediately killed 2 birds with 1 stone by getting a dehumidfier. I keep it at 30% humidity (it was easily 65-75% before), which means I can leave my house at 80F and it feels like 76F (all while still not being a static electicity issue). Even though my roof has been replaced, I find that I don't have nearly as much mold that grows in the bathrooms or kitchen between cleanings as a result. My house has the original A/C and I'm probably going replace it within the next year or two. I'm definitely getting one with a dehumidifier built-in -- it's well worth the cost. > It has a pretty modern CHAunit that I'm told is pretty efficient, > so I might get off easy. The original A/C unit I have is fairly efficient for a 1987 model when I bought the house in 1997. But it's eventually going to give. > One thing's for sure. My five boxes sure put off a ton of BTUs. My wife and I have our MicroATX cubes with sub-50W LCDs. My system is a 31-55W Athlon 64 3200+, although hers is an older Athlon XP2600+. Probably the kickers in our systems are the nVidia GeForce 7800GTX and 6800GT, respectively. But they seem to work fine on 300-400W PSes, and we take the down regularly. Firewall is an old Pentium MMX 166MHz in a NLX form-factor that probably doesn't use 100W. Have really no need to change that, other than to maybe put in redundant disks. My main server is a dual P3 750MHz ServerWorks IIILE with six (6) 80GB Maxtor/WD drives on a 3Ware Escalade 7800 currently. That might be temporarily replaced, still working on my plans. I just haven't had time. I still have that dual-Athlon MP2400+ system, it just sits unused. I also have seven (7) Seagate Baracuda ATA 7200.7 200GB drives just sitting around. Plus I also have that NetCell SR5000 card (64-bit, 66MHz, 5-port PATA w/128MB SDRAM). Probably my biggest power guzzler is a newly acquired Dell PowerEdge 6300 (old 4x Xeon 550MHz, i450NX chipset, 6xLVD-SCSI, 3x redundant PS) that I'm thinking of using temporarily when I reconfigure my server. Just haven't had time, but I wanted to grab it when it was out there on Craigslist -- especially since I have several Seagate 50GB LVD SCSI SCA disks that can be used in it. > You can seriously feel the difference in temperature, which the > A/C has to eventually cool down to a reasonable level for me. I have no such issues with my MicroATX boxen, and not even the P3 server. The dual-Athlon MP is a different story and that old PowerEdge 6300 is definitely a heat incubator. > I guess I could replace everything with lower power stuff, > but as Derek mentioned, that's a game of numbers and a potential > loser. Other than video card, I'm glad I did. My MicroATX cubes are so quiet and fairly low-heat, plus I can place them anywhere without worry about space. > When I do buy new stuff, I'll certainly keep an eye on power > efficiency, but I won't upgrade just to chase potential savings. > On that note, how bad does it hurt to have PSUs that have lower > efficiency ratings. What's that mean, anyway? For example the > ones I have now have something like 70% or 75% efficiency. It can make a little difference. The Seasonic S12 I'm using in my system is over 80% efficient, and I'm sure I'm only using half of its 500W rating. It doesn't get very warm at all. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Tue Feb 7 18:19:51 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> <43E9054D.30101@HiWAAY.net> Message-ID: <43E92B17.2000600@HiWAAY.net> Derek Konigsberg wrote: >> didn't really make much of a dent in my power bill. I recently >> unplugged 2 of my 3 SGI Octanes (power burps got 'em), thus quieting >> 2 750 watt (!!!!) PSU's, we'll see how much difference that makes >> :-). I'm holding out w/ the beer-fridge & chest freezer though .... > > > Having a 750W PSU does *not* mean you're actually using anywhere near > that much power. I've got a machine with a 670W PSU that only draws > 300W, and plenty of similar examples. The only way to know how much > power something uses is to measure it. > > -Derek > Absolutely true :-). I *did* have 2 GB of RAM & 3 largish (when purchased) SCSI HDD's on both of these, however .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From wam at HiWAAY.net Tue Feb 7 18:29:08 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602071645.12330.pberry2@cfl.rr.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D6E6@tlc001.tlcusa.thelimucompany.com> <200602071645.12330.pberry2@cfl.rr.com> Message-ID: <43E92D44.4020307@HiWAAY.net> Patrick wrote: > >Fans? Fans cool people, not rooms. A ceiling fan can run 300-400 watts per >hour. If you leave the room turn it off! > > Hmmmmm .... where did you get that number (300-400 W/h for ceiling fans), sounds quite a bit high to me, more like 1/3 - 1/2 that ? >Check out your Water Heater. Is the temp on both heating elements set to 120 >degrees? In the past ten years or so, most water heaters are set to about >120 degrees, from the factory. > > Most hot water heaters are pretty well insulated; mine is virtually ambient to the touch, & I run it at 180 degrees. Use less hot water per shower that way, and as you observe, they do stay warm quite a while once charged up .... >I installed a $40 timer clock, (in homes that I own!) and set my heater for >two heating cycles each day, as the water stays hot for up to about 8 hours, >and the heating cycles should be on for about 2 to 3 hours. REMEMBER! Only >Crazy Patrick plays with Electircity! If you don't know what you are doing, >then you need to hire a licensed electrician to install the clock, as it >switches 250 Volts AC!! That is LETHAL voltage, if you get 'bit'! And, be >safe and disable circuits, before working! > >The Dishwasher has a HOT water heater internally. It really uses some KW. >We don't run any special loads that do it. Normal cycle, and, we have a water >softener that uses Osmosis (salt process). > > Agree w/ everything else :-). -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From octo at logicprobe.org Tue Feb 7 18:43:31 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> Message-ID: <200602071843.32114.octo@logicprobe.org> For the run-down on my actual power usage of machines in the computer room, look here: http://www.logicprobe.org/~octo/stuff/power.txt For more detail about each of the listed machines/devices: http://www.logicprobe.org/systems/ Frankly, I find these numbers quite interesting. It goes to show that an Athon64 whizbanger uses a lot less power than you'd expect, and the same goes for my 6-drive file server. Also shows just how much power my network devices use. -- --------------------------- Derek Konigsberg octo@.logicprobe.org http://hecgeek.blogspot.com --------------------------- From philb at philb.us Tue Feb 7 18:52:39 2006 From: philb at philb.us (Phil Barnett) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <200602071657.10971.pberry2@cfl.rr.com> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> <200602071657.10971.pberry2@cfl.rr.com> Message-ID: <200602071852.39751.philb@philb.us> On Tuesday 07 February 2006 16:57, Patrick wrote: > On Tuesday 07 February 2006 12:30, Bryan J. Smith wrote: > > "William A. Mahaffey III" wrote: > > > Sounds like a Torx bit .... Sears & others have them, used in > > > automotive & apparently computers :-). > > > > Yep, many are T-10. > > > > A few vendors play games like putting a "key" in the center, so you > > can't use a standard Torx bit. > > Called a "security" or "Tamper-proof" torx. Most the drives and cpu > heatsinks I see are T8. start at $6.63 here: > http://www.wihatools.com/362trSF.htm > > and regular Torx are starting at $4.30 here: > http://www.wihatools.com/indexes/indxtorx.htm Every year at hamcation, there are at least 3 vendors that sell the security bit sets for $10 or less. (my last one was $8) I'd suggest getting one there. -- "Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next. From philb at philb.us Tue Feb 7 18:55:00 2006 From: philb at philb.us (Phil Barnett) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> Message-ID: <200602071855.00217.philb@philb.us> On Tuesday 07 February 2006 14:57, Jason Boxman wrote: > Anyone know just how much waste is created from these 'always-on' > appliances? I just realized my TV is 'always' on. The DVD player is more > obvious, since it has a standby light. What about CRT monitors in 'energy > saver' mode? > > Is there any sizable savings in screwing with this stuff, or is it not > worth the effort to calculate up how much power things draw? The guys who build solar electric houses call these loads 'Phantom Loads'. Google it. -- "Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next. From b.j.smith at ieee.org Tue Feb 7 18:55:01 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <200602071843.32114.octo@logicprobe.org> Message-ID: <20060207235501.66353.qmail@web34107.mail.mud.yahoo.com> Derek Konigsberg wrote: > Frankly, I find these numbers quite interesting. It goes to show > that an Athon64 whizbanger uses a lot less power than you'd expect, Why would you assume otherwise? Even though AMD lags Intel 12+ months in multi-billion dollar fabrication technology out of lack of fund reserves, their 1999 design is much more efficient than the NetBurst hack. Simple rule of series and scalability: - Intel i686 (PPro-PIII-PM) scales from 200-1.0GHz, 1.5GHz with first async refit, 2.1GHz with redesign (Pentium M). 7+1 pipelines (2 ALU, 2 FPU + 1 SSE in P3) with stages are typically 12 or less. - AMD K7-8 scales from 500MHz to 2.0GHz, 2.8+GHz with refits. 9 pipelines (3 ALU, 3 FPU) with stages typically 20 or less. - Intel NetBurst (P4) scales from 1.5GHz to 3.5GHz, 4.0+GHz with refits. 7+2 pipelines (2 ALU, 1+1 FPU + 2 SSE) with stages typically 40 or so. Intel i686 is more power efficient MHz for MHz than AMD K7-8, but the latter scales higher. The Intel NetBurst has a much higher MHz scalability, but absolutely sucks up both power and wastes stages. The concept that AMD uses more power than Intel _died_ once the NetBurst hack was introduced. Intel is _dropping_ NetBurst and going back to i686 for new products. They are clearly giving up the MHz crown (they've already lost the performance crown) for power efficiency. The current 2.1GHz Pentium-M products use as little as 21W, no more than around 45W at full load. The AMD Athlon 64 1.8-2.4GHz (3000-3700+) use as little as 31W, no more than 55W at full load. Intel NetBurst can't break less than 70W. Then again, NetBurst was designed rather quickly, and did not have a full development cycle -- unlike i686 or K7-8. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Tue Feb 7 18:59:43 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060207235501.66353.qmail@web34107.mail.mud.yahoo.com> References: <20060207235501.66353.qmail@web34107.mail.mud.yahoo.com> Message-ID: <200602071859.44123.octo@logicprobe.org> On Tuesday 07 February 2006 6:55 pm, Bryan J. Smith wrote: > Derek Konigsberg wrote: > > Frankly, I find these numbers quite interesting. It goes to show > > that an Athon64 whizbanger uses a lot less power than you'd > > expect, > > Why would you assume otherwise? > > Even though AMD lags Intel 12+ months in multi-billion dollar > fabrication technology out of lack of fund reserves, their 1999 > design is much more efficient than the NetBurst hack. > You totally missed my point :-) I wasn't implying that AMD made more power hungry chips than Intel, or vice versa. What I was trying to say, is that a nicely loaded PC desktop doesn't use anywhere near as many watts off the cord as you'd think. (especially given how the average PC tweakers seem to talk about their suped-up machines and high-end power supplies) -- --------------------------- Derek Konigsberg octo@.logicprobe.org http://hecgeek.blogspot.com --------------------------- From philb at philb.us Tue Feb 7 19:01:06 2006 From: philb at philb.us (Phil Barnett) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: References: Message-ID: <200602071901.06874.philb@philb.us> On Tuesday 07 February 2006 15:22, George Laiacona wrote: > E.g., I like to leave the 40W light above my stove on 24x7 on as it > provides just enough lumination throughout my house so I don't trip > if I get up in the middle of the night. ?I was replacing incandescent > blubs every 6 months or so. ?I replaced it with a 60W equivalent, 13W > actual, fluourescent light about 2 years ago. ?I haven't replaced it > since, and it provides even more light to reach the entire other end > of the house, while still being low enough. My first cf bulb was installed when I first bought the house here in Apopka, 17 years ago. It's in the hallway. It's one of those 4.5" diameter GE builbs. This one is old enough that it doesn't have an electric ballast and I hear the stater pinging it to live each time it starts. It's still burning after 17 years and it still gets plenty bright to keep us from tripping over things in the hobby room, kitchen, bedrooms and more. Quite amazing, actually. Since then, I've pretty much replaced all the filament bulbs with cf. You get a second benefit from them in that they don't product heat that the AC then has to remove. -- "Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next. From wam at HiWAAY.net Tue Feb 7 19:11:35 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <200602071901.06874.philb@philb.us> References: <200602071901.06874.philb@philb.us> Message-ID: <43E93737.1010901@HiWAAY.net> Phil Barnett wrote: >On Tuesday 07 February 2006 15:22, George Laiacona wrote: > > >>E.g., I like to leave the 40W light above my stove on 24x7 on as it >>provides just enough lumination throughout my house so I don't trip >>if I get up in the middle of the night. I was replacing incandescent >>blubs every 6 months or so. I replaced it with a 60W equivalent, 13W >>actual, fluourescent light about 2 years ago. I haven't replaced it >>since, and it provides even more light to reach the entire other end >>of the house, while still being low enough. >> >> > >My first cf bulb was installed when I first bought the house here in Apopka, >17 years ago. It's in the hallway. It's one of those 4.5" diameter GE builbs. >This one is old enough that it doesn't have an electric ballast and I hear >the stater pinging it to live each time it starts. > >It's still burning after 17 years and it still gets plenty bright to keep us >from tripping over things in the hobby room, kitchen, bedrooms and more. > >Quite amazing, actually. > >Since then, I've pretty much replaced all the filament bulbs with cf. You get >a second benefit from them in that they don't product heat that the AC then >has to remove. > > > Roger that last point !!!! That was my problem w/ the now-moribund SGI's, you walk into that room & it was a good 8-10 degrees warmer than the hall, so I was getting it twice, power the PSU's, then remove the heat. -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.matrixlist.com/pipermail/pc_support/attachments/20060207/99656609/attachment.html From pberry2 at cfl.rr.com Wed Feb 8 06:56:02 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <200602071852.39751.philb@philb.us> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> <200602071657.10971.pberry2@cfl.rr.com> <200602071852.39751.philb@philb.us> Message-ID: <200602080656.02208.pberry2@cfl.rr.com> On Tuesday 07 February 2006 18:52, Phil Barnett wrote: > On Tuesday 07 February 2006 16:57, Patrick wrote: > > On Tuesday 07 February 2006 12:30, Bryan J. Smith wrote: > > > "William A. Mahaffey III" wrote: > > > > Sounds like a Torx bit .... Sears & others have them, used in > > > > automotive & apparently computers :-). > > > > > > Yep, many are T-10. > > > > > > A few vendors play games like putting a "key" in the center, so you > > > can't use a standard Torx bit. > > > > Called a "security" or "Tamper-proof" torx. Most the drives and cpu > > heatsinks I see are T8. start at $6.63 here: > > http://www.wihatools.com/362trSF.htm > > > > and regular Torx are starting at $4.30 here: > > http://www.wihatools.com/indexes/indxtorx.htm > > Every year at hamcation, there are at least 3 vendors that sell the > security bit sets for $10 or less. (my last one was $8) > > I'd suggest getting one there. But, one caveat on the bit set, is that they insert into a driver and that is sometimes too large in diameter to reach through the heatsinks, to use on computer processors. In those instances, the Wiha Torx driver is great. -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From octo at logicprobe.org Wed Feb 8 07:06:27 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <200602080656.02208.pberry2@cfl.rr.com> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> <200602071852.39751.philb@philb.us> <200602080656.02208.pberry2@cfl.rr.com> Message-ID: <200602080706.28219.octo@logicprobe.org> > But, one caveat on the bit set, is that they insert into a driver and that > is sometimes too large in diameter to reach through the heatsinks, to use > on computer processors. In those instances, the Wiha Torx driver is great. I'm still amazed at how we got into this whole long discussion of torx screwdrivers, and no one actually bothered to mention that circuit-board-swapping may actually not work. The two drives may not be 100% identical. Even if they are, the electronics are bound to have some unique information on bad sectors, sector-remapping, and the like. In other words, the logical-to-physical mappings of the two drives may not be identical in any case. Oh, and I know there are places that specialize in recovering data from dead hard drives. These places are even qualified to open and fix the hard drives in a clean environment. I just don't know of any such places around here. (One of these places was located near where I used to live, back in high school.) -- --------------------------- Derek Konigsberg octo@.logicprobe.org http://hecgeek.blogspot.com --------------------------- From ae4ko at amsat.org Wed Feb 8 09:08:50 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <200602080706.28219.octo@logicprobe.org> References: <200602080656.02208.pberry2@cfl.rr.com> Message-ID: <43E9B522.15354.1D6C64@ae4ko.amsat.org> On 8 Feb 2006 at 7:06, Derek Konigsberg wrote: > I'm still amazed at how we got into this whole long discussion of torx > screwdrivers, and no one actually bothered to mention that > circuit-board-swapping may actually not work. The two drives may not > be 100% identical. Even if they are, the electronics are bound to > have some unique information on bad sectors, sector-remapping, and the > like. In other words, the logical-to-physical mappings of the two > drives may not be identical in any case. > And I'm aware of that. However, when a drive recovery place wants between $500 and $3000 to recover data, it's worth the time and effort to give it a try. From jasonb at edseek.com Wed Feb 8 11:40:07 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:38 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <200602080706.28219.octo@logicprobe.org> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> <200602080656.02208.pberry2@cfl.rr.com> <200602080706.28219.octo@logicprobe.org> Message-ID: <200602081140.08183.jasonb@edseek.com> On Wednesday 08 February 2006 07:06, Derek Konigsberg wrote: > I'm still amazed at how we got into this whole long discussion of torx > screwdrivers, and no one actually bothered to mention that > circuit-board-swapping may actually not work. The two drives may not be > 100% identical. Even if they are, the electronics are bound to have some > unique information on bad sectors, sector-remapping, and the like. In > other words, the logical-to-physical mappings of the two drives may not be > identical in any case. True, but it also may. I sent someone my old, dead Maxtor 120GB which he found had his exact serial # on my Web site. He later informed me that using its PCB he was able to successfully recover his data. He had tried other PCBs, but as you said there had been compatibility issues. So it's still worth a shot, even multiple times, if your data is really important, but you're on a budget or could never afford professional recovery services anyway. -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Wed Feb 8 13:35:21 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Advice request: Windows non-raid -> Windows RAID on 7006-2 In-Reply-To: <43EA3356.7030405@lubik.ca> Message-ID: <20060208183521.58664.qmail@web34109.mail.mud.yahoo.com> Ugo Bellavance wrote: > Hi Brian, > I have to do this task: > Take a workstation with Windows XP pro + other apps + data, that is > running on only one, 13-gig HDD and make it run on a RAID1, powered > by a 3ware 7006-2 and 2 WD800JB, without reinstalling windows. Is > it possible? Yes, with one huge caveat on NT5+ (2000+). > I thought of that as a potential solution: > 1- Install the 3ware drivers into windows, with or w/o the 3ware > card inside and/or the 2 disks. Then shutdown the system, create > the RAID1 array with the 2 WD, use ghost to copy the 13 GB drive > to the array, then remove the 13 GB, tell the BIOS to boot on the > controller, then hope it works. Unfortunately, that doesn't solve the huge caveat. To boot NTLDR, you need: A) The Registry settings for the first storage device to match the actual B) 3Ware storage controller driver installed as C:\NTBOOTDD.SYS (which is required when a stock ATA driver isn't used as the boot device) "B" can be handled by booting into Rescue Mode with the CD (and hitting F6 at boot to load the 3Ware driver), and running "fixboot" (possibly "fixmbr" as well, although it's probably not needed). "A" is the big issue. Installing the drivers before rebooting _may_ work, but it typically hasn't for me when the driver isn't just a simple ATA controller change. > I'll do a ghost image on an external USB HDD just in case... Ghost, Partimage, etc... isn't the issue. Although it will prevent the NT5+ (2000+) kernel from relabeling drive letters, it doesn't solve A or B. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Wed Feb 8 15:16:53 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Advice request: Windows non-raid -> Windows RAID on 7006-2 In-Reply-To: <43EA3BE8.5000807@lubik.ca> Message-ID: <20060208201653.42620.qmail@web34104.mail.mud.yahoo.com> Ugo Bellavance wrote: > So I have no choice but to re-install? Not exactly. 1) Again, installing the driver beforehand _might_ work, especially if you take the 3Ware .SYS file, copy it as C:\NTBOOTDD.SYS 2) Alternatively, you can install a "helper NT5+ install" (in an alternate directory on the new 3Ware volume as C:), read through it's registry (I can't remember the 3 keys set for the boot device, read through MSDN articles), then boot the Linux NT Password Recovery/Registry Editor CD and manually edit the registry to match. I've done this #2 personally. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From jasonb at edseek.com Wed Feb 8 16:05:05 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602071855.00217.philb@philb.us> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> <200602071855.00217.philb@philb.us> Message-ID: <49266.216.134.200.78.1139432705.squirrel@nebula.internal.foo> Phil Barnett said: > On Tuesday 07 February 2006 14:57, Jason Boxman wrote: >> Anyone know just how much waste is created from these 'always-on' >> appliances? I just realized my TV is 'always' on. The DVD player is more >> obvious, since it has a standby light. What about CRT monitors in 'energy >> saver' mode? >> >> Is there any sizable savings in screwing with this stuff, or is it not >> worth the effort to calculate up how much power things draw? > > The guys who build solar electric houses call these loads 'Phantom Loads'. > Google it. Is there any long term risk to continuously cycling these devices at the wall with a switch? Do the caps wear out sooner, or some such? I don't want to kill my five year old TV before I replace it some day... From b.j.smith at ieee.org Wed Feb 8 16:39:09 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <49266.216.134.200.78.1139432705.squirrel@nebula.internal.foo> Message-ID: <20060208213909.23810.qmail@web34103.mail.mud.yahoo.com> Jason Boxman wrote: > Is there any long term risk to continuously cycling these devices > at the wall with a switch? Do the caps wear out sooner, or some > such? I don't want to kill my five year old TV before I replace it > some day... I have this newer invention called the "remote control." They even have "universal remote controls." They remove the requirement to use a switch. (sorry, couldn't resist ;-) -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From whittake at sbaflorida.com Wed Feb 8 16:52:02 2006 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <200602071843.32114.octo@logicprobe.org> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602071843.32114.octo@logicprobe.org> Message-ID: <1139435522.807.75.camel@localhost.localdomain> On Tue, 2006-02-07 at 18:43 -0500, Derek Konigsberg wrote: > For the run-down on my actual power usage of machines in the computer room, > look here: > http://www.logicprobe.org/~octo/stuff/power.txt Assuming most these machines are on 24/7 how do you calculate the electric power they use per day? Homer > > For more detail about each of the listed machines/devices: > http://www.logicprobe.org/systems/ > > Frankly, I find these numbers quite interesting. It goes to show that an > Athon64 whizbanger uses a lot less power than you'd expect, and the same goes > for my 6-drive file server. Also shows just how much power my network > devices use. > From dmckenna at thelimucompany.com Wed Feb 8 17:35:46 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D75A@tlc001.tlcusa.thelimucompany.com> > I have this newer invention called the "remote control." > They even have "universal remote controls." > They remove the requirement to use a switch. > (sorry, couldn't resist ;-) I've also come across two interesting contraptions that are related to this: a "spouse" and "boffer weapons" - the combination is quite effective, I've found. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Wed Feb 8 18:02:02 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D75A@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060208230202.55272.qmail@web34103.mail.mud.yahoo.com> Damien McKenna wrote: > I've also come across two interesting contraptions that are > related to this: a "spouse" and "boffer weapons" - the > combination is quite effective, I've found. My spouse loses the remote in the couch all-the-time, so I can easily pretend it is missing when I want to. ;-> -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From ozz at ozz.is-a-geek.net Wed Feb 8 20:52:47 2006 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D75A@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D75A@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060208205247.1ead9832.ozz@ozz.is-a-geek.net> > > I have this newer invention called the "remote control." > > They even have "universal remote controls." > > They remove the requirement to use a switch. > > (sorry, couldn't resist ;-) Actually, these don't really work for what the original topic required. A remote places the device in 'stand-by' mode - it does not turn it off completely. Therefore, there is still some current being drawn, albeit usually quite small. Regards, Ozz. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.matrixlist.com/pipermail/pc_support/attachments/20060208/4b47f01c/attachment.bin From octo at logicprobe.org Wed Feb 8 22:22:44 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <1139435522.807.75.camel@localhost.localdomain> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602071843.32114.octo@logicprobe.org> <1139435522.807.75.camel@localhost.localdomain> Message-ID: >> For the run-down on my actual power usage of machines in the computer room, >> look here: >> http://www.logicprobe.org/~octo/stuff/power.txt > > Assuming most these machines are on 24/7 how do you calculate the > electric power they use per day? Well, electricity consumption (what you're billed for by the power company) is measured in Killowatt-hours. 1 kWh is equivalent to having something running for 1 hour, that consumes 1 kW of electricity. So you take the combined draw of my equipment in kW, multiply by the number of hours in a day, and you have the kWh per day. Of course this doesn't tell you the power used by the HVAC gear (my A/C) to compensate for the heat produced by the equipment. That is very hard to determine in a home environment since HVAC needs vary a lot with outdoor climate, airflow in the house, and human comfort needs. -Derek From jasonb at edseek.com Wed Feb 8 22:28:54 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> Message-ID: <200602082228.54276.jasonb@edseek.com> On Wednesday 08 February 2006 22:22, Derek Konigsberg wrote: > >> For the run-down on my actual power usage of machines in the computer > >> room, look here: > >> http://www.logicprobe.org/~octo/stuff/power.txt > > > > Assuming most these machines are on 24/7 how do you calculate the > > electric power they use per day? > > Well, electricity consumption (what you're billed for by the power > company) is measured in Killowatt-hours. 1 kWh is equivalent to having > something running for 1 hour, that consumes 1 kW of electricity. So you > take the combined draw of my equipment in kW, multiply by the number of > hours in a day, and you have the kWh per day. What equipment did you use to measure the actual power utilization for each system / device? -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From philb at philb.us Thu Feb 9 01:41:35 2006 From: philb at philb.us (Phil Barnett) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <20060208213909.23810.qmail@web34103.mail.mud.yahoo.com> References: <20060208213909.23810.qmail@web34103.mail.mud.yahoo.com> Message-ID: <200602090141.35542.philb@philb.us> On Wednesday 08 February 2006 16:39, Bryan J. Smith wrote: > Jason Boxman wrote: > > Is there any long term risk to continuously cycling these devices > > at the wall with a switch? Do the caps wear out sooner, or some > > such? I don't want to kill my five year old TV before I replace it > > some day... > > I have this newer invention called the "remote control." > They even have "universal remote controls." > They remove the requirement to use a switch. > > (sorry, couldn't resist ;-) Remote controls do not turn off Phantom Loads. -- "Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next. From mflang at bellsouth.net Wed Feb 8 17:36:18 2006 From: mflang at bellsouth.net (Max F Lang) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Help Request: HDD parts to revive dead drive In-Reply-To: <200602080706.28219.octo@logicprobe.org> References: <20060207173047.20931.qmail@web34109.mail.mud.yahoo.com> <200602080656.02208.pberry2@cfl.rr.com> <200602080706.28219.octo@logicprobe.org> Message-ID: <200602081736.18178.mflang@bellsouth.net> On Wednesday 08 February 2006 07:06, Derek Konigsberg wrote: > > But, one caveat on the bit set, is that they insert into a > I'm still amazed at how we got into this whole long discussion of > torx screwdrivers, and no one actually bothered to mention that > circuit-board-swapping may actually not work. The two drives may > not be 100% identical. Even if they are, the electronics are > bound to have some unique information on bad sectors, > sector-remapping, and the like. In other words, the > logical-to-physical mappings of the two drives may not be > identical in any case. Derek's right, even if you match the bad drive with another drive from the exact same model (even from the same lot), the swap still might not work. So don't be disappointed if it doesn't. I've done this kind of swap on at least two dozen drives, and had it fail in maybe a fourth of them. All you can do is try and hope for the best. "If you break it, you get to keep the pieces"... I've seen at least three different screw types on WD hard drives. Old drives seemed to use a tiny regular screw, newer ones used a bead head that a regular hex key worked on, and then the newest drives have a head with a star and a shallow indention. This last screw is the one I'm talking about. And the bit I was referring to is Torx-like, but isn't a Torx. It has a shallow point to it. Not flat, no very pointy, just enough that a regular Torx won't work. It's like the bit needs to exert a large degree of downward pressure on the head before you try to twist the screw off. I tried a number of bits, and wound up stripping the heads before they nudged. That's where I figured that a tiny ViseGrip wrench I got from Sears worked fine, and was cheaper than any bit set. From philb at philb.us Thu Feb 9 01:46:38 2006 From: philb at philb.us (Phil Barnett) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <49266.216.134.200.78.1139432705.squirrel@nebula.internal.foo> References: <55932.216.134.200.78.1139342232.squirrel@nebula.internal.foo> <200602071855.00217.philb@philb.us> <49266.216.134.200.78.1139432705.squirrel@nebula.internal.foo> Message-ID: <200602090146.38419.philb@philb.us> On Wednesday 08 February 2006 16:05, Jason Boxman wrote: > Phil Barnett said: > > On Tuesday 07 February 2006 14:57, Jason Boxman wrote: > >> Anyone know just how much waste is created from these 'always-on' > >> appliances? I just realized my TV is 'always' on. The DVD player is > >> more obvious, since it has a standby light. What about CRT monitors in > >> 'energy saver' mode? > >> > >> Is there any sizable savings in screwing with this stuff, or is it not > >> worth the effort to calculate up how much power things draw? > > > > The guys who build solar electric houses call these loads 'Phantom > > Loads'. Google it. > > Is there any long term risk to continuously cycling these devices at the > wall with a switch? Do the caps wear out sooner, or some such? I don't > want to kill my five year old TV before I replace it some day... Of course, there are risks to doing anything. A risk to never unplugging anything is lightning damage. In the past, full power cycling got a bad rap for several reasons. 1. Remote controls do not work. 2. The unit fully cools off. This makes the off to on temperature swing larger. 3. Large temperature swings have been associated with cold solder joints releasing and good solder joints becoming cracked. Most everything these days is put together with solder, so this risk still remains. HOWEVER, it is my opinion that the quality of automated soldering has vastly improved over the last 25 years when these 'old wives tales' originated. I don't think it's anywhere near the problem that it used to be, both based on what I read and what I have experienced, so I'd say that the risk of damage from completely removing power during down time for a great majority of all appliances no longer carries much risk. YMMV of course. -- "Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next. From b.j.smith at ieee.org Thu Feb 9 07:22:41 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602090141.35542.philb@philb.us> Message-ID: <20060209122241.28288.qmail@web34103.mail.mud.yahoo.com> Phil Barnett wrote: > Remote controls do not turn off Phantom Loads. Ahhh, yes, now I realize the original context. Okay, I guess I should have assumed you'd turn it off with a remote first, then kill all load with a swith. Okay (Doh!). -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Thu Feb 9 07:32:09 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <200602082228.54276.jasonb@edseek.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> <200602082228.54276.jasonb@edseek.com> Message-ID: > > What equipment did you use to measure the actual power utilization for each > system / device? > I used a device called Kill-A-Watt, which you can now buy from ThinkGeek: http://www.thinkgeek.com/gadgets/electronic/7657/ (it's dropped in price too, when I first got it, it was closer to $50, and wasn't on Think Geek) Basically, its a device you plug in-line between the device/load and the outlet. It then tells you all sorts of info from volts, to amps, to watts, power factor, AC frequency, etc. -Derek From b.j.smith at ieee.org Thu Feb 9 07:33:28 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] MicroATX nForce4 SLI coming to Monarch Computer ... Message-ID: <20060209123328.63551.qmail@web34105.mail.mud.yahoo.com> FYI, Monarch Computer is selling a MicroATX mainboard with nForce4 SLI in their new Hornet Pro series of cubes. I can't find much about the mainboard yet, but in their configurator, it's $199 as part of the assembled unit: http://www.monarchcomputer.com/Merchant2/merchant.mv?Screen=PROD&Store_Code=M&Product_Code=80999&Category_Code=Hornet When I find out who the manufacturer is, I'll let you'all know. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From pberry2 at cfl.rr.com Thu Feb 9 08:51:57 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <200602082228.54276.jasonb@edseek.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602082228.54276.jasonb@edseek.com> Message-ID: <200602090851.57691.pberry2@cfl.rr.com> On Wednesday 08 February 2006 22:28, Jason Boxman wrote: > On Wednesday 08 February 2006 22:22, Derek Konigsberg wrote: > > >> For the run-down on my actual power usage of machines in the computer > > >> room, look here: > > >> http://www.logicprobe.org/~octo/stuff/power.txt > > > > > > Assuming most these machines are on 24/7 how do you calculate the > > > electric power they use per day? > > > > Well, electricity consumption (what you're billed for by the power > > company) is measured in Killowatt-hours. 1 kWh is equivalent to having > > something running for 1 hour, that consumes 1 kW of electricity. So you > > take the combined draw of my equipment in kW, multiply by the number of > > hours in a day, and you have the kWh per day. > > What equipment did you use to measure the actual power utilization for each > system / device? I use a Digisnap DSA-1000, that I got at a garage sale. I think I paid $10, with charger. You can buy a very in-expensive cousin Cen Tech (Chinese made) Digital Volt - Ohm - Amp Meter with LCD screen, and data hold button. They are offered at HFT. $29.99 but, on sale now for $19.99. Note the "Peak hold" feature. http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=42397 or, this one, $18.99, note the "Data hold" feature: http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=42396 You'll want to buy a line splitter: $4.99 http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=92072 Be safe out there, ya hear? Work dry, wear protective gear, if you open up any electrical panels!! The Splitter makes it safe, for 120 VAC. I presume most of you are doing appliance/computer measurements. If you measure any refrigeration, freezer, or HVAC systems, be aware that when you shutdown or unplug, you want to wait a few minutes before turning it back on so the head pressure in the compressor can equalize. -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From b.j.smith at ieee.org Thu Feb 9 14:28:19 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Another 19", 8ms, DVI+VGA for sub-$220 AR ... Message-ID: <20060209192819.33229.qmail@web34112.mail.mud.yahoo.com> Just FYI, Rosewill R192E 19" display runs $259.99 - $40 rebate = $219.99 from NewEgg.COM. Typical 19", 1280x1024 res, 8ms response, DVI and VGA inputs, etc... http://dealnews.com/deals/Rosewill-R912-E-19-Monitor-w-DVI-for-220-after-rebate/109711.html -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Thu Feb 9 14:47:43 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] AGP Owners: nVidia has your 7800GS ... Message-ID: <20060209194743.32392.qmail@web34105.mail.mud.yahoo.com> Nice to see nVidia is still supporting AGP, and bringing the lower power per pixel in the G7x series thanx to their new 7800GS for around $300. Although only about 66-75% the performance of a $400+ GeForce 7800GTX PCIe, it's fairly close to its $300+ GeForce 7800GT PCIe brother in performance. Especially at a lower price/power standpoint (let alone better performance) than aging solutions like ATI's $400 X850XT and nVidia's previous GeForce 6800 Ultra product. Anand and Tom's Reviews ... http://www.anandtech.com/video/showdoc.aspx?i=2686 http://www.tomshardware.com/2006/02/02/nvidia_7800gs_confirms_agp_aint_dead_yet/ -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Thu Feb 9 15:51:22 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Gigabyte's $400 dual-S940, $60-100 mainboard+video rebate offer ... Message-ID: <20060209205122.49521.qmail@web34115.mail.mud.yahoo.com> For those of you who crave performance, Gigabyte is drumming up some noise with their new GA-2CEWH dual Socket-940 (_not_ the forthcoming AMD AM2 slot, but Opteron 2xx) mainboard: http://www.gigabyte-usa.com/Promos/2CEWH.html A very, _very_ modern design with not only the full nVidia nForce Pro 2200+2050 combination, but a full dual PCI-X 2.0 AMD8132 tunnel with two (2) slots at the full 133MHz (or slower, for backward compatibility, of course) for the ultimate in I/O capability. Best of all, the price undercuts most older designs at $400. Gigabyte is currently running an offer of a $60 rebate if you purchase either their $300-400 GeForce 6800GS or 7800GT (not even the full 7800 GTX), or $100 if you buy a pair for SLI. So overall this saves you 10% or so, if you were interested in the dual S940 solution. http://www.gigabyte-usa.com/Promos/2CEWH.html The cost of Opteron 2xx isn't much more than uniprocessor Athlon 64 or Opteron 1xx. The real kicker is the Registered ECC DDR SDRAM requirement, which isn't as commodity. But if you want the ultimate in power and I/O , this is a far better way to put the same amount of money that you might drop in an Athlon FX SLI solution into a serious and true dual Socket performer with a full 32 PCIe channels -- 16 on each socket. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Thu Feb 9 16:05:15 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] nVidia Quadro NVS 440 puts four (4) displays in a single PCIe x1 (yes, x1) card ... Message-ID: <20060209210515.30095.qmail@web34109.mail.mud.yahoo.com> nVidia has introduced a new member of the dual-GPU, quad-display Quadro NVS 400 series in the new NVS 440. But instead of the original NVS 400's PCI or AGP+PCI options, the new NVS 440 is a single PCIe x1 slot that uses only 31W (passive), allowing it to work on virtually any mainboard with a spare PCIe x1 slot. The PCIe x1 slot gives a dedicated 0.5GBps (0.25GBps bi-directional) I/O interconnect. With dual-NV43 (GeForce 6600[GT]) series GPUs, the quad-display is a full-height, half-length, passively-cooled card that takes up no additional space (just one slot). Connectors are two (2) DMS-59 with two (2) cables each converting DMS-59 into two (2) DVI-I (4xDVI total) and four (4) DVI-to-DB15 converters for optional analog. Memory is 128-bit, typically implemented as 256MiB per card (128MiB per GPU frame buffer) DDR or DDR3 (as the 6600 v. 6600GT). Analog is supported up to 2048x1536@75Hz and Digital is supported up to 1920x1200 on each display. nVidia NVS series pages: http://www.nvidia.com/page/quadronvs.html http://www.nvidia.com/object/IO_14605.html Leadtek NVS 440 product page: http://www.leadtek.com.tw/eng/workstation_graphics/overview.asp?lineid=2&pronameid=248 -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From ozz at ozz.is-a-geek.net Thu Feb 9 17:48:28 2006 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> <200602082228.54276.jasonb@edseek.com> Message-ID: <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> On Thu, 9 Feb 2006 07:32:09 -0500 (EST), Derek Konigsberg wrote: > > I used a device called Kill-A-Watt, which you can now buy from ThinkGeek: > http://www.thinkgeek.com/gadgets/electronic/7657/ > (it's dropped in price too, when I first got it, it was closer to $50, > and wasn't on Think Geek) Those things are pretty good - I use one myself. I picked mine up at my local RadioShack. Regards, Ozz. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.matrixlist.com/pipermail/pc_support/attachments/20060209/aff9659d/attachment.bin From ozz at ozz.is-a-geek.net Thu Feb 9 18:10:10 2006 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] MicroATX nForce4 SLI coming to Monarch Computer ... In-Reply-To: <20060209123328.63551.qmail@web34105.mail.mud.yahoo.com> References: <20060209123328.63551.qmail@web34105.mail.mud.yahoo.com> Message-ID: <20060209181010.528dd02b.ozz@ozz.is-a-geek.net> On Thu, 9 Feb 2006 04:33:28 -0800 (PST), "Bryan J. Smith" wrote: > > FYI, Monarch Computer is selling a MicroATX mainboard with nForce4 > SLI in their new Hornet Pro series of cubes. I can't find much about > the mainboard yet, but in their configurator, it's $199 as part of > the assembled unit: > http://www.monarchcomputer.com/Merchant2/merchant.mv?Screen=PROD&Store_Code=M&Product_Code=80999&Category_Code=Hornet > > When I find out who the manufacturer is, I'll let you'all know. One thing I have learned about Monarch is that for individual components their delivery rocks. For complete systems, their delivery sucks. However, the systems, when they finally do arrive, are very well put together. In that respect I'd recommend them to anyone. In the last few months I've bought my workstation (Dual-Opteron 246) and 4 1U rackmount servers (ranging from single Opteron 246 to Dual-Opteron 250, 1.6Tb RAID5) and been very happy with everything except delivery. They all took best part of a month to deliver. They were also very sporadic at updating their on-line order status. Sometimes they would update it several times a day. Others they could go a week or more without updates. On one occasion their website said my server was at Stage 4 (having the components installed in the case - a good week away from delivery) and yet I had not only already received that very box, but I'd installed Debian amd64 on it, installed it in the rack, and it was happily proxying all of our web traffic... Monarch just missed out on a $15k order for four more servers thanks to their delivery. We just couldn't afford to wait for Monarch. We are about to move to a new office, and will be buying a TON of new gear. I'd love to go Monarch, but I'm not sure we can risk their delivery times... Much as I hate Dell, I can place an order with Dell on Monday and have the hardware by Thursday... Regards, Ozz. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.matrixlist.com/pipermail/pc_support/attachments/20060209/7c5754d8/attachment.bin From jasonb at edseek.com Thu Feb 9 18:17:59 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> <200602082228.54276.jasonb@edseek.com> <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> Message-ID: <51953.216.134.200.78.1139527079.squirrel@nebula.internal.foo> Austin Denyer said: > > On Thu, 9 Feb 2006 07:32:09 -0500 (EST), Derek Konigsberg > wrote: >> >> I used a device called Kill-A-Watt, which you can now buy from ThinkGeek: >> http://www.thinkgeek.com/gadgets/electronic/7657/ >> (it's dropped in price too, when I first got it, it was closer to $50, >> and wasn't on Think Geek) > > Those things are pretty good - I use one myself. > > I picked mine up at my local RadioShack. My piece of junk ~$32 desktop computer MIC from RadioShack that I purchased in November simply confirms what I've had confirmed before. RadioShack sucks. :) From jasonb at edseek.com Thu Feb 9 18:22:42 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] MicroATX nForce4 SLI coming to Monarch Computer ... In-Reply-To: <20060209181010.528dd02b.ozz@ozz.is-a-geek.net> References: <20060209123328.63551.qmail@web34105.mail.mud.yahoo.com> <20060209181010.528dd02b.ozz@ozz.is-a-geek.net> Message-ID: <47947.216.134.200.78.1139527362.squirrel@nebula.internal.foo> Austin Denyer said: > > Monarch just missed out on a $15k order for four more servers thanks > to their delivery. We just couldn't afford to wait for Monarch. We are > about to move to a new office, and will be buying a TON of new gear. > I'd love to go Monarch, but I'm not sure we can risk their delivery > times... > > Much as I hate Dell, I can place an order with Dell on Monday and > have the hardware by Thursday... Other fun Monarch stuff. http://edseek.com/archives/2005/08/25/boo-on-monarch-computer-for-taking-my-cash/ Did finally work out, but shame on them for selling such a faulty brand. I get mail from people all over with the same issue I had two with XFX cards. Not sure I'd order from them for personal stuff again. From ozz at ozz.is-a-geek.net Thu Feb 9 18:34:32 2006 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <51953.216.134.200.78.1139527079.squirrel@nebula.internal.foo> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> <200602082228.54276.jasonb@edseek.com> <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> <51953.216.134.200.78.1139527079.squirrel@nebula.internal.foo> Message-ID: <20060209183432.64c6d957.ozz@ozz.is-a-geek.net> On Thu, 9 Feb 2006 18:17:59 -0500 (EST), "Jason Boxman" wrote: > > My piece of junk ~$32 desktop computer MIC from RadioShack that I purchased > in November simply confirms what I've had confirmed before. RadioShack > sucks. I'm generally not a RadioShack fan either - they're usually WAY overpriced - but I happened to need the thing in a hurry. Regards, Ozz. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.matrixlist.com/pipermail/pc_support/attachments/20060209/552c892a/attachment.bin From jasonb at edseek.com Thu Feb 9 18:50:49 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060209183432.64c6d957.ozz@ozz.is-a-geek.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> <200602082228.54276.jasonb@edseek.com> <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> <51953.216.134.200.78.1139527079.squirrel@nebula.internal.foo> <20060209183432.64c6d957.ozz@ozz.is-a-geek.net> Message-ID: <43563.216.134.200.78.1139529049.squirrel@nebula.internal.foo> Austin Denyer said: > > On Thu, 9 Feb 2006 18:17:59 -0500 (EST), "Jason Boxman" > wrote: >> >> My piece of junk ~$32 desktop computer MIC from RadioShack that I >> purchased >> in November simply confirms what I've had confirmed before. RadioShack >> sucks. > > I'm generally not a RadioShack fan either - they're usually WAY > overpriced - but I happened to need the thing in a hurry. The best part? I 'borrowed' it for a few weeks, figuring it would be crap anyway, then took it back. When asked why, I just said it didn't work with my PeeCee. The sales rep gave me a dumb look like I was an idiot and asked if I had a sound card. I repeated my statement. He credited my card and I left. From lists at brianrose.net Thu Feb 9 21:28:19 2006 From: lists at brianrose.net (Brian Rose) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602082228.54276.jasonb@edseek.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <1139435522.807.75.camel@localhost.localdomain> <200602082228.54276.jasonb@edseek.com> Message-ID: <43EBFA43.8070305@brianrose.net> Jason Boxman wrote: > > What equipment did you use to measure the actual power utilization for each > system / device? > A simple multimeter that reads AC current is fine. I have a $25 RadioShack meter that would work fine. The multimeter needs to be in series when measuring current, so you'll need to rig up a cable. ________________________ | | 120VAC + ---------- | Meter + Meter COM |------------- ------------------------ Socket Plug - ------------------------------------------------ Measure the current draw and multiply by 120V to get your power. For a frugal device that draws 8mA... mA = 0.008A * 120V = .96W ~= 1W = .001kW .001kW * 24 hours * 30 days = 0.720kWh Price for electricity is around 10 cents per kilowatt, so each standby watt costs you seven cents per month. Roughly. -- Brian From jasonb at edseek.com Thu Feb 9 21:42:10 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <43EBFA43.8070305@brianrose.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602082228.54276.jasonb@edseek.com> <43EBFA43.8070305@brianrose.net> Message-ID: <200602092142.10397.jasonb@edseek.com> On Thursday 09 February 2006 21:28, Brian Rose wrote: > Jason Boxman wrote: > > What equipment did you use to measure the actual power utilization for > > each system / device? > > A simple multimeter that reads AC current is fine. I have a $25 RadioShack > meter that would work fine. The multimeter needs to be in series when > measuring current, so you'll need to rig up a cable. > > ________________________ > > 120VAC + ---------- | Meter + Meter COM |------------- > ------------------------ Socket > Plug - ------------------------------------------------ > > > Measure the current draw and multiply by 120V to get your power. Cool. > For a frugal device that draws 8mA... > > mA = 0.008A * 120V = .96W ~= 1W = .001kW > > .001kW * 24 hours * 30 days = 0.720kWh > > Price for electricity is around 10 cents per kilowatt, so each standby watt > costs you seven cents per month. > > Roughly. Yeah, I pay around $0.09 I think, or there abouts. I wonder how much my TV draws in stand-by... -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Thu Feb 9 21:45:43 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: MicroATX nForce4 SLI coming to Monarch Computer ... In-Reply-To: <20060209181010.528dd02b.ozz@ozz.is-a-geek.net> Message-ID: <20060210024543.40735.qmail@web34107.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > ... selling a MicroATX mainboard with nForce4 SLI ... > I can't find much about the mainboard yet ... > When I find out who the manufacturer is, I'll let you'all know. "Austin (Ozz) Denyer" wrote: > One thing I have learned about Monarch ... cut ... Jason Boxman wrote: > Other fun Monarch stuff ... cut ... I posted this _not_ to recommend Monarch, but to confirm the existance of a MicroATX mainboard with SLI. That's why I posted the above verbage as highlighted in this e-mail. I haven't seen any manufacturer/vendor offering a SLI solution in MicroATX ... until now. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From jasonb at edseek.com Thu Feb 9 22:21:13 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: MicroATX nForce4 SLI coming to Monarch Computer ... In-Reply-To: <20060210024543.40735.qmail@web34107.mail.mud.yahoo.com> References: <20060210024543.40735.qmail@web34107.mail.mud.yahoo.com> Message-ID: <200602092221.13502.jasonb@edseek.com> On Thursday 09 February 2006 21:45, Bryan J. Smith wrote: > "Austin (Ozz) Denyer" wrote: > > One thing I have learned about Monarch ... cut ... > > Jason Boxman wrote: > > Other fun Monarch stuff ... cut ... > > I posted this _not_ to recommend Monarch, but to confirm the > existance of a MicroATX mainboard with SLI. That's why I posted the > above verbage as highlighted in this e-mail. > > I haven't seen any manufacturer/vendor offering a SLI solution in > MicroATX ... until now. Yes. And no one thinks you were doing so. But you did post a link to Monarch and sometimes people can't help but talk about experiences with a particular vendor when someone points a link to a product on said vendor's site. It just is. -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From tec at homemail.bjt.net Thu Feb 9 23:39:05 2006 From: tec at homemail.bjt.net (Thomas Carlson) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Palm Treo 650 - Must Do, Great Software, Hacks In-Reply-To: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> References: <2532.71.252.176.10.1139289815.squirrel@qtmail.dgnal.net> Message-ID: <1139546345.2661.2.camel@xpc.tecsplace> I like splashid software on the palm pdas and phones. http://www.splashdata.com/splashid/ On Mon, 2006-02-06 at 23:23 -0600, David Simmons wrote: > Bryan (and those on the list with a Treo 650), > > > Sent from my Treo > > The Lord (and my new company) has so blessed me with such a beast - any > suggestions on 'Must Have' software / Hacks / Sites / etc? > > Thanks in advance, > > dave > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support > From pberry2 at cfl.rr.com Fri Feb 10 07:52:21 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:39 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060209183432.64c6d957.ozz@ozz.is-a-geek.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <51953.216.134.200.78.1139527079.squirrel@nebula.internal.foo> <20060209183432.64c6d957.ozz@ozz.is-a-geek.net> Message-ID: <200602100752.21837.pberry2@cfl.rr.com> On Thursday 09 February 2006 18:34, Austin Denyer wrote: > On Thu, 9 Feb 2006 18:17:59 -0500 (EST), "Jason Boxman" > > wrote: > > My piece of junk ~$32 desktop computer MIC from RadioShack that I > > purchased in November simply confirms what I've had confirmed before. > > RadioShack sucks. > > I'm generally not a RadioShack fan either - they're usually WAY > overpriced - but I happened to need the thing in a hurry. > > Regards, > Ozz. When I need some weird fuse, or a quick grab of a part to fix something, it usually is a Radio Shack. Last night I went to the Dollar Tree for a lookee to see if they had any more BSD 6foot extensions, Cat5e 6' cables, and other goodies. The Radio Shack is next door, in Butler Plaza, on 436 at Howell Branch Road. Anyway, I met a couple guys who are 'web page developers' and hadn't seen or tried out NVU and Seamonkey, yet. Gave them a copy of PCLinuxOS. Went to the parking lot, noticed two guys and two girls puzzling over Daddy's Cheerokee that wouldn't start. I diagnosed it as a tired starter solenoid, bumped it with the handle of a screwdriver, it started, and they drove away with a PCLinuxOS Live CD, and the knowledge of how easy it will be for them to replace the starter, on Saturday, after the Golug.org installfest at SCC. (Ron Weidner had to move it up due to scheduleing conflicts). But, I mentioned the Kill-A-Watt, and the radio Shack folks all knew what a neat gadget it was. #63-1152 was dropped in 2003. Home Depot web page says it is out of stock. Found it for $21.49 (NIP!) on Amazon.com mail is $6.64. Yes, "if the light bulb were invented today, it would be illegal to sell due to conservation rules" was stated on the radio (WDBO-540Khz, just yesterday, 9 Feb 2006!), by one of the "Screaming guys on air in the afternoons". All but five ( two in the 'cosmetic make-up' bathroom), and those in my oven, refrigerator, freezer, and Microwave oven, are the new fluorescent replacement bulbs. -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From pberry2 at cfl.rr.com Fri Feb 10 08:02:02 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602092142.10397.jasonb@edseek.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <43EBFA43.8070305@brianrose.net> <200602092142.10397.jasonb@edseek.com> Message-ID: <200602100802.02246.pberry2@cfl.rr.com> On Thursday 09 February 2006 21:42, Jason Boxman wrote: > On Thursday 09 February 2006 21:28, Brian Rose wrote: > > Jason Boxman wrote: > > > What equipment did you use to measure the actual power utilization for > > > each system / device? > > > > A simple multimeter that reads AC current is fine. I have a $25 > > RadioShack meter that would work fine. The multimeter needs to be in > > series when measuring current, so you'll need to rig up a cable. > > > > ________________________ > > > > 120VAC + ---------- | Meter + Meter COM |------------- > > ------------------------ Socket > > Plug - ------------------------------------------------ > > > > > > Measure the current draw and multiply by 120V to get your power. > > Cool. > > > For a frugal device that draws 8mA... > > > > mA = 0.008A * 120V = .96W ~= 1W = .001kW > > > > .001kW * 24 hours * 30 days = 0.720kWh > > > > Price for electricity is around 10 cents per kilowatt, so each standby > > watt costs you seven cents per month. > > > > Roughly. > > Yeah, I pay around $0.09 I think, or there abouts. > > I wonder how much my TV draws in stand-by... Was in the Harbor Freight Tools at Butler Plaza, last afternoon, Feb. 9, 2006, and the Digital Clamp Amp meter was on sale for $11.99!!! China made Cen Tech. Right up front on the shelves at the cash registers. There is a larger Harbor Freight on SOBT 17-92, near Lancaster road. -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From pberry2 at cfl.rr.com Fri Feb 10 08:08:46 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: Reducing Electricity Costs -- bulbs first and foremost ... In-Reply-To: <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <20060209174828.5597ee7f.ozz@ozz.is-a-geek.net> Message-ID: <200602100808.46667.pberry2@cfl.rr.com> On Thursday 09 February 2006 17:48, Austin Denyer wrote: > On Thu, 9 Feb 2006 07:32:09 -0500 (EST), Derek Konigsberg > > wrote: > > I used a device called Kill-A-Watt, which you can now buy from ThinkGeek: > > http://www.thinkgeek.com/gadgets/electronic/7657/ > > (it's dropped in price too, when I first got it, it was closer to $50, > > and wasn't on Think Geek) > > Those things are pretty good - I use one myself. > > I picked mine up at my local RadioShack. > > Regards, > Ozz. Radio Shack discontinued Kill A Watt #63-1152 in 2003. Ordered one for $21.49 through Amazon.com with $6.64 shipping. -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From b.j.smith at ieee.org Fri Feb 10 09:03:02 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Mastering Regular Expressions -- WAS: [LeapList] Scanning files for patterns. Message-ID: <20060210140302.80896.qmail@web34102.mail.mud.yahoo.com> George Laiacona wrote: > I'm new to shell scripting, and Perl for that matter. How would I > scan all the files in a directory looking for a certain word, for > instance? What I'm trying to do is read a bunch of text files looking > for run numbers, in the following format YY00XXXXa where YY is the 2 > digit year, 06, XXXX is a number, and a is a sub-unit, like a, b, c, > etc. For example 06004512a. All I want to know is what file contains > that string. > Is this possible? It's certainly tedious opening dozens of files by > hand and trying to extract this information. I'm going cross-eyed > trying. You can liberally use the "legacy wildcards" asterisk (*) and question (?) to represent zero or more "any" characters and exactly 1 "any" character, respectively, in a filename passed to most UNIX shells. That _includes_ using the asterisk (*) in the middle of a filename. You'll find UNIX shells to be _more_powerful_ in this application than DOS COMMAND.COM or OS/2-NT CMD.EXE. About the only limitation is the GNU/UNIX "mv" command can be more limiting than DOS-OS/2-NT's "rename" command when it comes to wildcards. Although mc (Midnight Commander, a Norton Commander-like console shell for UNIX) can do some very flexible copying/renaming with just these "legacy wildcards." - Regular Expressions (regex) Otherwise, you're largely talking about the greater concept of using "Regular Expressions" (regex) -- the infinite, built-in power of GNU/UNIX systems. They can differ slightly in implementation. E.g., the GNU/UNIX commands' regex and Perl's regex differ slightly. A _great_ book for regular expressions is O'Reilly's: http://www.amazon.com/gp/product/0596002890/ Chapter 1 answers 80% of your basic questions. It then dives deeper into the variations of regex's across many implementations. I can write very basic regex's to get simple things done, and know how to deal with common pitfalls. But other people are masters, and if you want to be one, this is an _excellent_ book. Knowing regex's is a _powerful_ art. I really need to devote a full week to learning them sometime myself. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From lists at brianrose.net Fri Feb 10 09:13:41 2006 From: lists at brianrose.net (Brian Rose) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602100802.02246.pberry2@cfl.rr.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <43EBFA43.8070305@brianrose.net> <200602092142.10397.jasonb@edseek.com> <200602100802.02246.pberry2@cfl.rr.com> Message-ID: <43EC9F95.3070406@brianrose.net> Patrick wrote: > > Was in the Harbor Freight Tools at Butler Plaza, last afternoon, Feb. 9, 2006, > and the Digital Clamp Amp meter was on sale for $11.99!!! China made Cen > Tech. Right up front on the shelves at the cash registers. There is a larger > Harbor Freight on SOBT 17-92, near Lancaster road. > The problem with the "clamp" type ammeters is that they have a minimum current of 1-3 amps. This is well above the standby current for most home electronics, which is typically in the milliamp range. On the plus side, they can generally measure hundreds of amps, where a typical mulitmeter is only good up to around 10A. The clamps would be good for figuring the current draw on your major appliances. You just clamp them around the wires and the meter reads the current by induction. No need to unplug them, which is not an option for many appliances. -- Brian From ae4ko at amsat.org Fri Feb 10 09:23:16 2006 From: ae4ko at amsat.org (Aaron Morrison) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Mastering Regular Expressions -- WAS: [LeapList] Scanning files for patterns. In-Reply-To: <20060210140302.80896.qmail@web34102.mail.mud.yahoo.com> Message-ID: <43EC5B84.7885.3193B9@ae4ko.amsat.org> This site might help as well: http://weitz.de/regex-coach/ Abstract Made with LispThe Regex Coach is a graphical application for Windows and Linux/x86 which can be used to experiment with (Perl-compatible) regular expressions interactively. It has the following features: * It shows whether a regular expression matches a particular target string. * It can also show which parts of the target string correspond to captured register groups or to arbitrary parts of the regular expression. * It can "walk" through the target string one match at a time. * It can simulate Perl's split and s/// (substitution) operators. * It tries to describe the regular expression in plain English. * It can show a graphical representation of the regular expression's parse tree. * It can single-step through the matching process as performed by the regex engine. * Everything happens in "real time", i.e. as soon as you make a change somewhere in the application all other parts are instantly updated. Very handy for experimenting and learning regex's --am On 10 Feb 2006 at 6:03, Bryan J. Smith wrote: > George Laiacona wrote: > > I'm new to shell scripting, and Perl for that matter. How would I > > scan all the files in a directory looking for a certain word, for > > instance? What I'm trying to do is read a bunch of text files > looking > > for run numbers, in the following format YY00XXXXa where YY is the > 2 > > digit year, 06, XXXX is a number, and a is a sub-unit, like a, b, > c, > > etc. For example 06004512a. All I want to know is what file > contains > > that string. > > Is this possible? It's certainly tedious opening dozens of files by > > hand and trying to extract this information. I'm going cross-eyed > > trying. > > You can liberally use the "legacy wildcards" asterisk (*) and > question (?) to represent zero or more "any" characters and exactly 1 > "any" character, respectively, in a filename passed to most UNIX > shells. That _includes_ using the asterisk (*) in the middle of a > filename. > > You'll find UNIX shells to be _more_powerful_ in this application than > DOS COMMAND.COM or OS/2-NT CMD.EXE. About the only limitation is the > GNU/UNIX "mv" command can be more limiting than DOS-OS/2-NT's "rename" > command when it comes to wildcards. Although mc (Midnight Commander, > a Norton Commander-like console shell for UNIX) can do some very > flexible copying/renaming with just these "legacy wildcards." > > - Regular Expressions (regex) > > Otherwise, you're largely talking about the greater concept of using > "Regular Expressions" (regex) -- the infinite, built-in power of > GNU/UNIX systems. They can differ slightly in implementation. E.g., > the GNU/UNIX commands' regex and Perl's regex differ slightly. > > A _great_ book for regular expressions is O'Reilly's: > http://www.amazon.com/gp/product/0596002890/ > > Chapter 1 answers 80% of your basic questions. It then dives deeper > into the variations of regex's across many implementations. I can > write very basic regex's to get simple things done, and know how to > deal with common pitfalls. But other people are masters, and if you > want to be one, this is an _excellent_ book. > > Knowing regex's is a _powerful_ art. I really need to devote a full > week to learning them sometime myself. > > > > -- > Bryan J. Smith Professional, Technical Annoyance > b.j.smith@ieee.org http://thebs413.blogspot.com > ---------------------------------------------------- > *** Speed doesn't kill, difference in speed does *** > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support From pberry2 at cfl.rr.com Fri Feb 10 10:19:17 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <43EC9F95.3070406@brianrose.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602100802.02246.pberry2@cfl.rr.com> <43EC9F95.3070406@brianrose.net> Message-ID: <200602101019.17198.pberry2@cfl.rr.com> On Friday 10 February 2006 09:13, Brian Rose wrote: > Patrick wrote: > > Was in the Harbor Freight Tools at Butler Plaza, last afternoon, Feb. 9, > > 2006, and the Digital Clamp Amp meter was on sale for $11.99!!! China > > made Cen Tech. Right up front on the shelves at the cash registers. > > There is a larger Harbor Freight on SOBT 17-92, near Lancaster road. > > The problem with the "clamp" type ammeters is that they have a minimum > current of 1-3 amps. This is well above the standby current for most home > electronics, which is typically in the milliamp range. On the plus side, > they can generally measure hundreds of amps, where a typical mulitmeter is > only good up to around 10A. > > The clamps would be good for figuring the current draw on your major > appliances. You just clamp them around the wires and the meter reads the > current by induction. No need to unplug them, which is not an option for > many appliances. But, the splitter ($4.99, #92072) at Harbor Freight Tools, has a X1 and an X10 hole! So, internally, it is wired for X10 magnification of the current flowing on one side of the AC circuit! -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From lists at brianrose.net Fri Feb 10 10:29:36 2006 From: lists at brianrose.net (Brian Rose) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <200602101019.17198.pberry2@cfl.rr.com> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602100802.02246.pberry2@cfl.rr.com> <43EC9F95.3070406@brianrose.net> <200602101019.17198.pberry2@cfl.rr.com> Message-ID: <43ECB160.4010005@brianrose.net> Patrick wrote: > > > But, the splitter ($4.99, #92072) at Harbor Freight Tools, has a X1 and an X10 > hole! > > So, internally, it is wired for X10 magnification of the current flowing on > one side of the AC circuit! I'm guessing the X10 increases the sensitivity of the circuit by 10 times. Even so, you still have a minumum readable current of at least 100-300mA, which is 12-36W. Anything less than this would a challenge for the device to display accurately. -- Brian From b.j.smith at ieee.org Fri Feb 10 10:40:55 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] CMV 8ms and FujiPlus 12ms 19" LCDs with DVI for $199 AR ... Message-ID: <20060210154055.72689.qmail@web34114.mail.mud.yahoo.com> Just a pair of deals on 19" LCDs with both DVI and VGA, 1280x1024 resolution. The CMV CT-934D has an 8ms refresh and the FujiPlus FP-988D has a 12ms refresh. http://dealnews.com/deals/CMV-CT-934-D-19-LCD-Monitor-for-200-after-rebate/109752.html http://dealnews.com/deals/Fuji-Plus-FP-988-D-19-LCD-Monitor-w-DVI-for-200-after-rebate/109577.html -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Fri Feb 10 10:43:48 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Mastering Regular Expressions -- WAS: [LeapList] Scanning files for patterns. In-Reply-To: <43EC5B84.7885.3193B9@ae4ko.amsat.org> Message-ID: <20060210154348.73698.qmail@web34114.mail.mud.yahoo.com> Aaron Morrison wrote: > This site might help as well: > http://weitz.de/regex-coach/ Nice! I had previously tried ActiveState's built-in Komodo solution awhile back, but wasn't too impressed. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Fri Feb 10 10:48:39 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] EFF: Don't use Google desktop ... Message-ID: <20060210154839.66064.qmail@web34105.mail.mud.yahoo.com> If the EFF has finally piped up, it's probably very valid: http://www.eweek.com/article2/0,1895,1925064,00.asp -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From pberry2 at cfl.rr.com Fri Feb 10 12:34:23 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Reducing Electricity Costs In-Reply-To: <43ECB160.4010005@brianrose.net> References: <20060207220649.36559.qmail@web34105.mail.mud.yahoo.com> <200602101019.17198.pberry2@cfl.rr.com> <43ECB160.4010005@brianrose.net> Message-ID: <200602101234.23464.pberry2@cfl.rr.com> On Friday 10 February 2006 10:29, Brian Rose wrote: > Patrick wrote: > > But, the splitter ($4.99, #92072) at Harbor Freight Tools, has a X1 and > > an X10 hole! > > > > So, internally, it is wired for X10 magnification of the current flowing > > on one side of the AC circuit! > > I'm guessing the X10 increases the sensitivity of the circuit by 10 times. > Even so, you still have a minumum readable current of at least 100-300mA, > which is 12-36W. Anything less than this would a challenge for the device > to display accurately. I will keep that in mind, which is one reason that I ordered the P3 International product, the KillAWatt P4400, from an Amazon.com vendor, for $21.49 plus $6.64 shipping and handling... But, my A.W. Sperry DSA 1000 clamp Amp V-O-M can read any service mains voltages, like 250VAC, 477 VAC, with the inductive clamps. I think the one selling for $11.99 at Harbor Freight Tools, can, also! -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From justinkz at gmail.com Fri Feb 10 14:08:25 2006 From: justinkz at gmail.com (Justin M. Keyes) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] EFF: Don't use Google desktop ... In-Reply-To: <20060210154839.66064.qmail@web34105.mail.mud.yahoo.com> References: <20060210154839.66064.qmail@web34105.mail.mud.yahoo.com> Message-ID: <53b562310602101108j36d8458dl226c7e90e1273d28@mail.gmail.com> On 2/10/06, Bryan J. Smith wrote: > If the EFF has finally piped up, it's probably very valid: > http://www.eweek.com/article2/0,1895,1925064,00.asp After reading TFA, it seems they are only recommending against the specific "Share Across Computers" feature of Google Desktop, which is optional. Of course, I planned to never use that feature before the EFF recommended so ;) -- Justin M. Keyes From b.j.smith at ieee.org Sat Feb 11 09:56:11 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] EFF: Don't use Google desktop ... In-Reply-To: <53b562310602101108j36d8458dl226c7e90e1273d28@mail.gmail.com> References: <20060210154839.66064.qmail@web34105.mail.mud.yahoo.com> <53b562310602101108j36d8458dl226c7e90e1273d28@mail.gmail.com> Message-ID: <1139669771.4755.1.camel@bert64.oviedo.smithconcepts.com> On Fri, 2006-02-10 at 14:08 -0500, Justin M. Keyes wrote: > After reading TFA, it seems they are only recommending against the > specific "Share Across Computers" feature of Google Desktop, which is > optional. Of course, I planned to never use that feature before the > EFF recommended so ;) Yeah, I turn that off as well on my parent's and wife's computers. I'm considering uninstalling the whole thing though. I just don't trust Google. Or I should rephrase that to say I don't trust Google thinks things through on privacy. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From b.j.smith at ieee.org Sun Feb 12 15:24:46 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] nVidia GeForce 6 and 7 series tabled, analyzed, sub-$100, $100+, $200+ recommendations ... Message-ID: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> Okay, I've really gotten TIRED of the mainstream enthusiast sites NOT putting up good COMPARISONS of all the endless nVidia variants. So lacking the hardware to do equivalent testing, I've put together a set of tables of ALL the GeForce 6 and 7 series -- NV44, NV43, NV40/41/42/45, G70, G72 and forthcoming/rumored G71/72/73 products: http://thebs413.blogspot.com/2006/02/geforce-6-and-7-series-variants-nuts.html I've also updated my recommendations based on an analysis at sub-$100 ($50-70), $100+ and $200+ in the same -- for _both_ PCIe and AGP. -- Bryan P.S. This, of course, assumes you're not considering ATI. ATI does have some interesting options at sub-$100 these days -- like the X1300 (ignoring the X300) that seems to best the GeForce 6600LE and 7300GS for PCIe at around the $75 price-point. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From jasonb at edseek.com Sun Feb 12 15:40:39 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] nVidia GeForce 6 and 7 series tabled, analyzed, sub-$100, $100+, $200+ recommendations ... In-Reply-To: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> References: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> Message-ID: <200602121540.39690.jasonb@edseek.com> On Sunday 12 February 2006 15:24, Bryan J. Smith wrote: > Okay, I've really gotten TIRED of the mainstream enthusiast sites NOT > putting up good COMPARISONS of all the endless nVidia variants. > > So lacking the hardware to do equivalent testing, I've put together a > set of tables of ALL the GeForce 6 and 7 series -- NV44, NV43, > NV40/41/42/45, G70, G72 and forthcoming/rumored G71/72/73 products: > > http://thebs413.blogspot.com/2006/02/geforce-6-and-7-series-variants-nuts.h >tml > > I've also updated my recommendations based on an analysis at sub-$100 > ($50-70), $100+ and $200+ in the same -- for _both_ PCIe and AGP. Sweet. Sounds like my next upgrade will be to a PCIe board for my gaming box and I'll trade the Socket 754 system I have now down to my workstation. Then, I can even run a 64-bit kernel if I want or try a 64-bit compilation of Debian. -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Sun Feb 12 16:00:20 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Upgrading this summer -- WAS: nVidia GeForce 6 and 7 series In-Reply-To: <200602121540.39690.jasonb@edseek.com> References: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> <200602121540.39690.jasonb@edseek.com> Message-ID: <1139778020.4727.22.camel@bert64.oviedo.smithconcepts.com> On Sun, 2006-02-12 at 15:40 -0500, Jason Boxman wrote: > Sweet. Sounds like my next upgrade will be to a PCIe board for my gaming box > and I'll trade the Socket 754 system I have now down to my workstation. > Then, I can even run a 64-bit kernel if I want or try a 64-bit compilation of > Debian. I've been holding off on an upgrade from my current nForce Standard mainboard with my Athlon 64 3200+, awaiting the AMD M2 socket. I figure I'll upgrade around the July-August timeframe, giving the market a good 3 months of "shakedown" in the platform, as well as see the next crop of video cards released. I'm planning on doing the following ... 1) My system: A) M2-S940 MicroATX w/GbE on-board B) M2-S940 dual-core CPU (at least 2x2.2GHz and/or 2x1MiB L2) C) 4GiB DDR2 SDRAM (probably 2x2GiB DDR2-533) D) New PCIe card (optional**) 2) Wife's system: A) New S939 MicroATX w/4xDDR slots, GbE on-board B) Existing S939 Athlon 64 3200+ (2.0GHz, 512KiB L2) C) Existing 4GiB DDR400 SDRAM (2x1GiB from mine + 2x1GiB from wife's -- all DDR400 2.5-3-3-6) D) nVidia GeForce 7800GTX (optional**) **NOTE: I'm either going to get a GeForce 7900GT[X] or whatever is the latest, assuming I can for $300-400. If not, and the new cards are $500, I might just buy a $200 GeForce 6800GS or $300 7800GT and put that in my wife's system, keeping and moving the GeForce 7800GTX for my new system. 3) That would leave the following available for sale/donation: A) S939 MicroATX nForce4 Standard B) S462 MicroATX ViA KM400+VT8235 _and_ AMD Athlon XP2600+ (2.13GHzm 256KiB L2) C) nVidia GeForce 6800GT 256MiB GDDR3 AGP I'd probably give 3A away for free or for small trade. I'd probably sell 3B for $50 or so (it's no slouch) I'd probably sell 3C for $150 or less. Although I'm tempted to keep it as a "backup AGP card" because it's a NV40 and 1.5V tolerant, so it works with any AGP 2.0 (and 3.0) mainboard. The 6800GT still kicks a lot of butt -- even though it's 18 months old. -- Bryan P.S. Although I _am_ going GbE in my house very shortly. All 802.3x at least, Jumbo frames capable (although probably disabled) -- although I'm tempted to get a L3 switch for a few hundred bucks with 4+ GbE ports. Putting in 0.5GBps PCI64-66 (servers) and 0.5Gbps PCIe x1 cards (workstations), although my wife's current, old system will only get a 32-bit PCI card (until I upgrade her this summer). I'm the bandwidth pig around there. ;-ppp -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From jvsmith at digitalmatter.us Sun Feb 12 19:20:29 2006 From: jvsmith at digitalmatter.us (Jason Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Upgrading this summer -- WAS: nVidia GeForce 6 and 7 series In-Reply-To: <1139778020.4727.22.camel@bert64.oviedo.smithconcepts.com> References: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> <200602121540.39690.jasonb@edseek.com> <1139778020.4727.22.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1139790030.2515.29.camel@athlon.bedroom.lan> On Sun, 2006-02-12 at 16:00 -0500, Bryan J. Smith wrote: > I've been holding off on an upgrade from my current nForce Standard > mainboard with my Athlon 64 3200+, awaiting the AMD M2 socket. Bryan, Can you elaborate on the new AMD M2 socket. I haven't heard anything about this particular socket. I'm hopefully going to get a new PC sometime this year and it would be nice to know whether I should wait for this particular socket to come out. Thanks, Jason -- Jason Smith jvsmith at digitalmatter dot us http://www.digitalmatter.us From b.j.smith at ieee.org Sun Feb 12 22:35:57 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Upgrading this summer -- WAS: nVidia GeForce 6 and 7 series In-Reply-To: <1139790030.2515.29.camel@athlon.bedroom.lan> References: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> <200602121540.39690.jasonb@edseek.com> <1139778020.4727.22.camel@bert64.oviedo.smithconcepts.com> <1139790030.2515.29.camel@athlon.bedroom.lan> Message-ID: <1139801757.4727.28.camel@bert64.oviedo.smithconcepts.com> On Sun, 2006-02-12 at 18:20 -0600, Jason Smith wrote: > Can you elaborate on the new AMD M2 socket. Actually, to correct myself, it is "AM2" now (no longer just M2). > I haven't heard anything about this particular socket. AMD is replacing Socket-939 with this new "AM2" Socket-940. It adds several features, as well all AM2 processors will have DDR2 support. It's not backward compatible with previous Socket-940 processors. The Opteron is getting a new, 1200 or so pin socket. > I'm hopefully going to get a new PC sometime this year and it would > be nice to know whether I should wait for this particular socket > to come out. Yes, I'd wait until summer. Unless you plan on reusing your existing DDR memory, then I'd said for the AM2 and other changes in the leading edge AMD platform. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From whittake at sbaflorida.com Mon Feb 13 11:10:17 2006 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Upgrading this summer -- WAS: nVidia GeForce 6 and 7 series In-Reply-To: <1139778020.4727.22.camel@bert64.oviedo.smithconcepts.com> References: <1139775886.4727.3.camel@bert64.oviedo.smithconcepts.com> <200602121540.39690.jasonb@edseek.com> <1139778020.4727.22.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1139847017.12156.9.camel@localhost.localdomain> On Sun, 2006-02-12 at 16:00 -0500, Bryan J. Smith wrote: > > I'm planning on doing the following ... > 3) That would leave the following available for sale/donation: > A) S939 MicroATX nForce4 Standard > B) S462 MicroATX ViA KM400+VT8235 > _and_ AMD Athlon XP2600+ (2.13GHzm 256KiB L2) Dibs on the AMD Atholon for $50. 1st: I'm queer for amd's and secondly I have a XP2600 and it is truly a jewel. It would be a nice replacement for my K7S5A which is getting old. Homer > C) nVidia GeForce 6800GT 256MiB GDDR3 AGP > > I'd probably give 3A away for free or for small trade. > > I'd probably sell 3B for $50 or so (it's no slouch) > > I'd probably sell 3C for $150 or less. Although I'm tempted to keep it > as a "backup AGP card" because it's a NV40 and 1.5V tolerant, so it > works with any AGP 2.0 (and 3.0) mainboard. The 6800GT still kicks a > lot of butt -- even though it's 18 months old. > > -- Bryan > > P.S. Although I _am_ going GbE in my house very shortly. All 802.3x at > least, Jumbo frames capable (although probably disabled) -- although I'm > tempted to get a L3 switch for a few hundred bucks with 4+ GbE ports. > Putting in 0.5GBps PCI64-66 (servers) and 0.5Gbps PCIe x1 cards > (workstations), although my wife's current, old system will only get a > 32-bit PCI card (until I upgrade her this summer). I'm the bandwidth > pig around there. ;-ppp > > From b.j.smith at ieee.org Mon Feb 13 14:45:22 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Portable Apps (for Windows) ... Message-ID: <20060213194522.74651.qmail@web34111.mail.mud.yahoo.com> Definitely something to keep on your USB thumb drive: http://portableapps.com/ There are also other projects out there that offer such capabilities, like Cygwin/X. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 13 15:12:46 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] If you haven't switched ... unpatched Firefox v. MS IE ... Message-ID: <20060213201246.94786.qmail@web34113.mail.mud.yahoo.com> Firefox has bugs and _can_ be hit by spyware by less savvy users as well ... ... sent their crawlers to 45,000 Web sites ... 1.6 percent of the domains infected the first IE configuration, the one mimicking a na?ve user blithely clicking 'Yes;' ... In the same kind of configurations, Firefox survived relatively unscathed. Only .09 percent of domains infected the Mozilla Corp. browser when it was set, like IE, to act as if the user clicked through security dialogs ... But when the user _denies_ downloads ... about a third as many domains (0.6 percent) did drive-by downloads by planting spyware even when the user rejected the installations ... no domain managed to infect the Firefox- equipped PC in a drive-by download attack. Let's read that again, out of approximately 45,000 web sites NO DOMAIN MANAGED TO INFECT THE FIREFOX-EQUIPPED PC IN A 'DRIVE-BY DOWNLOAD ATTACK'. In other words, even _unpatched_ Firefox 1.0 results in *0* spyware infections when you are not prompted out-of-the-box. Not *1* site out of 45,000! "Locked down" MS IE _still_ results in 0.6% of sites infecting you -- literally 270 sites out of the approximately 45,000 -- 1 out of every 163 sites you hit, even when MS IE has security turned _way_up_! -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From dmckenna at thelimucompany.com Mon Feb 13 16:13:29 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] AnandTech on Intel DuoCore vs USB Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D861@tlc001.tlcusa.thelimucompany.com> Intel DuoCore vs USB, Anand is the referee: http://www.anandtech.com/mobile/showdoc.aspx?i=2693 -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Mon Feb 13 18:48:27 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] AnandTech on Intel DuoCore vs USB In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D861@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060213234827.21516.qmail@web34103.mail.mud.yahoo.com> Damien McKenna wrote: > Intel DuoCore vs USB, Anand is the referee: > http://www.anandtech.com/mobile/showdoc.aspx?i=2693 The problem _always_ has been and _always_ will be that USB _lacks_ standards. It was designed by Intel (IC) and Microsoft (software host) to put _all_ of the brains in the end-device. That means the "host controller" has virtually _no_control_ over how the device behaves. Without those standards and without that control, everything is pretty arbitrary. Left to the device driver provided by the end device. That results in conflicts, incompatibility, total lack of the host being able to pre-empt the device, etc... Now there have been _some_ post-USB release standards created. But most are not followed, or only a small subset are. So this doesn't surprise me one damn bit. By the same token that Linux does _not_ work with a lot of USB devices, for those USB devices Linux _does_ work with, it's _well_implemented_. This is typically because A) the hardware must conform to many post-USB release standards/APIs and/or B) the drivers are written by the Linux USB team/developers, which _include_ advantaged management features. This Linux v. Windows 2000/XP on USB is a _repeat_ of the Linux v. Windows 9x on PCMCIA/CardBus issue. For Windows, Microsoft just relied on vendors for PCMCIA/CardBus controller and end device drivers -- and they often _conflicted_ with each other (before NT standardized many things). The Linux world built a _proper_ set of standard host interfaces and drivers were written meeting those host interfaces for both the controller and end devices. So, again, it doesn't surprise me one damn bit that USB devices basically _control_ Windows, and not vice-versa. Ironically enough, Apple (who has a BSD licensed kernel and can sign NDAs -- a major plus over the Linux community) is far better of an example than Microsoft on how to "tame" USB. Until Microsoft starts taking responsibility for building a _proper_ USB subsystem, instead of just leaving drivers to the end-devices, this will continue. Just my $0.02 ... -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Tue Feb 14 17:08:22 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard .... Message-ID: <43F254D6.3050705@HiWAAY.net> .... I just got finished building a box for the boss, Asus P4P800 Mbd, 4 X 1 GB Corsair DDR400 RAM sticks, various other goodies. Installed SuSE 9.2 w/ only 1 GB of RAM onboard (longstanding bug w/ SuSE (& maybe other distro) installers), then put the rest in & tried to reboot. It hung partway through the OS boot (past POST successfully). I took 2 GB out (left DIMMs in slots 1 & 3 for dual channel), rebooted, & viola, everything runs smooth as silk, no problems whatsoever. I suspected bad RAM or bad slots, so I booted down, removed the RAM from slots 1 & 3, & installed the other 2 DIMMs in slots 2 & 4 & booted memtest86 v3.2. It is running overnight, but when I left, it had gone about 30 minutes silently, making me think the RAM & mbd are in fact OK. I *think* there are some arguments to use at boot-time (booting w/ GRUB, BTW) when you are using > 2 GB of RAM, but they elude me at the moment (I *am* RTFM'ing furiously). Does this ring any bells w/ anyone ? C'mon, *GIVE* :-) .... TIA -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From b.j.smith at ieee.org Tue Feb 14 17:55:35 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- JEDEC spec, 3.6GiB, limits of S478 In-Reply-To: <43F254D6.3050705@HiWAAY.net> Message-ID: <20060214225535.49394.qmail@web34115.mail.mud.yahoo.com> [ I'm not on LEAPLIST, so feel free to forward this to LEAPLIST. Please don't ask me to resubscribe to LEAPLIST, because I really tire of anything I say or, worse yet, recommend being questioned to death. ] Summary of considerations: 1. Asus P4P800 is a known mainboard with _major_issues_ 2. JEDEC specs for PC3200 call for *1* DIMM per channel 3. Server/Workstation BIOSes offer a "memory hole" at 3.6GiB 4. Intel Socket-478 _lacks_ Process Address Extensions (PAE) support We'll get to each in my bottom posting ... "William A. Mahaffey III" wrote: > .... I just got finished building a box for the boss, > Asus P4P800 Mbd, 1. Danger! Danger Will Robinson! The Asus P4P800 is a well known and _poor_ design. Asus took the P8C800 with the i875 chipset and replaced it with the i865 to create the P4P800. The i865 and i875 rolls off the _same_ fab lines, _but_ the i865 units are so labeled because they _fail_ the tolerances for the i875 branding. Because Asus developed and tested the P8C800 PCB and microelectronics against the i875 and its better tolerances, _expect_ a higher rate of issues with P4P800 products -- especially as the microelectronics age. You've been warned. ;-> > 4 X 1 GB Corsair DDR400 RAM sticks, 2. JEDEC spec for PC3200 recommend 1 DIMM per channel JEDEC standards on PC3200 (DDR400) recommend only 2 [32-bit] banks (1 [64-bit] DIMM) per PC3200 channel for non-registered/unbuffered (double for registered/buffered). You can use 4 banks (2 DIMMs) for PC2100/2700 (DDR266/333), and 6 banks (3 DIMMs) for PC1600 (DDR200) SDRAM (again, double each for registered/buffered). Most mainboard vendors don't bother to document this at all, leaving it to the system integrator (or end-assembly enthusiast ;-) to know this. BTW, _only_ Socket-939/940 are _true_ "dual-channel" -- direct 368-trace to "glueless" 128-bit wide memory bus. Socket-603/604 and LGA-775 are 128-bit wide busses via the Memory Controller Hub (MCH), so they are virtually dual-channel. Socket-462 (EV6 Athlon) and Socket-478 (AGTL+ Pentium 4) are actually 64-bit wide memory switch (EV6) or hub (AGTL+) interleaved between two (2) 64-bit DDR channels -- so really single channel, but use interleaving to reduce latency. > various other goodies. Installed SuSE 9.2 w/ only 1 GB of RAM > onboard (longstanding bug w/ SuSE (& maybe other distro) > installers), then put the rest in & tried to reboot. It > hung partway through the OS boot (past POST successfully). FYI, there _numerous_issues_ with putting 4GiB in a 32-bit platform. 3. Server/Workstation BIOSes offer a "memory hole" at 3.6GiB Because 32-bit platforms need to map _all_ addresses to _under_ 4GiB, you need to "reserve" a good 384MiB or so for ROM and memory mapped I/O. On more server/workstation BIOSes, there is typically a "memory hole" option at 3.6-4.0GiB. Do _not_ confuse this with the legacy 24-bit ISA 15-16MiB "memory hole" (which many BIOSes still offer as well). _Unfortunately_, you typically _only_ find this on Intel 7200/7500 (which are based on the RCC/ServerWorks ServerSet IV), and _rarely_ on the "cheaper/desktop" i865 (although sometimes you'll find it on hte i875). If you don't have the option, then this is a _major_ issue as the Intel APIC registers need to be configured at POST-time. [ SIDE NOTE: If you research "bounce buffers" -- you'll quickly find this is a "necessary evil" and _platform_-level limitation of even "alleged 64-bit" Intel IA-32e/EM64T processors, and something that AMD Athlon 64/Opteron (and even the 32-bit Athlon MP) does _not_ have an issue with thanx to its true 40-bit EV6 interconnect, NUMA/HyperTransport design and I/O Memory Management Units (MMUs) for even "legacy" 32-bit hardware. ] Which brings us to yet another, related issue ... 4. Intel Socket-478 _lacks_ Process Address Extensions (PAE) support Intel Socket-478 (IIRC) does _not_ have Processor Address Extensions (PAE) with 36-bit/64GiB addressing support. You have to be running Socket-8 (Pentium Pro), Slot-2 (P2/P3 Xeon), Socket-423 (P4 DRDRAM -- i.e., Rambus), Socket-603/604 (P4 Xeon) or LGA-775 (P4 128-bit MCH/PCIe). PAE support (also known retroactively as PAE 36-bit, because AMD x86-64's "Long Mode" is PAE 52-bit, which is PAE 36-bit as well as 32/24/20-bit compatible) is basically _required_ to run the Linux kernel in 4G/4G mode that would let you access the full 3.6GiB. Otherwise, you're stuck like Windows -- limited to largely 2GiB mode max (or the various 3GiB hacks for both the NTLDR/kernel.exe or Linux). This, in a nutshell, hits you _hard_ on Socket-478. Even AMD "32-bit" Socket-462 is _much_, _much_ better and has _full_ PAE support (and even 40-bit addressing support internally, long story). > I took 2 GB out (left DIMMs in slots 1 & 3 for dual channel), > rebooted, & viola, everything runs smooth as silk, no problems > whatsoever. I suspected bad RAM or bad slots, so I booted down, > removed the RAM from slots 1 & 3, & installed the other 2 DIMMs > in slots 2 & 4 & booted memtest86 v3.2. > It is running overnight, but when I left, it had gone about 30 > minutes silently, making me think the RAM & mbd are in fact OK. Again, understand the Socket-478 platform was _never_ designed for more than 4GiB, much less "very compatible" at more than 2GiB. You're also violating JEDEC specifications on PC3200 (DDR400). > I *think* there are some arguments to use at boot-time (booting w/ > GRUB, BTW) when you are using > 2 GB of RAM, but they elude me at > the moment (I *am* RTFM'ing furiously). Does this ring any bells w/ > anyone ? C'mon, *GIVE* :-) .... TIA Append "mem=" to your boot line. _Unfortunately_ that does _not_ address the larger issue of attempting to use the _entire_ 4GiB space on a machine that much map _all_ ROM, memory mapped I/O, kernel space and other things under that 4GiB. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 14 18:01:33 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- mem=3584M In-Reply-To: <20060214225535.49394.qmail@web34115.mail.mud.yahoo.com> Message-ID: <20060214230133.82267.qmail@web34103.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > Append "mem=" to your boot line. You can _try_ "mem=3584M", however ... > _Unfortunately_ that does _not_ address the larger issue of > attempting to use the _entire_ 4GiB space on a machine that much > map _all_ ROM, memory mapped I/O, kernel space and other things > under that 4GiB. It's very, very, _very_likely_ that without the BIOS option to create a "memory hole" at 3.6-4.0GiB, your ROM and memory mapped I/O will _not_ be mapped correctly in the system's APIC registers, and you will _still_ have issues. And that's assuming it's not the fact that you're well beyond JEDEC PC3200 standards on the DDR400 signaling with 8 banks (4 DIMMs). ;-> -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Tue Feb 14 18:33:10 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- mem=3584M In-Reply-To: <20060214230133.82267.qmail@web34103.mail.mud.yahoo.com> References: <20060214230133.82267.qmail@web34103.mail.mud.yahoo.com> Message-ID: <43F268B6.8060600@HiWAAY.net> Bryan J. Smith wrote: >"Bryan J. Smith" wrote: > > >>Append "mem=" to your boot line. >> >> > >You can _try_ "mem=3584M", however ... > > > >>_Unfortunately_ that does _not_ address the larger issue of >>attempting to use the _entire_ 4GiB space on a machine that much >>map _all_ ROM, memory mapped I/O, kernel space and other things >>under that 4GiB. >> >> > >It's very, very, _very_likely_ that without the BIOS option to create >a "memory hole" at 3.6-4.0GiB, your ROM and memory mapped I/O will >_not_ be mapped correctly in the system's APIC registers, and you >will _still_ have issues. > >And that's assuming it's not the fact that you're well beyond JEDEC >PC3200 standards on the DDR400 signaling with 8 banks (4 DIMMs). ;-> > > Thanks for the detailed response, we are *quite* aware of the limitations of 4 GB address space max, minus whatever the kernel/BIOS want, we are looking for a bit over 3 GB for the application. Where do I put the "mem=3584M" line (in what file, we don't want to have to remember it every time we reboot) ? The man pages (at least the default ones) are notably sparse on details of GRUB ('apropos grub' only shows 3 entries, all quite truncated, virtually useless). TIA .... As an aside/digression, we are also aware of the problems/limitations of using a relatively old OS for this box, but the client in this case is using predominantly SuSE 9.3 on desktops and a similar vintage SLE on their cluster, and we wanted as few problems/compatibility-surprises w/ the software we are supposed to deliver as possible :-). Once SuSE 9.2 was settled on, we deliberately avoided newer hardware (S775 Pentiums, for example) over hardware-compatibility concerns w/ the older SuSE. *Not* my decision making, BTW, I suggested an Opteron :-) .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.matrixlist.com/pipermail/pc_support/attachments/20060214/4b5bcaeb/attachment.html From b.j.smith at ieee.org Tue Feb 14 18:56:06 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- mem=3584M In-Reply-To: <43F268B6.8060600@HiWAAY.net> Message-ID: <20060214235606.87324.qmail@web34101.mail.mud.yahoo.com> "William A. Mahaffey III" wrote: > Thanks for the detailed response, we are *quite* aware of the > limitations of 4 GB address space max, minus whatever the > kernel/BIOS want, Er, I don't think you realize the _full_details_ of such. 1. It's pretty much the "common rule" to reserve the last 384MiB for ROM, memory mapped I/O and other _hardware_ usage. This _must_ be setup at POST time IIRC. 2. The _kernel_ space usage is _another_ detail. Using the _standard_ i486 TLB ... The Linux kernel offers 1G/3G user/kernel (960MiB userspace), and options for 2G/2G user/kernel (1984MiB userspace) and I think there is even a hack for 3G/1G (or maybe it was 2.5G/1[.5]G?) user/kernel. I've _only_ seen the former two in action -- _never_ anything beyond 2GiB userspace without using PAE (see below). Even under Workstation/Server (non "Advanced Server") NTKernel.exe NT4.0, 5.0 (2000) and 5.1 (XP/2003), it is 2G/2G user/kernel (1984MiB userspace). On NT 4.0 Enterprise, 5.0 (2000) Advanced/Datacenter Server and any 5.1 (XP/2003), you _can_ pass an option of /3GB in the BOOT.INI to give 3GiB userspace, although programs have to know about the 3GiB option to use it. E.g., Oracle server is one such program. Now if you enable the PAE of i686 TLB -- _assuming_ you have a PAE capable platform -- you can get different models ... The common Linux kernel approach (and default in 2.6 IIRC) is the 4G/4G user/kernel model, where the userspace gets 1 or more 4GiB userspace sets, and the kernel gets its own 4GiB -- up to the maximum of 64GiB for 36-bit PAE. AMD's "Long Mode" 52-bit PAE is different, although it's 36-bit PAE compatible. In Windows NT 4.0 Enterprise, NT 4.0 Advanced/Datacenter Server and any 5.1 (XP/2003), multiple 512MiB user space pages are mapped under 4GiB as needed, up to 64GiB for 36-bit PAE. Again, IIRC, _no_ Socket-478 platform processor offers PAE support. It's the primary reason why I adopted either Socket-462 Athlon or Socket-603/604 P4-Xeon _instead_ -- because _both_ platforms _always_s support PAE. I could be mistaken though. Maybe I'm thinking that the Socket-478 only supports 3.6GiB (and not 36-bit/64GiB), _but_ it still has PAE so it can use the 4G/4G kernels. I can't remember for sure. > we are looking for a bit over 3 GB for the application. Unless you use the 4G/4G kernel model, you're going to be limited to 2GiB userspace. > Where do I put the "mem=3584M" line (in what file, we don't want to > have to remember it every time we reboot) ? At the end of the kernel line in your grub.conf or lilo.conf. > The man pages (at least the default ones) are notably sparse on > details of GRUB ('apropos grub' only shows 3 entries, all quite > truncated, virtually useless). TIA .... $ info grub > As an aside/digression, we are also aware of the > problems/limitations of using a relatively old OS for this box, Even age-old SuSE Linux [Enterprise Server] 7.x _offers_ the old 64GiB model. I know 9.x uses the newer 4G/4G model and I want to say 8.x also. *BUT* those modes _require_ you to have a PAE capable processor. One quick way to find out is to: # cat /proc/cpuinfo The 4G/4G mode _might_ work on some Socket-478 processors. Again, I'm not sure how PAE works on Socket-478 -- I just know it does _not_ support more than 4GiB of _physical_ RAM. I just tired of the whole non-sense of Intel and went EV6 -- even AMD Socket-462 -- which was physically a 40-bit/1TiB platform. Not tricks, no non-sense, pure 40-bit. > but the client in this case is using predominantly SuSE 9.3 on > desktops and a similar vintage SLE on their cluster, and we wanted > as few problems/compatibility-surprises w/the software we are > supposed to deliver as possible :-). Once SuSE 9.2 was settled on, > we deliberately avoided newer hardware (S775 Pentiums, for example) > over hardware-compatibility concerns w/ the older SuSE. Consider SL[ES] 9.2 x86-64 to _kill_ the issue altogether. > *Not* my decision making, BTW, I suggested an Opteron :-) .... Well, you _should_ have gone with an _enterprise_ platform like E7200/7500 Socket-604 P4-Xeon. Or at least considered the P8C800 with the i875 -- the P8P800 with the i865 is a _bastard_. I'd really push you to consider _at_least_ a i875, if not an E7200 at a _bare_minimum_. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Tue Feb 14 19:05:46 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- short answer ... In-Reply-To: <20060214235606.87324.qmail@web34101.mail.mud.yahoo.com> Message-ID: <20060215000546.97184.qmail@web34104.mail.mud.yahoo.com> "William A. Mahaffey III" wrote: > we deliberately avoided newer hardware (S775 Pentiums, for > example) over hardware-compatibility concerns w/ the older SuSE. Short answer ... You _should_ have gone Socket-604 (pre-DDR2/PCIe). Socket-604 is the _professional_/_enterprise_ version of Socket-478, _more_reliable_, _more_compatible_, etc... Pre-PCIe E7200/7500 chipsets are basically existing, _reliable_ ServerWorks ServerSet IV APIC, then paired with Intel ESB/ICH I/O Controller Hubs (ICH) for peripherals. Why you chose Socket-478? You tell me. Why you chose the "consumer"/"*FAILED* i875 quality" i865? You tell me. ;-> Not trying to pick on you, but I don't understand the logic here. If I walked into a client that needed >>3GiB of userspace, but recommended Socket-478 with the i865 chipset -- I'd expect them to hit me with a 2x4. You do _not_ have to go AMD or LGA-775 to get >>3GiB. You can adopt _legacy_ Socket-604 (or even Socket-603) for that matter and get _better_ compatibility _and_ reliability. Socket-603/604 solutions are _cheaper_ than you think. ;-> -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From wam at HiWAAY.net Tue Feb 14 19:19:12 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- mem=3584M In-Reply-To: <20060214235606.87324.qmail@web34101.mail.mud.yahoo.com> References: <20060214235606.87324.qmail@web34101.mail.mud.yahoo.com> Message-ID: <43F27380.3020000@HiWAAY.net> Bryan J. Smith wrote: >"William A. Mahaffey III" wrote: > > >>Thanks for the detailed response, we are *quite* aware of the >>limitations of 4 GB address space max, minus whatever the >>kernel/BIOS want, >> >> > >Er, I don't think you realize the _full_details_ of such. > >1. It's pretty much the "common rule" to reserve the last 384MiB for >ROM, memory mapped I/O and other _hardware_ usage. This _must_ be >setup at POST time IIRC. > >2. The _kernel_ space usage is _another_ detail. > >Using the _standard_ i486 TLB ... > > > > >I could be mistaken though. Maybe I'm thinking that the Socket-478 >only supports 3.6GiB (and not 36-bit/64GiB), _but_ it still has PAE >so it can use the 4G/4G kernels. I can't remember for sure. > > I have written small programs just to allocate & access (do arithmetic) on arbitrarily large amounts of RAM, you *can* address over 3 GB in user-space w/ an Intel I865 chipset & S478 CPU under SuSE 9.2 (& SuSE 8.2, kernel 2.4.20-4GB for that matter), I have done it 1st hand. *SLOW* as soon as it starts to page, but that's a separate issue (those boxen only had 2 GB RAM or less, this one has 4 GB) > > >>we are looking for a bit over 3 GB for the application. >> >> > >Unless you use the 4G/4G kernel model, you're going to be limited to >2GiB userspace. > > > >>Where do I put the "mem=3584M" line (in what file, we don't want to >>have to remember it every time we reboot) ? >> >> > >At the end of the kernel line in your grub.conf or lilo.conf. > > > >>The man pages (at least the default ones) are notably sparse on >>details of GRUB ('apropos grub' only shows 3 entries, all quite >>truncated, virtually useless). TIA .... >> >> > > $ info grub > > Did I mention I *loath* info :-) ? I think this will get us underway, thanks all :-). -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.matrixlist.com/pipermail/pc_support/attachments/20060214/5d75a87e/attachment.html From b.j.smith at ieee.org Wed Feb 15 06:55:17 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:40 2006 Subject: [Pc_Support] Re: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard -- virtual paging != physical tables In-Reply-To: <43F27380.3020000@HiWAAY.net> References: <20060214235606.87324.qmail@web34101.mail.mud.yahoo.com> <43F27380.3020000@HiWAAY.net> Message-ID: <1140004517.4723.13.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-02-14 at 18:19 -0600, William A. Mahaffey III wrote: > I have written small programs just to allocate & access (do > arithmetic) on arbitrarily large amounts of RAM, you *can* address > over 3 GB in user-space w/ an Intel I865 chipset & S478 CPU under SuSE > 9.2 (& SuSE 8.2, kernel 2.4.20-4GB for that matter), I have done it > 1st hand. *SLOW* as soon as it starts to page, but that's a separate > issue (those boxen only had 2 GB RAM or less, this one has 4 GB) Again, you're _confusing_ physical RAM with virtual. I'm not talking about "usable virtual," I'm talking about "usable _physical_." I probably should have mention that when I say "paging" I mean changing "_memory_ page _tables_". Entirely _different_ concepts. IIRC, the (non-x86-64) kernel uses 39-bit _software_ addressing. That's how it does virtual memory up to 128TiB. That's in 100% _software_. The kernel does _not_ use the processors page tables for virtual memory (at the most, it _only_ uses its virtual 48-bit addressing). When it comes to _physical_ memory, you're at the _mercy_ of the _hardware_ TLB. The i486 TLB is only capable of normalizing to 32-bit values for _any_ physical RAM, ROM, memory mapped I/O, etc... -- period. Unless the kernel enables the PAE hardware to give it 36-bit page tables. Again, do _not_ confuse the concepts of _virtual_ memory and paging with _physical_ page tables. The former is OS/software, the latter is enable by the OS (if hardware supported) but done in _hardware_. > Did I mention I *loath* info :-) ? I don't like it either, but it _does_ have a _lot_ more info for GRUB. I liked the old Galeon web browser because it had a native "info://" interface. I haven't seen it with Epiphany or Firefox. Both just try to launch info in a terminal. Wonder if Konqueror does it? ... let's see ... Yep! Very nice! -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From b.j.smith at ieee.org Wed Feb 15 07:41:59 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Understanding _physical_ memory models -- WAS: LUNA: How to get SuSE 9.2 to boot w/ 4 GB of RAM onboard In-Reply-To: <1140004517.4723.13.camel@bert64.oviedo.smithconcepts.com> References: <20060214235606.87324.qmail@web34101.mail.mud.yahoo.com> <43F27380.3020000@HiWAAY.net> <1140004517.4723.13.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1140007319.4723.56.camel@bert64.oviedo.smithconcepts.com> On Wed, 2006-02-15 at 06:55 -0500, Bryan J. Smith wrote: > When it comes to _physical_ memory, you're at the _mercy_ of the > _hardware_ TLB. The i486 TLB is only capable of normalizing to 32-bit > values for _any_ physical RAM, ROM, memory mapped I/O, etc... -- period. > Unless the kernel enables the PAE hardware to give it 36-bit page > tables. - The i486 TLB If you're not seeing a repeat theme here, the i486 Translation Lookaside Buffer (TLB) is pretty much the "staple" of 20-bit/1MiB (Real86) segmented and 24/16MiB (Protected286) and 32-bit/4GiB (Protected386 and Virtual86) flat memory model compatibility in a 48-bit/256TiB virtual table. All 48-bit/256TiB virtual addresses _must_ "normalize" into a 20/24/32-bit _physical_ address. No matter what the kernel does, it _must_ make the _physical_ memory appear as a 32-bit/4GiB space with all the i486 TLB addressing functions. - i486 4K v. i586+ 4M page tables You may have heard of the concept of i486 4K[iB] and i586/Pentium 4M[iB] "page tables." That's the TLB they are talking about. _All_ Intel processors have very _weak_ TLBs in comparison to AMD's Athlon cores (including 64/Opteron as well as 32-bit) design, which is why AMD made a very quick transition to a new, 52-bit PAE model (see below), although they _under-tested_ 4M page tables because it's just as fast at 4K page tables for large memory (whereas i586/Pentium and i686/Pentium Pro get a good 10% boost out of 4M page tables). - Virtual memory = 100% software OS, Page tables = 100% _hardware_ TLB This is _completely_different_ than virtual memory. Virtual memory is 100% _software_ paging. The kernel "pages" memory in and out of different parts of the disk as memory to give you more. The kernel has its own 39-bit (IIRC for x86, even with 36-bit PAE), so you _can_ have up to 128TiB of _virtual_ memory. But this is completely _separate_ from the i486 TLB and its _physical_ 32-bit/48-bit page tables (although the kernel virtual memory _might_ use points to non-normalized 48-bit virtual addressing outside of the 32-bit/4GiB space, but just for its _own_ use -- not sure on that one and how the 39-bit _software_ addressing of the kernel works exactly). - Getting around 32-bit addressing w/Processor Address Extensions (PAE) Processor Address Extensions (PAE) let you map in different _physical_ pages below 32-bit/4GiB, so you can use up to 36-bit/64GiB for i686 PAE capable processors. It also allows for different, protected memory models of kernel and user-space, even when you have 4GiB or less. It uses the 48-bit/256TiB virtualization in _hardware_ very differently. It's a long, long, involved story -- but it allows additional 4K or 4M page tables to be used. Programs can _also_ be written to the 36- bit/64GiB model -- they basically understand how the OS' kernel is doing PAE paging (e.g., old 64G or new 4G/4G model in Linux, or 512M page tables in Win32PAE) so they can use them directly. [ Knowing how this works is _very_, _very_important_ for applications like Oracle for 32-bit OSes. E.g., Oracle services written for the older 64G kernel model typically won't work on the newer 4G/4G model in Linux. ] - AMD 64-bit "Long Mode" is actually 48-bit (with 52-bit PAE) The first generation of AMD x86-64 uses a 7-layer, 52-bit/1PiB PAE, that is both PAE 36-bit/64GiB compatible and still gives you full 20/24/32- bit addressing capability. In a nutshell, instead of using the 32-bit offset register as a "flat" 32-bit model, and then the 16-bit segment register for page tables with the TLB -- the 16-bit segment and 32-bit offset are combined into a _physical_ 48-bit address register (hence the name "Long Mode"). So the first generation AMD x86-64 uses _flat_ 48-bit/256TiB addressing (_not_ 64-bit). The actual page table registers in a x86-64 (which the kernel sets up/uses) are 52-bit to accommodate the Processor Address Extensions (PAE) that are the 7-layer TLB that does make completely compatible with both 36-bit PAE applications and the old 20/24/32-bit memory models. If AMD did _not_ offer the "Long Mode" approach with a superset PAE design, you could _not_ have x86-64 kernels that were memory model compatible with PAE 36-bit, flat 32-bit, 24-bit and segmented 20-bit programs. - AMD x86-64 is physically 40-bit, due to legacy Athlon EV6 platform 32-bit Athlon was designed for the 40-bit/1TiB Alpha (a 64-bit processor) EV6 "switch" platform. AMD's use of the EV6 interconnect was the first time anyone took a non-GTL interconnect and virtualized the PC's GTL "hub" architecture over a _wholly_incompatible_ interconnect. As you may have seen me mention in other threads, had Intel not created the APIC design in i586+ -- this might have not been possible (long story). The Athlon 64 and Opteron are basically the _same_ "core design" as the 32-bit Athlon. But now just "open up" the _full_ 40-bit addressing with new 52-bit PAE register / 48-bit flat model support in the TLB, as well as tunnel that EV6 addressing over the "glueless" HyperTransport concept. The addressing foundations of EV6 are still at the heart of the Athlon64/Opteron. - Side Note: I/O Memory Management above 32-bit/4GiB AMD goes one step further and adds what's known as an I/O Memory Management Unit (MMU). This is a unit that uses page tables for memory mapped I/O. Why would you want that? Because 32-bit hardware (and even some 64-bit hardware that only offers 32-bit addressing -- especially in old firmware/drivers) that is _unable_ to address programs/services above 32-bit/4GiB needs to be accommodated. In the "old days," the kernel used "bounce buffers" below 32-bit/4GiB so all programs/services talked to memory mapped I/O and did Direct Memory Address (DMA) transfers below 32-bit/4GiB. This is why you have the separate user/kernel space (as well as the 384MiB "memory hole") -- because the kernel needs to _reserve_physical_ memory (be it real with the i486 TLB, or physical pages mapped by the page table under 32- bit/4GiB with PAE36 in either 64G or 4G/4G models in Linux) for these "bounce buffers." It's a massive _waste_ of both memory and resources, and _severely_limits_ the amount of memory mapped I/O that can be used. The I/O MMU of the Athlon64/Opteron (as well as 32-bit Athlon MP, long story) handles page tables that exist above 32-bit/4GiB and make them appear below 32-bit/4GiB for I/O operations that do not support the flat 48-bit model. It a _major_speed_boost_. Small Hack: The Linux kernel actually has a "special mode" for "specially configured" 32-bit Athlon MPs. The Athlon MP actually has the predecessor of the Athlon 64/Opteron's I/O MMU in its on-board AGPgart. Intel [A]GTL[+] "hub" architecture forces _all_ operations to happen 1 at a time through a _single_ point -- so Intel just puts the AGPgart and other I/O memory management in the single Memory Controller Hub (MCH, the "northbridge" of the chipset) for consistency. EV6 is a 16-point "switch," Alpha 264 and Athlon MP processors, memory and I/O act _independently_ from each other, with _multiple_ transactions occurring simultaneously -- so the AGPgart _must_ be in the end processors who track consistency independently. So, in a nutshell, there are Linux kernels which can enable this AGPgart as a greater I/O MMU -- because it's capable of tracking _any_ type of I/O DMA as well as AGP's DiME (Direct, in-Memory Execution -- i.e., AGP acts like a peer processor that can _use_ memory, _not_ an I/O card that just transfers to/from memory) The _lack_ of an I/O MMU in _all_ Intel IA-32e/EM64T processors is a _major_ performance (and reliability) issue. The Linux kernel _always_ enables "bounce buffers" because it can't trust 32-bit hardware on Intel's "alleged 64-bit" Pentium 4s at all. Red Hat has tried and they just can't guarantee drivers will do it for old 32-bit hardware, so "bounce buffers" mean you're cool Intel 64-bit processor virtually acts like a 32-bit processor from the standpoint of I/O performance (which is the reality of _any_ I/O-oriented server ;-). -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From jasonb at edseek.com Wed Feb 15 10:29:34 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 3ware corruption on K7S5A Message-ID: I saw this twice myself, the first time under Windows (I was dual booting, but my OS files for Linux were on a SCSI array) where I had Windows on my RAID 1 array on a 6200 shared with my Linux /home. The second time with a different 3Ware a year later (6200 again) I tried straight up RAID 1 for my Debian install (dual booting had gone away) and had the same issues as Windows. Random segfaults, corruption, death. It was immediate and continuous such that the machine wasn't usuable. Not a Windows issue after all. Months later, I thought I'd Google today for the heck of it and found this where someone claims confirmation from 3Ware back in 2002 that this is a problem with the K7S5A. http://www.geek.com/news/geeknews/2001nov/bch20011129009082.htm "data corruption and incompatability (2:30pm EST Fri Jul 19 2002) "Using the K7S5A with 3ware RAID cards can cause data corruption (and is on my machine). I will have to return the K7S5A and move to a better board. I have confirmed this with 3ware tech support. "While the board may work for some- I would not use it for any advanced applications. - by frustrated" Can anyone confirm this, before I consider trying again. I was hoping it was just a random PCI slot and I needed to just try another or something, but probably not... Thanks. From b.j.smith at ieee.org Wed Feb 15 12:01:26 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 3ware corruption on K7S5A In-Reply-To: Message-ID: <20060215170126.58750.qmail@web34112.mail.mud.yahoo.com> Jason Boxman wrote: > Months later, I thought I'd Google today for the heck of it and > found this where someone claims confirmation from 3Ware back in > 2002 that this is a problem with the K7S5A. > http://www.geek.com/news/geeknews/2001nov/bch20011129009082.htm I stopped using advanced storage and other multi-function/bridged devices on the SiS735 chipset after I found APIC/I2C comaptibility issues. E.g., I had issues with PCI-to-CardBus controllers. But I still liked the K7S5A as a general desktop. I even ran it with a 3Ware Escalade 6410 card for awhile. Kinda wondering now if that's why I lost that old XFS 1.0 version /var filesystem years ago -- the combination of a 3Ware not flushing a buffer and the XFS bug's inability to recover such an inconsistent filesystem (fixed in XFS 1.1). -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From dmckenna at thelimucompany.com Wed Feb 15 18:11:50 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Fasttrack100TX2 vs Adaptec 7896N? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D95E@tlc001.tlcusa.thelimucompany.com> I'm reinstalling a web server here at work and have come to realize that what was touted to me as a good server really is, well, a piece of monkey.Dooh. Its an Intel 440GX+ mobo with onboard Adaptec 7896N chip that can power LVD drives, yet someone decided to put in a desktop-level Promise Fasttrack100TX2 in it and two IBM Deskstar drives rather than some SCSI goodness. Curious. So, given that its only a dual 1GHz system running an Intel chipset, would it be worth spending some money on getting a SCSI drive or two, or should I just deal with the IDE drives? Thanks. -- Damien McKenna From b.j.smith at ieee.org Wed Feb 15 18:33:06 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Fasttrack100TX2 vs Adaptec 7896N? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D95E@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060215233306.25408.qmail@web34112.mail.mud.yahoo.com> Damien McKenna wrote: > I'm reinstalling a web server here at work and have come to > realize that what was touted to me as a good server really is, > well, a piece of monkey. How dare you question Muppet Labs! (a quote from someone else, I believe on this list?) > Dooh. Its an Intel 440GX+ mobo Glorified i440BX with a little more SDRAM technology support and Slot-2 support. Still piece'o schnike 32-bit@33MHz PCI. > with onboard Adaptec 7896N chip that can power LVD drives, yet > someone decided to put in a desktop-level Promise Fasttrack100TX2 > in it and two IBM Deskstar drives rather than some SCSI goodness. Geez, a Fastrack "FRAID"! Does the mainboard have a Zero-channel SCSI RAID slot? I seriously doubt it (since no 64-bit PCI or PCI-X), but I have to ask. > Curious. So, given that its only a dual 1GHz system running an > Intel chipset, would it be worth spending some money on getting a > SCSI drive or two, or should I just deal with the IDE drives? Unless it has a Zero-channel SCSI RAID slot, you'd be better off replacing the card with a $100 3Ware Escalade 7006-2 instead. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From pberry2 at cfl.rr.com Wed Feb 15 20:14:35 2006 From: pberry2 at cfl.rr.com (Patrick) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Fasttrack100TX2 vs Adaptec 7896N? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D95E@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D95E@tlc001.tlcusa.thelimucompany.com> Message-ID: <200602152014.35167.pberry2@cfl.rr.com> On Wednesday 15 February 2006 18:11, Damien McKenna wrote: > I'm reinstalling a web server here at work and have come to realize that > what was touted to me as a good server really is, well, a piece of > monkey.Dooh. Its an Intel 440GX+ mobo with onboard Adaptec 7896N chip > that can power LVD drives, yet someone decided to put in a desktop-level > Promise Fasttrack100TX2 in it and two IBM Deskstar drives rather than > some SCSI goodness. Curious. > > So, given that its only a dual 1GHz system running an Intel chipset, > would it be worth spending some money on getting a SCSI drive or two, or > should I just deal with the IDE drives? > > Thanks. The SCA/LVD drives I have been getting are 18GB and 50Gb Seagate Cheetah and Barracuda drives, bidding out on ebay, used, at about $0.20 to $0.55 per Gb, delivered. My advice is this: You'll want some LVD drives anyway, right? So, get a couple, used or new, that you really want. Then, if you still think it is so bad (when it starts doing file transfers at 160 Mb/sec, or 320Mb/sec) and you want to upgrade, you can sell the dual cpu system at a profit., on ebay, and buy something you really want! Anything dual cpu really bids up fast on ebay! Go look at it! -- http://livecdlist.com http://distrowatch.com http://www.pclinuxonline.com/pclos http://yolinux.com http://safeharbordome.com http://minidome.net http://monolithicdome.com From damien at mc-kenna.com Wed Feb 15 20:57:19 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Fasttrack100TX2 vs Adaptec 7896N? In-Reply-To: <20060215233306.25408.qmail@web34112.mail.mud.yahoo.com> References: <20060215233306.25408.qmail@web34112.mail.mud.yahoo.com> Message-ID: <43F3DBFF.3050706@mc-kenna.com> Bryan J. Smith wrote: >> Dooh. Its an Intel 440GX+ mobo >> > Glorified i440BX with a little more SDRAM technology support and > Slot-2 support. Still piece'o schnike 32-bit@33MHz PCI. > .. and dual CPU. It does actually have an extended PCI slot, never seen one before, not sure what its for, but all of the on-mobo PCI slots are covered by the generic 32bit PCI riser card anyway. If I'm working it out correct, it was a server thrown together from spare parts. Oh, and it has an ISA slot! >> with onboard Adaptec 7896N chip that can power LVD drives, yet >> someone decided to put in a desktop-level Promise Fasttrack100TX2 >> in it and two IBM Deskstar drives rather than some SCSI goodness. >> > Geez, a Fastrack "FRAID"! > Ye-haw! > Does the mainboard have a Zero-channel SCSI RAID slot? I seriously > doubt it (since no 64-bit PCI or PCI-X), but I have to ask. > No idea, what is that? > Unless it has a Zero-channel SCSI RAID slot, you'd be better off > replacing the card with a $100 3Ware Escalade 7006-2 instead. > I'll mention that to them. -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From b.j.smith at ieee.org Thu Feb 16 06:30:21 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Fasttrack100TX2 vs Adaptec 7896N? In-Reply-To: <43F3DBFF.3050706@mc-kenna.com> References: <20060215233306.25408.qmail@web34112.mail.mud.yahoo.com> <43F3DBFF.3050706@mc-kenna.com> Message-ID: <1140089421.5941.18.camel@bert64.oviedo.smithconcepts.com> On Wed, 2006-02-15 at 20:57 -0500, Damien McKenna wrote: > .. and dual CPU. So could i440BX or i440LX for that matter. So what, you've got a 1.0GBps (64-bit @ 133MHz) FSB that can _only_ use a 0.125GBps (32-bit @ 33MHz) I/O. Can you say ... "bottleneck"? ;-> > It does actually have an extended PCI slot, never seen one before, That's probably the Zero-channel SCSI RAID slot. What is the mainboard make/model and I'll look it up. Alternatively, I've seen some Slot-1 and Slot-2 combo i440GX mainboards where you could use _either_ processor. But that's typically only on a single CPU i440GX. Lastly, it could be some other proprietary riser. > not sure what its for, but all of the on-mobo PCI slots are > covered by the generic 32bit PCI riser card anyway. If I'm working it > out correct, it was a server thrown together from spare parts. > No idea, what is that? It allows you to plug in a board that has a microcontroller with DRAM for hardware RAID using the on-board SCSI channels. Hence the name "Zero-channel" SCSI RAID slot/card. > I'll mention that to them. Again, if you give me a make/model I can find out for sure. With 64-bit PCI and PCI-X, they formalized the Zero-channel option far better so you can use a number of different cards. Those earlier ones were pretty vendor-specific. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From dmckenna at thelimucompany.com Thu Feb 16 09:38:51 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Fasttrack100TX2 vs Adaptec7896N? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D96A@tlc001.tlcusa.thelimucompany.com> The board is a BOXL440GXG revision AA 721242-020, a generic Intel mobo. The riser sites *over* the extended PCI slot, and it's a 2U case its pretty much unusable. It just really feels like they threw together any old cheap bits they had to spare, but I can't fault them too much, they were in a bind at the time. There's a manual for the board at: ftp://download.intel.com/support/motherboards/server/l440gx/254151-003.p df It says that the extended PCI slot thingy is an "Adaptec ARO-1130U2 connector / PCI slot 4", which appears to be the zero-channel slot you mentioned. There's also a thing called the Intelligent Chassis Management Bus (ICMB) header and an Isolated Server Management (ISOL) IMB connector, whatever they are. Lastly, four of the PCI slots are 33MHz while two are 66MHz. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Thu Feb 16 10:31:14 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Adaptec RAIDport III In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D96A@tlc001.tlcusa.thelimucompany.com> Message-ID: <20060216153114.56867.qmail@web34115.mail.mud.yahoo.com> Damien McKenna wrote: > The board is a BOXL440GXG revision AA 721242-020, a generic Intel > mobo. The riser sites *over* the extended PCI slot, and it's a 2U > case its pretty much unusable. It just really feels like they > threw together any old cheap bits they had to spare, but I can't > fault them too much, they were in a bind at the time. Yes, there are a couple of notes about different version of the board at Intel. E.g.: http://www.intel.com/support/motherboards/server/sb/cs-000937.htm > There's a manual for the board at: > ftp://download.intel.com/support/motherboards/server/l440gx/254151-003.pdf I found the products home page here: http://support.intel.com/support/motherboards/server/L440GX/ And this (updated?) manual: ftp://download.intel.com/support/motherboards/server/l440gx/72207707.pdf developer.intel.com is your friend. > It says that the extended PCI slot thingy is an "Adaptec ARO-1130U2 > connector / PCI slot 4", which appears to be the zero-channel slot > you mentioned. Yep, the old Adaptec "RAIDport" option -- "RAIDport III" to be exact: http://www.adaptec.com/worldwide/support/techspecs.jsp?cat=%2FProduct%2FARO-1130U2&prodkey=ARO-1130U2 Drivers are here (NT5.0/2000 drivers should work on NT5.1/XP/2003): http://www.adaptec.com/worldwide/support/drivers_by_product.jsp?cat=%2FProduct%2FARO-1130U2&prodkey=ARO-1130U2 Froogle found a refurbished version for $39: http://www.antonline.com/p_ARO-1130U2-F_187364.htm If you've got spare SCSI drives lying around, you might want to consider it. > There's also a thing called the Intelligent Chassis Management Bus > (ICMB) header and an Isolated Server Management (ISOL) IMB > connector, whatever they are. Management modules are typically their own computers. Probably an Intel i960-base microcontroller and embedded firmware. I haven't read through the manual to check though. > Lastly, four of the PCI slots are 33MHz while two are 66MHz. Yeah, I just noted that. Normally the i440BX/GX provides aAGP 1.0 (x1/x2) on PCI bus 1. Remember, AGP is little more than a glorified, bridged 32-bit PCI bus with DDR signaling (QDR in AGP 3.0???) with special "CPU-like" memory access called Direct in-Memory Execution (D[i]ME). Instead, they decided to break out PCI bus 1 into two (2) 66MHz PCI slots. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From dmckenna at thelimucompany.com Thu Feb 16 11:16:20 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D983@tlc001.tlcusa.thelimucompany.com> Well, after getting our wireless working again yesterday morning when I had the laptop sitting right beside the router, I brought the laptop out to the living room and the signal strength dropped to about 30%. Last night I turned it off and when I turned it on this morning the card was completely unable to make a connection. I didn't have time to fiddle with it so I'm not sure if it's the distance or if the drivers gave up. I was wondering whether y'all thought that getting an 802.11g card might improve things? The current card is only 802.11b and as mentioned has signal problems back to the router, would an 11g card do any better? Thanks. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From jasonb at edseek.com Thu Feb 16 11:18:26 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D983@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D983@tlc001.tlcusa.thelimucompany.com> Message-ID: On Thu, 16 Feb 2006, Damien McKenna wrote: > I was wondering whether y'all thought that getting an 802.11g card might > improve things? The current card is only 802.11b and as mentioned has > signal problems back to the router, would an 11g card do any better? If you happen to get a higher quality card than whatever your 11b is, it's certainly a possibility. It really depends on how awful your existing card and AP solution is as to whether a quality or even marginal 11g solution will beat it out. Good luck! From dmckenna at thelimucompany.com Thu Feb 16 11:43:42 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D988@tlc001.tlcusa.thelimucompany.com> This is the router I have: http://catalog.belkin.com/IWCatProductPage.process?Product_Id=136493 Actually, now that I look at it I need to verify whether it really supports 802.11g or if its 54g. It has the latest BIOS revision so I suspect its compatible... I'll look tonight. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From jasonb at edseek.com Thu Feb 16 11:43:16 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D988@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D988@tlc001.tlcusa.thelimucompany.com> Message-ID: On Thu, 16 Feb 2006, Damien McKenna wrote: > This is the router I have: > http://catalog.belkin.com/IWCatProductPage.process?Product_Id=136493 > Actually, now that I look at it I need to verify whether it really > supports 802.11g or if its 54g. It has the latest BIOS revision so I > suspect its compatible... I'll look tonight. What's 54g? There's 802.11b, 802.11a, and 802.11g at present. Maybe you meant 54Mbps, which is the rated speed for regular 11g. Vendors tend to advertise 'turbo' speeds that are vendor specific of 104Mbps or so. From dmckenna at thelimucompany.com Thu Feb 16 11:56:42 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Web server install - Adaptec RAIDport III Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D98B@tlc001.tlcusa.thelimucompany.com> > > The board is a BOXL440GXG revision AA 721242-020, a generic Intel > > mobo. The riser sites *over* the extended PCI slot, and it's a 2U > > case its pretty much unusable. It just really feels like they > > threw together any old cheap bits they had to spare, but I can't > > fault them too much, they were in a bind at the time. > > Yes, there are a couple of notes about different version of the board > at Intel. E.g.: > http://www.intel.com/support/motherboards/server/sb/cs-000937.htm > I found the products home page here: > http://support.intel.com/support/motherboards/server/L440GX/ I found that too and have printed off a load of documentation on it. It has some potential, as far as five-year-old entry-level mobos go. > developer.intel.com is your friend. Indeed. > Yep, the old Adaptec "RAIDport" option -- "RAIDport III" to be exact: > Froogle found a refurbished version for $39: > http://www.antonline.com/p_ARO-1130U2-F_187364.htm Nice deal, other places have them for $300?!? > If you've got spare SCSI drives lying around, you might want to > consider it. We don't, and as mentioned it's a 2U case so it won't take vertical PCI cards unless they're half-height. > > There's also a thing called the Intelligent Chassis Management Bus > > (ICMB) header and an Isolated Server Management (ISOL) IMB > > connector, whatever they are. > > Management modules are typically their own computers. Probably an > Intel i960-base microcontroller and embedded firmware. I haven't > read through the manual to check though. Not something I'm going to fuss over too much. > > Lastly, four of the PCI slots are 33MHz while two are 66MHz. > > Yeah, I just noted that. Normally the i440BX/GX provides aAGP 1.0 > (x1/x2) on PCI bus 1. Remember, AGP is little more than a glorified, > bridged 32-bit PCI bus with DDR signaling (QDR in AGP 3.0???) with > special "CPU-like" memory access called Direct in-Memory Execution > (D[i]ME). Instead, they decided to break out PCI bus 1 into two (2) > 66MHz PCI slots. Too bad they put it in a 2U case, pretty much makes most of its expansion capability moot. So what would you recommend for a basic 33MHz/32bit PCI IDE RAID card to replace the existing POS? I'd like to do mirroring, we don't have too much data so reliability is my main concern. I'm honestly almost tempted to ask if they'd buy a single IDE 36gb Raptor, then just have a daily backup routine onto our spare server (1U DL320). Whaddya think? -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From dmckenna at thelimucompany.com Thu Feb 16 12:05:56 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D98C@tlc001.tlcusa.thelimucompany.com> > What's 54g? http://www.belkin.com/support/kb/kb.asp?a=2552 -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From jasonb at edseek.com Thu Feb 16 15:57:06 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D98C@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D98C@tlc001.tlcusa.thelimucompany.com> Message-ID: On Thu, 16 Feb 2006, Damien McKenna wrote: > > What's 54g? > > http://www.belkin.com/support/kb/kb.asp?a=2552 Sounds like marketing to me. What's more, if '54g' equipment can't renegotiate itself to lower levels as interference dictates, it's pretty useless. Also, they indicate supposedly compatibility with 11g. If your 11g stuff can do 54Mbps given the local operating conditions it will. From dmckenna at thelimucompany.com Thu Feb 16 16:36:55 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D9CB@tlc001.tlcusa.thelimucompany.com> > > > What's 54g? > > http://www.belkin.com/support/kb/kb.asp?a=2552 > > Sounds like marketing to me. It's the same as all the "pre-N" stuff on the market today, k-flex vs USR's v90 vs ITU's v90, etc - its all the competing precursor "standards" before official ITU / ISO / IEEE standards are ratified. > What's more, if '54g' equipment can't renegotiate itself to lower > levels as interference dictates, it's pretty useless. I think a core problem there are the drivers more than anything. Also I think its running in the 2400MHz range, same as most wireless phones. While we don't have a wireless phone our neighbor does and could be causing some interference there. > Also, they indicate supposedly compatibility with 11g. I saw that, I need to verify which edition my router is. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Thu Feb 16 18:28:27 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Philips 30PW8402/37B 30" flat CRT HDTV for $312.30 shipped Message-ID: <20060216232827.57689.qmail@web34115.mail.mud.yahoo.com> Okay, I _personally_ own one of these, and have since early summer 2005. I would _avoid_ any sub-$600 HDTV CRT in the stores, but this one is definitely a good design -- other than being a refurb (mine's still perfect). I talk about the specs below. http://dealnews.com/deals/Refurbished-Philips-30-HD-Ready-Widescreen-Tube-TV-w-HDMI-for-312-shipped/110456.html - Preface: Why I decided to buy a 2nd 30PW8402/37B today I've been watching for deals on 26", 27", 30" or 32" LCDs for months now, but I haven't found what I want. The closest I came besides a 26" "loss leader" (the LEAPBS fiasco) was a 30" Sceptre for $699 at Cosco.COM (that was over $820 after tax/shipping). I've really been looking for a 1280x768. Not 1280x720 (720p) which wouldn't do 4:3 1024x768 easily, especially since I'd like a VGA/computer in (retro Windows gaming box). And not 1366x768 (16:9 768p) since 1366 is a PITA for VGA let alone I only want 720p _except_ for 4:3 1024x768. I was starting to get impatient and almost just dropped $1,000 the other day. But I'm glad I didn't. I really like this "cheap" CRT HDTV because it's A) Got HDMI (which can do DVI with a $15 cable), B) Got 2 YPrPb inputs and C) does a good 1080i for Xbox/360 usage, as well as _good_ 480p/540p for everything else. It also let's me wait until I find a "good deal" on a 1280x768 LCD for $600 or so. - Specifications of the 30PW8402/37B It does 1080i and 480p/540p (_no_ 720p). It has 1000 lines which is typical of a low-cost, 50+" projection CRT (although some are _only_ a measly 800!), whereas some more "costly" 50+" project CRTs have 1200-1280 (1280 being the same as 720p LCD/DLP models), 1440-1600 (some of the high-end ones) or same new models have the full 1920 lines (let alone new 1080p LCD/DLP models). But it's much, much sharper with better viewing angles than projection CRT. So if you're looking for ultra-sharp 1920x1080i for desktop usage -- it's not quite. But for Xbox 360 games at 1080i (and the few for the old Xbox at 1080i) -- _damn_ straight, it looks _damn_good_! The 8402 model (_unlike_ the 8420) has five (5) sets of inputs -- (1) HDMI (and I've tested with a DVI converter), (2) YPrPb component (I've tested one at 1080i, not sure about the other), (2) s-video/composite. So you can put your HDMI/DVI cable box or computer on it, then both your Xbox or PSx and your DVD player and then a couple of legacy systems. A nice touch is the fact that one YPrPb also has s-video/component so if you have a DVD/VCR combo, it can handle both inputs to the same set. Being a CRT, it's not light. It's 100lbs., and a LCD equivalent will be half that. At the same time, it only uses 150W, which is not much more than an equivalent LCD (and _less_ than a plasma) and it will _not_ "heat up your room." Good design IMHO. The only "negative" is that you'll see bands of white at the top when the picture background is dark. This was an oversight in engineering design. They are caused by the backlight of the buttons at the top, and some of the light "leaks through" and is noticable when the background is black. Otherwise, you'll _never_ notice them. I don't find it's _any_ issue except during the credits of a movie or the flash between commercials. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From damien at mc-kenna.com Thu Feb 16 20:55:46 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Firefox weirdness - in-page cursor? Message-ID: <43F52D22.9070909@mc-kenna.com> You know how normally in Firefox when you scroll up/down with the keys that it moves the entire page as a whole, unlike say a text editor where it just moves the cursor around? Well, my Firefox 1.5.0.1 install on Winders XP has decided to take on the text editor paradigm and its driving me batty. I've no idea how it started doing this, I don't see anything on Google, and its getting quite annoying. Any ideas folks? Thanks. -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From damien at mc-kenna.com Thu Feb 16 21:14:55 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 802.11b vs g - signal distance & strength In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1C1D9CB@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1C1D9CB@tlc001.tlcusa.thelimucompany.com> Message-ID: <43F5319F.3080707@mc-kenna.com> OK, my router's control interface specifically says 11b and 11g, which is good to know. It has an option for an "extended range" feature which is currently enabled and I'm getting a link quality of 40-55% and a signal strength of 55%. So what do you guys think of my chances of boosting the connection if I upgraded to an 11g NIC? Also, have any of you tried wpa_supplicant? (http://hostap.epitest.fi/wpa_supplicant/) Its a n OSS WPA/WPA2 daemon that runs on Linux, BSD and Windows. Looks interesting, I think I'll give it a try myself. -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From dmckenna at thelimucompany.com Fri Feb 17 10:26:34 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Firefox weirdness - in-page cursor? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1D9FC@tlc001.tlcusa.thelimucompany.com> While cleaning my desk at work I found what the problem was, I had turned on caret browsing by hitting F7. Go figure. I knew it'd be something simple. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From b.j.smith at ieee.org Fri Feb 17 11:21:55 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] eVGA GeForce 6800GS PCIe 256MiB DDR for $169 - $15 Rebate Message-ID: <20060217162155.93400.qmail@web34114.mail.mud.yahoo.com> Okay, here's a good $150 deal on an eVGA GeForce 6800GS PCIe for those with PCIe: http://www.newegg.com/Product/Product.asp?Item=N82E16814130267&CMP=BAC-dealmac As you'll note from my recent nVidia GeForce 6/7 series blog: http://www.evga.com/products/moreinfo.asp?pn=256-P2-N386-AX&family=17 This is a 425MHz NV42 "PCIe native" core (unlike only 350MHz for AGP NV40 core) with 12/5 pixel/shader pipelines and 256-bit 1000MHz DDR3 memory. That clearly blows away the 6600 series, even GT, with 33-40% less pipes and 128-bit memory, while saving you only a little. NOTICE: There is someone on NewEgg.COM claiming he got a 350MHz 6800GS, and not a 425MHz version. That should _only_ be the NV40 AGP version, _not_ the NV42 PCIe version. I'm curious if people are putting their reviews on the wrong product, or if this really happened. Although a few others are complaining about DoAs too, and that's typically been atypical for eVGA (although it could be the NV42 itself), so I leave it to you. BTW, I checked the vendor page for the PCIe version (256-P2-N386-AX) and it's supposed to be 425MHz: http://www.evga.com/products/moreinfo.asp?pn=256-P2-N386-AX&family=17 They even have new "overclocked" versions at 450+MHz. Only the AGP version (256-A8-N387-TX) is 350MHz: http://www.evga.com/products/moreinfo.asp?pn=256-A8-N387-TX&family=17 It's just a good deal IMHO, and 6800GS is _now_ the nVidia "price/performance" leader for PCIe without stooping down to a 6200/7300 at $50-75. Virtually _no_reason_ to go 6600GT IMHO, not even on PCIe -- unless there is something wrong with NV42 itself. -- Bryan P.S. It also means I probably won't hesitate to get a 2nd PCIe video card when I upgrade my wife this summer if 6800GS are already $150 now. So I'll probably sell my 1.5 year old GeForce 6800GT AGP currently in use by my wife at that time. Although I'm tempted to use it as a late release gaming rig dedicated to a TV -- maybe just keep it in its existing Athlon XP2600+ cube. So many computers, so many components, so much to over-use and over-stock into systems that are rarely used (and could benefit others far more than my limited 2-4 hours/month gaming budget can afford). -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Fri Feb 17 12:51:56 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] eVGA GeForce 6800GS PCIe 256MiB DDR for $169 - $15 Rebate In-Reply-To: <20060217162155.93400.qmail@web34114.mail.mud.yahoo.com> References: <20060217162155.93400.qmail@web34114.mail.mud.yahoo.com> Message-ID: On Fri, 17 Feb 2006, Bryan J. Smith wrote: > P.S. It also means I probably won't hesitate to get a 2nd PCIe video > card when I upgrade my wife this summer if 6800GS are already $150 > now. So I'll probably sell my 1.5 year old GeForce 6800GT AGP > currently in use by my wife at that time. Although I'm tempted to > use it as a late release gaming rig dedicated to a TV -- maybe just > keep it in its existing Athlon XP2600+ cube. So many computers, so > many components, so much to over-use and over-stock into systems that > are rarely used (and could benefit others far more than my limited > 2-4 hours/month gaming budget can afford). Aha! Caught you saying "when I upgrade my wife this summer". I knew it! You're "wife" is a computer! This explains a lot :-) Well, at least she'll be getting a new GeForce :-) (j/k, obviously) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From dmckenna at thelimucompany.com Fri Feb 17 13:01:36 2006 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] eVGA GeForce 6800GS PCIe 256MiB DDR for $169 - $15Rebate Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1C1DA12@tlc001.tlcusa.thelimucompany.com> > Aha! Caught you saying "when I upgrade my wife this summer". > I knew it! You're "wife" is a computer! This explains a lot :-) He had her name changed from Cray to Betsy so neighbors would stop giving him strange glances we he talked about "the wife, Cray". -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From readg at nfl.jaguars.com Fri Feb 17 14:01:42 2006 From: readg at nfl.jaguars.com (Read, Greg) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Philips 30PW8402/37B 30" flat CRT HDTV for $312.30shipped Message-ID: Thanks for posting this! I was looking at this yesterday and giving it some consideration. I was thinking about a MythTV box and wondering if this monitor would work well with it? Now for some rapid fire questions, I'm not sure where to look for answers for these: If I got this TV monitor, would it be better to pay Comcast more and get HDTV digital cable? OR Get OTA HDTV using a MythTV box and a tuner? Would one of the newer 61x0 MB be good for a MythTV box? http://www.pcstats.com/artvnl.cfm?articleID=1865 http://www.pcstats.com/articleview.cfm?articleID=1900 Can I connect my cheap DVD player (progressive scan) and VCR to this HDTV? TIA, G > -----Original Message----- > From: pc_support-bounces@matrixlist.com [mailto:pc_support- > bounces@matrixlist.com] On Behalf Of Bryan J. Smith > Sent: Thursday, February 16, 2006 6:28 PM > To: This is the PC Support list. > Subject: [Pc_Support] Philips 30PW8402/37B 30" flat CRT HDTV for > $312.30shipped > > Okay, I _personally_ own one of these, and have since early summer > 2005. I would _avoid_ any sub-$600 HDTV CRT in the stores, but this > one is definitely a good design -- other than being a refurb (mine's > still perfect). I talk about the specs below. > > http://dealnews.com/deals/Refurbished-Philips-30-HD-Ready-Widescreen-Tub e- > TV-w-HDMI-for-312-shipped/110456.html > > - Preface: Why I decided to buy a 2nd 30PW8402/37B today > > I've been watching for deals on 26", 27", 30" or 32" LCDs for months > now, but I haven't found what I want. The closest I came besides a > 26" "loss leader" (the LEAPBS fiasco) was a 30" Sceptre for $699 at > Cosco.COM (that was over $820 after tax/shipping). I've really been > looking for a 1280x768. Not 1280x720 (720p) which wouldn't do 4:3 > 1024x768 easily, especially since I'd like a VGA/computer in (retro > Windows gaming box). And not 1366x768 (16:9 768p) since 1366 is a > PITA for VGA let alone I only want 720p _except_ for 4:3 1024x768. > > I was starting to get impatient and almost just dropped $1,000 the > other day. But I'm glad I didn't. I really like this "cheap" CRT > HDTV because it's A) Got HDMI (which can do DVI with a $15 cable), B) > Got 2 YPrPb inputs and C) does a good 1080i for Xbox/360 usage, as > well as _good_ 480p/540p for everything else. It also let's me wait > until I find a "good deal" on a 1280x768 LCD for $600 or so. > > - Specifications of the 30PW8402/37B > > It does 1080i and 480p/540p (_no_ 720p). It has 1000 lines which is > typical of a low-cost, 50+" projection CRT (although some are _only_ > a measly 800!), whereas some more "costly" 50+" project CRTs have > 1200-1280 (1280 being the same as 720p LCD/DLP models), 1440-1600 > (some of the high-end ones) or same new models have the full 1920 > lines (let alone new 1080p LCD/DLP models). But it's much, much > sharper with better viewing angles than projection CRT. So if you're > looking for ultra-sharp 1920x1080i for desktop usage -- it's not > quite. But for Xbox 360 games at 1080i (and the few for the old Xbox > at 1080i) -- _damn_ straight, it looks _damn_good_! > > The 8402 model (_unlike_ the 8420) has five (5) sets of inputs -- (1) > HDMI (and I've tested with a DVI converter), (2) YPrPb component > (I've tested one at 1080i, not sure about the other), (2) > s-video/composite. So you can put your HDMI/DVI cable box or > computer on it, then both your Xbox or PSx and your DVD player and > then a couple of legacy systems. A nice touch is the fact that one > YPrPb also has s-video/component so if you have a DVD/VCR combo, it > can handle both inputs to the same set. > > Being a CRT, it's not light. It's 100lbs., and a LCD equivalent will > be half that. At the same time, it only uses 150W, which is not much > more than an equivalent LCD (and _less_ than a plasma) and it will > _not_ "heat up your room." Good design IMHO. > > The only "negative" is that you'll see bands of white at the top when > the picture background is dark. This was an oversight in engineering > design. They are caused by the backlight of the buttons at the top, > and some of the light "leaks through" and is noticable when the > background is black. Otherwise, you'll _never_ notice them. I don't > find it's _any_ issue except during the credits of a movie or the > flash between commercials. > > > -- > Bryan J. Smith Professional, Technical Annoyance > b.j.smith@ieee.org http://thebs413.blogspot.com > ---------------------------------------------------- > *** Speed doesn't kill, difference in speed does *** > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support From b.j.smith at ieee.org Fri Feb 17 14:44:10 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] RIP/OSPF router software for XP Embedded ... Message-ID: <20060217194410.40695.qmail@web34104.mail.mud.yahoo.com> I've been looking around for RIP/OSPF router services software for XP Embedded. I'm not talk about listening for RIP/OSPF updates for the ntkernel's routing table. I'm talking about an actual service that basically makes XP Embedded a dynamic router. Ideally this would already have HTML/ASP/whatever pages for administration, but I can write those if needbe. Does anyone know of any such router service software for XP/XP Embedded? I'm tempted to search what has been ported with MinGW (if not Cygwin), maybe Zeus, and build my own XPE Component. But if there is something else out there (or someone has already done so with open source), please point it out. All I'm finding is BSD/Linux/UNIX-only daemons. Thanx! -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Fri Feb 17 15:11:19 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Philips 30PW8402/37B 30" flat CRT HDTV for $312.30shipped In-Reply-To: Message-ID: <20060217201119.55747.qmail@web34109.mail.mud.yahoo.com> "Read, Greg" wrote: > Thanks for posting this! For just over $310 (about $330 with tax), it's a great "tide me over for 6-12 months and LCDs come down a bit more" solution. No 720p, twice as heavy as a LCD, but it's a fine solution. I have one and I like it for the price I paid ($430) compared to the lesser quality $400-600 CRTs in the stores. And now I've ordered my 2nd. > I was looking at this yesterday and giving it some consideration. > I was thinking about a MythTV box and wondering if this monitor > would work well with it? Yes, yes, yes and no. Yes for 1080i if you have either DVI out (which you can convert to its HDMI in). Yes for 1080i, if you have component out. Yes for 480p/480i, if you have s-video/composite and don't care about getting 1080i. But no if you don't have any of those. > Now for some rapid fire questions, I'm not sure where to look for > answers for these: If I got this TV monitor, would it be better > to pay Comcast more and get HDTV digital cable? Here's the best recommendations for _now_: 1. If you want an over-the-air HDTV ATSC tuner that doesn't honor the DRM/broadcast flag, go ahead and buy a HDTV3000-whatever for Linux and build a Mythbox. You can always use it in the future for over-the-air. 2. But if you're going to rely on someone else -- cable, satellite, etc... -- _avoid_ buying equipment if you can. There are new standards for cable coming out. Expanding upon #2, I'll tell you what _I'm_ doing. A) PREFACE: I *ALMOST* bought a $700 Hughest HD DirectTiVO when I got the TV about 9 months ago. Damn I'm glad I did _not_. They've not only dropped to $400 last fall, but I don't think they will work with DirecTV after 2006. Whew, glad I did _not_ make that purchase. B) I'm currently admist dropping DirecTV and going back to BrightHouse. It's $10/month for their HD DVR. I will use this solution for 1-2 years. I want to _avoid_ buying equipment right now because ... C) 2-way CableCARD hasn't been adopted yet, and the equipment is still being developed. 1-way CableCARD just lets you decode the stream at the TV or other device, and the FCC now requires it for all cable providers, *BUT* it doesn't let you "control" the feed. I want 2-way CableCARD so I can buy a generic DVR (or possible assemble one) and control my cable feed with 1 box -- period. That's what I'm waiting for (hence why I'm doing "B" for now). BTW, CableCARD is _only_ for terrestrial carriers. I.e., you won't see it for satellite (possibly never due to how the law is written?). > OR Get OTA HDTV using a MythTV box and a tuner? > Would one of the newer 61x0 MB be good for a MythTV box? > http://www.pcstats.com/artvnl.cfm?articleID=1865 > http://www.pcstats.com/articleview.cfm?articleID=1900 Get one with HDTV out if you do. Otherwise, spend a little money ($70) for a low-profile GeForce 7300 PCIe with HDTV out. In all honesty, the GeForce Ti4x00 (NV25/28) series is the _best_ for the Linux HDTV3000 card. It's designed for it's built-in acceleration. > Can I connect my cheap DVD player (progressive scan) and VCR to > this HDTV? Yes. They actually did an _outstanding_job_ for those of us with those dual-DVD/VCR decks (mine's actually a DVD-RAM/HiFi-VCR) where the VCR can_not_ use the component video in (CVI). The inputs on my TV are as follows: AV1/CVI: VCR (svideo-composite) / DVD (component) AV2: TiVO (std. def) AV3: Retro-gaming (3dfx V3 svideo) AV4: HDMI (DVI-to-HDMI when I want to display my cube) AV5: XBox 360 (and formerly Xbox) When I put it on AV1 input, I get either the VCR or DVD, but only 480i (svideo-composite). When I change the "channel" to CVI input, then I get my DVD in 480p (component). Never tried the CVI at 1080i, could only be 480p. AV2 and AV3 are svideo-composite. The svideo definitely looks sharp -- enough that 640x480 and even 800x600 out from the Voodoo3 tuner just on svideo is the best I've ever seen on a TV (although writing letters would get old). AV4 took a few tries, but I got my nVidia drivers in Windows as well as Xorg lines in Linux to do 1920x1080 at 29.97Hz (59.94Hz interlaced). AV5 is Xbox, and it comes in at both 480p and full 1080i. All of my Xbox 360 games come in at beautiful 1080i. You clearly do NOT want to use the 1080i for regular PC desktop usage -- but I can read the Xbox 360 management screens without any issue. -- Bryan P.S. Brian Ashe might have more input on this than I. Brian, my original post is here: http://lists.leap-cf.org/pipermail/pc_support/2006-February/001829.html -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From hescominsoon at emmanuelcomputerconsulting.com Fri Feb 17 18:04:36 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Reliable Opteron Vendor Message-ID: <43F65684.60603@emmanuelcomputerconsulting.com> I am in the market for an Opteron based server. I need a dual cpu(not dual core) with 1 gig of ram per processor, dual 80 gig drives, and 3 years onsite service(next day is fine). My budget is $2500. am i not going to be able to get this at this budget? I don't want to go Intel for this but if the opteron box won't price Intel is my fallback. -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From ozz at ozz.is-a-geek.net Fri Feb 17 20:26:11 2006 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Reliable Opteron Vendor In-Reply-To: <43F65684.60603@emmanuelcomputerconsulting.com> References: <43F65684.60603@emmanuelcomputerconsulting.com> Message-ID: <20060217202611.2eeee231.ozz@ozz.is-a-geek.net> On Fri, 17 Feb 2006 18:04:36 -0500, William Warren wrote: > > I am in the market for an Opteron based server. I need a dual cpu(not > dual core) with 1 gig of ram per processor, dual 80 gig drives, and 3 > years onsite service(next day is fine). My budget is $2500. am i not > going to be able to get this at this budget? I don't want to go Intel > for this but if the opteron box won't price Intel is my fallback. Depends how urgently you need it. I have a Dual-Opteron 246, 2 gigs of RAM, 2x200Gb SATA HDD - I got it back in September, and paid ~$2100 from Monarch Computers. The machine ROCKS, but Monarch's delivery sucks. It took them best part of a month to get it to me. They are also very slow at answering pre-sales questions unless you corner them on the phone. Again, I love their hardware, it is very well put together, but if the rack servers I ordered recently are anything to go by, by the time they arrive I've forgotten what I ordered them for. They have just lost a $25k server order from me because of delivery and pre-sales sluggishness - I needed the boxen too quickly, and ended up paying more money to a different supplier for (in my opinion) inferior machines. Basically, if you know what you want, and don't mind waiting for it, then Monarch are pretty good - great bang for the buck. Otherwise, don't bother. Regards, Ozz. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.matrixlist.com/pipermail/pc_support/attachments/20060217/90b6d97a/attachment.bin From hescominsoon at emmanuelcomputerconsulting.com Fri Feb 17 21:28:49 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Reliable Opteron Vendor In-Reply-To: <20060217202611.2eeee231.ozz@ozz.is-a-geek.net> References: <43F65684.60603@emmanuelcomputerconsulting.com> <20060217202611.2eeee231.ozz@ozz.is-a-geek.net> Message-ID: <43F68661.9000604@emmanuelcomputerconsulting.com> how about onsite technical support? Austin (Ozz) Denyer wrote: > On Fri, 17 Feb 2006 18:04:36 -0500, William Warren > wrote: >> I am in the market for an Opteron based server. I need a dual cpu(not >> dual core) with 1 gig of ram per processor, dual 80 gig drives, and 3 >> years onsite service(next day is fine). My budget is $2500. am i not >> going to be able to get this at this budget? I don't want to go Intel >> for this but if the opteron box won't price Intel is my fallback. > > Depends how urgently you need it. > > I have a Dual-Opteron 246, 2 gigs of RAM, 2x200Gb SATA HDD - I got it > back in September, and paid ~$2100 from Monarch Computers. > > The machine ROCKS, but Monarch's delivery sucks. It took them best > part of a month to get it to me. They are also very slow at answering > pre-sales questions unless you corner them on the phone. > > Again, I love their hardware, it is very well put together, but if the > rack servers I ordered recently are anything to go by, by the time they > arrive I've forgotten what I ordered them for. > > They have just lost a $25k server order from me because of delivery and > pre-sales sluggishness - I needed the boxen too quickly, and ended up > paying more money to a different supplier for (in my opinion) inferior > machines. > > Basically, if you know what you want, and don't mind waiting for it, > then Monarch are pretty good - great bang for the buck. Otherwise, > don't bother. > > Regards, > Ozz. > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From jasonb at edseek.com Fri Feb 17 23:03:42 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] 3Ware 6200, Linux 2.6.15, and AEN errors Message-ID: Any chance it's just a bad dream? I've got the latest firmware on a 6200 series 3Ware card. FE6X 1.02.28.053 ME6X 1.01.00.028 BE6X 1.07.02.005 ME6X 1.01.00.028 After a few hours of running without incident with a 3Ware 7500-4 on a separate 64/33 bus than the 6200 is on, I ended up with these fun, fatal errors: 3w-xxxx: scsi0: Reset sequence failed. 3w-xxxx: tw_chrdev_ioctl(): Reset failed for card 0. 3w-xxxx: scsi0: Character ioctl (0x1f) timed out, resetting card. 3w-xxxx: scsi0: AEN drain failed, retrying. 3w-xxxx: scsi0: AEN drain failed, retrying. 3w-xxxx: scsi0: AEN drain failed, retrying. 3w-xxxx: scsi0: Controller errors, card not responding, check all cabling. 3w-xxxx: scsi0: Reset sequence failed. 3w-xxxx: tw_chrdev_ioctl(): Reset failed for card 0. sd 0:0:0:0: rejecting I/O to offline device EXT3-fs error (device sda5): ext3_find_entry: reading directory #93328 offset 0 ... I'm checking the drive now for errors with MaxBlaster, Maxtor's tool, but I'm coming up clean so far... I fear it's the controller. Anyone have any thoughts? I'd rather replace the Maxtor, as hard drives are cheap and easy, but there's no point in replacing what's not actually broken (likely the hard disk in this case). Thanks. From damien at mc-kenna.com Sat Feb 18 07:37:01 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Reliable Opteron Vendor In-Reply-To: <43F65684.60603@emmanuelcomputerconsulting.com> References: <43F65684.60603@emmanuelcomputerconsulting.com> Message-ID: <43F714ED.80101@mc-kenna.com> I looked at Sun servers this past week and they charge a *lot* for onsite support, but you can get servers from them fairly readily. You might want to take a look at them anyway. -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From hescominsoon at emmanuelcomputerconsulting.com Sat Feb 18 08:05:19 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:41 2006 Subject: [Pc_Support] Reliable Opteron Vendor In-Reply-To: <43F714ED.80101@mc-kenna.com> References: <43F65684.60603@emmanuelcomputerconsulting.com> <43F714ED.80101@mc-kenna.com> Message-ID: <43F71B8F.3030604@emmanuelcomputerconsulting.com> onsite is a requirement as the server is going to be colo'ed on the other side of the country from me. Damien McKenna wrote: > I looked at Sun servers this past week and they charge a *lot* for > onsite support, but you can get servers from them fairly readily. You > might want to take a look at them anyway. > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From b.j.smith at ieee.org Sat Feb 18 10:24:07 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Reliable Opteron Vendor In-Reply-To: <43F65684.60603@emmanuelcomputerconsulting.com> References: <43F65684.60603@emmanuelcomputerconsulting.com> Message-ID: <1140276247.5443.29.camel@bert64.oviedo.smithconcepts.com> On Fri, 2006-02-17 at 18:04 -0500, William Warren wrote: > I am in the market for an Opteron based server. I need a dual cpu(not > dual core) with 1 gig of ram per processor, dual 80 gig drives, and 3 > years onsite service(next day is fine). My budget is $2500. am i not > going to be able to get this at this budget? I don't want to go Intel > for this but if the opteron box won't price Intel is my fallback. It's very cost effective to go with a single Opteron 165/175 dual-core these days, including ECC (Non-Registered) and PCI-X channels for storage/NIC. I don't see why you couldn't get one for $1,000-1,500. If you want dual-Opteron 2xx (single core, maybe dual), then the added costs of Registered ECC memory as well as the board peripherals do start to add up. It really depends on what your storage requirements are. -- Bryan P.S. Everyone seems to be moving to Areca SATA RAID solutions now with the Intel IOP33x XScale superscalar microcontroller. They have both PCI-X and PCIe versions in 4, 8, 12 and 16 port. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From jasonb at edseek.com Sat Feb 18 17:56:51 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3ware corruption on K7S5A In-Reply-To: <20060215170126.58750.qmail@web34112.mail.mud.yahoo.com> References: <20060215170126.58750.qmail@web34112.mail.mud.yahoo.com> Message-ID: <200602181756.51703.jasonb@edseek.com> On Wednesday 15 February 2006 12:01, Bryan J. Smith wrote: > Jason Boxman wrote: > > Months later, I thought I'd Google today for the heck of it and > > found this where someone claims confirmation from 3Ware back in > > 2002 that this is a problem with the K7S5A. > > http://www.geek.com/news/geeknews/2001nov/bch20011129009082.htm > > I stopped using advanced storage and other multi-function/bridged > devices on the SiS735 chipset after I found APIC/I2C comaptibility > issues. E.g., I had issues with PCI-to-CardBus controllers. > > But I still liked the K7S5A as a general desktop. I even ran it with > a 3Ware Escalade 6410 card for awhile. Kinda wondering now if that's > why I lost that old XFS 1.0 version /var filesystem years ago -- the > combination of a 3Ware not flushing a buffer and the XFS bug's > inability to recover such an inconsistent filesystem (fixed in XFS 1.1). Heh. Turns out the 3Ware board in question seems to be dead. However, it fails differently than it did in the K7S5A. What's more, it failed identically to another 6200 board that is known good in a different machine for over a year. So it seems this 6200 was probably good in my K7S5A, but unusable thanks to ECS. Now, it's just plain dead. Sigh. Now I'm accessing my ATA disk over the ServerWorks IIILE directly and it's so slow at the mdma 2 max. ;) /dev/hda: Timing cached reads: 812 MB in 2.00 seconds = 406.00 MB/sec Timing buffered disk reads: 42 MB in 3.07 seconds = 13.68 MB/sec -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From hescominsoon at emmanuelcomputerconsulting.com Sat Feb 18 21:06:32 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Reliable Opteron Vendor In-Reply-To: <43F65684.60603@emmanuelcomputerconsulting.com> References: <43F65684.60603@emmanuelcomputerconsulting.com> Message-ID: <43F7D2A8.8090203@emmanuelcomputerconsulting.com> Brain, There was a server vendor probably last month or before that you said you liked. Who was that again? William Warren wrote: > I am in the market for an Opteron based server. I need a dual cpu(not > dual core) with 1 gig of ram per processor, dual 80 gig drives, and 3 > years onsite service(next day is fine). My budget is $2500. am i not > going to be able to get this at this budget? I don't want to go Intel > for this but if the opteron box won't price Intel is my fallback. > > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From jasonb at edseek.com Sat Feb 18 23:12:42 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port Message-ID: <200602182312.42471.jasonb@edseek.com> Anyone have any thoughts on this? http://tinyurl.com/dtvca "ONE of the ports of the 4 ports is non-operational. The card has THREE working ports. The bad port does not slow down the card, you basically have a 3 port card instead of the 4." I don't even have any SATA drives, but more and more drives available on the cheap after rebates are SATA and I have no SATA 3Ware cards. What's more, I doubt I'll ever be running four drive RAID 1+0, but probably three drive RAID 5 on it with larger SATA drives. Seems hard to close for cheap. But it still has 14 hours left, so it might go for >$100 even with a dead port. Either way, I think I'll bid low. -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Sun Feb 19 00:35:00 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port In-Reply-To: <200602182312.42471.jasonb@edseek.com> References: <200602182312.42471.jasonb@edseek.com> Message-ID: <1140327300.4744.42.camel@bert64.oviedo.smithconcepts.com> On Sat, 2006-02-18 at 23:12 -0500, Jason Boxman wrote: > Anyone have any thoughts on this? > http://tinyurl.com/dtvca > "ONE of the ports of the 4 ports is non-operational. The card has THREE > working ports. The bad port does not slow down the card, you basically have a > 3 port card instead of the 4." > I don't even have any SATA drives, but more and more drives available on the > cheap after rebates are SATA and I have no SATA 3Ware cards. What's more, I > doubt I'll ever be running four drive RAID 1+0, but probably three drive RAID > 5 on it with larger SATA drives. > Seems hard to close for cheap. But it still has 14 hours left, so it might go > for >$100 even with a dead port. Either way, I think I'll bid low. 3Ware cards have a 3-year warranty. I don't think the 9000S series is quite 3 years old yet, but I could be wrong. I'd buy it and see if I could get it replaced under warranty. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From b.j.smith at ieee.org Sun Feb 19 11:29:19 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: MicroATX nForce4 SLI coming to Monarch Computer -- eVGA 131-K8-NF44-AX mainboard In-Reply-To: <20060209123328.63551.qmail@web34105.mail.mud.yahoo.com> References: <20060209123328.63551.qmail@web34105.mail.mud.yahoo.com> Message-ID: <1140366559.4744.50.camel@bert64.oviedo.smithconcepts.com> On Thu, 2006-02-09 at 04:33 -0800, Bryan J. Smith wrote: > FYI, Monarch Computer is selling a MicroATX mainboard with nForce4 > SLI in their new Hornet Pro series of cubes ... > When I find out who the manufacturer is, I'll let you'all know. Okay, the mainboard is the eVGA P/N 131-K8-NF44-AX. There is virtually _no_ information on this mainboard. I've looked and looked and looked. Monarch only seems to let you buy it as part of a Hornet Pro barebones or full system. Not ideal since their Hornet Pros seem to have drastically _underrated_ power supplies (400W max -- not sure it's SLI- rated either), let alone space issues. But at least their bundled price has now dropped to $169.99. I have sent e-mails to both Monarch and eVGA about getting the board directly, and choosing a well designed nVidia SLI-certified power supply like the Seasonic S12 600W (let alone a better enclosure IMHO). -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From jasonb at edseek.com Sun Feb 19 13:56:12 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port In-Reply-To: <1140327300.4744.42.camel@bert64.oviedo.smithconcepts.com> References: <200602182312.42471.jasonb@edseek.com> <1140327300.4744.42.camel@bert64.oviedo.smithconcepts.com> Message-ID: <200602191356.12498.jasonb@edseek.com> On Sunday 19 February 2006 00:35, Bryan J. Smith wrote: > 3Ware cards have a 3-year warranty. I don't think the 9000S series is > quite 3 years old yet, but I could be wrong. I'd buy it and see if I > could get it replaced under warranty. It went for >$100. I figured it would. ;) A guy has to try, right? Good thing I slept through it, though. I find for me Ebay is more about "Screw you, I won't let you have it that cheap!" than actually trying to win auctions. I hate it when people get better deals than I do. I saw a 7xxx go for <$100 by far once that didn't have any brackets on it for mounting, a simple fix. I was quite annoyed I let it go so easily even though I didn't even need it. *shrug* -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Sun Feb 19 15:03:00 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port In-Reply-To: <200602191356.12498.jasonb@edseek.com> References: <200602182312.42471.jasonb@edseek.com> <1140327300.4744.42.camel@bert64.oviedo.smithconcepts.com> <200602191356.12498.jasonb@edseek.com> Message-ID: <1140379380.4744.59.camel@bert64.oviedo.smithconcepts.com> On Sun, 2006-02-19 at 13:56 -0500, Jason Boxman wrote: > It went for >$100. I figured it would. ;) Just won't it was _not_ me. ;-> -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From b.j.smith at ieee.org Sun Feb 19 15:06:44 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port In-Reply-To: <1140379380.4744.59.camel@bert64.oviedo.smithconcepts.com> References: <200602182312.42471.jasonb@edseek.com> <1140327300.4744.42.camel@bert64.oviedo.smithconcepts.com> <200602191356.12498.jasonb@edseek.com> <1140379380.4744.59.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1140379604.4744.61.camel@bert64.oviedo.smithconcepts.com> On Sun, 2006-02-19 at 15:03 -0500, Bryan J. Smith wrote: > Just won't it was _not_ me. ;-> Ugh! Just KNOW it was not me. Duh! (I is a college graduate -- sometimes I be wonder though) -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ Overworked IT Professional #52: Your wife can only reach you via e-mail, but it is filtered out because it says ... "I Love You." From b.j.smith at ieee.org Sun Feb 19 19:54:03 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> Message-ID: <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-02-07 at 09:44 -0800, Bryan J. Smith wrote: > It has a standard 3-prong A/C out, no external AC/DC adapter, which > is a very nice touch for the price. > http://dealnews.com/deals/ABS-8-Port-Gigabit-Ethernet-Switch-for-32-after-rebate/109438.html On Tue, 2006-02-07 at 09:55 -0800, Bryan J. Smith wrote: > There is also a 5-port version, same specs: > http://dealnews.com/deals/ABS-5-Port-Gigabit-Ethernet-Switch-for-12-after-rebate/109024.html More information on these switches: http://thebs413.blogspot.com/2006/02/vitesse-vsc738x-gbe-breaks-5port.html -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ ****** Speed doesn't kill. Difference in speed does! ****** From mflang at bellsouth.net Sun Feb 19 19:57:00 2006 From: mflang at bellsouth.net (Max F Lang) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port In-Reply-To: <1140379604.4744.61.camel@bert64.oviedo.smithconcepts.com> References: <200602182312.42471.jasonb@edseek.com> <1140379380.4744.59.camel@bert64.oviedo.smithconcepts.com> <1140379604.4744.61.camel@bert64.oviedo.smithconcepts.com> Message-ID: <200602191957.00420.mflang@bellsouth.net> On Sunday 19 February 2006 15:06, Bryan J. Smith wrote: > On Sun, 2006-02-19 at 15:03 -0500, Bryan J. Smith wrote: > > Just won't it was _not_ me. ;-> > > Ugh! Just KNOW it was not me. Duh! > (I is a college graduate -- sometimes I be wonder though) Sicks munts ago I cud not spel injuneere... From damien at mc-kenna.com Mon Feb 20 11:32:47 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> Message-ID: <43F9EF2F.6040305@mc-kenna.com> Bryan J. Smith wrote: > More information on these switches: > Lots of good stuff there, Bryan, thanks! -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From octo at logicprobe.org Mon Feb 20 12:06:18 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <43F9EF2F.6040305@mc-kenna.com> References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> Message-ID: Often I find myself looking for what higher-end stuff can be had for good prices on eBay, rather than bargain newer stuff. For example, got myself a GigE switch within the past year. It is an Intel NetStructure 470F (8-port, fiber). I was lucky enough to snag it for $100, though the going price on eBay was $400'ish. (all my existing GigE cards were fiber, and my 100Mbit switches had fiber uplinks, so it offered more flexibility to me) Of course the fiber patch cables themselves are one of those odd items thats super-expensive retail, but super-cheap on eBay. In the 10/100 (1000-uplink-capable) department, I've been very happy with my BayStack 450-24T switches. They're fully managed with all the bells and whistles, and have become really cheap. (I paid $200'ish for my first one, $100'ish for the second, no idea how low they are now) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- On Mon, 20 Feb 2006, Damien McKenna wrote: > Bryan J. Smith wrote: >> More information on these switches: >> > Lots of good stuff there, Bryan, thanks! > > -- > Damien McKenna, husband, father, geek. > damien@mc-kenna.com - http://www.mc-kenna.com/ > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support > From jasonb at edseek.com Mon Feb 20 12:14:37 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> Message-ID: <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> Derek Konigsberg said: > Often I find myself looking for what higher-end stuff can be had for good > prices on eBay, rather than bargain newer stuff. For example, got myself > a GigE switch within the past year. It is an Intel NetStructure 470F > (8-port, fiber). I was lucky enough to snag it for $100, though the going > price on eBay was $400'ish. (all my existing GigE cards were fiber, and > my 100Mbit switches had fiber uplinks, so it offered more flexibility to > me) Of course the fiber patch cables themselves are one of those odd > items thats super-expensive retail, but super-cheap on eBay. > > In the 10/100 (1000-uplink-capable) department, I've been very happy with > my BayStack 450-24T switches. They're fully managed with all the bells > and whistles, and have become really cheap. (I paid $200'ish for my first > one, $100'ish for the second, no idea how low they are now) So, having no GbE yet, would there be increased value in getting Fibre cards and a Fibre switch as opposed to going copper? I suppose I lose out on copper GbE that's on-board on newer chipsets. Every time I try GbE I end up getting crappy NICs and plug them into a poor switch or just into each other and things don't quite work out. I really need to buy some quality NICs for GbE. Of course, my backup server can't push 10.5MB/s over an SSH connection with arcfour encryption -- let alone 3DES or even Blowfish --, so the point is probably moot until I retask systems... From octo at logicprobe.org Mon Feb 20 12:36:53 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> Message-ID: > So, having no GbE yet, would there be increased value in getting Fibre cards > and a Fibre switch as opposed to going copper? I suppose I lose out on > copper GbE that's on-board on newer chipsets. > > Every time I try GbE I end up getting crappy NICs and plug them into a poor > switch or just into each other and things don't quite work out. I really > need to buy some quality NICs for GbE. Of course, my backup server can't > push 10.5MB/s over an SSH connection with arcfour encryption -- let alone > 3DES or even Blowfish --, so the point is probably moot until I retask > systems... The advantage of fiber cards is that they came out first (in any meaningful way). As such, older server-grade cards and switch uplink ports are much more likely to be fiber than copper. So if you want older "high-end" stuff from the likes of eBay, chances are you're dealing with fiber. Today, for newer stuff, copper is much easier to find. The biggest difference I believe is cable length limitations, where fiber always gives you longer length options. The limit on copper gets really short once you go GigE, but it is still much longer than what you're likely to run from the "access switch". On my x86 machines, my GigE fiber NICs are Netgear GA620's. (I managed to get two of these with a patch cable off some random eBay auction a while back) I've been quite happy with them. I've also got a Sun GigaSwift, and the uplink ports of my BS 450-24T 10/100 switches. --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 12:46:26 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question Message-ID: <43FA0072.8070401@emmanuelcomputerconsulting.com> I have a PE SC430 running SBS2k3 Standard. I have dual 160 gig SATA hdd's that i want to run in RAID-1. Which card would you all recommend i track down for this machine? The pci slots are 32-bit 33mhz. i have 1-1x 1-4x and 1-8x pci-e available. -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From jasonb at edseek.com Mon Feb 20 12:56:39 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> Message-ID: <49279.24.73.78.146.1140458199.squirrel@nebula.internal.foo> Derek Konigsberg said: > On my x86 machines, my GigE fiber NICs are Netgear GA620's. (I managed to > get two of these with a patch cable off some random eBay auction a while > back) I've been quite happy with them. I've also got a Sun GigaSwift, > and the uplink ports of my BS 450-24T 10/100 switches. What speeds do you see out of them? Are they crossover'd or are they hooked into a switch? I've heard of the GA620s, but didn't end up buying any last year. Which driver did you use? I might have to look into a pair, maybe. It's somewhat questionable I have any real need for GbE. What do you do with yours? Besides being able to backup the 300GB+ on my fileserver (which rarely changes by more than 2 or 3 GB a day) I can't think of much to do with it. From jasonb at edseek.com Mon Feb 20 12:57:17 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] PSU quality review Message-ID: <49302.24.73.78.146.1140458237.squirrel@nebula.internal.foo> I found this link hidden in a Slashdot article. Looks nice. http://www.pcpro.co.uk/custompc/labs/26/psus/introduction.html :) From jasonb at edseek.com Mon Feb 20 12:59:40 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <43FA0072.8070401@emmanuelcomputerconsulting.com> References: <43FA0072.8070401@emmanuelcomputerconsulting.com> Message-ID: <49437.24.73.78.146.1140458380.squirrel@nebula.internal.foo> William Warren said: > I have a PE SC430 running SBS2k3 Standard. I have dual 160 gig SATA > hdd's that i want to run in RAID-1. Which card would you all recommend > i track down for this machine? > > The pci slots are 32-bit 33mhz. i have 1-1x 1-4x and 1-8x pci-e > available. Good question. Bryan's been looking for/at PCIe based ATA/SATA solutions. I don't believe 3Ware offers one, yet. The performance might be okay in a 32/33 for a 3Ware, but for the cost of a two port 3Ware SATA card you might want find a native PCIe solution, which you probably were anyway. From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 13:14:36 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <49437.24.73.78.146.1140458380.squirrel@nebula.internal.foo> References: <43FA0072.8070401@emmanuelcomputerconsulting.com> <49437.24.73.78.146.1140458380.squirrel@nebula.internal.foo> Message-ID: <43FA070C.5080005@emmanuelcomputerconsulting.com> if not 3ware is there another brand that offers hardware raid one native pci-e? Jason Boxman wrote: > William Warren said: >> I have a PE SC430 running SBS2k3 Standard. I have dual 160 gig SATA >> hdd's that i want to run in RAID-1. Which card would you all recommend >> i track down for this machine? >> >> The pci slots are 32-bit 33mhz. i have 1-1x 1-4x and 1-8x pci-e >> available. > > Good question. Bryan's been looking for/at PCIe based ATA/SATA solutions. > > I don't believe 3Ware offers one, yet. > > The performance might be okay in a 32/33 for a 3Ware, but for the cost of a > two port 3Ware SATA card you might want find a native PCIe solution, which > you probably were anyway. > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From b.j.smith at ieee.org Mon Feb 20 13:26:57 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: Message-ID: <20060220182657.24387.qmail@web34108.mail.mud.yahoo.com> Derek Konigsberg wrote: > Often I find myself looking for what higher-end stuff can be had > for good prices on eBay, rather than bargain newer stuff ... > For example, got myself a GigE switch within the past year. > ... In the 10/100 (1000-uplink-capable) department, I've been > very happy with my BayStack 450-24T switches. Those are only layer-2 managed switches, and aged at that. I typically look for a layer-3 these days or just go dumb. Especially when NetGear's layer-3 switches start at under $500 now -- and sometimes refurbished units can be had for 33% list. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Mon Feb 20 13:30:45 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <49279.24.73.78.146.1140458199.squirrel@nebula.internal.foo> References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> <49279.24.73.78.146.1140458199.squirrel@nebula.internal.foo> Message-ID: > What speeds do you see out of them? Are they crossover'd or are they hooked > into a switch? All my GigE interfaces terminate at that Intel NetStructure 470F that I mentioned, so they are hooked through a switch. Benchmarking can be tricky, because "real world performance" can vary greatly from a "pure network benchmark". And even pure network benchmarks can be affected a lot by the machines in question. For example, I just did some quick netperf runs: Tritainum(GA620) --> Photon(GA620): 159.41Mbps Hyperion(Sun GigaSwift) --> Photon(GA620): 433.30Mbps I'll also mention that the card in Tritainium is plugged into a 32-bit PCI slot (and I'm only using the Gig card there because it was the only good VLAN-capable card I had on-hand at the time). Meanwhile, both Photon and Hyperion have their cards in 64-bit PCI slots. Actual file transfers are much worse than these benchmarks, probably due to plenty of other factors including the fact that I'm using default 1500byte MTUs on all the links. > It's somewhat questionable I have any real need for GbE. What do you do > with yours? Besides being able to backup the 300GB+ on my fileserver (which > rarely changes by more than 2 or 3 GB a day) I can't think of much to do > with it. For me, it was to bring the backbone of my network together. Even if I don't realize full performance, it is still a bit faster than 100Mbit. I also make heavy use of the VLAN features, which are well supported on all my cards. However, most people often don't even fully utilize 100Mbit ethernet. (though I'll easily exceed the speed of 10Mbit ethernet) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From b.j.smith at ieee.org Mon Feb 20 13:30:51 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Copper v. Fibre -- WAS: ABS 8-Port, 5-Port GbE Switch In-Reply-To: <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> Message-ID: <20060220183051.88426.qmail@web34105.mail.mud.yahoo.com> Jason Boxman wrote: > So, having no GbE yet, would there be increased value in getting > Fibre cards and a Fibre switch as opposed to going copper? > I suppose I lose out on copper GbE that's on-board on newer > chipsets. Immunity to EMI at the cable is the big one -- especially in the closet. Anytime you concentrate more than a dozen Cat-5 350MHz lines into a confined area, you're going to generate a lot of EMF. 1000Base-SX Multi Mode NICs/cabling are relatively inexpensive (at least compared to 1000Base-LX Single Mode). > Every time I try GbE I end up getting crappy NICs and plug them > into a poor switch or just into each other and things don't quite > work out. I really need to buy some quality NICs for GbE. It's also probably an issue with your switch too. > Of course, my backup server can't push 10.5MB/s over an SSH > connection with arcfour encryption -- let alone 3DES or even > Blowfish --, so the point is probably moot until I retask > systems... Or just use NFS. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Mon Feb 20 13:33:38 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <20060220182657.24387.qmail@web34108.mail.mud.yahoo.com> References: <20060220182657.24387.qmail@web34108.mail.mud.yahoo.com> Message-ID: > Derek Konigsberg wrote: >> Often I find myself looking for what higher-end stuff can be had >> for good prices on eBay, rather than bargain newer stuff ... >> For example, got myself a GigE switch within the past year. >> ... In the 10/100 (1000-uplink-capable) department, I've been >> very happy with my BayStack 450-24T switches. > > Those are only layer-2 managed switches, and aged at that. > > I typically look for a layer-3 these days or just go dumb. > Especially when NetGear's layer-3 switches start at under $500 now -- > and sometimes refurbished units can be had for 33% list. > Yes, they are only layer-2 managed, and older ones. But, I only need layer-2 switches. Also, they work just perfectly fine for most needs. I do all my layer-3 stuff on other devices, and just run VLANs through the switches. I also tend to prefer managed switches whenever I have the option, so I can do things like VLANs. (and manual port settings, SNMP stuff, etc.) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From b.j.smith at ieee.org Mon Feb 20 13:36:46 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: Message-ID: <20060220183646.57361.qmail@web34104.mail.mud.yahoo.com> Derek Konigsberg wrote: > The biggest difference I believe is cable length limitations, where > fiber always gives you longer length options. Not really. 1000Base-SX Multi Mode fiber is not quite 3x 1000Base-T in distance. You have to go 1000Base-LX Single Mode fiber to get length. > The limit on copper gets really short once you go GigE, Huh? Unless you mean EMI resistance at the cable, 1000Base-T _is_ rated for 100m runs, just like 100Base-T and 10Base-T. There are far greater issues with ASIC overhead and 1500 byte frames. > On my x86 machines, my GigE fiber NICs are Netgear GA620's. I several of those to Intel 510 and Bay 450 switch stacks in the late '90s. They have 512KiB SRAM buffer, although that was rather small for the time (1-2MiB SRAM was typical). No 802.3x support IIRC. Today I wouldn't put anything but a wire-speed layer-3 switch, 20+Mbps fabric and use Jumbo Frames for anything GbE or that is possible of supporting it. Running with old managed layer-2 switches typically offer little benefit over today's commodity switch ASICs. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 13:39:01 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <43FA0072.8070401@emmanuelcomputerconsulting.com> Message-ID: <20060220183901.47238.qmail@web34115.mail.mud.yahoo.com> William Warren wrote: > I have a PE SC430 running SBS2k3 Standard. I have dual 160 gig > SATA hdd's that i want to run in RAID-1. Which card would you all > recommend i track down for this machine? > The pci slots are 32-bit 33mhz. i have 1-1x 1-4x and 1-8x pci-e > available. For just two (2) SATA drives, either use software RAID-1 or a 3Ware Escalade 8006-2 -- although the latter is least ideal with a legacy 32-bit @ 33MHz card. For four or more (4+) SATA drives, consider an Areca ARC-12xx series PCIe x8 card if you have an available PCIe x8 or x16 slot. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 13:42:18 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] PSU quality review In-Reply-To: <49302.24.73.78.146.1140458237.squirrel@nebula.internal.foo> Message-ID: <20060220184218.33085.qmail@web34109.mail.mud.yahoo.com> Jason Boxman wrote: > I found this link hidden in a Slashdot article. > Looks nice. > http://www.pcpro.co.uk/custompc/labs/26/psus/introduction.html Hmmm, kinda mirrors the Tom's Hardware review from about a year or so back. Wonder how original that information is? Or maybe they just found similar results. In any case, the Seasonic S12 series seems to be the most efficient for the price, hands down. I have the 500W (purchased on-sale for $99) which does have dual-SLI PCIe connectors, but is not nVidia rated for SLI. The 600W is nVidia certified for SLI. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 13:43:50 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <43FA070C.5080005@emmanuelcomputerconsulting.com> Message-ID: <20060220184350.95861.qmail@web34113.mail.mud.yahoo.com> William Warren wrote: > if not 3ware is there another brand that offers hardware raid one > native pci-e? Areca ARC-12xx series. They use the Intel IOP332/333 X-Scale processor which has an internal PCI-X to PCIe bridge (i.e., the logic is still actually PCI-X-based, but internal to the ASIC/microcontroller). But there is _no_ 2-port SATA version, _only_ 4+ port. And they aren't cheap, that's for sure. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 13:51:39 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: Message-ID: <20060220185139.51456.qmail@web34102.mail.mud.yahoo.com> Derek Konigsberg wrote: > Actual file transfers are much worse than these benchmarks, > probably due to plenty of other factors including the fact that > I'm using default 1500byte MTUs on all the links. Using a layer-3 switch with wire-speed forwarding of separate subnets/VLANs would let you transparently use 9000 byte jumbo frames. Around 400Mbps "real-world performance" becomes possible with NFS UDP then. > However, most people often don't even fully utilize 100Mbit > ethernet. (though I'll easily exceed the speed of 10Mbit ethernet) And I've seen some trashy, 802.3x-less setups where GbE is _slower_ than 100Mbps. ;-> At least you're using some quality (if 8 year old) hardware -- both concentrator and NIC. Without the SRAM size on the GA620, you'd probably run into performance issues. Derek Konigsberg wrote: > Yes, they are only layer-2 managed, and older ones. But, I only > need layer-2 switches. Also, they work just perfectly fine for > most needs. I do all my layer-3 stuff on other devices, and just > run VLANs through the switches. Not if you want wire-speed performance. I like switches where I can plug in a node and it is in one VLAN/segment if it uses 1500 byte MTUs and another if it uses 9000 byte MTUs and the switch does direct switching. > I also tend to prefer managed switches whenever I have the option, > so I can do things like VLANs. (and manual port settings, SNMP > stuff, etc.) And that's why I use layer-3 switches, taken to another level with wire-speed routing of VLANs/different MTUs. I'll then downlink to a GbE desktop switch for cost. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 13:52:31 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <20060220183901.47238.qmail@web34115.mail.mud.yahoo.com> References: <20060220183901.47238.qmail@web34115.mail.mud.yahoo.com> Message-ID: <43FA0FEF.2060700@emmanuelcomputerconsulting.com> will there be a performance hit using the 8k 3ware on the pci slot? If the performance hit is unacceptable I am getting an error when i try to setup software raid1 on the server which i'll post in another thread if the 3ware option is unacceptable performance wise. Bryan J. Smith wrote: > William Warren wrote: >> I have a PE SC430 running SBS2k3 Standard. I have dual 160 gig >> SATA hdd's that i want to run in RAID-1. Which card would you all >> recommend i track down for this machine? >> The pci slots are 32-bit 33mhz. i have 1-1x 1-4x and 1-8x pci-e >> available. > > For just two (2) SATA drives, either use software RAID-1 or a 3Ware > Escalade 8006-2 -- although the latter is least ideal with a legacy > 32-bit @ 33MHz card. > > For four or more (4+) SATA drives, consider an Areca ARC-12xx series > PCIe x8 card if you have an available PCIe x8 or x16 slot. > > > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From octo at logicprobe.org Mon Feb 20 13:54:18 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <20060220183646.57361.qmail@web34104.mail.mud.yahoo.com> References: <20060220183646.57361.qmail@web34104.mail.mud.yahoo.com> Message-ID: >> On my x86 machines, my GigE fiber NICs are Netgear GA620's. > > I several of those to Intel 510 and Bay 450 switch stacks in the late > '90s. They have 512KiB SRAM buffer, although that was rather small > for the time (1-2MiB SRAM was typical). No 802.3x support IIRC. Yes, they *do* have 802.3x support. At least according to what my Intel 470F switch tells me, and what's written in this PDF of the install guide: http://www.netgearinc.co.jp/support/manual/ga620t/installguide.pdf (page 2, upper-right corner) > Today I wouldn't put anything but a wire-speed layer-3 switch, > 20+Mbps fabric and use Jumbo Frames for anything GbE or that is > possible of supporting it. Running with old managed layer-2 switches > typically offer little benefit over today's commodity switch ASICs. First of all, why would you pay for a layer-3 switch if you have absolutely no intention/desire/need to do routing inside that switch? Second of all, while Jumbo frames do many sense if you're trying to maximize GigE performance, normal-size frames also make sense for interoperability. One setup I could forsee would involve a jumbo-frame VLAN for back-end high speed communications, and a normal-frame VLAN for communications with 10/100Mbps end hosts. With my setup, the only reason I'm really not yet doing jumbo frames is that I'm still mostly 100Mbps. Also, I'm using some of my "in the server" GigE cards more for their VLAN capabilites than raw performance at the moment. (based on what hardware I had at the time when I set things up) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From b.j.smith at ieee.org Mon Feb 20 14:06:37 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <43FA0FEF.2060700@emmanuelcomputerconsulting.com> Message-ID: <20060220190637.69956.qmail@web34104.mail.mud.yahoo.com> William Warren wrote: > will there be a performance hit using the 8k 3ware on the pci slot? Yes. The 3Ware Escalade 8006-2 is a 64-bit @ 66MHz (0.5GBps) capable card. Of course, the maximum you'll ever is 2 x single drive DTR, so forget more than 150MBps. But still, you don't want to be saturating all of your 32-bit @ 33MHz (0.125GBps) PCI I/O. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 14:11:44 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: Message-ID: <20060220191144.60761.qmail@web34115.mail.mud.yahoo.com> Derek Konigsberg wrote: > Yes, they *do* have 802.3x support. At least according to what my > Intel 470F switch tells me ... cut ... Sorry, I mixed my response. I meant the GA620 card. > First of all, why would you pay for a layer-3 switch if you have > absolutely no intention/desire/need to do routing inside that > switch? So I can mix 1500 byte MTU and 9000 byte MTU clients. You can_not_ mix the two on the same layer-2 network. ;-> > Second of all, while Jumbo frames do many sense if you're trying to > maximize GigE performance, normal-size frames also make sense for > interoperability. See previous answer. ;-> Jumbo can make a _serious_ performance boost -- especially as it drastically cuts down on the throttling 802.3x has to do. > One setup I could forsee would involve a jumbo-frame VLAN for > back-end high speed communications, and a normal-frame > VLAN for communications with 10/100Mbps end hosts. But what if you have a half-dozen or more workstations using a few servers for NFS when 300+Mbps performance is _expected_? That's why I like to have a 8-16 port layer-3 backbone for such servers/workstations, then "dumb" desktop switches for regular users (to a GbE uplink on that backbone switch). _Cheaper_ than most layer-2 switch stack setups -- even at refurb/used costs. ;-> > With my setup, the only reason I'm really not yet doing jumbo > frames is that I'm still mostly 100Mbps. Also, I'm using some > of my "in the server" GigE cards more for their VLAN capabilites > than raw performance at the moment. I'm still kicking myself for not buying a GSM7312 refurb when I saw it on sale for a few hundred bucks. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 14:12:41 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <20060220190637.69956.qmail@web34104.mail.mud.yahoo.com> References: <20060220190637.69956.qmail@web34104.mail.mud.yahoo.com> Message-ID: <43FA14A9.7090600@emmanuelcomputerconsulting.com> Wouldn't i be facing pci bus saturation with the 2 160 sata in software Raid-1 ? Bryan J. Smith wrote: > William Warren wrote: >> will there be a performance hit using the 8k 3ware on the pci slot? > > Yes. The 3Ware Escalade 8006-2 is a 64-bit @ 66MHz (0.5GBps) capable > card. Of course, the maximum you'll ever is 2 x single drive DTR, so > forget more than 150MBps. But still, you don't want to be saturating > all of your 32-bit @ 33MHz (0.125GBps) PCI I/O. > > > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From b.j.smith at ieee.org Mon Feb 20 14:14:36 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <43FA14A9.7090600@emmanuelcomputerconsulting.com> Message-ID: <20060220191436.45739.qmail@web34109.mail.mud.yahoo.com> William Warren wrote: > Wouldn't i be facing pci bus saturation with the 2 160 sata in > software Raid-1 ? Not if the SATA channels are connected to a dedicated PCIe x1 channel or the HyperTransport itself. But in all seriousness, the 8006-2 will be fine for 2 drives. You're not going to get much of a performance hit -- just don't put anything else on the 32-bit PCI bus you expect to get performance out of. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From whittake at sbaflorida.com Mon Feb 20 14:19:13 2006 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> Message-ID: <1140463153.28943.17.camel@localhost.localdomain> What in the world do these $100/$200 items do? I noted the 1000 uplink capability but why is that better than a good $10 switch? I am not attempting to be smart, I would truly like to know what they would provide for that much money. I bought some Cisco units, 4 for $25 and I do not see any difference between them and an el-cheapo 8 port unit purchased a BigLots for less than $7.00. In fact two of the Cisco units have crapped out during the past year or so. Homer Whittaker On Mon, 2006-02-20 at 12:06 -0500, Derek Konigsberg wrote: > Often I find myself looking for what higher-end stuff can be had for good > prices on eBay, rather than bargain newer stuff. For example, got myself > a GigE switch within the past year. It is an Intel NetStructure 470F > (8-port, fiber). I was lucky enough to snag it for $100, though the going > price on eBay was $400'ish. (all my existing GigE cards were fiber, and > my 100Mbit switches had fiber uplinks, so it offered more flexibility to > me) Of course the fiber patch cables themselves are one of those odd > items thats super-expensive retail, but super-cheap on eBay. > > In the 10/100 (1000-uplink-capable) department, I've been very happy with > my BayStack 450-24T switches. They're fully managed with all the bells > and whistles, and have become really cheap. (I paid $200'ish for my first > one, $100'ish for the second, no idea how low they are now) > > --------------------------- > Derek Konigsberg > octo@logicprobe.org > http://hecgeek.blogspot.com > --------------------------- > > On Mon, 20 Feb 2006, Damien McKenna wrote: > > > Bryan J. Smith wrote: > >> More information on these switches: > >> > > Lots of good stuff there, Bryan, thanks! > > > > -- > > Damien McKenna, husband, father, geek. > > damien@mc-kenna.com - http://www.mc-kenna.com/ > > > > _______________________________________________ > > Pc_support mailing list > > Pc_support@matrixlist.com > > http://lists.matrixlist.com/mailman/listinfo/pc_support > > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support From octo at logicprobe.org Mon Feb 20 14:24:05 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <20060220191144.60761.qmail@web34115.mail.mud.yahoo.com> References: <20060220191144.60761.qmail@web34115.mail.mud.yahoo.com> Message-ID: > Derek Konigsberg wrote: >> Yes, they *do* have 802.3x support. At least according to what my >> Intel 470F switch tells me ... cut ... > > Sorry, I mixed my response. I meant the GA620 card. Yes, and I responded with info showing that the GA620 had 802.3x, if you read after where you put the "... cut ...". >> First of all, why would you pay for a layer-3 switch if you have >> absolutely no intention/desire/need to do routing inside that >> switch? > > So I can mix 1500 byte MTU and 9000 byte MTU clients. > You can_not_ mix the two on the same layer-2 network. ;-> But with a proper managed layer-2 switch, you can have multiple VLANs which could use different MTUs. The whole point is that you can break the switch into multiple layer-2 networks. I seriously think you're getting your layers mixed up, since none of this requires *any* layer-3 functionality on the switch. (where layer-3 implies something like IP, and this is all in ethernet) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From octo at logicprobe.org Mon Feb 20 14:28:46 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:42 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <1140463153.28943.17.camel@localhost.localdomain> References: <20060207174455.38170.qmail@web34115.mail.mud.yahoo.com> <1140396843.4744.68.camel@bert64.oviedo.smithconcepts.com> <43F9EF2F.6040305@mc-kenna.com> <1140463153.28943.17.camel@localhost.localdomain> Message-ID: > What in the world do these $100/$200 items do? I noted the 1000 uplink > capability but why is that better than a good $10 switch? > I am not attempting to be smart, I would truly like to know what they > would provide for that much money. > I bought some Cisco units, 4 for $25 and I do not see any difference > between them and an el-cheapo 8 port unit purchased a BigLots for less > than $7.00. In fact two of the Cisco units have crapped out during the > past year or so. > Homer Whittaker They do a number of things that an el-cheapo switch can't do: - Mount in a 19" equipment rack - Have many more ports (24-ports in these) - Have a management interface (serial or telnet) - Support SNMP (you can query statistics and info from external software) - Support VLANs and VLAN trunking (multiple logical ethernet networks) - Provide the ability to configure any aspect of switch operation - and I could go on and on In other words, lots of features totally unnecessary if you're "plugging 5 PCs together", but lots of interesting and useful stuff if you're building/managing a larger network (or are just a network geek that likes to tinker with such things on a smaller scale) --------------------------- Derek Konigsberg octo@logicprobe.org http://hecgeek.blogspot.com --------------------------- From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 14:29:55 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <20060220191436.45739.qmail@web34109.mail.mud.yahoo.com> References: <20060220191436.45739.qmail@web34109.mail.mud.yahoo.com> Message-ID: <43FA18B3.1060109@emmanuelcomputerconsulting.com> *wink* this is a dell..Wrong HT.. I am not sure what the sata controllers or hte onboard gig-e are hooked into. Bryan J. Smith wrote: > William Warren wrote: >> Wouldn't i be facing pci bus saturation with the 2 160 sata in >> software Raid-1 ? > > Not if the SATA channels are connected to a dedicated PCIe x1 channel > or the HyperTransport itself. > > But in all seriousness, the 8006-2 will be fine for 2 drives. You're > not going to get much of a performance hit -- just don't put anything > else on the 32-bit PCI bus you expect to get performance out of. > > > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From jasonb at edseek.com Mon Feb 20 14:48:01 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Copper v. Fibre -- WAS: ABS 8-Port, 5-Port GbE Switch In-Reply-To: <20060220183051.88426.qmail@web34105.mail.mud.yahoo.com> References: <47841.24.73.78.146.1140455677.squirrel@nebula.internal.foo> <20060220183051.88426.qmail@web34105.mail.mud.yahoo.com> Message-ID: <54563.24.73.78.146.1140464881.squirrel@nebula.internal.foo> Bryan J. Smith said: > Or just use NFS. For Dirvish backups over SSH? I can't. I could setup NFS on every single machine I run and do a 'local' backup, but that makes little sense too. Plus, you kind of lose out on that security thing unless you use NFSv4 or SFS. From jasonb at edseek.com Mon Feb 20 14:54:30 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <20060220191436.45739.qmail@web34109.mail.mud.yahoo.com> References: <43FA14A9.7090600@emmanuelcomputerconsulting.com> <20060220191436.45739.qmail@web34109.mail.mud.yahoo.com> Message-ID: <54910.24.73.78.146.1140465270.squirrel@nebula.internal.foo> Bryan J. Smith said: > William Warren wrote: >> Wouldn't i be facing pci bus saturation with the 2 160 sata in >> software Raid-1 ? > > Not if the SATA channels are connected to a dedicated PCIe x1 channel > or the HyperTransport itself. > > But in all seriousness, the 8006-2 will be fine for 2 drives. You're > not going to get much of a performance hit -- just don't put anything > else on the 32-bit PCI bus you expect to get performance out of. For software RAID 1 you'd have to send writes down the bus for each drive, right? From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 15:01:54 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <54910.24.73.78.146.1140465270.squirrel@nebula.internal.foo> References: <43FA14A9.7090600@emmanuelcomputerconsulting.com> <20060220191436.45739.qmail@web34109.mail.mud.yahoo.com> <54910.24.73.78.146.1140465270.squirrel@nebula.internal.foo> Message-ID: <43FA2032.2080306@emmanuelcomputerconsulting.com> AFAIK that's correct. Jason Boxman wrote: > Bryan J. Smith said: >> William Warren wrote: >>> Wouldn't i be facing pci bus saturation with the 2 160 sata in >>> software Raid-1 ? >> Not if the SATA channels are connected to a dedicated PCIe x1 channel >> or the HyperTransport itself. >> >> But in all seriousness, the 8006-2 will be fine for 2 drives. You're >> not going to get much of a performance hit -- just don't put anything >> else on the 32-bit PCI bus you expect to get performance out of. > > For software RAID 1 you'd have to send writes down the bus for each drive, > right? > > > > _______________________________________________ > Pc_support mailing list > Pc_support@matrixlist.com > http://lists.matrixlist.com/mailman/listinfo/pc_support > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From b.j.smith at ieee.org Mon Feb 20 17:30:10 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <54910.24.73.78.146.1140465270.squirrel@nebula.internal.foo> Message-ID: <20060220223010.76949.qmail@web34111.mail.mud.yahoo.com> Jason Boxman wrote: > For software RAID 1 you'd have to send writes down the bus for each > drive, right? Correct. It actually should _not_ saturate the bus too bad during writes because of that. But it _will_ saturate the bus during reads because it will interleave them. If you're using software RAID to a SATA PCIe x1 channel, then you should avoid this -- despite the redundant write. The problem is the legacy PCI I/O, not software v. hardware. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From ozz at ozz.is-a-geek.net Mon Feb 20 17:38:21 2006 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] 3Ware 9500s-4LP SATA 1 dead port In-Reply-To: <200602191957.00420.mflang@bellsouth.net> References: <200602182312.42471.jasonb@edseek.com> <1140379380.4744.59.camel@bert64.oviedo.smithconcepts.com> <1140379604.4744.61.camel@bert64.oviedo.smithconcepts.com> <200602191957.00420.mflang@bellsouth.net> Message-ID: <20060220173821.4fe55224.ozz@ozz.is-a-geek.net> On Sun, 19 Feb 2006 19:57:00 -0500, Max F Lang wrote: > > On Sunday 19 February 2006 15:06, Bryan J. Smith wrote: > > On Sun, 2006-02-19 at 15:03 -0500, Bryan J. Smith wrote: > > > Just won't it was _not_ me. ;-> > > > > Ugh! Just KNOW it was not me. Duh! > > (I is a college graduate -- sometimes I be wonder though) > > Sicks munts ago I cud not spel injuneere... Ewe cane knot trussed yore spell chequer two fined awl mistakes. Regards, Ozz. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.matrixlist.com/pipermail/pc_support/attachments/20060220/adc79722/attachment.bin From jasonb at edseek.com Mon Feb 20 17:48:17 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] 3Ware question In-Reply-To: <20060220223010.76949.qmail@web34111.mail.mud.yahoo.com> References: <54910.24.73.78.146.1140465270.squirrel@nebula.internal.foo> <20060220223010.76949.qmail@web34111.mail.mud.yahoo.com> Message-ID: <32905.24.73.78.146.1140475697.squirrel@nebula.internal.foo> Bryan J. Smith said: > Jason Boxman wrote: >> For software RAID 1 you'd have to send writes down the bus for each >> drive, right? > > Correct. > > It actually should _not_ saturate the bus too bad during writes > because of that. But it _will_ saturate the bus during reads because > it will interleave them. > > If you're using software RAID to a SATA PCIe x1 channel, then you > should avoid this -- despite the redundant write. The problem is the > legacy PCI I/O, not software v. hardware. I completely didn't follow. What's the performance penalty for running software RAID 1 over hardware RAID 1? I thought it was purely duplicate data being written to each disk instead of only once to the controller to mirror to each drive itself? It sounds like software RAID 1 reads the same data from both drives for a read, but ends up throwing one of the reads away? I don't follow? From b.j.smith at ieee.org Mon Feb 20 19:41:08 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: Message-ID: <20060221004108.5967.qmail@web34104.mail.mud.yahoo.com> Derek Konigsberg wrote: > Yes, and I responded with info showing that the GA620 had 802.3x, > if you read after where you put the "... cut ...". Hmmm, I guess I had earlier revisions or a different model. Good to know. > But with a proper managed layer-2 switch, you can have multiple > VLANs which could use different MTUs. Setting up VLANs is really no different between layer-2 and layer-3 switches, although more advanced switches can associate VLAN membership by attributes (such as frame size). The _difference_ is in the fact that the VLAN doesn't have to leave to the layer-3 switch to be routed. Once the switch associates > The whole point is that you can break the switch into multiple > layer-2 networks. I seriously think you're getting your layers > mixed up, since none of this requires *any* layer-3 functionality > on the switch. (where layer-3 implies something like IP, and > this is all in ethernet) But how do you communicate between _different_ VLANs? That is the key to everything I'm saying. ;-> -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 19:50:01 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: 3Ware question -- context is everything In-Reply-To: <32905.24.73.78.146.1140475697.squirrel@nebula.internal.foo> Message-ID: <20060221005002.96591.qmail@web34106.mail.mud.yahoo.com> Jason Boxman wrote: > I completely didn't follow. Yes, because you're taking the discussion out of the _original_ context. The original context was: 3Ware on 133MBps shared PCI32@33 v. software on bi-250MBps (500MBps effective) PCIe x1 I said given a 75MBps DTR hard drive, you will get ... For an effective 75MBps write: - 75MBps write using a 133MBps PCI32@33 - 150MBps write using a bi-250MBps PCIe x1 For an effective 150MBps interleaved read: - 150MBps read using a 133MBps PCI32@33 (ugh!) - 150MBps write using a bi-250MBps PCIe x1 So given these facts, using a 3Ware card in a PCI32@33 channel makes _less_sense_ -- especially since you will _saturate_ your PCI bus (at least with modern disks and a good, interleaved set of reads). But _only_ if you can guarantee the SATA channels in the software RAID are using a dedicated PCIe x1 channel. Otherwise, if you use SATA channels on the same PCI32@33, or any PATA channels (most are connected on the legacy PCI32@33), there is no dedicated PCIe x1 with 250MBps in each direction. > What's the performance penalty for running software RAID 1 over > hardware RAID 1? > I thought it was purely duplicate data being written to > each disk instead of only once to the controller to mirror > to each drive itself? Again, these _not_ part of the context. ;-> -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From hescominsoon at emmanuelcomputerconsulting.com Mon Feb 20 19:58:23 2006 From: hescominsoon at emmanuelcomputerconsulting.com (William Warren) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: 3Ware question -- context is everything In-Reply-To: <20060221005002.96591.qmail@web34106.mail.mud.yahoo.com> References: <20060221005002.96591.qmail@web34106.mail.mud.yahoo.com> Message-ID: <43FA65AF.7080504@emmanuelcomputerconsulting.com> so basically I ahve to see about this server's chipset and figure out what hte sata ports are hooked into. If they are pci-e then use software raid..if they are pci go with hardware raid..:) Bryan J. Smith wrote: > Jason Boxman wrote: >> I completely didn't follow. > > Yes, because you're taking the discussion out of the _original_ > context. > > The original context was: > 3Ware on 133MBps shared PCI32@33 v. > software on bi-250MBps (500MBps effective) PCIe x1 > > I said given a 75MBps DTR hard drive, you will get ... > > For an effective 75MBps write: > - 75MBps write using a 133MBps PCI32@33 > - 150MBps write using a bi-250MBps PCIe x1 > > For an effective 150MBps interleaved read: > - 150MBps read using a 133MBps PCI32@33 (ugh!) > - 150MBps write using a bi-250MBps PCIe x1 > > So given these facts, using a 3Ware card in a PCI32@33 channel makes > _less_sense_ -- especially since you will _saturate_ your PCI bus (at > least with modern disks and a good, interleaved set of reads). > > But _only_ if you can guarantee the SATA channels in the software > RAID are using a dedicated PCIe x1 channel. Otherwise, if you use > SATA channels on the same PCI32@33, or any PATA channels (most are > connected on the legacy PCI32@33), there is no dedicated PCIe x1 with > 250MBps in each direction. > >> What's the performance penalty for running software RAID 1 over >> hardware RAID 1? >> I thought it was purely duplicate data being written to >> each disk instead of only once to the controller to mirror >> to each drive itself? > > Again, these _not_ part of the context. ;-> > > > -- My "Foundation" verse: Isa 54:17 No weapon that is formed against thee shall prosper; and every tongue that shall rise against thee in judgment thou shalt condemn. This is the heritage of the servants of the LORD, and their righteousness is of me, saith the LORD. -- carpe ductum -- "Grab the tape" CDTT (Certified Duct Tape Technician) Linux user #322099 Machines: 206822 256638 276825 http://counter.li.org/ From b.j.smith at ieee.org Mon Feb 20 20:00:56 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: 3Ware question -- context is everything In-Reply-To: <43FA65AF.7080504@emmanuelcomputerconsulting.com> Message-ID: <20060221010056.88677.qmail@web34112.mail.mud.yahoo.com> William Warren wrote: > so basically I ahve to see about this server's chipset and figure > out what hte sata ports are hooked into. If they are pci-e then > use software raid..if they are pci go with hardware raid..:) Unless you want to spend $400 on a 4 SATA channel Areca ARC-1210, yes. For 2 drives, this would be overkill. Of course, if you want the piece-of-mind of 3Ware, then get the $125 Escalade 8006-2. A little bus saturation on reads is tolerable, depending. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Mon Feb 20 20:10:59 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <20060221004108.5967.qmail@web34104.mail.mud.yahoo.com> References: <20060221004108.5967.qmail@web34104.mail.mud.yahoo.com> Message-ID: <200602202011.00278.octo@logicprobe.org> On Monday 20 February 2006 7:41 pm, Bryan J. Smith wrote: > > The whole point is that you can break the switch into multiple > > layer-2 networks. I seriously think you're getting your layers > > mixed up, since none of this requires *any* layer-3 functionality > > on the switch. (where layer-3 implies something like IP, and > > this is all in ethernet) > > But how do you communicate between _different_ VLANs? > That is the key to everything I'm saying. ;-> Of course you need a router (or layer-3 switch) for that purpose. What I'm saying is that most of the switches in your network have no need for MLS (multi-layer switching / layer-3 switching / routing / whatever you want to call it). However, such L2-only switches often still have a need for features like VLANs (port-based and trunking) that tend to not exist on dumb switches. In Cisco-speak, we'd only consider L3 switches at the "distribution layer", but we still want VLAN capabilities at the L2-only "access layer". In fact, I think my favorate product that illustrates a demand for such a capability is the 3Com NJ220. It is a 5-port switch (1 is an uplink) with VLAN capabilities, built into a WALL PLATE. The whole idea is that your office can have 4 ethernet ports on 4 different networks, when only one physical cable is running down the wall. -- --------------------------- Derek Konigsberg octo@.logicprobe.org http://hecgeek.blogspot.com --------------------------- From jasonb at edseek.com Mon Feb 20 20:17:51 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: 3Ware question -- context is everything In-Reply-To: <20060221005002.96591.qmail@web34106.mail.mud.yahoo.com> References: <20060221005002.96591.qmail@web34106.mail.mud.yahoo.com> Message-ID: <200602202017.51932.jasonb@edseek.com> On Monday 20 February 2006 19:50, Bryan J. Smith wrote: > Jason Boxman wrote: > > I completely didn't follow. > > Yes, because you're taking the discussion out of the _original_ > context. And I interjected a question in a relevant thread. > The original context was: > 3Ware on 133MBps shared PCI32@33 v. > software on bi-250MBps (500MBps effective) PCIe x1 > > I said given a 75MBps DTR hard drive, you will get ... > > For an effective 75MBps write: > - 75MBps write using a 133MBps PCI32@33 > - 150MBps write using a bi-250MBps PCIe x1 > > For an effective 150MBps interleaved read: > - 150MBps read using a 133MBps PCI32@33 (ugh!) > - 150MBps write using a bi-250MBps PCIe x1 Makes sense. > So given these facts, using a 3Ware card in a PCI32@33 channel makes > _less_sense_ -- especially since you will _saturate_ your PCI bus (at > least with modern disks and a good, interleaved set of reads). > > But _only_ if you can guarantee the SATA channels in the software > RAID are using a dedicated PCIe x1 channel. Otherwise, if you use > SATA channels on the same PCI32@33, or any PATA channels (most are > connected on the legacy PCI32@33), there is no dedicated PCIe x1 with > 250MBps in each direction. If your I/O is such that you rarely will do your DTR per drive under your I/O load, would it matter which solution you deploy? It seems to be an issue of which bottleneck you're actually going to reach. My drives only have DTRs of ~ 50MB/s anyway. -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Mon Feb 20 20:18:03 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Finally! GeForce 7300GS benchmarks! Message-ID: <20060221011803.9979.qmail@web34115.mail.mud.yahoo.com> More follow-up to my recent Blog entry here: http://thebs413.blogspot.com/2006/02/geforce-6-and-7-series-variants-nuts.html High Quality: http://www.anandtech.com/video/showdoc.aspx?i=2699&p=3 Medium Quality: http://www.anandtech.com/video/showdoc.aspx?i=2699&p=4 As most predicted, the 7300GS kicks more butt than the 6200TC or X300, but it's not quite as good as the X1300 (at least in Windows). With HDTV-out commonplace on sub-$75 boards, the 7300GS makes an interesting option if you don't want to spend $100 for a 6600 PCIe or $150 for a 6800GS PCIe -- especially one with HDTV-out. Although it would have been nice to benchmark the difference from the 6600 PCIe (non-GT). The 7600 series offerings will prove interesting to say the least. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From b.j.smith at ieee.org Mon Feb 20 20:32:51 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: ABS 8-Port, 5-Port GbE Switch -- VSC7388, VSC7385 (Blog article) In-Reply-To: <200602202011.00278.octo@logicprobe.org> Message-ID: <20060221013251.26638.qmail@web34103.mail.mud.yahoo.com> Derek Konigsberg wrote: > Of course you need a router (or layer-3 switch) for that purpose. Yes. And instead of putting in a dedicated router that will have serious performance issues at >>10Mbps speeds -- let alone must leave the switch fabric to do such -- I'd rather have wire-speed forwarding in a $300-800 device. Instead of $100 here, $200 there, router here, etc... > What I'm saying is that most of the switches in your network have > no need for MLS (multi-layer switching / layer-3 switching / > routing / whatever you want to call it). Which brings us back to our discussion with regards VLANs based on 1500 byte frames and 9000 byte frames. _Transparent_, [near] _full_wire_speed_ communication between nodes of 1500 byte and 9000 byte frames. ;-> > However, such L2-only switches often still have a need for > features like VLANs (port-based and trunking) that tend to not > exist on dumb switches. I'm talking about: A) 8-24 ports GbE (possibly with 100Mbps ports) layer-3 backbone with 20+Gbps switch fabric B) Servers and performance workstations directly on that layer-3 backbone, largely talking 9000 byte jumbo frames C) Drop 5-8 port dumb switches for a room/cube-concentration for general network usage with standard 1500 byte jumbo frames > In Cisco-speak, Oh boy ... > we'd only consider L3 switches at the "distribution layer", First off, that's _legacy_ '90s Cisco-speak. ;-> Secondly, Cisco's new model is based on VoIP at the desktop, which brings in QoS and layer-3/4 management at the desktop. > but we still want VLAN capabilities at the L2-only "access layer". Again, _legacy_ '90s Cisco-speak. ;-> Cisco's _new_ model is, again, layer-3/4 at the desktop. ;-> But regardless of all that, I'm talking about _desktop_ switches. Something I drop in between 4-7 systems/cubes, but still manage on a _dedicated_ layer-3 switch port. > In fact, I think my favorate product that illustrates a demand for > such a capability is the 3Com NJ220. It is a 5-port switch (1 is > an uplink) with VLAN capabilities, built into a WALL PLATE. The > whole idea is that your office can have 4 ethernet ports on 4 > different networks, when only one physical cable is running down > the wall. But how do you manage the wall-plate? Furthermore, do you _know_ how much those things are?! $30 per 100Mbps/port -- YIKES!!! Listen, I'm _all_ for a _fully_ switched stack, but it is _cost_prohibitive_ in many SMB setups of 50 nodes or so. Not only is layer-3 almost as cheap as layer-2 today, but you do _not_ need managed layer-2 (or 3 for that matter) throughout your network. That will _always_ be a repeat case of _overkill_ for SMB setups of 50 nodes or so. In a SMB of 50 nodes or so, you put layer-3 in your 20+Gbps switched backbone, removing the need for _any_ other layer-3 device. You then connect servers and power workstations directly to that backbone. You can then use localized desktop switches as layout/expansion dictates for regular users. Cost is _sub-$1,000_ for 20+Gbps aggregate performance. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From jasonb at edseek.com Mon Feb 20 20:38:41 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Bus Bottleneck Crash, No Bandwidth Left? Message-ID: <200602202038.41135.jasonb@edseek.com> With all the talk of bus bottlenecking, I'm curious what exactly happens when you hit a bottleneck in the PCI bus, be it legacy PCI or PCIe? Does it simply limit your peak performance or do you actually get worse performance when bottlenecking than if you peaked at 90% or whatever? Should bottlenecking be avoided at all costs, or is it non fatal if you can live with the bus being your limiting factor? A related question: Can a device be sending on the PCI bus while another card receives or is the bus simple utilized or not utilized at any given moment? Put another way, since PCIe is often compared to a layer two switch, is the legacy PCI bus like a full duplex or half duplex HUB or something else? -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Mon Feb 20 20:53:28 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Bus Bottleneck Crash, No Bandwidth Left? In-Reply-To: <200602202038.41135.jasonb@edseek.com> Message-ID: <20060221015328.23403.qmail@web34115.mail.mud.yahoo.com> Jason Boxman wrote: > With all the talk of bus bottlenecking, I'm curious what exactly > happens when you hit a bottleneck in the PCI bus, be it legacy > PCI or PCIe? Have you ever gotten "choppy audio" in the old days of ISA? Or today with cheap PCI audio codecs that need lots of bus access -- especially for 3D spatial? You've experienced bus saturation. Especially when raw block I/O, like disk or (to a lesser extent) network, fights host software-driven DSP operations like audio, communications, etc... > Does it simply limit your peak performance or do you actually > get worse performance when bottlenecking than if you peaked > at 90% or whatever? Depends on your use of the term "peak performance"? If you're thinking of terms of networking, where you have CSMA/CD -- carrier sense, multiple access, collision avoidance -- aka "Ethernet," no. You typically will _not_ get "diminished returns" like you will due to collisions or lack of flow control. The shared PCI peripheral bus is far more "controlled." You have priorities, interrupts, etc... But you can still saturate the bus so both latency and throughput are limited. > Should bottlenecking be avoided at all costs, or is it non fatal > if you can live with the bus being your limiting factor? Depends. You should typically try to segment off disk and, to a lesser extent, network operations from the legacy PCI32@33 bus. PCIe does this with far, far less traces than another PCI bus like PCI64 or PCI-X. > A related question: Can a device be sending on the PCI bus while > another card receives or is the bus simple utilized or not utilized > at any given moment? Depends on what you are asking. No, whatever device controls the PCI bus for its access (such as a Direct Memory Access transfer to system memory), it holds the bus. It _could_ possibly deny other devices access to the bus -- especially during a prolonged burst. Yes, the bus _is_ "there" whether it's being used or not. The maximum data transfer rate (DTR) is _only_ if the bus is being used _continually_ -- not including overhead. Furthermore, if a transfer is not buffered, just because you bridge one bus through a higher speed bus does _not_ mean it's not "wasting" the other busses' time. This is the #1 reason why Opteron kicks the living crap out of Xeon. If you have some low-DTR PCI DMA process going, it can tie up your _entire_ Xeon's MCH to memory. With multiple HyperTransport links and localized memory, and proper processor affinity for I/O support in the OS, AMD will only use the path the transfer takes. Furthermore, the HyperTransport is a _true_ system interconnect -- not a peripheral interconnect, and bursts/buffers as appropriate at the bridges as capable/designed. But no, again, it's not CSMA/CD Ethernet. You're not going to have "collisions" or "overflows" -- at least not in the Ethernet sense. > Put another way, since PCIe is often compared to a layer two > switch, Huh? That's an Intel over-simplification. PCIe allows simple segementation of PCI busses at a very low trace count -- something that bridged PCI64/PCI-X does not. PCIe is actually a full, 4-layer stack. > is the legacy PCI bus like a full duplex or half duplex HUB > or something else? It's a full duplex channel, whereas PCI/PCI-X is half duplex. But unlike Intel's marketing, there _are_ "switched" PCI/PCI-X designs. They're called HyperTransport tunnels/bridges. In fact, HyperTransport PCIe tunnels are more "true switched" PCIe designs than Intel's new MCH/ICH which use PCIe channels as a poor-man's system interconnect. -- Bryan J. Smith Professional, Technical Annoyance b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------- *** Speed doesn't kill, difference in speed does *** From octo at logicprobe.org Mon Feb 20 21:27:15 2006 From: octo at logicprobe.org (Derek Konigsberg) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Bus Bottleneck Crash, No Bandwidth Left? In-Reply-To: <20060221015328.23403.qmail@web34115.mail.mud.yahoo.com> References: <20060221015328.23403.qmail@web34115.mail.mud.yahoo.com> Message-ID: <200602202127.15372.octo@logicprobe.org> > But unlike Intel's marketing, there _are_ "switched" PCI/PCI-X > designs. They're called HyperTransport tunnels/bridges. In fact, > HyperTransport PCIe tunnels are more "true switched" PCIe designs > than Intel's new MCH/ICH which use PCIe channels as a poor-man's > system interconnect. Or, they're called "you didn't buy a PC, and got a machine with multiple PCI busses plugged into a crossbar interconnect" :-) (assuming pre-HyperTransport days, of course) Speaking of this subject area, anyone know if the limited IRQ stupidity has gone away with newer EFI-based machines (such the new MacTel boxes)? (yes, I know about IRQ sharing, but it still feels like an ugly hack, like the BIOS, and most of what has been needed to maintain x86 legacy compatability on modern x86 machines) -- --------------------------- Derek Konigsberg octo@.logicprobe.org http://hecgeek.blogspot.com --------------------------- From b.j.smith at ieee.org Tue Feb 21 06:53:19 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Bus Bottleneck Crash, No Bandwidth Left? In-Reply-To: <200602202127.15372.octo@logicprobe.org> References: <20060221015328.23403.qmail@web34115.mail.mud.yahoo.com> <200602202127.15372.octo@logicprobe.org> Message-ID: <1140522799.4721.19.camel@bert64.oviedo.smithconcepts.com> On Mon, 2006-02-20 at 21:27 -0500, Derek Konigsberg wrote: > Or, they're called "you didn't buy a PC, and got a machine with multiple PCI > busses plugged into a crossbar interconnect" :-) > (assuming pre-HyperTransport days, of course) No, that's Intel marketing. The Memory Controller Hub (MCH) is a hub. Despite Intel's marketing, it's not a switch. Everything is a bridged 32-bit or 64-bit PCI bus from that -- AGP (built into MCH), ESB, ICH, etc... ServerWorks' Intel P3/P4 designs are 2-4 chips. There are 1-3 "Inter Module Bus" (IMB) chips that bridge between each other and then 1 "South Bridge" with a Thin-IMB. Or were you referring to Athlon/MP with a 4-5 node EV6 crossbar? > Speaking of this subject area, anyone know if the limited IRQ stupidity has > gone away with newer EFI-based machines (such the new MacTel boxes)? > (yes, I know about IRQ sharing, but it still feels like an ugly hack, like the > BIOS, and most of what has been needed to maintain x86 legacy compatability > on modern x86 machines) Why would it go away? You _still_ need to emulate a pair of legacy 8259 Interrupt controllers. _How_ you do it is dependent on the architecture. Intel still virtually uses the same, legacy PC/AT logic at its MCH with the Gunning Transistor Logic (GTL) design. Now Intel offers Advanced Programmable Interrupt Controllers (APIC) to deal with more and more system devices, interrupts between processors, etc..., but it still seems like a major hack. AMD _kicked_ the legacy GTL bus _off_ the board with EV6 Athlon/MP. Although this is only an assumption, I believe EV6 is still the interconnect inside of A64/Opteron. In _any_ case, legacy 8259/GTL interrupts are completely virtualized over EV6/HyperTransport, and extremely well tamed -- all while emulating an Intel APIC. Case-in-point: I _never_ had interrupt issues on Athlon or later. I cannot say the same for Intel, not even Socket-603/604 or LGA-775. About the only thing I ever have an issue on AMD platform is when I have 7+ buses like with the nFPro2200+nFPro2050+AMD813x combination. I can run into an install-time BIOS v. Linux PCI enumeration issue, because the OS never expects that many buses on a true Intel platform (because it's impossible). Newer BIOSes or Linux kernel patches the issue though. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ ****** Speed doesn't kill. Difference in speed does! ****** From b.j.smith at ieee.org Tue Feb 21 06:59:10 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: Bus Bottleneck Crash, No Bandwidth Left? -- APIC: Intel (1) v. AMD (virtual) In-Reply-To: <1140522799.4721.19.camel@bert64.oviedo.smithconcepts.com> References: <20060221015328.23403.qmail@web34115.mail.mud.yahoo.com> <200602202127.15372.octo@logicprobe.org> <1140522799.4721.19.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1140523150.4721.24.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-02-21 at 06:53 -0500, Bryan J. Smith wrote: > Why would it go away? > You _still_ need to emulate a pair of legacy 8259 Interrupt controllers. > _How_ you do it is dependent on the architecture. > Intel still virtually uses the same, legacy PC/AT logic at its MCH with > the Gunning Transistor Logic (GTL) design. Now Intel offers Advanced > Programmable Interrupt Controllers (APIC) to deal with more and more > system devices, interrupts between processors, etc..., but it still > seems like a major hack. FYI, Microsoft has a WHDC article on Intel's APIC approach: http://www.microsoft.com/whdc/system/sysperf/apic.mspx In a nutshell, Intel puts in 1 APIC for a system which allows up to 24 IRQs. This has _0_ to do with the BIOS. The BIOS, at best, only allows the mapping of the 2 legacy 8259 interrupt controllers into the legacy 15 IRQs. > AMD _kicked_ the legacy GTL bus _off_ the board with EV6 Athlon/MP. > Although this is only an assumption, I believe EV6 is still the > interconnect inside of A64/Opteron. In _any_ case, legacy 8259/GTL > interrupts are completely virtualized over EV6/HyperTransport, and > extremely well tamed -- all while emulating an Intel APIC. > Case-in-point: I _never_ had interrupt issues on Athlon or later. I > cannot say the same for Intel, not even Socket-603/604 or LGA-775. AMD virtualizes >>32 IRQs. In fact, I think the APIC is per-CPU starting with A64/Opteron, possibly Athlon MP as well. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ ****** Speed doesn't kill. Difference in speed does! ****** From b.j.smith at ieee.org Tue Feb 21 08:51:24 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] I'm leaving the list ... Message-ID: <1140529884.9003.29.camel@bert64.oviedo.smithconcepts.com> For reasons unspecified. I'll reconsider in the future. Thanx for putting up with all my verbosity over all these years. I just tire of any post of mine on things like "dumb" switches turning into a debate that just escalates out-of-context where no one wants to admit to disagree. As I always say, I don't start'em, but I have a nasty habit of not stopping. I'll readily admit that. But the problem is the other guy won't admit he's just as bad as me (let alone should review the chronology of why it went there). -- Bryan P.S. The problem with credentials, education and experience is that someone else always has more. I always keep that in mind. I wish some other people would before quoting things like half-decade out-of-date Cisco design. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ ****** Speed doesn't kill. Difference in speed does! ****** From jasonb at edseek.com Tue Feb 21 10:24:50 2006 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] I'm leaving the list ... In-Reply-To: <1140529884.9003.29.camel@bert64.oviedo.smithconcepts.com> References: <1140529884.9003.29.camel@bert64.oviedo.smithconcepts.com> Message-ID: <200602211024.50709.jasonb@edseek.com> On Tuesday 21 February 2006 08:51, Bryan J. Smith wrote: > For reasons unspecified. > I'll reconsider in the future. > Thanx for putting up with all my verbosity over all these years. Thanks for the information these past years. :) -- Jason Boxman http://edseek.com/ - Linux and FOSS stuff From damien at mc-kenna.com Tue Feb 21 22:02:35 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Microsoft extends VB6 support in Vista for 6-7 years Message-ID: <43FBD44B.2060007@mc-kenna.com> http://www.devsource.com/article2/0,1895,1929552,00.asp Basically Microsoft has caved in to pressure from 3rd parties (i.e. developers) to retain compatibility in Windows Vista for the VB6 runtime. Apparently we'll find out more in a month-ish. -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From damien at mc-kenna.com Tue Feb 21 22:05:53 2006 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] nVidia Geforce 7600 details emerge Message-ID: <43FBD511.7030001@mc-kenna.com> http://www.dailytech.com/article.aspx?newsid=908 Some basic details of the Geforce 7600 (GT) are out: * PCIe native * 560MHz core frequency * 128-bit memory interface * 20.2GB/sec. memory bandwidth * 7B pixels/sec. fill rate * 730M vertices/sec. * 12 pixels per cycle * Built in dual-link DVI support for 2560x1600 resolution -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From b.j.smith at ieee.org Wed Feb 22 01:06:14 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Re: Microsoft extends VB6 support in Vista for 6-7 years In-Reply-To: <43FBD44B.2060007@mc-kenna.com> References: <43FBD44B.2060007@mc-kenna.com> Message-ID: <1140588374.4747.0.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-02-21 at 22:02 -0500, Damien McKenna wrote: > http://www.devsource.com/article2/0,1895,1929552,00.asp > Basically Microsoft has caved in to pressure from 3rd parties (i.e. > developers) to retain compatibility in Windows Vista for the VB6 > runtime. Apparently we'll find out more in a month-ish. Translation: They're still using VB6 internally at Muppet Labs. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------ ****** Speed doesn't kill. Difference in speed does! ****** From wam at HiWAAY.net Thu Feb 23 23:46:46 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue Oct 31 13:17:43 2006 Subject: [Pc_Support] Bytewize .... Message-ID: <43FE8FB6.9030807@HiWAAY.net> .... anyone have any 1st-hand experience, yea or nay, w/ these folks (correctly spelled) ? TIA :-) .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton