From dsimmons at powersmiths.com Tue Aug 2 08:37:28 2005 From: dsimmons at powersmiths.com (David Simmons) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Fastest Hard Drive Access for Large (Video) Files Message-ID: <1122986248.7053.14.camel@suse.something.com> Guys, I have a box that does video work...so most of it's files are very large and I've noticed that most of the time my CPU usage is low, but hard-drive light is flashing like crazy - so, my bottle-neck is just moving this data around. Root Question: How do I create the (within a normal budget) fastest way to move large files around? Considering that my 'work area' could be as small as 10Gb. Thoughts: 1). I know that a RAM drive would probably be the fastest...but given that I'm looking for 10Gb, I don't know of 'normal' motherboards going that high...and I'm sure dedicated RAM-Drive cards are way out of the price-range 2). Have been reading some good results using SATA-300 or SATA-II. But given that I have the old standard PCI bus, will that be the limiting factor? (as I'll have to add a SATA-II card to the system). 3). Any comments on the WD Raptor? 4). Instead of a single drive - do I go RAID0 (striping) - if so, any recommendations on setup? (ie. go ATA/IDE or SATA? (Again, will the PCI bus be the limit?) Thanks in advance, dave From dmckenna at thelimucompany.com Tue Aug 2 08:58:50 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Fastest Hard Drive Access for Large (Video) Files Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1858F37@tlc001.tlcusa.thelimucompany.com> > 3). Any comments on the WD Raptor? Its still the fastest SATA drive on the market, though the SATA-II drives are finally getting close. -- 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 Aug 2 09:10:48 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files In-Reply-To: <1122986248.7053.14.camel@suse.something.com> References: <1122986248.7053.14.camel@suse.something.com> Message-ID: <1122988248.4407.19.camel@bert64.mobile.smithconcepts.com> On Tue, 2005-08-02 at 07:37 -0500, David Simmons wrote: > 2). Have been reading some good results using SATA-300 or SATA-II. But > given that I have the old standard PCI bus, will that be the limiting > factor? (as I'll have to add a SATA-II card to the system). Can you guys posts some links/reference info on this? Not that I question it, I just haven't seen it. Although the speed of the interface should not differ because the drive is the same (and 150MBps is more than 2x of today's drive DTR), SATA-300/II uses a twisted pair cable that might reduce loss and retransmits, therefore, increasing performance. > 3). Any comments on the WD Raptor? Hitachi 10,000rpm Ultra320 SCSI "enterprise" drive gone SATA. Lower platter density, but much faster seek. If you're drive light is constantly on and you hear the drive seeking way too much, definitely this your baby! > 4). Instead of a single drive - do I go RAID0 (striping) - if so, any > recommendations on setup? (ie. go ATA/IDE or SATA? 3Ware Escalade 8506-4LP will definitely kick @$$ for less than $300 to start. Throw in four (4) 73GB Raptors for RAID-10 and your life is great for about a grand total. ;-> > (Again, will the PCI bus be the limit?) Yeah, that's the kicker. Who cares if you have a powerful disk controller if you're putting it in a legacy, shared 32-bit @ 33MHz slot. There is no PCIe x4/8 SATA RAID card yet, only the PCIe x8 LSI Logic MegaRAID 320-2E (2-channel U320). -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------------ DISCLAIMER: Assume all statements above are not factual and made only for any possible entertainment or other non-redeeming value. From dsimmons at powersmiths.com Tue Aug 2 09:08:24 2005 From: dsimmons at powersmiths.com (David Simmons) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files In-Reply-To: <1122988248.4407.19.camel@bert64.mobile.smithconcepts.com> References: <1122986248.7053.14.camel@suse.something.com> <1122988248.4407.19.camel@bert64.mobile.smithconcepts.com> Message-ID: <1122988104.7053.21.camel@suse.something.com> On Tue, 2005-08-02 at 08:10 -0500, Bryan J. Smith wrote: > Can you guys posts some links/reference info on this? Not that I > question it, I just haven't seen it. googling and reading sites like Anandtech had alot of the info I was looking for...basically 300 MB/sec bursts > > 3). Any comments on the WD Raptor? > > Hitachi 10,000rpm Ultra320 SCSI "enterprise" drive gone SATA. Lower > platter density, but much faster seek. If you're drive light is > constantly on and you hear the drive seeking way too much, definitely > this your baby! Ok...so given that my motherboard already has a SATA connection...and (hopefully) it's on some kinda 'dedicated' or internal bus...then this seems to be the best cost/performance contender? (thanks Damien / Bryan) dave From dmckenna at thelimucompany.com Tue Aug 2 09:23:19 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1858F3B@tlc001.tlcusa.thelimucompany.com> > On Tue, 2005-08-02 at 07:37 -0500, David Simmons wrote: > > 2). Have been reading some good results using SATA-300 or > > SATA-II. > > Can you guys posts some links/reference info on this? Not that I > question it, I just haven't seen it. http://www.anandtech.com/storage/showdoc.aspx?i=2454 http://www.anandtech.com/storage/showdoc.aspx?i=2450 -- 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 Aug 2 09:45:07 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files In-Reply-To: <1122988104.7053.21.camel@suse.something.com> References: <1122986248.7053.14.camel@suse.something.com> <1122988248.4407.19.camel@bert64.mobile.smithconcepts.com> <1122988104.7053.21.camel@suse.something.com> Message-ID: <1122990307.4407.24.camel@bert64.mobile.smithconcepts.com> On Tue, 2005-08-02 at 08:08 -0500, David Simmons wrote: > googling and reading sites like Anandtech had alot of the info I was > looking for...basically 300 MB/sec bursts But the drives are no where near capable of that. They are all under 80MBps these days. So a SATA-150 can handle it easily. Also note that 300MBps SATA is now known as SATA-IO or SATA-300, and _no_longer_ SATA-II. Just like USB 2.0, it's become a marketing gimmick and doesn't necessarily mean that performance. > Ok...so given that my motherboard already has a SATA connection...and > (hopefully) it's on some kinda 'dedicated' or internal bus... Many chipsets use one (1) PCIe channel with a dedicated, bi-directional 250MBps bus for the 2-4 SATA channels. > then this seems to be the best cost/performance contender? (thanks > Damien / Bryan) _If_ you need lots of random seek. If it is just raw, linear throughput with sequential reads, then sometimes density is better. It all depends. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------------ DISCLAIMER: Assume all statements above are not factual and made only for any possible entertainment or other non-redeeming value. From b.j.smith at ieee.org Tue Aug 2 09:48:50 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] RE: Fastest Hard Drive Access for Large (Video) Files In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1858F3B@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1858F3B@tlc001.tlcusa.thelimucompany.com> Message-ID: <1122990530.4407.28.camel@bert64.mobile.smithconcepts.com> On Tue, 2005-08-02 at 09:23 -0400, Damien McKenna wrote: > http://www.anandtech.com/storage/showdoc.aspx?i=2454 Note the "multimedia" creation, which is basically a sequential write operation. That's where the lower density of the Raptor is not as good. You want to go with high density commodity disk. But I agree, for most operations with lots of simultaneous tasks and response time, Raptor is best. > http://www.anandtech.com/storage/showdoc.aspx?i=2450 Actually, that article does a good job of saying what I've always said. Drives cannot even break 150MBps right now. Plus the fact that the 300MBps channel is now a gimmick. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ------------------------------------------------------------------ DISCLAIMER: Assume all statements above are not factual and made only for any possible entertainment or other non-redeeming value. From dsimmons at powersmiths.com Tue Aug 2 09:54:29 2005 From: dsimmons at powersmiths.com (David Simmons) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files In-Reply-To: <1122990307.4407.24.camel@bert64.mobile.smithconcepts.com> References: <1122986248.7053.14.camel@suse.something.com> <1122988248.4407.19.camel@bert64.mobile.smithconcepts.com> <1122988104.7053.21.camel@suse.something.com> <1122990307.4407.24.camel@bert64.mobile.smithconcepts.com> Message-ID: <1122990869.7053.47.camel@suse.something.com> On Tue, 2005-08-02 at 08:45 -0500, Bryan J. Smith wrote: > > Ok...so given that my motherboard already has a SATA connection...and > > (hopefully) it's on some kinda 'dedicated' or internal bus... > > Many chipsets use one (1) PCIe channel with a dedicated, bi-directional > 250MBps bus for the 2-4 SATA channels. Maybe on newer motherboards...but again, this is not one of them...just PCI bus with SATA built-on....It's a Gigabyte...hmmm...GA-7N400VPro I think. > _If_ you need lots of random seek. If it is just raw, linear throughput > with sequential reads, then sometimes density is better. It all > depends So...that would mean a RAID-0 setup? dave From b.j.smith at ieee.org Tue Aug 2 11:18:34 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files In-Reply-To: <1122990869.7053.47.camel@suse.something.com> Message-ID: <20050802151834.42330.qmail@web34107.mail.mud.yahoo.com> David Simmons wrote: > Maybe on newer motherboards...but again, this is not one of > them...just PCI bus with SATA built-on....It's a > Gigabyte...hmmm...GA-7N400VPro I think. nForce2. Yeah, it's probably on the PCI bus. IIRC, only the NIC in the nForce2, 3, 4/Pro series are a native HyperTransport peripheral. > So...that would mean a RAID-0 setup? RAID-0 and RAID-3 (including NetCell RAID-XL) are definitely geared towards sequential reads/writes. So in that setup, look to the density of commodity disk. RAID-1 and RAID-10 are great for random reads, and if the files are smaller, RAID-4 is even better. All are fairly good at writes, especially larger writes. RAID-5 is best for lots of random reads and small writes. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Tue Aug 2 11:38:25 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: Fastest Hard Drive Access for Large (Video) Files -- 2200/2050 and HT-1000 mainboards In-Reply-To: <20050802151834.42330.qmail@web34107.mail.mud.yahoo.com> Message-ID: <20050802153825.25372.qmail@web34101.mail.mud.yahoo.com> David Simmons wrote: > Maybe on newer motherboards...but again, this is not one > of them...just PCI bus with SATA built-on....It's a > Gigabyte...hmmm...GA-7N400VPro I think. "Bryan J. Smith" wrote: > nForce2. Yeah, it's probably on the PCI bus. For entry-level workstations and servers, I have been recommending the following mainboard for systems: http://www.foxconnchannel.com/products_motherboard_2.cfm?pName=NFPIK8AA-8EKRS Which runs about $240, and add in $140-200 for an Opteron 140-146), and some for the registered memory (typically ECC, which is good in a production workstation/server). It is a nForce Pro 2200+2050, with (2) PCIe x16 slots, (1) PCIe x4 slot and (2) PCIe x1 slots. It has dual-1000Mbps NICs (HyperTransport peripheral), one on each of the 2200 and 2050, and eight (8) SATA channels, four (4) on each of the 2200 and 2050 (PCIe x1 channel peripheral). This gives you some "raw" SATA bandwidth now (although no NCQ or antyhing), and room to grow as intelligent PCIe SATA RAID cards come about very soon (probably with Broadcom's briding PCIe x8/PCI-X SATA/SAS chip). For now, only the LSI Logic 320-2E (PCIe x8 dual-channel U320 SCSI) is about the only major RAID controller I've seen. NOTE: I have _not_ personally deployed this board. Broadcom's ServerWorks does have an "entry-level" PCI-X chipset for A64/Opteron in the HT-2000 and HT-1000. The HT-1000 on its own would be a cost-effective, entry-level server with a single PCI-X channel (1 slot at 133MHz, 2 slots at 100MHz, 3-5 slots at 66MHz), especially for Opteron 1xx. But so far, I've only seen the HT-2000+HT-1000, which is costly. -- Bryan P.S. There is a $300 dual-Opteron 2xx mainboard out there that does _not_ offer PCIe or PCI-X at all (AMD8151+8111). _Avoid_ it, it has serious I/O bandwidth issues. I've seen people recommend it for workstations/servers and I want to scream. @-OOO -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Wed Aug 3 23:50:44 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: ATT: Bryan Smith - Ram compatibility with i440BX in Toshiba Satellite 4090XCDT In-Reply-To: References: Message-ID: <1123127444.4508.55.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-04 at 03:19 +0000, Dale Barber wrote: > Hi Bryan, > I found a posting by you back in 2002 about the i440BX memory capability. I > hope you don't mind me mailing you directly but I figured you could throw > some light on my problem. > I've also posted this request on the forum at www.wimsbios.com > My Toshiba Satellite 4090XCDT is severely memory challenged even with the > maximum manufacturers specified memory installed, this being 64MB internal + > 128MB sodimm = 192MB total. > Since the i440BX chipset in this notebook is theoretically capable of > handling MUCH more memory ie 4 x 256MB, I've tried installing a number of > different types of 256MB sodimms, including a Micron MT16LSDF3264HG-10E > module which is a 16 chip low density module with 4K refresh, which is > essential according to Intel's i440BX spec sheet, and uses a max of > 128Mb=16Mx8 bit 4 bank chip technology in two module ranks....but still the > laptop won't boot. 64/128MB [SO-]DIMM configurations are all that will work with the 8-bit wide ICs. The i440BX is not capable of 256Mbit technology, hence why the 8-bit wide ICs won't do. You have to use 32 chip x 4-bit wide ICs in a _registered_ DIMM configuration for 256MB. AFAIK, nobody makes those other than DIMMs for servers (and they are typically 36x4-bit with ECC). Which is where the real irony comes in. i440BX and i810/815 use completely _different_ 256MB DIMMs. 32x4-bit (36x4-bit for ECC) for i440BX, 16x8 for i810 or 16x8 or 8x16 for i815. > Can you advise me here? Is the BIOS possibly limiting the maximum memory to > 192MB....if so, do 3rd party patched BIOSes exist to increase this to > 64+256=320MB ? I'm currently running Toshiba's latest BIOS V8.2. > I've consulted the "i440BX RAM Compatibility FAQ" but still I'm stuck at > 192MB ! > Do you know of anyone who has successfully installed more memory? I'd really > appreciate any light you can throw on this issue. > Cheers...Dale -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Thu Aug 4 00:36:15 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: ATT: Bryan Smith - Ram compatibility with i440BX in ToshibaSatellite 4090XCDT In-Reply-To: References: Message-ID: <1123130175.4508.101.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-04 at 04:18 +0000, Dale Barber wrote: > Did you mean "the i440BX is not capable of 256MByte technology"....because > the Micron 256MByte sodimm that I tried uses 128Mbit chips (=16Mx8) and it > has 16 of them, giving 16 x 16M x 8 = 256MB ? So why doesn't this work? Actually, 128Mbit (16MB @ 8-bit) IC -> 256MByte technology. The whole terminology gets a bit ambiguous. The _only_ 16M chips you can use _must_ be 4-bit ICs, resulting in 8MB ICs. In a 32 chip "registered" 128-bit configuration (36 chip for ECC/144-bit), that will be a 256MB Registered DIMM. BTW, just because a chipset supports a particular technology size / IC width doesn't mean it will work in all combinations. You'd have to design a memory controller to understand the exponential increase in transistor count and, ultimately, additional wait state prevents a lot of flexibility. > I've been told that the limitation with the i440BX was its 4K refresh cycle, > and only the 256MByte sodimms configured as 16 x 16M x 8 had this...all > other 256MByte modules require an 8K refresh cycle and hence definitely > won't work! That could be true, I honestly don't know. I just know the IC size, width and chip arrangement configurations that the i440BX supports. The only 256MB modules supported are 32/36-chip with 4-bit wide ICs in a registered configuration. The "main problem" is that most 256MB modules in the i8xx series have IC that are _not_ 4-bit, but 8-bit wide. No i8xx chipset supported 4-bit wide ICs -- leading to a "no man's land" of conflicting 256MB DIMMs. In a nutshell, the more "commodity" i8xx 256MB was far more commonplace than the "registered" i440BX 256MB DIMM, hence why they are hard to find. Especially for notebook SO-DIMMs. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Thu Aug 4 00:42:58 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: ATT: Bryan Smith - Ram compatibility with i440BX in ToshibaSatellite 4090XCDT In-Reply-To: <1123130175.4508.101.camel@bert64.mobile.smithconcepts.com> References: <1123130175.4508.101.camel@bert64.mobile.smithconcepts.com> Message-ID: <1123130578.4508.102.camel@bert64.mobile.smithconcepts.com> On Wed, 2005-08-03 at 23:36 -0500, Bryan J. Smith wrote: > The "main problem" is that most 256MB modules in the i8xx series have IC > that are _not_ 4-bit, but 8-bit wide. No i8xx chipset supported 4-bit > wide ICs -- leading to a "no man's land" of conflicting 256MB DIMMs. In > a nutshell, the more "commodity" i8xx 256MB was far more commonplace > than the "registered" i440BX 256MB DIMM, hence why they are hard to > find. > Especially for notebook SO-DIMMs. And even if you did find one, if your on-board 64MB is not a "registered" DIMM configuration (which I safely assume is not), then you wouldn't be able to use it anyway. I think your quest is an impossible one. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Thu Aug 4 01:39:04 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: ATT: Bryan Smith - Ram compatibility with i440BX inToshibaSatellite 4090XCDT In-Reply-To: References: Message-ID: <1123133944.4508.145.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-04 at 05:16 +0000, Dale Barber wrote: > Hey Bryan, thanks again. > I don't understand why 4 bit wide chips or 8 bit wide chips wide is so > significant to the controller here. Surely by conncecting two 4 bit wide > chips side by side on a sodimm, the controller basically sees an 8 bit wide > memory device? Not when it results in extra tri-state buffers and logic. ;-> > And why is registered memory necessary in this case...I thought it was only > needed in the case of too much electrical load on the data and address buses > when more than 3 sodimms are connected to the controller? The 64MB memory on > my Toshiba motherboard is not registered memory so you suggest that a > registered sodimm wouldn't work with it? There is a bit width total of 128-bit, double the 64-bit DIMM, which requires extra logic on-DIMM. This adds buffer and states which is not ideal for Synchronous timing. > Interestingly, the ASUS L8400 C notebook uses the same i440BX chipset and it > will accept a 256MB 16 chip sodimm, giving a maximum memory of 320MB or 384 > MB, depending on whether there is 64MB or 128MB on the motherboard > respectively. Are you sure it is the i440BX and not the i440GX or possibly the i815? *OR* are you sure it's not a 256MB EDO SO-DIMM? Many early SDRAM chipsets also supported EDO DIMMs at double the size because there is not the synchronous issues in using more ICs/widths. This includes the i440BX. > This leads me to suspect a BIOS limitation! Do you think Toshiba could write > a limit into their BIOS to limit the maximum size of the sodimm to 128MB or > the maximum total memory to 192MB? Have you heard of this being done before > in the BIOS? I know that desktop/server motherboards are a lot more > flexible, but maybe laptop manufacturers don't want to offer this kind of > flexibility in their BIOS! -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Thu Aug 4 09:49:59 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Re: ATT: Bryan Smith - Ram compatibility with i440BXinToshibaSatellite 4090XCDT In-Reply-To: References: Message-ID: <1123163399.4508.163.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-04 at 07:01 +0000, Dale Barber wrote: > Hey Bryan, > here's a link to Crucial Memory's upgrade site for an ASUS L8400 C series > which definitely uses the i440BX chipset: > http://support.crucial.com/store/listparts.asp?Mfr%2BProductline=ASUS%2B+Notebooks&mfr=ASUS&tabid=CR&model=L8400+C+Series&submit=Go > Finally, the Dell Inspiron 2100 also uses the i440BX chipset and here's an > attached specs html which indicates i440BX chipset and max memory of 1 > sodimm of 256MB sdram. > Thanks for your feedback. Hey, I'm glad you found something. It's been a bit since I researched this. Intel might have rev'd the chipset so it supported higher memory capacities. From the looks of it, it looks like a 32Mx16, but I could be wrong. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Thu Aug 4 10:04:56 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] Adaptec 2100-S (was: Re: ATT: Bryan Smith - Ram compatibility with i440BXinToshibaSatellite 4090XCDT) In-Reply-To: <1123163399.4508.163.camel@bert64.mobile.smithconcepts.com> References: <1123163399.4508.163.camel@bert64.mobile.smithconcepts.com> Message-ID: <200508041004.56815.jasonb@edseek.com> On Thursday 04 August 2005 09:49, Bryan J. Smith wrote: > Intel might have rev'd the chipset so it supported higher memory > capacities. From the looks of it, it looks like a 32Mx16, but I could > be wrong. Speaking of other fun chipset compatibilities, I actually bought a 128MB SIMM module that claimed identical specs to a supported 128MB SIMM for an old Adaptec 2100-S, but no joy. The card reported a LED code that wasn't listed in the error manual. The original 32MB SIMM works fine. The recommended SIMM was some OEM'd module that I couldn't find anymore. I guess the specs were the same, but the IC configuration was still different enough that the 2100-S wouldn't play nice. (Needless to say, we are finally getting a new server with an Escalade 9000 series card and 7 74GB WD Raptors.) From dmckenna at thelimucompany.com Thu Aug 4 20:55:50 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] AMD goodies Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D185900A@tlc001.tlcusa.thelimucompany.com> Lots of AMD news. First off, a new lower-cost dual-core Athlon-64 that out-paces everything Intel has at the price-point, retailing for around $350. http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2484 A review of a Tyan's K8WE NF4 Pro SLI dual-CPU motherboard based on the nVidia nForce Professional chipsets and retailing for around $430-460: http://www.hardcoreware.net/reviews/review-296-1.htm An upcoming motherboard from Iwill supports 64 *gigabytes* of RAM on a dual-CPU Opteron board: http://www.digitimes.com/mobos/a20050729PR208.html A review of ATI's new Athlon-64 chipset: http://www.anandtech.com/mb/showdoc.aspx?i=2481 AMD's Opterons CPUs are in use at ILM and LucasArt's new headquarters in San Francisco: http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~100323 ,00.html AMD has launched what seems to be an entry-level Opteron CPU series: http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~100324 ,00.html Fun stuff. -- 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 Aug 5 16:09:52 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] No excuse: LG GSA-4163 is now $42.99 everyday at NewEgg ... Message-ID: <20050805200952.76403.qmail@web34101.mail.mud.yahoo.com> NewEgg.com's standard price on the beige LG GSA-4163BB is now only $42.99. It is the "do everything" drive, including 5x DVD-RAM, 4x DVD+R DL, 16x DVD-R(G) and DVD+R, etc... AnandTech timed it as one of the fastest 16x DVD-R(G) and DVD+R drives, period, with low error rates. http://www.newegg.com/Product/Product.asp?Item=N82E16827136046 There's no excuse not to pay $10 more to put one of these in than a standard CD-RW or DVD-ROM drive. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dmckenna at thelimucompany.com Fri Aug 5 16:12:34 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:10 2006 Subject: [Pc_Support] No excuse: LG GSA-4163 is now $42.99 everyday atNewEgg ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859022@tlc001.tlcusa.thelimucompany.com> > NewEgg.com's standard price on the beige LG GSA-4163BB is now > only $42.99. Wow, nice price! And at work they just got one for me from Sam's Club. -- 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 Aug 5 16:30:51 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] No excuse: LG GSA-4163 is now $42.99 everyday atNewEgg ... In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859022@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050805203051.43513.qmail@web34113.mail.mud.yahoo.com> Damien McKenna wrote: > Wow, nice price! And at work they just got one for me from > Sam's Club. Those are typically DVD+RW based, at least last time I checked. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dmckenna at thelimucompany.com Fri Aug 5 16:41:27 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] No excuse: LG GSA-4163 is now $42.99 everydayatNewEgg ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859023@tlc001.tlcusa.thelimucompany.com> > > And at work they just got one for me from Sam's Club. > > Those are typically DVD+RW based, at least last time I checked. No, no, its the GSA-4163, I've been bugging them for six months to get me one. Of course they paid $80 for it, but that's their fault for waiting so long. -- 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 Aug 5 16:55:06 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] No excuse: LG GSA-4163 is now $42.99 everydayatNewEgg ... In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859023@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050805205506.17814.qmail@web34112.mail.mud.yahoo.com> Damien McKenna wrote: > No, no, its the GSA-4163, I've been bugging them for six > months to get me one. Of course they paid $80 for it, but > that's their fault for waiting so long. Oh, cool! I had only seen 2 different models at Sam's Club, both DVD+RW based. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Sat Aug 6 22:57:31 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] SFF Update: Foxconn NF4K8MC-EKRS, prices drop to $174 for the case+PS+mainboard Message-ID: <1123383451.31913.28.camel@bert64.mobile.smithconcepts.com> A couple of the guys over at SLUUG were drooling over my semi-SFF system when I presented at two SLUUG meetings last month. I carried in my Notebook (bag w/strap over shoulder), 19" LCD (box w/handle in my left hand -- although I could have carried it in my its stable base without its outer box) and SFF box (box w/handle in my right hand) and a wireless KB/mouse box (with a couple of cables in it, between my notebook bag straps) all by myself. Anyhoo, I just wrote a post on the system here per inquiry: http://www.sluug.org/~archive/discuss/200508/msg00038.shtml.gz [ NOTE: Archives username is "discuss" password is "freely" ] I wanted to point out that the 10lbs. Chenming 118 w/300W ATX2.0 PS has dropped to $75 now: http://www.newegg.com/Product/Product.asp?Item=N82E16811125485 And a buck more ($76) if you want the plastic windows (adds a tad more weight, maybe 11-12lbs., at least the Aspire X-QPack's did): http://www.newegg.com/Product/Product.asp?Item=N82E16811125484 Of course there are always the Aspire X-QPacks as well with a 420W, but only ATX1.0, not ATX2.0 with the split-12V. I also wanted to point out Foxconn has updated my mainboard with a "K" which now uses a Broadcom PHY to the nVidia MAC so full GbE is provided (even though it's still the nForce4 standard chipset), and the price has dropped to $79 (a full $10). I don't know how that works, but it should mean the MAC is _still_ directly connected to the HyperTransport and using the forcedeth (or nvnet) driver, and not the legacy PCI bus. Here was my aged board (Foxconn NFS4K8MC-ERS): http://www.foxconnchannel.com/products_motherboard_2.cfm?pName=NF4K8MC-ERS And the newer version (Foxconn NFS4K8MC-EKRS): http://www.foxconnchannel.com/products_motherboard_2.cfm?pName=NF4K8MC-EKRS And it's now $10 cheaper ($79) than when I bought the former in April: http://www.newegg.com/Product/Product.asp?Item=N82E16813186066 Of course, my board is still here if you want to save a buck ($78, $11 off from when I bought it back in April): http://www.newegg.com/Product/Product.asp?Item=N82E16813186052 -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From dmckenna at thelimucompany.com Sun Aug 7 17:25:21 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] White noise from speakers? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859048@tlc001.tlcusa.thelimucompany.com> If I've got my PC speakers turned on, no audio playing but I'm hearing white noise, is that the soundcard or the speakers at fault? -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From pberry2 at cfl.rr.com Sun Aug 7 19:29:26 2005 From: pberry2 at cfl.rr.com (Wise Linux User Patrick) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] White noise from speakers? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859048@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1859048@tlc001.tlcusa.thelimucompany.com> Message-ID: <200508071929.27197.pberry2@cfl.rr.com> On Sunday 07 August 2005 05:25 pm, Damien McKenna wrote: > If I've got my PC speakers turned on, no audio playing but I'm hearing > white noise, is that the soundcard or the speakers at fault? Do you have a TV or video capture card, or, an FM Radio card? I found out that sometimes, when I minimize a window that was playing sounds, they stop, but the amp remains on, making some white noise in the speakers! Check to see if for instance, you have one application using ALSA sound, while another tries to use another sound program! The white noise is usually, in my experience, from some program that I opened, and then shut down, but, didn't correctly stop the sound server that had been used for that program. -- Check these out: http://knopper.net/knoppix http://livecdlist.com http://distrowatch.com http://yolinux.com http://safeharbordome.com From b.j.smith at ieee.org Sun Aug 7 21:20:51 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] White noise from speakers? In-Reply-To: <200508071929.27197.pberry2@cfl.rr.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1859048@tlc001.tlcusa.thelimucompany.com> <200508071929.27197.pberry2@cfl.rr.com> Message-ID: <1123464051.4575.14.camel@bert64.mobile.smithconcepts.com> On Sun, 2005-08-07 at 19:29 -0400, Wise Linux User Patrick wrote: > Do you have a TV or video capture card, or, an FM Radio card? > I found out that sometimes, when I minimize a window that was playing sounds, > they stop, but the amp remains on, making some white noise in the speakers! It's up to the application to turn on/off the line-in/aux/whatever input you configure it to use for incoming audio. I know most of the Windows software out there is really bad about doing this (even when they offer an option to set the input audio interface, it doesn't always close -- let alone it doesn't do it when minimized). -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Mon Aug 8 01:42:46 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] SFF Update: Foxconn NF4K8MC-EKRS, prices drop to $174 for the case+PS+mainboard In-Reply-To: <1123383451.31913.28.camel@bert64.mobile.smithconcepts.com> References: <1123383451.31913.28.camel@bert64.mobile.smithconcepts.com> Message-ID: <1455.216.77.197.48.1123479766.squirrel@localhost> Bryan J. Smith said: > I also wanted to point out Foxconn has updated my mainboard with a "K" > which now uses a Broadcom PHY to the nVidia MAC so full GbE is provided > (even though it's still the nForce4 standard chipset), and the price has > dropped to $79 (a full $10). I don't know how that works, but it should > mean the MAC is _still_ directly connected to the HyperTransport and > using the forcedeth (or nvnet) driver, and not the legacy PCI bus. Okay, that's making me second think Ebaying an older nForce2 Ultra for $40-$60 and buying one of these instead. Of course, I'll need the new PSU and a compatible CPU, too. I'd be happy with a cheaper Semapron (too late to look up spelling, sigh) CPU. From b.j.smith at ieee.org Mon Aug 8 18:35:10 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] BS Detector Off-the-Scale! Micrsoft says Monad was cut for security ... Message-ID: <20050808223510.71657.qmail@web34113.mail.mud.yahoo.com> Given the fact that Monad was going to be a _true_ .NET environment with actual UNIX-like security, this rates 10 on the "BS Scale" for retroactive PR non-sense. The sooner something like Monad is available, the sooner various, _standard_ Windows services get a more secure environment for processing -- from Exchange's ESMTP to Windows scripting. It's really a complement to Indigo, which is the .NET sandboxed (atop of Win32) run-time of Longhorn Server, and more par to what Java already has on anything (Win32, as well as UNIX). Right now God knows that you've got all sorts of _core_ scripting in Win32 which was written for Chicago and _by_passes_ even Win32's limited security model. And a lot of that is already tied into the MSIE client. So, pardon my French, this is utter and total BS and it's PR non-sense that I can't stand. Story: http://www.pcworld.com/news/article/0,aid,122145,00.asp -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From pberry2 at cfl.rr.com Mon Aug 8 19:39:58 2005 From: pberry2 at cfl.rr.com (Wise Linux User Patrick) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] VISTA is an Acronym! Message-ID: <200508081939.59144.pberry2@cfl.rr.com> "Cringefan Ken M. points out that VISTA is really an acronym for: Viruses, Infections, Spyware, Trojans, and Adware. That's clarity for you." Consumer Reports has a very full magazine out, this Sept. 2005 issue, (got ours today in mail) covering several pages about adware, virus, spybots, trojans, preventative measures, and an entire article on internet access providers, plus computer comparisons (purchasing of desktops, or laptops). The cost per incident is a bit inflated, to my mind, at $265 per "spyware", $312 per "virus", and $395 per 'phishing' incident. There was one comparison mention of MAC OSX. I have heard that the cleaning of a windows computer, in a shop, runs over $75, now, in the Orlando area. Wow! Other than that, there were NO MENTIONS of the OTHER 450+ virus free, fairly secure, NON-Microsoft OSes. Well, at least the 20 % of the world who don't use the vulnerable OS are knowledgeable of the flaws mentioned in the articles. And, I suspect that we run stand-alone firewalls, and, GNU/Linux as the OS. Little to No mention of firewalls in stand alone routers, in the wifi or Ethernet works. Very scarey, how the users, having purchased a computer with an installed proprietary OS from the world's biggest multiple convicted felon, simply wanting to check email, or surf for bargains, are so greatly deceived! -- Check these out: http://knopper.net/knoppix http://livecdlist.com http://distrowatch.com http://yolinux.com http://safeharbordome.com From b.j.smith at ieee.org Mon Aug 8 23:26:02 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] nVidia launches SLI X16 for AMD, Intel, Dell marks first non-Intel move ... Message-ID: <1123557962.4675.27.camel@bert64.mobile.smithconcepts.com> If you haven't heard, nVidia is launching a new chipset, the nForce4 SLI X16. What is it? It's basically 2 IC version of their nForce4, the SPP being a full 8GBps HyperTransport tunnel to the MCP. Despite the references to legacy chipset that typically only had a 1GBps connection, the MCP is basically a full nForce4 series core and peripherals as well as the SPP (sans peripherals). It's going to be available in both an AMD HyperTransport and Intel AGTL+ flavor. The Intel AGTL+ SPP will have a dual-channel DDR2 interface. Even more shocking is that Dell is going to split on going with only Intel solutions, and start carrying the AGTL+ solution. I'm still scratching my head on that one, unless Dell feels that they need more of a "gamers solution" in small quantities than Intel will provide (and won't mind them carrying). http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2493 -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Mon Aug 8 23:55:45 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] SilverStone SG01 adds its MicroATX/ATX-PS combo SFF ... Message-ID: <1123559745.4675.39.camel@bert64.mobile.smithconcepts.com> I think mainboard manufacturers need to start paying attention. If my Chenming 118 and Aspire X-QPack purchases say anything, consumers like the size and portability of these new breed of small form-factor (SFF) cases using _standard_ MicroATX mainboards and ATX power supplies. In a desktop, most people just need four (4) mainboard slots with today's on-board peripherals, while only two (2) 5.25" bays are enough with (2) 3.5" internal bays for hard drives (JBOD, RAID-0 or RAID-1). But at the same time, with today's CPUs and video cards, they still need a full ATX power supply (as most MicroATX PSes that can handle the load are _rare_ -- my 460W MicroATX2.0 was clearly a "long" MicroATX PS ;-). SilverStone has added their own solution in the SG01. It is about 1" x 1" shorter and narrower than the Chenming/Aspire, while being about 1" longer. What does this mean versus the Chenming/Aspire? Positives / Nice Touches: 1. The 5.25" bays have more room for cabling/length (1" longer) 2. The 5.25" bays are offset to the right, more room/easier access on the card slot side 3. The internal 3.25" bays for hard drives are sideways mounted and easily accessed 4. The internal 3.25" bays have their own fan blowing right over Negatives: 1. No external 1" floppy (1" shorter) 2. The ATX PS is above the CPU, not the slots (1" shorter), so CPU fansink height is more limited 3. The fan is above the slots and only 80mm, instead of 120mm 4. No handle (at least I couldn't see one / didn't see a spec) 5. Cost: $190 without a CPU! Nothing I'd buy for the price, but I figured it's a sign of more to come from other vendors. If the SilverStone TJ-06 was any indicator, SilverStone licenses their designs from others, so clones will follow (at a cheaper price point): http://www.tomshardware.com/howto/20050720/index.html -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 9 00:00:51 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] AMD releases 64-bit Sempron 3300+/3400+ for Socket-754 ... Message-ID: <1123560051.4675.43.camel@bert64.mobile.smithconcepts.com> I guess I missed this: http://www.tomshardware.com/cpu/20050801/index.html Performance wasn't too bad at the $125 Sempron 3300+ versus AMD's Socket-939 Athlon64 solutions. At this point, with the nForce4 Standard/Ultra Socket-939 mainboards so cheap, PCIe clearly the future and the Athlon64 3000+ around the same price, I can't recommend the Sempron unless you already have a Socket-754 mainboard and/or AGP video card you are satisfied with. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 9 00:08:47 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Be wary on the Hitachi DeskStar 7K500 (500GB) drives = 5 platters! Message-ID: <1123560528.4675.48.camel@bert64.mobile.smithconcepts.com> They use 100GB platters so they have 5 platters! Most drives today are only 1-3 sometimes 4 platters, largely for reliability in a commodity disk drive capacity. The last time a company put 5 platters in a capacity disk drive capacity was IBM's DeskStar 75GXP (15GB/platter). Most people with the 30-45GB platters had few failure rates, but those with the 60-75GB, especially 75GB, had high failure rates. Now HD cooling was little exposed back then because 7200rpm wasn't commonplace. And the new materials and fabrication processes in use by Hitachi and Seagate are now rated for 60 degrees C operation instead of the former 40 degrees C. But the vibration and other operational tolerances are still limited, so be wary of how you deploy these drives. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 9 01:35:48 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] The "Small Enough" Form-Factor PC ... Message-ID: <1123565748.4675.51.camel@bert64.mobile.smithconcepts.com> Updated my blog with a general ramble on what I believe should be the new, commodity "box" enclosure using MicroATX mainboards and full ATX power supplies (typically 10-11" x 8-9" x 14-15" dimensions) since it addresses about 95%+ of any expansion, cooling and power needs these days. http://thebs413.blogspot.com/2005/08/small-enough-form-factor-pc.html -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Wed Aug 10 12:03:58 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Backing up 1-4TiB: LTO-3, SDLT or SAIT? Message-ID: <20050810160358.9328.qmail@web34110.mail.mud.yahoo.com> I have 1TiB of potential storage that may quickly increase to up to 4TiB in the future. The system is a standalone system of systems (largely Windows 2000 and HP/UX 11), so it requires its own, local backup solution. In the past, the company standard has been DLT for production areas, but LTO has been increasingly deployed. I think it's more of a vendor decision (HP) than anything (although I have already interjected that we might want to start looking at Copan System's MAID/VTL as an intermedia, near-line storage buffer between systems and tape). If I was rolling out tape for a production area, I'd definitely stick with DLT. But in this case, I have a leading-edge, standalone system, and I really could use a leading-edge performing backup solution. As such, I'm really looking to LTO, namely LTO-3. LTO-3 reads LTO-1 and reads/writes LTO-2 and LTO-3. LTO-4 is planned. It is 400GB native with a native transfer rate of 80MBps (double each on average using hardware compression), pretty much the leader right now. Tape cartridge media is one of the cheapest and reliability is typical for the industry (to get any higher, you'd need to go with an IBM proprietary). LTO is a consortium of 3 companies, Certance, HP and IBM, although that's no guarantee of full compatibility. But the multivendor nature does explain popularity. About 2/3rd of all new enterprise drive/media sales are LTO according to a 2003 study by SAIC, one of our core partners. The other options are largely HP SDLT and Sony SAIT. As I mentioned, we have DLT in use in production areas, but the cost, capacity and performance are not what I'd like. Since this is a standalone system, and we have deployed LTO-1/2 in other areas, I'm really trying to think ahead. Since the LTO-3 drives can write LTO-2, it seems like a no brainer at this point, and a good move for the future since LTO-4 should read (if not write) LTO-3 as well. On the SDLT front, 300GiB is the latest capacity, but cost and performance aren't still there. Sony has its SAIT at a whopping 500GiB native, but the performance is still lackluster to LTO-3 and newer SDLT. Not sure about reliability, although I would be interesting in considering it if reliability is better. But the vendor issue might not be doable. As such, does anyone have any pros/cons they've personally dealt with on the formats? Is there any reason not to go with LTO? BTW, I'm considering the HP 1/8 Autoloader with an Ultrium 960 drive, so native capacity is 3.2GiB with a 80GBps transfer rate over Ultra320 SCSI. I'm still researching the hardware support in Symantec (fka Veritas) NetBackup, which is our standard, but I believe I won't have any issues with supporting Windows 2000 and HP/UX 11 (although I might have to use Windows 2000 as the NetBackup buffering server for both hardware and temporary disk considerations). SAIC Study (Sponsored by USGS): http://wgiss.ceos.org/archive/archive.doc/FY03MediaTradeStudy.doc -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Wed Aug 10 12:11:41 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Re: Backing up 1-4TiB: LTO-3, SDLT or SAIT? -- corrections/addendeum In-Reply-To: <20050810160358.9328.qmail@web34110.mail.mud.yahoo.com> Message-ID: <20050810161142.30762.qmail@web34115.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > BTW, I'm considering the HP 1/8 Autoloader with an Ultrium > 960 drive, so native capacity is 3.2GiB with a 80GBps > transfer rate over Ultra320 SCSI. Er, that should read "native 3.2*T*iB [aggregate] capacity" with a "80*M*Bps [native] transfer rate" (doh!). Typical performance with 2:1 compression results in around 160MBps, which is enough to saturate even the 1Gbps GbE "out-of-band" links I'm planning on using between the NetBackup server and its agent systems. Although I plan on using buffering and delaying committal to tape. In fact, I think another benefit of LTO is that it can do variable write speeds, so it can "slow down" to accommodate the incoming stream. I haven't researched AIT, but I don't think DLT can do that. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Thu Aug 11 01:58:16 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? Message-ID: <3208.216.77.194.206.1123739896.squirrel@localhost> I have decided to stick with old, 32-bit technology for my next upgrade. Should I bother with getting an expensive Athlon 3200+ so I can run at 200MHz FSB, or get a lessly expensive 166MHz FSB Athlon and run my existing PC2700 RAM or upgrade to plentiful PC3200 RAM. (Matched pair 'dual channel' packages of less expensive RAM is available often.) I can pick up a MSI K7N2 Delta2 Platinum (nForce2 400 Ultra) for $68 shipped from ZZFly. Obviously, the cheapest FSB200 Athlon is expensive at $108. The FSB166 CPUs are a bit better and PC3200 is more plentiful than PC4000. I suspect things will move to DDR2 before PC4000 ever becomes widely available. Thanks. (fyi upgrading from ECS K7S5A running an XP 1800+ @ 133FSB w PC2700 @ 133MHz.) From b.j.smith at ieee.org Thu Aug 11 09:01:50 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Re: Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <3208.216.77.194.206.1123739896.squirrel@localhost> References: <3208.216.77.194.206.1123739896.squirrel@localhost> Message-ID: <1123765310.6152.25.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-11 at 01:58 -0400, Jason Boxman wrote: > I have decided to stick with old, 32-bit technology for my next upgrade. > Should I bother with getting an expensive Athlon 3200+ so I can run at > 200MHz FSB, or get a lessly expensive 166MHz FSB Athlon and run my existing > PC2700 RAM or upgrade to plentiful PC3200 RAM. If you go Athlon64, it doesn't matter, you can use whatever you want -- PC1600 (DDR200) to PC3200 (DDR400). > (Matched pair 'dual channel' packages of less expensive RAM is available often.) Just know that Socket-462, 478 and 754 only have a _single_ DDR channel for a 64-bit path to CPU. They use "interleaving" and call it "dual channel." In fact, the only processor that has a _true_ 128-bit DDR channel to CPU is Socket-939/940, totally glueless and direct to CPU. Even LGA-775 and Socket-640 use some muxed lines. > I can pick up a MSI K7N2 Delta2 Platinum (nForce2 400 Ultra) for $68 shipped > from ZZFly. Obviously, the cheapest FSB200 Athlon is expensive at $108. $68-108??? You're talking brand new nForce4 Standard/Ultra mainboard prices for Socket-939! AnandTech had a good price guide here (seems to be down right now): http://www.anandtech.com/guides/showdoc.aspx?i=2490 They have a new rev of my MicroATX nForce4 Standard mainboard with GbE on-board for $79 now. And some of the full ATX nForce4 Ultra mainboards are around $90 now too, $105 for a good, quality board in the AnandTech price guide. Brand new, Rev. E (31W PowerNow, SSE3) Athlon64 3000+ start at $136. And you can reuse your memory, _no_ CPU-I/O slowdown. You might be able to find some older nForce3 chipset mainboards for even cheaper. > The FSB166 CPUs are a bit better and PC3200 is more plentiful than PC4000. > I suspect things will move to DDR2 before PC4000 ever becomes widely > available. > Thanks. > (fyi upgrading from ECS K7S5A running an XP 1800+ @ 133FSB w PC2700 @ 133MHz.) Just don't know if it's worth buy old Athlon. Heck, even the new Socket-754 Sempron 3300+ (and 3400+) are now 64-bit, around $100 and perform comparable to an Athlon[64] 3000-3200+. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From dmckenna at thelimucompany.com Thu Aug 11 09:29:06 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859133@tlc001.tlcusa.thelimucompany.com> > Should I bother with getting an expensive Athlon 3200+ so I can > run at 200MHz FSB, or get a lessly expensive 166MHz FSB Athlon and > run my existing PC2700 RAM or upgrade to plentiful PC3200 RAM. Simply put, if you're considering spending a money on anything other than second-hand / cheap parts then you should spend the extra $30 to get a Socket 939 / Athlon64 system, and as Bryan said you can then reuse your existing memory, its just wasted money. -- 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 Aug 11 13:20:12 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859133@tlc001.tlcusa.thelimucompany.c om> References: <5C9DC445A45FEC4185D272DAF6AF37D1859133@tlc001.tlcusa.thelimucompany.com> Message-ID: <1133.216.78.64.62.1123780812.squirrel@localhost> Damien McKenna said: >> Should I bother with getting an expensive Athlon 3200+ so I can >> run at 200MHz FSB, or get a lessly expensive 166MHz FSB Athlon and >> run my existing PC2700 RAM or upgrade to plentiful PC3200 RAM. > > Simply put, if you're considering spending a money on anything other > than second-hand / cheap parts then you should spend the extra $30 to > get a Socket 939 / Athlon64 system, and as Bryan said you can then reuse > your existing memory, its just wasted money. Where is everyone buying their Athlon 64s that I am not looking? They're starting at $150! And, my old PSU is probably insufficient, so that's another $150 for a quality PSU. Now I'm spending about $450 instead of $180. Oops. If I was sure it was worth it, I might take the plunge, but if the least expensive A64 stuff now isn't that good and you really need to spend even a bit more than $450, that's way out of my range. From jasonb at edseek.com Thu Aug 11 14:05:56 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <1133.216.78.64.62.1123780812.squirrel@localhost> References: <5C9DC445A45FEC4185D272DAF6AF37D1859133@tlc001.tlcusa.thelimucompany.com> <1133.216.78.64.62.1123780812.squirrel@localhost> Message-ID: <1219.216.78.64.62.1123783556.squirrel@localhost> Jason Boxman said: > Where is everyone buying their Athlon 64s that I am not looking? > > They're starting at $150! > > And, my old PSU is probably insufficient, so that's another $150 for a > quality PSU. Now I'm spending about $450 instead of $180. Oops. > > If I was sure it was worth it, I might take the plunge, but if the least > expensive A64 stuff now isn't that good and you really need to spend even a > bit more than $450, that's way out of my range. What's more I have a $150 AGP card I bought a few weeks ago. I could go for store credit or resell it on Ebay, but that's just going to recoup a portion of the cost and further increase the cost of an AMD64 foray. Nope, I think my fate is sealed. If I build a new dedicated workstation I'll move to 64, but for this box it appears I'll be in 32 land. Or I'll need $150 PSU, $150 CPU, $100 board, and $150 VC (-$100 Ebay'd value) versus $65 for a board, $100 for a CPU, and ~ $70 for a deal on paired 512MBx2 PC3200. As I've probably made clear before, I am very cheap. I'll probably never even notice the difference between A64 and an older highend A32 setup for the activities I perform anyway. From dmckenna at thelimucompany.com Thu Aug 11 14:28:18 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D185914E@tlc001.tlcusa.thelimucompany.com> Food for thought: CPUs: http://labs.anandtech.com/products.php?sfilter=134&price=yes AGP compatible Motherboards: http://www.newegg.com/Product/ProductList.asp?Manufactory=&PropertyCodeV alue=709%3A7495&PropertyCodeValue=734%3A7577&description=&MinPrice=&MaxP rice=&SubCategory=22&Submit=Property What PSU do you have? I'd be surprised if it didn't work with these motherboards. It may not do eight harddrives and five DVDR drives but it should work. -- 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 Aug 11 15:08:44 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D185914E@tlc001.tlcusa.thelimucompany.c om> References: <5C9DC445A45FEC4185D272DAF6AF37D185914E@tlc001.tlcusa.thelimucompany.com> Message-ID: <1468.216.78.64.62.1123787324.squirrel@localhost> Damien McKenna said: > Food for thought: > > CPUs: > http://labs.anandtech.com/products.php?sfilter=134&price=yes > > AGP compatible Motherboards: > http://www.newegg.com/Product/ProductList.asp?Manufactory=&PropertyCodeV > alue=709%3A7495&PropertyCodeValue=734%3A7577&description=&MinPrice=&MaxP > rice=&SubCategory=22&Submit=Property > > What PSU do you have? I'd be surprised if it didn't work with these > motherboards. It may not do eight harddrives and five DVDR drives but > it should work. You make a compelling argument. I had never heard of E-Wiz. I'll have to check out resellerratings. The current PSU is CoolMax 325W. You can't buy it anymore on Newegg, but I posted the ratings on this list about a month ago. I think it was 12A or 15A on the +12V. Probably the former. I think an upgrade is strongly suggested if I'm going to run a 6600GT. It may run, but who knows if I'll have stability issues are the PSU ages and such. I think I will look again at the nForce3 boards, though. With the ~ $118 price for the cheapest A64 and AGP 3.0 support on the nForce3 boards, going A64 is starting to appear more realistic now. Thanks. From b.j.smith at ieee.org Thu Aug 11 17:35:18 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <1133.216.78.64.62.1123780812.squirrel@localhost> Message-ID: <20050811213518.4181.qmail@web34111.mail.mud.yahoo.com> Jason Boxman wrote: > Where is everyone buying their Athlon 64s that I am not > looking? They're starting at $150! But after you figure mainboard and you're debating memory, you're already up to spending $80 more. It's far better to go for the Socket-939 Athlon64 3000+ or even 3200+ on a newer Socket-939 nForce4 mainboard for the _same_price_. Furthermore, you can get a nForce3 Socket-754 mainboard for $50 these days. The Socket-754 Sempron 2800+ is under $80. $130 bucks right there, $80 cheaper, and the _same_price_ as a nForce4 Socket-462 and Athlon. Because AMD had to match Intel adding EM64T to the Celeron, the Sempron for Socket-754 _now_ has a 64-bit upgrade option (the new 3300+/3400+ are 64-bit). > And, my old PSU is probably insufficient, so that's another > $150 for a quality PSU. Not so. All Sempron and all new 90mm Winchester/Venice (Rev.D/E) Athlon64's use the same or _less_ power than the last generation of Athlons. Most of the $50 nForce3 Socket-754 mainboards are ATX1.0. So no power supply upgrade required. You can also continue to use your AGP cards, and the slots on the nForce3 are _definitely_ AGP3.0/8x and 0.8V capable. And even if you go Socket-939, as long as you don't get an ultra-power-sucking PCIe card, you don't need ATX2.0. In fact, the only reason you need ATX2.0 is because PCIe provides the same 50-100W that AGP"Pro" used to on workstation boards. If your video card don't suck the juice, then you don't need it as long as you're putting in an Athlon64 90mm 3000+, 3200+ or 3500+. > Now I'm spending about $450 instead of $180. Oops. I said it last year, unless you can get a _killer_ deal on a Socket-462 mainboard -- do _not_ buy a new Socket-462. Go Socket-754 instead for the _same_price. Or consider Socket-939 for the future, especially since some mainboards are ATX1.0 tolerant. > If I was sure it was worth it, It has *0* to do with 64-bit. It has to do with the fact that memory and I/O kicks ass on Socket-754 and, even more so, 939. Then add in the fact that Sempron/Athlon64 do SSE2 and the new Rev.E Venice cores do SSE3, while even the latest rev. Athlon XP-"M" only does SSE. Now consider the fact that Athlon XP is _dead_, AMD _stopped_ producing them months ago. All you have is Sempron, and the only way to 64-bit is the new Sempron on Socket-754 _only_ -- no Socket-462 option planned. > I might take the plunge, but if the least expensive A64 > stuff now isn't that good That's changed since the Sempron 3300+/3400+ were introduced. And even the Sempron 2800+ bests the Athlon XP2800+ in many applications. > and you really need to spend even a bit more than $450, > that's way out of my range. $100-150 gets you Socket-754, and you can even recycle your memory. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Thu Aug 11 17:41:34 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <1219.216.78.64.62.1123783556.squirrel@localhost> Message-ID: <20050811214135.15729.qmail@web34115.mail.mud.yahoo.com> Jason Boxman wrote: > Or I'll need $150 PSU, ?!?!?! Are you buying an SLI setup with dual-GeForce 6800Ultra or 7800GTX cards ?!?!?! You only need to spend $150 for a P.S. with triple or quad +12V rails and (2) 6-pin WS/PCIe connectors. If not, then try about $50 for quality, split-+12V P.S. If you absolutely want the 6-pin WS/PCIe connector, then maybe $75. I'm running with an Athlon64 3200+, 1GB RAM, GeForce 6800GT 256MB DDR3 (100W!), (2) hard drives, (1) DVD-RAM/RW/+RW with a _measly_ 300W power supply (around ~345W max) that is an ATX2.0 with split +12V lines (only 15A and 10A). > $150 CPU, Okay, I agree there. > $100 board, Try $80 these days. And with the SLI X16, the nForce4 Standard/Ultra mainboards are going to drop to $60 very soon. And if you go nForce3, you can use AGP instead. > and $150 VC (-$100 Ebay'd value) versus $65 for a board, > $100 for a CPU, and ~ $70 for a deal on paired 512MBx2 > PC3200. Why are you upgrading your memory??? Didn't we discuss this? > As I've probably made clear before, I am very cheap. > I'll probably never even notice the difference between A64 > and an older highend A32 setup for the activities I perform > anyway. Why not Socket-754 because Socket-462 is _dead_?! Socket-754 is still very much AGP. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Thu Aug 11 17:47:12 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] RE: Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <1468.216.78.64.62.1123787324.squirrel@localhost> Message-ID: <20050811214712.7581.qmail@web34111.mail.mud.yahoo.com> Jason Boxman wrote: > I think an upgrade is strongly suggested if I'm going to > run a 6600GT. NV43 (6600[GT]) sucks up a lot less juice than NV40 (6800[GT|Ultra]). > I think I will look again at the nForce3 boards, though. > With the ~ $118 price for the cheapest A64 and AGP 3.0 > support on the nForce3 boards, going A64 is starting to > appear more realistic now. Don't forget Socket-754! You can get a Socket-754 Sempron 2800+ for under $80, and Socket-754 now has an upgrade path to 64-bit thanx to the new Sempron models (to compete with the new EM64T Celerons). I only said that Socket-754 wasn't worth it IMHO versus Socket-939 for $100 cheaper. But Socket-754 is _definitely_ worth it over Socket-462 for the _same_price_. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Thu Aug 11 18:03:05 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] NewEgg: Socket-754 nForce3 250 mainboards + Sempron 64 = $120-140! Message-ID: <20050811220305.36731.qmail@web34107.mail.mud.yahoo.com> I note 4 choices from $50-60: http://www.newegg.com/Product/ProductList.asp?submit=PROPERTY&SubCategory=22&propertycodevalue=709:7494,719:7522&bop=and All the latest/last-gen AGP3.0/x8 with 0.8V support. All but 1 looked standard ATX1.0, and I bet the 1 that wasn't takes an ATX1.0 anyway (because unless it's an AGP"Pro" slot, it can't send more than 25W). And here are the Socket-754 from $65 upward: http://www.newegg.com/Product/ProductList.asp?submit=PROPERTY&SubCategory=343&propertycodevalue=517:7438&bop=and Do you see what I see? You can get the Sempron 2800+ for $69.99, _or_ you can splurge for the Sempron _64_ 2800+ for $76. Better yet, $13 more = $89, will get you the Sempron _64_ 3000+ (and the non-64 for the same price). _All_ of those processors are 90nm Rev.D (SSE2). -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Thu Aug 11 19:37:50 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] NewEgg: Socket-754 nForce3 250 mainboards + Sempron 64 = $120-140! In-Reply-To: <20050811220305.36731.qmail@web34107.mail.mud.yahoo.com> References: <20050811220305.36731.qmail@web34107.mail.mud.yahoo.com> Message-ID: <1343.216.77.197.232.1123803470.squirrel@localhost> Bryan J. Smith said: > I note 4 choices from $50-60: > > > Do you see what I see? You can get the Sempron 2800+ for > $69.99, _or_ you can splurge for the Sempron _64_ 2800+ for > $76. Better yet, $13 more = $89, will get you the Sempron > _64_ 3000+ (and the non-64 for the same price). _All_ of > those processors are 90nm Rev.D (SSE2). Wow, sounds like a winner (for me). I guess I'll need to research the boards and pick one. I notice L2 is either 128K or 256K on the Semprons. Is that just a newegg typo or do they really differ? From jasonb at edseek.com Thu Aug 11 19:27:53 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <20050811214135.15729.qmail@web34115.mail.mud.yahoo.com> References: <1219.216.78.64.62.1123783556.squirrel@localhost> <20050811214135.15729.qmail@web34115.mail.mud.yahoo.com> Message-ID: <1322.216.77.197.232.1123802873.squirrel@localhost> Bryan J. Smith said: > Jason Boxman wrote: >> Or I'll need $150 PSU, > > ?!?!?! Are you buying an SLI setup with dual-GeForce > 6800Ultra or 7800GTX cards ?!?!?! You only need to spend > $150 for a P.S. with triple or quad +12V rails and (2) 6-pin > WS/PCIe connectors. > > If not, then try about $50 for quality, split-+12V P.S. If > you absolutely want the 6-pin WS/PCIe connector, then maybe > $75. My bad. I know PSUs like the Antec True Power series are rather pricey, but are supposedly of superior quality. When I looked for decent PSUs two years ago the one I bought for $50 was about entry level I think. It would seem prices are more favorable now. > I'm running with an Athlon64 3200+, 1GB RAM, GeForce 6800GT > 256MB DDR3 (100W!), (2) hard drives, (1) DVD-RAM/RW/+RW with > a _measly_ 300W power supply (around ~345W max) that is an > ATX2.0 with split +12V lines (only 15A and 10A). Nice. >> $150 CPU, > > Okay, I agree there. > >> $100 board, > > Try $80 these days. But does that get you anything interesting? Granted I don't have any FireWire devices, SATA seems standard now, and I have few effective uses for GbE. The later would be nice to have, though, as GbE becomes more common place. I recently saw a switch, unmanaged, that claimed support for jumbo (9KiB) frames for about $60 AR. Then again, it may have been crap. Overtime I suspect the quality of 'consumer' GbE switches will increase, though. > And with the SLI X16, the nForce4 Standard/Ultra mainboards > are going to drop to $60 very soon. > > And if you go nForce3, you can use AGP instead. I'll have to. I already dropped the cash for the VC thinking that would be all I was going to buy, my current board being an AGP board. But then I realized one game in particular is rather I/O intensive and is too boring to play on my old system. (It takes 30 minutes, realtime to get anywhere on the world map playing Silent Hunter III, a subsim, using full in-game time compression. That makes it quite boring. It's much faster for players who own modern systems, though.) >> and $150 VC (-$100 Ebay'd value) versus $65 for a board, >> $100 for a CPU, and ~ $70 for a deal on paired 512MBx2 >> PC3200. > > Why are you upgrading your memory??? > Didn't we discuss this? We did. I would have upgraded the memory on an nForce2 Ultra 400 so I could run the CPU and RAM in sync. I understand that with an nForce3 or nForce4 solution the memory speed no longer matters relative to the CPU, since the concept of the FSB is no longer there. >> As I've probably made clear before, I am very cheap. >> I'll probably never even notice the difference between A64 >> and an older highend A32 setup for the activities I perform >> anyway. > > Why not Socket-754 because Socket-462 is _dead_?! > Socket-754 is still very much AGP. Mostly, because I am out of touch with modern hardware and the initial price of the A64 3000+ is quite a huge sticker shock to someone who's never paid more than $60 for a CPU, save my Cyrix P150+ for about $150 back in '97. What you've said makes sense and is reassuring. I am definitely going to move to either 754 or 939. I may do the former and get the absolute least expensive quality mainboard and lowest end CPU I can find, perhaps one of the 64-bit Semprons. I found this Sempron to be pretty inexpensive. The newegg listings indicate you can buy both 128K and 256K L2 cache models of 64-bit Semprons. I'm shooting for the 256K variants, of course. AMD Sempron 64 2800+ Palermo 800MHz FSB Socket 754 Processor Model SDA2800BXBOX - Retail - $76 I am getting the impression I shouldn't settle for a mainboard with any chipset other than nForce3 (for 754) or nForce4 (or better, for 939)? I see the prices on the 754 stuff is very, very inexpensive as you said. I'm definitely moving to 754 on the cheap then. 754 is about to be dead though, too, isn't it? ;) From ozz at ozz.is-a-geek.net Thu Aug 11 20:04:28 2005 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <1133.216.78.64.62.1123780812.squirrel@localhost> References: <5C9DC445A45FEC4185D272DAF6AF37D1859133@tlc001.tlcusa.thelimucompany.com> <1133.216.78.64.62.1123780812.squirrel@localhost> Message-ID: <20050811200428.31678e62.ozz@ozz.is-a-geek.net> On Thu, 11 Aug 2005 13:20:12 -0400 (EDT), "Jason Boxman" wrote: > > And, my old PSU is probably insufficient, so that's another $150 for a > quality PSU. Now I'm spending about $450 instead of $180. Oops. You shouldn't have to spend anything like that on a PSU. Earlier this year I got a 535W PSU (Enermax EG565P-FMA REV.2.0 ATX 2.0 w/SLI Support) from Monarch for $99, and that sucker was major overkill. The main reason I got it was because I'm running a lot of additional stuff (lotsa drives, etc.). Remember too that the AMD64 often draws less power than the older Athlons. My AMD64 3000+ draws about 25% of the power my 1.3GHz Athlon used to pull. Regards, Ozz. From dmckenna at thelimucompany.com Thu Aug 11 20:21:53 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859179@tlc001.tlcusa.thelimucompany.com> [$80 mobo] > But does that get you anything interesting? Shop around. You'll tend to get Firewire, SATA, USB 2.0 High Speed, etc, all the good basics. > I am getting the impression I shouldn't settle for a > mainboard with any chipset other than nForce3 (for 754) or > nForce4 (or better, for 939)? Looking at the reviews it seems this is true. > 754 is about to be dead though, too, isn't it? ;) Yeah, Socket 939 looks to become the new low-end next year when they bring out *another* new socket type. Still, with a socket 939 you can start off with a base CPU now and upgrade to a faster dual-core chip later, which is my intention. Probably late next year I'll upgrade a few bits (CPU, gfx, maybe RAID), for now I'm just going to get my backup drive going and we'll be grand :) -- 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 Aug 11 20:35:50 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] NewEgg: Socket-754 nForce3 250 mainboards + Sempron 64 = $120-140! In-Reply-To: <1343.216.77.197.232.1123803470.squirrel@localhost> Message-ID: <20050812003550.74821.qmail@web34107.mail.mud.yahoo.com> Jason Boxman wrote: > I guess I'll need to research the boards and pick one. Yeah, look at your choices. Remember, nForce is extremely Linux compatible as long as you are running kernels 2.4.23/2.6.5 or greater. > I notice L2 is either 128K or 256K on the Semprons. Is > that just a newegg typo or do they really differ? That's why the Semprons are cheaper, they have less cache. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Thu Aug 11 20:37:40 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859179@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050812003741.32042.qmail@web34108.mail.mud.yahoo.com> Damien McKenna wrote: > Yeah, Socket 939 looks to become the new low-end next year > when they bring out *another* new socket type. ??? Only DDR2 _might_ change this. Socket-754 is the "new Duron" platform. It's not going away anytime soon. My point was and continues to be, Socket-939 is _best_, but do _not_ even look at Socket-462 -- it died last year. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dmckenna at thelimucompany.com Thu Aug 11 20:44:55 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D185917A@tlc001.tlcusa.thelimucompany.com> > Damien McKenna wrote: > > Yeah, Socket 939 looks to become the new low-end next year > > when they bring out *another* new socket type. > > ??? Only DDR2 _might_ change this. http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2476 > Socket-754 is the "new Duron" platform. > It's not going away anytime soon. My guess is that it'll be pushed towards the low-end OEM market (eMachines et al) while Socket 939 gets reduced in price even more to the point where the retail difference between 754 and 939 is negligible, and I'm guessing that'll start happening this year. Just my caffeine-overdose-induced chicken-entrails guess though. -- 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 Aug 11 20:57:26 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Staying with 32-bit: PC3200 or PC4000, 166 or 200MHz Athlon? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D185917A@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050812005726.87428.qmail@web34101.mail.mud.yahoo.com> Damien McKenna wrote: > http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=2476 Then that's what it is, a new socket. They are basically introducing a new socket to prevent people from blowing their memory. DDR and DDR2 signaling is not compatible, hence why you can't use the same processor. Understand the DDR channels are _direct_pin-outs_ on the Socket-754, 939 and 940 -- 184-pins/true 64-bit DDR channel, hence 754+184=938. > My guess is that it'll be pushed towards the low-end OEM > market (eMachines et al) They've done that already by dropping the Athlon 64. Again, I have been advocating Socket-939 since last year. And I said do _not_ buy a new Socket-462 platform. If you were really cheap, just get a Socket-754. > while Socket 939 gets reduced in price even more to > the point where the retail difference between 754 and > 939 is negligible, and I'm guessing that'll start happening > this year. I see M2 replacing Socket-939 from Anand's roadmap. It's clear that AMD is only going to offer a few parts for it. And AMD had already said they'd eventually come out with Sempron for Socket-939. From the roadmap, I see both. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From tim at mcdonough.net Fri Aug 12 11:40:03 2005 From: tim at mcdonough.net (Tim McDonough) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Email Troubleshooting Message-ID: <42FCC2D3.1050108@mcdonough.net> I'm having some email trouble when users are sending mail between two domains and am looking for thoughts on how to troubleshoot and resolve the problem. Here's the situation: There are two domains which are hosted by two different ISPs who also provide the email services for each location. When anyone@domainA.com sends an email to anyone@domainB.com it can take hours or even days for the message to be delivered. anyone@domainA.com can send emails to other domains and they are typically delivered within seconds/minutes. anyone@domainB.com can send email to anyone@domainA.com and it is delivered withi seconds/minutes. These big delays only occur when email is sent from Domain A to Domain B and, as far as we can determine, not any other time. I'm not familiar with the ins and outs of email systems, I'm just looking for things to check and the right questions to ask the providers. Of course, initially oth ISPs have said "everything is okay on our end". Fixing this is somewhat of a priority--one of the company's owners is at Domain A! Thanks in advance for any pointers. -- Tim From work at sprynet.com Fri Aug 12 13:41:34 2005 From: work at sprynet.com (J.T. Hayden) Date: Tue Oct 31 13:17:11 2006 Subject: [Pc_Support] Email Troubleshooting In-Reply-To: <42FCC2D3.1050108@mcdonough.net> Message-ID: <02c901c59f65$15623c80$650aa8c0@xpmaster> Sounds like a filering peer issue, if they are not "peered" correctly there is a lag in the send/receive of the mail (they are sent to the bottom of the send/receive que) plus any spam/virii checks at that point are delayed also slowing delivery J.T. Hayden -----Original Message----- From: pc_support-bounces@matrixlist.com [mailto:pc_support-bounces@matrixlist.com] On Behalf Of Tim McDonough Sent: Friday, August 12, 2005 11.40 To: This is the PC Support list. Subject: [Pc_Support] Email Troubleshooting I'm having some email trouble when users are sending mail between two domains and am looking for thoughts on how to troubleshoot and resolve the problem. Here's the situation: There are two domains which are hosted by two different ISPs who also provide the email services for each location. When anyone@domainA.com sends an email to anyone@domainB.com it can take hours or even days for the message to be delivered. anyone@domainA.com can send emails to other domains and they are typically delivered within seconds/minutes. anyone@domainB.com can send email to anyone@domainA.com and it is delivered withi seconds/minutes. These big delays only occur when email is sent from Domain A to Domain B and, as far as we can determine, not any other time. I'm not familiar with the ins and outs of email systems, I'm just looking for things to check and the right questions to ask the providers. Of course, initially oth ISPs have said "everything is okay on our end". Fixing this is somewhat of a priority--one of the company's owners is at Domain A! Thanks in advance for any pointers. -- Tim _______________________________________________ Pc_support mailing list Pc_support@matrixlist.com http://lists.matrixlist.com/mailman/listinfo/pc_support From tim at mcdonough.net Fri Aug 12 14:40:39 2005 From: tim at mcdonough.net (Tim McDonough) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Email Troubleshooting In-Reply-To: <02c901c59f65$15623c80$650aa8c0@xpmaster> References: <02c901c59f65$15623c80$650aa8c0@xpmaster> Message-ID: <42FCED27.2000809@mcdonough.net> Any suggestion as to which ISP may have the settings "sub-optimal"? J.T. Hayden wrote: > Sounds like a filering peer issue, if they are not "peered" correctly > there is a lag in the send/receive of the mail (they are sent to the > bottom of the send/receive que) plus any spam/virii checks at that point > are delayed also slowing delivery > > J.T. Hayden > > -----Original Message----- > From: pc_support-bounces@matrixlist.com > [mailto:pc_support-bounces@matrixlist.com] On Behalf Of Tim McDonough > Sent: Friday, August 12, 2005 11.40 > To: This is the PC Support list. > Subject: [Pc_Support] Email Troubleshooting > > > I'm having some email trouble when users are sending mail between two > domains and am looking for thoughts on how to troubleshoot and resolve > the problem. Here's the situation: > > There are two domains which are hosted by two different ISPs who also > provide the email services for each location. When anyone@domainA.com > sends an email to anyone@domainB.com it can take hours or even days > for the message to be delivered. > > anyone@domainA.com can send emails to other domains and they are > typically delivered within seconds/minutes. > > anyone@domainB.com can send email to anyone@domainA.com and it is > delivered withi seconds/minutes. > > These big delays only occur when email is sent from Domain A to Domain > B and, as far as we can determine, not any other time. > > I'm not familiar with the ins and outs of email systems, I'm just > looking for things to check and the right questions to ask the > providers. Of course, initially oth ISPs have said "everything is okay > on our end". > > Fixing this is somewhat of a priority--one of the company's owners is > at Domain A! > > Thanks in advance for any pointers. > -- Tim From work at sprynet.com Fri Aug 12 15:56:52 2005 From: work at sprynet.com (J.T. Hayden) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Email Troubleshooting In-Reply-To: <42FCED27.2000809@mcdonough.net> Message-ID: <02f601c59f77$fde009d0$650aa8c0@xpmaster> Well they both have them optimal for the services they are peered to, what has to happen IF you can arrange it is have THEM confer on optimization between themselves J.T. Hayden -----Original Message----- From: pc_support-bounces@matrixlist.com [mailto:pc_support-bounces@matrixlist.com] On Behalf Of Tim McDonough Sent: Friday, August 12, 2005 14.41 To: work@mindspring.com; This is the PC Support list. Subject: Re: [Pc_Support] Email Troubleshooting Any suggestion as to which ISP may have the settings "sub-optimal"? J.T. Hayden wrote: > Sounds like a filering peer issue, if they are not "peered" correctly > there is a lag in the send/receive of the mail (they are sent to the > bottom of the send/receive que) plus any spam/virii checks at that > point are delayed also slowing delivery > > J.T. Hayden > > -----Original Message----- > From: pc_support-bounces@matrixlist.com > [mailto:pc_support-bounces@matrixlist.com] On Behalf Of Tim McDonough > Sent: Friday, August 12, 2005 11.40 > To: This is the PC Support list. > Subject: [Pc_Support] Email Troubleshooting > > > I'm having some email trouble when users are sending mail between two > domains and am looking for thoughts on how to troubleshoot and resolve > the problem. Here's the situation: > > There are two domains which are hosted by two different ISPs who also > provide the email services for each location. When anyone@domainA.com > sends an email to anyone@domainB.com it can take hours or even days > for the message to be delivered. > > anyone@domainA.com can send emails to other domains and they are > typically delivered within seconds/minutes. > > anyone@domainB.com can send email to anyone@domainA.com and it is > delivered withi seconds/minutes. > > These big delays only occur when email is sent from Domain A to Domain > B and, as far as we can determine, not any other time. > > I'm not familiar with the ins and outs of email systems, I'm just > looking for things to check and the right questions to ask the > providers. Of course, initially oth ISPs have said "everything is okay > on our end". > > Fixing this is somewhat of a priority--one of the company's owners is > at Domain A! > > Thanks in advance for any pointers. > -- Tim _______________________________________________ Pc_support mailing list Pc_support@matrixlist.com http://lists.matrixlist.com/mailman/listinfo/pc_support From b.j.smith at ieee.org Sun Aug 14 00:10:45 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: I told you so.. In-Reply-To: <001801c5a083$4fa1f060$0203a8c0@hbco> Message-ID: <20050814041045.29977.qmail@web34114.mail.mud.yahoo.com> Bill Smith wrote: > Hi Brian, > This is likely old news to you, but ran across this > article. You have been predicting this for some time. > Kudos to you! > http://apnews.myway.com/article/20050812/D8BU2UI00.html Well, I'm waiting on more technical details. I.e., Let's see if it's the "Yamhill2" I have been predicting. If it's Intel's first x86[-64] redesign since Pentium Pro (i686) of 1994, then that's good news for Intel. I have predicted the EM64T design team codenamed as "Yamhill" was up to more than simple some extensions for Pentium 4, and that would take 36+ months like a traditional, full design. If os, I believe they have the first, full redesign since Pentium Pro. The P4 was just a 18-month "hack" known as "NetBurst" architecture market-wise, "extend the pipes and clock" reality-wise, with most performance gains coming from the 2nd, "lossy" SSE pipe. But don't think AMD is just lying down either. Their 1999 Athlon core might be only 6 years old, but that's the typical lifespan before a redesign. With AMD's purposely limiting the Athlon (including A64/Opteron, which use the same, 40-bit core) to only 1MB L2 -- even though the EV6 logic they are based on supports upto 8MB L2 -- is a sure sign that they have a full redesign in the work too. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Sun Aug 14 00:21:20 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] AOpen F90GS 19" LCD for $240 -- AOpen LCD same, portable design as Sceptre? Message-ID: <20050814042120.51114.qmail@web34106.mail.mud.yahoo.com> DealNews just passed on that NewEgg.com has the AOpen F90GS 19" (1280x1024, DVI/VGA, 12ms response) is $299.99 - $10 instant - $50 mail-in, for $239.99 total. http://dealnews.com/deals/AOpen-F90-GS-19-LCD-Display-for-240-after-rebate/93280.html?ref=dndaily In looking at the pictures for the monitor: http://www.newegg.com/Product/ShowImage.asp?image=24-174-029-01.jpg,24-174-029-03.jpg,24-174-029-04.jpg,24-174-029-02.jpg,24-174-029-05.jpg,24-174-029-06.jpg&CurImage=24-174-029-01.jpg&Description=AOpen%20F90GS%20Black%2019%22%2012ms%20LCD%20Monitor%20-%20Retail I want to say it looks like the same, ultra-portable ("fold base back and put in box") design as the Sceptre 19" I have. I really want a 12ms response LCD to replace my 25ms Sceptre for video/gaming, while being the exact same resolution (and close to same height) so it can be the primary head with the Sceptre as the secondary head (which would be off for gaming). I thought someone else here bought an AOpen recently? Or was that another (Acer?) vendor? Let me know because the deal is up today (Sunday). -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From m9u35g at gmail.com Sun Aug 14 22:54:56 2005 From: m9u35g at gmail.com (Justin M. Keyes) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Change CPU clock multiplier IN LINUX? Message-ID: <46f680d05081419541eb681f6@mail.gmail.com> Hi Guys, I got an AMD Athlon XP Mobile 2600+ 266MHz processor for a motherboard that supports the non-mobile version of the same spec. >From what I have read [1], the "mobile" series works in desktops, "but the problem is you must have a board that lets you recognize and change the multipliers, since the mobile starts up initially at its lowest multiplier and you must change it by bios to get up to its max speed." I put the mobile CPU in my Biostar M7VIW and it works just fine, except, as the guy said, it starts on its lowest multiplier, which means I only get 800MHz. I upgraded to the latest BIOS but unfortunately this board does _not_ allow you to change the clock multiplier! HOWEVER, there are several different Windows programs that allow you to change these settings from within Windows! Some of the most popular ones: http://crystalmark.info/software/CrystalCPUID/index-e.html http://www.cpuheat.wz.cz/html/CPUMSR_main.htm My question: can I do this from linux? Thanks! [1] http://forums.amd.com/lofiversion/index.php/t22692.html -- Justin Keyes From m9u35g at gmail.com Mon Aug 15 13:56:53 2005 From: m9u35g at gmail.com (Justin M. Keyes) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] AOpen F90GS 19" LCD for $240 -- AOpen LCD same, portable design as Sceptre? In-Reply-To: <20050814042120.51114.qmail@web34106.mail.mud.yahoo.com> References: <20050814042120.51114.qmail@web34106.mail.mud.yahoo.com> Message-ID: <46f680d05081510565ba3c1b8@mail.gmail.com> On 8/14/05, Bryan J. Smith wrote: > I want to say it looks like the same, ultra-portable ("fold > base back and put in box") design as the Sceptre 19" I have. > I really want a 12ms response LCD to replace my 25ms Sceptre > for video/gaming, while being the exact same resolution (and > close to same height) so it can be the primary head with the > Sceptre as the secondary head (which would be off for > gaming). > > I thought someone else here bought an AOpen recently? I bought the 16ms model. No problems with it so far, except, as I mentioned originally, it doesn't seem to cooperate with power management--when linux turns it off for power management, the monitor detects it and turns back _on_ to tell me that there is no signal -- then linux turns it off again... then it turns itself back on again... etc. I haven't tried switching the BIOS power management type (because I didn't think linux listened to the BIOS about anything anyways), but for now I just turn it off manually whenever I leave ;( Here is the previous thread: http://lists.matrixlist.com/pipermail/pc_support/2005-June/000415.html -- Justin Keyes From b.j.smith at ieee.org Mon Aug 15 16:30:37 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! Message-ID: <20050815203037.83270.qmail@web34104.mail.mud.yahoo.com> Wife is in the market for a new color printer. We've stretched our ink jets on long enough, and we really tire of the combination of the cost of ink cartridges combined with the reality that we're not doing photo-quality 99% of the time. A few weeks ago, Dealnews had the HP Color LaserJet 2600N on sale for $339.98: http://dealnews.com/deals/HP-Color-Laser-Jet-2600-N-Laser-Printer-for-340-shipped/92782.html Although it has a network interface, it was clearly as "host" driven printer designed only for the Windows Graphical Display Interface (GDI) and the fixed memory amount left much to be desired (and to network traffic). I mean, who cares if it does dozens of pages per minute in color if you can't feed it anything of substance (i.e., the communications is the bottleneck)? But today, lo'n behold, the Dell 3100cn Color Laser was on sale for $329.40, about the same price as its 3000cn brotheren. http://dealnews.com/deals/Dell-3100-cn-Color-Laser-Printer-for-329-shipped/93496.html?ref=dndaily The 3100cn, unlike the 3000cn, comes with a host of additions -- beyond just the paper trays, FULL POSTSCRIPT LEVEL 3 SUPPORT! Driven with an on-board 300MHz RISC processor and 64MB of RAM (expandable to 576MB), this is a serious printer for a low price. Yes, that's MacOS X and Linux support out-of-the-box, offering Line Printer Daemon (LPD) support (IPP and other options are available for a $99 add-on). In reading reviews, the color for a Laser is the best you'll get short of a leading edge photo-quality inkjet, and is better than the latter when you need to mix text and graphics (e.g., publication). Most people said skip the optional ($300) duplexer, it's got alignment issues that seem to be inherent to the design. A few others complained about Dell failing to ship the unit with internal packaging, and that it could result in it being damaged upon arrival. And we all know how India's--er, I mean Dell's support is. ;-> Hmmm, with that said, I think I'm going to chance it. I could really use a network printer for home ever since my IBM NP17 had sensor issues. My wife's current HP LaserJet 1200se eats through cartridges at a good cost (several cents/page) and the Dell 3100cn at just over a cent/page block seems to best even my aged Lexmark E310 at price-per back'n white (let alone our inkjet printers at colors). -- Bryan P.S. Just to be complete, there is the Dell 5100cn which DealNews mentioned is also on-sale for $699.30 shipped: http://dealnews.com/deals/Dell-5100-cn-Color-Laser-Printer-for-699-shipped/93498.html?ref=dndaily It is much, much faster at color, comes with the duplexer built-in and has other goodies (some standard, some still optional -- like the IPP/NCP/SMB support?). I also saw some notes that its color wasn't as good, but that could have been a dated review. IMHO, I'd rather spend half as much, especially since the warranty is still 1 year. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From brianashelist at yahoo.com Mon Aug 15 17:22:08 2005 From: brianashelist at yahoo.com (Brian Ashe) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <20050815203037.83270.qmail@web34104.mail.mud.yahoo.com> Message-ID: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> > Wife is in the market for a new color printer. Funny, I've been looking recently myself. (Note: I am not Bryan's wife.) I was looking at a $500 model and a $300 one. IIRC, one was some flavor of HP 2550. Had networking but I didn't look at RAM, etc. I think it was PostScript. Looking online, toner for 2550s is $100 per color. For $500, it probably ships with "starter rolls"--aka, half-empty. (Or half-full, depending on your POV. :-) ) The $300 model was, I think, from Brother, on sale at Office Depot. (Not saying these are better, just giving you some background.) Hadn't yet gotten 'round to checking it out, just saw it in the Sunday paper sales flyers. > But today, lo'n behold, the Dell 3100cn Color Laser > was on > sale for $329.40, about the same price as its 3000cn > brotheren. Sweet! Are these re-badged HPs? Where do you get consumables? Dell? HP? Searching CompUSA.com for 'hp 3100' I see only black: http://www.compusa.com/products/product_info.asp?product_code=50170019&pfp=SEARCH The same search at CDW comes up empty. I can't get through to dell.com right now. Have you priced consumables? What does it work out to, price-per-page? > And > we all > know how India's--er, I mean Dell's support is. ;-> FYI, I recently had the displeasure of going a couple rounds with some Apple customer (not tech) support people, all of whom had mild British accents. :-p Same everywhere, I suppose. (Though Dell brought their Corporate support people back to TX, IIRC. Baby steps...) ____________________________________________________ Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs From b.j.smith at ieee.org Mon Aug 15 18:06:08 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> Message-ID: <20050815220608.51260.qmail@web34103.mail.mud.yahoo.com> Brian Ashe wrote: > Funny, I've been looking recently myself. > (Note: I am not Bryan's wife.) Now that's a thought I didn't want. @-p > I was looking at a $500 model and a $300 one. > IIRC, one was some flavor of HP 2550. Reviews say the Dell is a Fuji-Xerox design. The 2550 lists a Motorola (PowerPC I assume, if not MCore) 264MHz and Postscript Level 3. I think some of the 300-400MHz in other products are either ARM variants, possibly Intel XScale (Superscalar ARM). I want to say Xerox is XScale now, but I could be wrong. HP has traditionally been Moto -- especially for the cheapy GDI approach/support (i.e., "host driven") -- but they flop regularly between products. I was just happy to see something under $400 with a beefy intelligence, amount of RAM and Postscript Level 3 in the unit itself. Dell is not shy, they put Linux compatibility right the specification page. They also state that the network does LPD -- there is a $99 upgrade for printer-based IPP, IPX/NCP and SMB support. I'll just use LPD and handle queuing at my real server if I need it. > Had networking but I didn't look at RAM, etc. As long as they work with MacOS X, they can't be "host-based" -- at least not easily. Most that do Mac also do Linux, at least in the Lasers because they at least have PCL, if not Postscript. > I think it was PostScript. Looking online, toner for 2550s > is $100 per color. Yeah, the toner costs gets up there. But thinking to ink, it's not really any worse, and typically much better/more economical. > For $500, it probably ships with "starter rolls"--aka, > half-empty. (Or half-full, depending on your POV. :-) ) The Dell 3x00cn series uses 2,000 and 4,000 sheet BCYM toner cartrides. I bet they are only 2,000 as shipped, at least for the CYM. $95 for the CYM, $45 for the B. The 5x00cn seems to cut costs more with 9,000 B and 8,000 CYM, specially on the black costs. > The $300 model was, I think, from Brother, on sale at Office > Depot. (Not saying these are better, just giving you some > background.) Hadn't yet gotten 'round to checking it > out, just saw it in the Sunday paper sales flyers. Yeah, I've been looking at DealNews regularly. I mentioned the HP Color LaserJet 2600N a few weeks ago and she wasn't interested. Then she came back later and said she was, but after the sale was up. I'm glad I waited because this Dell is PS3/LPD with a real, on-board intelligence, whereas the 2600N was a host-based with limited/fixed RAM that probably killed the high rate of color PPM. > Sweet! Are these re-badged HPs? Where do you get > consumables? Dell? HP? Searching CompUSA.com for 'hp > 3100' I see only black: >http://www.compusa.com/products/product_info.asp?product_code=50170019&pfp=SEARCH > The same search at CDW comes up empty. I can't get > through to dell.com right now. Have you priced > consumables? What does it work out to, price-per-page? Although I haven't "kept up to date" on all the latest models, the price-per-page of just the black'n white versus my current lasers was the "final selling point." The black is 1.1 cents/page, add another half-cent for the drum, for 1.6 cents/page. The color is 8.2 cents/page, add another 1.7 cents/page for dum, for 9.9 cents/page total. Not bad at all for that price point (with Dell's more expensive 5100cn knocking a half-cent off black, about 2.5 off color). My current best is my 6-year old Lexmark E310 (Postscript Level 2), which gets 10,000 pages per $160 extended toner, or 1.6 cents/page not including drum. The drum is rated around 20,000 pages, and I've just about crossed it. Not sure how much it costs. My wife's current printer is a 3.5 year-old HP LaserJet 1220se (print/scan/copy), which gets about 3.1 cents/page, not great at all. Sure, the toners only cost $65, but I only get 2,000 sheets. Haven't hit the drum life, which is around 20,000 as well IIRC. So this printer is definitely an improvement, just for black'n white. ;-> > FYI, I recently had the displeasure of going a couple > rounds with some Apple customer (not tech) support > people, all of whom had mild British accents. :-p > Same everywhere, I suppose. (Though Dell brought their > Corporate support people back to TX, IIRC. Baby > steps...) Were they British? Or were they Hong Kong-like British? -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Mon Aug 15 18:14:27 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! -- OT comment In-Reply-To: <20050815220608.51260.qmail@web34103.mail.mud.yahoo.com> Message-ID: <20050815221427.99697.qmail@web34107.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > My current best is my 6-year old Lexmark E310 (Postscript > Level 2), which gets 10,000 pages per $160 extended toner, > or 1.6 cents/page not including drum. The drum is rated > around 20,000 pages, and I've just about crossed it. Not > sure how much it costs. BTW, just an OT comment, this printer was a Godsend when I was a teacher. I couldn't get a copier card (seniority type BS, only worse), so I used to either get the electronic forms (typically PDF) of the materials (typically for science), or generate my own (for math via LyX). My printer was running day'n night when I was a teacher and the sucker never hiccuped, and maybe jammed no more than 3-4 times. I know I reeled off a good 10,000 pages just as a teacher, about 250 pages every other day on average. I know a lot of teachers were envious that I had my own notebook and printer, especially the younger teachers who were in the same "lack of seniority" boat I was, but didn't have the financial resources/existing computer hardware that I did. I used to help them out all I could, and generated math worksheets with LyX and graphs with GNUPlot and other tools. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From mflang at bellsouth.net Mon Aug 15 19:12:38 2005 From: mflang at bellsouth.net (Max F Lang) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> References: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> Message-ID: <200508151912.38782.mflang@bellsouth.net> On Monday 15 August 2005 17:22, Brian Ashe wrote: > $300 one. IIRC, one was some flavor of HP 2550. Had > networking but I didn't look at RAM, etc. I think it > was PostScript. Looking online, toner for 2550s is > $100 per color. For $500, it probably ships with > "starter rolls"--aka, half-empty. (Or half-full, > depending on your POV. :-) ) The $300 model was, I Brian, we use the LJ2550 at work for printing color lab reports in physician offices. This is a printer I'm coming to know well... The built-in print server is an onboard DirectJet 625. You can't remove it, but it's exactly the same thing firmware-wise. Does the usual IP, IPX, IPP and AppleTalk, with both an easy browser and telnet session based interface. They come with 32M ram standard. The toners included in the box are exactly the same carts you buy. They are rated for 6K pages in color, and the black cart is rated for 7K. And that's about what we get. They're not noobie carts at all, and yes, replacements are $$$, even at our corporate discount. The 2550 is a dumptruck of a printer: big, heavy, and awful slow to start. It usually takes about 20 seconds to start printing. But just like a dumptruck, once it starts going, it doesn't want to stop. In fact, the printer is best for long print jobs, where it just keeps printing along. For those clients who just print out a couple pages at a time, we usually will just put in a Lexmark Optra color inkjet, since it will usually finish a small job before the 2550 is even fully warmed up to start a job. Also like a dumptruck, the 2550 can be terribly dirty. After the first toner cart replacement, I was shocked how colorful the insides of the printer were, and yet it kept printing so nicely. I've had other models that a little spilled color toner, and it was an afternoon of cleaning the guts. I had a client "drop" a 2550. I picked it up, vacuumed out the excess toner, set it back up, and it was printing like always. The average color Oki would have been ruined, ready to toss in the garbage. Beware, though: we are somthing of a Citrix shop, and the 2550 just plain does not work with Citrix. In those rare cases, the only color LJ that works consistently is the earlier color 2500. Which you can still buy from HP new, they just don't advertise it and will try to deny at first. Max. From whittake at sbaflorida.com Mon Aug 15 19:22:02 2005 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> References: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> Message-ID: <4301239A.7080106@sbaflorida.com> Brian Ashe wrote: >>Wife is in the market for a new color printer. >> >> > >Funny, I've been looking recently myself. (Note: I am >not Bryan's wife.) I was looking at a $500 model and a >$300 one. IIRC, one was some flavor of HP 2550. Had >networking but I didn't look at RAM, etc. I think it >was PostScript. Looking online, toner for 2550s is >$100 per color. For $500, it probably ships with >"starter rolls"--aka, half-empty. (Or half-full, >depending on your POV. :-) ) The $300 model was, I >think, from Brother, on sale at Office Depot. (Not >saying these are better, just giving you some >background.) Hadn't yet gotten 'round to checking it >out, just saw it in the Sunday paper sales flyers. > > > >>But today, lo'n behold, the Dell 3100cn Color Laser >>was on >>sale for $329.40, about the same price as its 3000cn >>brotheren. >> >> > >Sweet! Are these re-badged HPs? Where do you get >consumables? Dell? HP? Searching CompUSA.com for 'hp >3100' I see only black: >http://www.compusa.com/products/product_info.asp?product_code=50170019&pfp=SEARCH >The same search at CDW comes up empty. I can't get >through to dell.com right now. Have you priced >consumables? What does it work out to, price-per-page? > > > >>And >>we all >>know how India's--er, I mean Dell's support is. ;-> >> >> > >FYI, I recently had the displeasure of going a couple >rounds with some Apple customer (not tech) support >people, all of whom had mild British accents. :-p >Same everywhere, I suppose. (Though Dell brought their >Corporate support people back to TX, IIRC. Baby >steps...) > > > > >____________________________________________________ >Start your day with Yahoo! - make it your home page >http://www.yahoo.com/r/hs > >_______________________________________________ >Pc_support mailing list >Pc_support@matrixlist.com >http://lists.matrixlist.com/mailman/listinfo/pc_support > > > Brian: I suggest that you look at the Lexmark printers. They are Professional grade when they print! HOWEVER, and I say it again, however you should examine their tech sheets to make sure that your flavor of Linux is supported. If not you do not get to utilize all of their excellent programming. I get dissed off at them because of their limited linux support but they are good! Homer Whittaker From ozz at ozz.is-a-geek.net Mon Aug 15 22:01:08 2005 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <20050815220608.51260.qmail@web34103.mail.mud.yahoo.com> References: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> <20050815220608.51260.qmail@web34103.mail.mud.yahoo.com> Message-ID: <20050815220108.45657455.ozz@ozz.is-a-geek.net> On Mon, 15 Aug 2005 15:06:08 -0700 (PDT), "Bryan J. Smith" wrote: > > Brian Ashe wrote: > > > FYI, I recently had the displeasure of going a couple > > rounds with some Apple customer (not tech) support > > people, all of whom had mild British accents. :-p > > Same everywhere, I suppose. (Though Dell brought their > > Corporate support people back to TX, IIRC. Baby > > steps...) > > Were they British? Or were they Hong Kong-like British? And no - I was not one of them #;-D Regards, Ozz. From jasonb at edseek.com Tue Aug 16 00:22:37 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] XFX DoA Issues In-Reply-To: <200507310030.25277.jasonb@edseek.com> References: <200507310030.25277.jasonb@edseek.com> Message-ID: <1447.192.168.0.13.1124166157.squirrel@localhost> Jason Boxman said: > fyi, I am finding a lot of threads about dead XFX cards now that I'm in a > situation where I need to RMA my card. It seems XFX is known to be one of > the worst when it comes to DoA problems and other failures. I've read a few > accounts of people going through multiple RMAs before receiving a working > card. > > Out of the two dozen different companies selling knock off NVidias, I guess > I > picked the worst. > > ;) ROFL. My RMA'd card is dead too. Getting another RMA. Going to see if Monarch will let me get an eVGA 6600GT instead. Same symptoms as before. I tried all the same resolutions. Going to give XFX's 800 # a try tomorrow during the day to see if they have anything useful to contribute. Sigh. From b.j.smith at ieee.org Tue Aug 16 09:38:26 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <4301239A.7080106@sbaflorida.com> References: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> <4301239A.7080106@sbaflorida.com> Message-ID: <1124199506.4679.22.camel@bert64.mobile.smithconcepts.com> On Mon, 2005-08-15 at 19:22 -0400, Homer Whittaker wrote: > Brian: I suggest that you look at the Lexmark printers. They are > Professional grade when they print! I know, I have one, the old Lexmark Optra E310. 6 years-old, it works like a champ at almost 20,000 pages. I take it everywhere with me, having both IEEE1284 and USB ports. I suspect the drum is nearing replacement though. > HOWEVER, and I say it again, however you should examine their tech > sheets to make sure that your flavor of Linux is supported. If not you > do not get to utilize all of their excellent programming. I only buy Postscript printers that do the Postscript at the device-end (with on-board intelligence) for maximum compatibility. Both my current Lexmark Optra E310 and HP LaserJet 1220se do Postscript Level 2 and have 100% Linux compatibility. The LaserJet 1220se is even supported for scanning under Linux: http://hpoj.sourceforge.net/suplist.shtml This Dell 3100cn is a Postscript Level 3 printer. > I get dissed off at them because of their limited linux support but they > are good! -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 16 09:41:28 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <200508151912.38782.mflang@bellsouth.net> References: <20050815212209.99800.qmail@web34012.mail.mud.yahoo.com> <200508151912.38782.mflang@bellsouth.net> Message-ID: <1124199688.4679.26.camel@bert64.mobile.smithconcepts.com> On Mon, 2005-08-15 at 19:12 -0400, Max F Lang wrote: > Brian, we use the LJ2550 at work for printing color lab reports in > physician offices. This is a printer I'm coming to know well... > The built-in print server is an onboard DirectJet 625. You can't > remove it, but it's exactly the same thing firmware-wise. Does the > usual IP, IPX, IPP and AppleTalk, with both an easy browser and > telnet session based interface. They come with 32M ram standard. > The toners included in the box are exactly the same carts you buy. > They are rated for 6K pages in color, and the black cart is rated > for 7K. And that's about what we get. They're not noobie carts at > all, and yes, replacements are $$$, even at our corporate discount. I think this Dell 3100cn is different. It's a Fuji-Xerox engine, cartridges are 4,000 sheets/each. > The 2550 is a dumptruck of a printer: big, heavy, and awful slow to > start. It usually takes about 20 seconds to start printing. But > just like a dumptruck, once it starts going, it doesn't want to > stop. In fact, the printer is best for long print jobs, where it > just keeps printing along. For those clients who just print out a > couple pages at a time, we usually will just put in a Lexmark Optra > color inkjet, since it will usually finish a small job before the > 2550 is even fully warmed up to start a job. I think most Lasers are like that in general. 15+ seconds to first page, a few break 10 seconds, but rarely. Most colors are 20+ seconds to first page. I'm still going to have my Lexmark Optra E310 connected to my PC and my wife's HP LaserJet 1220se connected to her PC, so those devices will still be options. The new Dell 3100cn will be located in between our two rooms (connected to the network) so either of us can grab things off of it. > Beware, though: we are somthing of a Citrix shop, and the 2550 just > plain does not work with Citrix. In those rare cases, the only > color LJ that works consistently is the earlier color 2500. Which > you can still buy from HP new, they just don't advertise it and > will try to deny at first. Worst case I will use WTS, but not Citrix, at home. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 16 10:00:24 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Dell 3100cn Color LaserJet with PostScript and LPD for $329.40 shipped! In-Reply-To: <20050815203037.83270.qmail@web34104.mail.mud.yahoo.com> References: <20050815203037.83270.qmail@web34104.mail.mud.yahoo.com> Message-ID: <1124200824.4679.39.camel@bert64.mobile.smithconcepts.com> On Mon, 2005-08-15 at 13:30 -0700, Bryan J. Smith wrote: > The 3100cn, unlike the 3000cn, comes with a host of additions > -- beyond just the paper trays, FULL POSTSCRIPT LEVEL 3 > SUPPORT! Driven with an on-board 300MHz RISC processor and > 64MB of RAM (expandable to 576MB), this is a serious printer > for a low price. In doing some research, I found this printer takes a standard PC133 SO- DIMM like most Dell notebooks. So instead of paying Dell $60, 70 or 150 + for their 128, 256 or 512MiB, respectively, I found an aftermarket dealer who specializes in 1:1 Dell parts. They offere the same for $15, 30 and 75, respectively. I grabbed the $30 256MiB module, although they clearly make up for it on shipping ($12 for UPS Ground!). Still, can't beat the price overall. I heard you need at least 128MiB to print a 600dpi 8x10" photo. That sounds about right just in raw pixels: (8in)(600/in) * (10in)(600/in) * [(24b) / (8b/B)] = 86.4MB I'm sure the Postscript overhead puts its at least twice as much. According to the reviews, 8x10" was doable with 192MiB. I'll have 320MiB. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 16 12:28:54 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] More on Intel's shift back to i686 ... Message-ID: <20050816162855.14752.qmail@web34109.mail.mud.yahoo.com> More technical details are to follow at IDF. Looks like the last P4-design will be the 65nm "Pressler." eWeek Story: http://www.eweek.com/article2/0,1895,1848474,00.asp?kc=ewnws081605dtx1k0000599 -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Tue Aug 16 12:50:20 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] XFX DoA Issues In-Reply-To: <1447.192.168.0.13.1124166157.squirrel@localhost> References: <200507310030.25277.jasonb@edseek.com> <1447.192.168.0.13.1124166157.squirrel@localhost> Message-ID: <1973.192.168.0.13.1124211020.squirrel@localhost> Jason Boxman said: > Jason Boxman said: >> fyi, I am finding a lot of threads about dead XFX cards now that I'm in a >> situation where I need to RMA my card. It seems XFX is known to be one of >> the worst when it comes to DoA problems and other failures. I've read a >> few >> accounts of people going through multiple RMAs before receiving a working >> card. >> >> Out of the two dozen different companies selling knock off NVidias, I >> guess >> I >> picked the worst. >> >> ;) > > ROFL. > > My RMA'd card is dead too. Getting another RMA. Going to see if Monarch > will let me get an eVGA 6600GT instead. > > Same symptoms as before. I tried all the same resolutions. Going to give > XFX's 800 # a try tomorrow during the day to see if they have anything > useful to contribute. Sigh. I called Monarch and they were well aware of the issues with the XFX cards. I'm getting a store credit, which I'll use to buy an eVGA which has a rebate currently. With shipping back RMA'd cards twice it'll end up costing the same $155 I got my borked XFX card. XFX: Never again. http://edseek.com/archives/2005/08/16/xfx-video-cards-6600gt-dont-waste-your-money/ From b.j.smith at ieee.org Tue Aug 16 12:58:32 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Microsoft to invade Adobe's stronghold, tying XAML to Avalon ... Message-ID: <20050816165832.76022.qmail@web34102.mail.mud.yahoo.com> Microsoft Next Target: Adobe http://thebs413.blogspot.com/2005/08/microsofts-next-target-adobe.html -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Tue Aug 16 13:04:30 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <1973.192.168.0.13.1124211020.squirrel@localhost> Message-ID: <20050816170430.96236.qmail@web34105.mail.mud.yahoo.com> Jason Boxman wrote: > I called Monarch and they were well aware of the issues > with the XFX cards. > I'm getting a store credit, which I'll use to buy an eVGA > which has a rebate currently. With shipping back RMA'd > cards twice it'll end up costing the same $155 I got my > borked XFX card. XFX: Never again. Sorry about that. I've yet to buy a XFX video card (only my NetCell SR5000, which I'm still playing with under Linux for my A/V server -- although 2.6.12.x releases now have the ATA support). I have bought nVidia cards from BFG (my parent's current GF4Ti4200), eVGA (my current GF6800GT PCIe), MSI (my A/V server's current GF4Ti4400-VIVO), PNY (my wife's current GF6800GT AGP) and several others. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Tue Aug 16 13:19:57 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <20050816170430.96236.qmail@web34105.mail.mud.yahoo.com> References: <20050816170430.96236.qmail@web34105.mail.mud.yahoo.com> Message-ID: <200508161319.57628.jasonb@edseek.com> On Tuesday 16 August 2005 13:04, Bryan J. Smith wrote: > Sorry about that. I've yet to buy a XFX video card (only my > NetCell SR5000, which I'm still playing with under Linux for > my A/V server -- although 2.6.12.x releases now have the ATA > support). > > I have bought nVidia cards from BFG (my parent's current > GF4Ti4200), eVGA (my current GF6800GT PCIe), MSI (my A/V > server's current GF4Ti4400-VIVO), PNY (my wife's current > GF6800GT AGP) and several others. I just ordered the eVGA 6600GT. The BFG was quite a bit more. This time I'm going to take the free shipping and just wait a week. I am no longer in any hurry. I've gotten bored playing with video cards and lost interest in playing SH3 and BF2. -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Tue Aug 16 17:07:04 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <200508161319.57628.jasonb@edseek.com> Message-ID: <20050816210704.3558.qmail@web34103.mail.mud.yahoo.com> Jason Boxman wrote: > I've gotten bored playing with video cards and lost > interest in playing SH3 and BF2. Well, I should have waited 3 months for the GeForce 7800GT PCIe, which is only $50 more than what I paid for my 6800GT PCIe in April. At 1280x1024 in 4xAA, BF2 can be 2x as fast as my 6800GT. I'm very tempted to upgrade and sell my 6800GT PCIe for $250, if anyone wants it for that price. Or I might just upgrade my wife to PCIe once and for all, and sell her 6800GT AGP. Or I might just be content for now. |-< -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Tue Aug 16 17:13:57 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <20050816210704.3558.qmail@web34103.mail.mud.yahoo.com> References: <20050816210704.3558.qmail@web34103.mail.mud.yahoo.com> Message-ID: <200508161713.57449.jasonb@edseek.com> On Tuesday 16 August 2005 17:07, Bryan J. Smith wrote: > Jason Boxman wrote: > > I've gotten bored playing with video cards and lost > > interest in playing SH3 and BF2. > > Well, I should have waited 3 months for the GeForce 7800GT > PCIe, which is only $50 more than what I paid for my 6800GT > PCIe in April. At 1280x1024 in 4xAA, BF2 can be 2x as fast > as my 6800GT. Why would it matter, unless you sync frames to your refresh rate? I'll never understand the insatiable question for faster and faster framerates. I know in Quake 3 it was 'spoitable for higher jumping, if you pegged to some key framerate, but the lowest I think was 43 FPS, which isn't that high. > I'm very tempted to upgrade and sell my 6800GT PCIe for $250, > if anyone wants it for that price. Or I might just upgrade > my wife to PCIe once and for all, and sell her 6800GT AGP. > > Or I might just be content for now. |-< Ouch, that's expensive. ;) From b.j.smith at ieee.org Tue Aug 16 17:24:01 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <200508161713.57449.jasonb@edseek.com> Message-ID: <20050816212401.40347.qmail@web34107.mail.mud.yahoo.com> Jason Boxman wrote: > Why would it matter, unless you sync frames to your refresh > rate? I'll never understand the insatiable question for > faster and faster framerates. I'm talking about the difference between 20fps and 40fps! Dude, when you start jacking up the visual quality, you start taking hits in the FPS area. ;-> > Ouch, that's expensive. ;) The GeForce 6800GT is typically still over $350 for PCIe (maybe as low as $300 for AGP -- so I'd probably sell my wife's AGP at $200). The GeForce 7800GT is available for around $400 for PCIe (no AGP, and probably won't be for some time). -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Tue Aug 16 17:25:51 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <20050816212401.40347.qmail@web34107.mail.mud.yahoo.com> References: <20050816212401.40347.qmail@web34107.mail.mud.yahoo.com> Message-ID: <200508161725.52042.jasonb@edseek.com> On Tuesday 16 August 2005 17:24, Bryan J. Smith wrote: > Jason Boxman wrote: > > Why would it matter, unless you sync frames to your refresh > > rate? I'll never understand the insatiable question for > > faster and faster framerates. > > I'm talking about the difference between 20fps and 40fps! > Dude, when you start jacking up the visual quality, you start > taking hits in the FPS area. ;-> I always lag behind in the monitor department. Only had a 17" CRT for about two years now. I might get a 19" LCD in December, if things go well. If they're not going well, I won't. I usually game at 800x600, even on my 17". ;) I consider it an upgrade, as I used to game on a 486 at 640x480, and later at the same resolution on a 14" on a Matrox G400. Probably best I don't know how good visual quality is on a 19" flat panel and a 7xxx series NVidia card. I'm sure I could never go back. From b.j.smith at ieee.org Tue Aug 16 17:33:50 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <200508161725.52042.jasonb@edseek.com> Message-ID: <20050816213350.70768.qmail@web34115.mail.mud.yahoo.com> Jason Boxman wrote: > I usually game at 800x600, even on my 17". ;) Ahh, well, with a LCD, you're stuck at resolution, or exactly 1/2 resolution, or things look ugly. So I game at 1280x1024 (or 1280x960), and sometimes at 640x480 if I can't push it that hard at 1280x1024. I really don't like going to 640x480, so I like to have a good card. > Probably best I don't know how good visual quality is on a > 19" flat panel and a 7xxx series NVidia card. I'm sure I > could never go back. Don't come to the next LEAP Installfest then. ;-> Although I'll only have my 25ms (40Hz) refresh LCD on me, I'll show you how good a GeForce 6800GT looks at 4xAA. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Tue Aug 16 17:36:45 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <20050816213350.70768.qmail@web34115.mail.mud.yahoo.com> References: <20050816213350.70768.qmail@web34115.mail.mud.yahoo.com> Message-ID: <200508161736.45276.jasonb@edseek.com> On Tuesday 16 August 2005 17:33, Bryan J. Smith wrote: > Don't come to the next LEAP Installfest then. ;-> > Although I'll only have my 25ms (40Hz) refresh LCD on me, > I'll show you how good a GeForce 6800GT looks at 4xAA. I'm in Gainesville, so I probably can't make it. Never been to one. By the time I get up it's usually over. ;) From jasonb at edseek.com Tue Aug 16 17:39:09 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Case w/ good airflow and PSU with good +12V? Message-ID: <200508161739.09512.jasonb@edseek.com> Since I'm going to buy a new mainboard and CPU, I think I'm just going to make it a system. Any thoughts on a quality case with lots of airflow? My current case completely sucks for airflow, so it's not really appropriate for my upgrade anyway. (It tends to run cooler with the case side off, a sure sign of problems.) I'm also going to snarf up a new PSU. While I wouldn't mind a quieter PSU, one of average noise is fine since the whole system will be in a noisy server room (relatively speaking) which I can't quiet down in any reasonable way. Thanks. From b.j.smith at ieee.org Tue Aug 16 18:04:00 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Re: Case w/ good airflow and PSU with good +12V? In-Reply-To: <200508161739.09512.jasonb@edseek.com> Message-ID: <20050816220400.66035.qmail@web34110.mail.mud.yahoo.com> Jason Boxman wrote: > Since I'm going to buy a new mainboard and CPU, I think I'm > just going to make it a system. Any thoughts on a quality > case with lots of airflow? My current case completely > sucks for airflow, so it's not really appropriate for my > upgrade anyway. (It tends to run cooler with the case side > off, a sure sign of problems.) You know my opinion on boxes and airflow: http://thebs413.blogspot.com/2005/08/small-enough-form-factor-pc.html Lots of MicroATX mainboards can be had for cheap (<$50) for older Socket-462 platforms. Otherwise, I listed my favorite MicroATX for Socket-939 ($79) -- one of the few available right now. > I'm also going to snarf up a new PSU. While I wouldn't > mind a quieter PSU, one of average noise is fine since the > whole system will be in a noisy server room (relatively > speaking) which I can't quiet down in any reasonable way. Other fans are typically more of an issue than the PS' fan. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dmckenna at thelimucompany.com Wed Aug 17 12:00:33 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859290@tlc001.tlcusa.thelimucompany.com> We've got an Active Directory setup on a Windows 2000 network. We've got custom OUs for managing the directory structure. I've tried adding some new security groups that in terms of the organization structure of the company are the superior of the next one, e.g. customer service manager is the parent of the customer service supervisors, etc. I've got e.g. the Supervisor object and I've tried adding it to the Manager's "members" list but absolutely no user groups are shown! Is there a flag set wrong somewhere to not allow me do this, or is it a limitation of Windows 2000? Thanks. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From dmckenna at thelimucompany.com Wed Aug 17 12:04:30 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859291@tlc001.tlcusa.thelimucompany.com> OK, to solve my own problem, to do this the group has to be created as a "domain local" group and not a "global" group. Sure, that makes sense :-\ -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From dmckenna at thelimucompany.com Wed Aug 17 12:13:45 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859292@tlc001.tlcusa.thelimucompany.com> > OK, to solve my own problem, to do this the group has to be > created as a "domain local" group and not a "global" group. > Sure, that makes sense :-\ OK, that doesn't seem to be it either. Argh. Anyone have any ideas before I go postal and install Netware 4.11 :-P -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From glaiacona at aikencountysc.gov Wed Aug 17 12:18:33 2005 From: glaiacona at aikencountysc.gov (George Laiacona) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: Go Netware 6.5 on the Linux core. AD is giving me fits too!. George. >>> dmckenna@thelimucompany.com 08/17/05 12:13 PM >>> > OK, to solve my own problem, to do this the group has to be > created as a "domain local" group and not a "global" group. > Sure, that makes sense :-\ OK, that doesn't seem to be it either. Argh. Anyone have any ideas before I go postal and install Netware 4.11 :-P -- 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 b.j.smith at ieee.org Wed Aug 17 12:24:23 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859290@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050817162423.27406.qmail@web34102.mail.mud.yahoo.com> Damien McKenna wrote: > We've got an Active Directory setup on a Windows 2000 > network. We've got custom OUs for managing the directory > structure. I've tried adding some new security groups that > in terms of the organization structure of the company are > the superior of the next one, e.g. customer service > manager is the parent of the customer service supervisors, > etc. I've got e.g. the Supervisor object and I've tried > adding it to the Manager's "members" list but absolutely > no user groups are shown! Is there a flag set wrong > somewhere to not allow me do this, or is it a limitation > of Windows 2000? Thanks. Are you talking a "parent OU" or something else? I guess what I'm trying to figure out is if you are trying to delegate authority over an _ADS_ subtree to a user, or merely add a user to a group so they can manage files in a NTFS/share subtree? If the former, you delegate authority using the appropriate Wizard. If the latter, you setup Domain Local groups on the NTFS/share subtree, and then assign Users to Global groups which are assigned to Domain Local groups. Damien McKenna wrote: > OK, to solve my own problem, to do this the group has to be > created as a "domain local" group and not a "global" > group. Sure, that makes sense :-\ Hold on, if you don't know the difference between "Domain Local" and "Global" groups, you should read up on them. 1. Users should _always_ be placed in Global groups 2. Domain Local groups should be applied to service objects (e.g., share, filesystem, etc...) 3. Global groups should then be assigned to Domain Local In other words: 1. _Never_ assign Global groups to service objects (e.g., share, filesystem, etc...) 2. _Avoid_ assigning Users to Domain Local groups Microsoft's official line revolves around the fact that people used different "resource" and "user" groups back in the CIFS days so you could assign access to a local domain object for another domain's users/groups. In reality, this is a limitation when you have multiple domains, which "Domain Local" now solves nicely. Technical insight: Domain Local group ensures SIDs are always _local_ to the server -- i.e., the SIDs on ACL of shares/files of a NTFS volume are _local_ domain SIDs. If not, NTFS can be corrupted because of its _flawed_ design (which is why CairoFS, now WinFS, is still very much needed, even if still quite vaporware). That's why you _always_ assign Domain Local groups of the local domain to the service resources of servers local to that domain. They you assign Global groups, which could be from any domain, to that Domain Local group. Again, _avoid_ directly assigning Global groups directly to service resources such as shares and files. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Wed Aug 17 12:37:38 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: <20050817163738.46657.qmail@web34115.mail.mud.yahoo.com> "Bryan J. Smith" wrote: > Hold on, if you don't know the difference between "Domain > Local" and "Global" groups, you should read up on them. > 1. Users should _always_ be placed in Global groups > 2. Domain Local groups should be applied to service > objects > (e.g., share, filesystem, etc...) > 3. Global groups should then be assigned to Domain Local > In other words: > 1. _Never_ assign Global groups to service objects (e.g., > share, filesystem, etc...) > 2. _Avoid_ assigning Users to Domain Local groups Once again, I will recommend what I call the "Universal Resource" for Windows 2000 Network operations: The Exam 70-218 Self-Paced Training Kit: http://www.amazon.com/exec/obidos/tg/detail/-/0735617767/ Chapter 8 has ~40 pages that gives you the "low down" on how groups work, basic limitations (nesting, local/global, universal, etc...), etc... If you really want to get uber-anal and be Microsoft'ized into "I am an Active Directory engineer," you can go find some 70-217 and/or 70-219 books for Active Directory implementation and design, respectively (or the 2003 equivalents). And you can even go through the 70-216 and/or 70-221 for Windows Networking implementation and design, respectively. But for "real world" usage, this "single book" for 70-218 designed as the "single exam" for MCSA (instead of the above 4 for MCSE) is the "ultimate intro" to everything, and those 40 pages on ADS groups will help you out tremendously, along with all the other chapers of the book on other details. It's basic stuff that you've "gotta know" to maintain ADS 2000. Unfortunately, Microsoft seems to have not created such an excellent single exam/guide for Windows Server 2003. Too bad, the 70-218 was the only exam/guide I think anyone needs. -- Bryan P.S. I have loaned out both of my copies, sorry. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dmckenna at thelimucompany.com Wed Aug 17 12:41:11 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859293@tlc001.tlcusa.thelimucompany.com> > Once again, I will recommend what I call the "Universal > Resource" for Windows 2000 Network operations: I'll take a look, I might have a book for that at home. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From dmckenna at thelimucompany.com Wed Aug 17 12:48:00 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:12 2006 Subject: [Pc_Support] Nested group security in Active Directory (Win2000)? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859294@tlc001.tlcusa.thelimucompany.com> Some more detail, might help explain what I want to do. We have the following OUs: Root OU=Our Company -> OU=User Accounts -> OU=Departments -> OU=Customer Service -> User=Joe User -> User=Jo User -> OU=Operations -> OU=Technology -> User=Damien McKenna -> OU=Aliases -> alias email addresses.. -> OU=Groups -> OU=Distribution -> email distribution addresses -> OU=Security -> OU=Departments -> Group=Customer Service -> Special Access -> Group=Customer Service Supervisors -> Group=Customer Service Managers What I'd like to do is make Group=Customer Service Managers a member of Group=Customer Service Supervisors so that e.g. they can read all of the supervisor's email, access their files, etc. So what do I need? -- 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 Aug 17 13:03:42 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] RE: Nested group security in Active Directory (Win2000)? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859294@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050817170342.39073.qmail@web34102.mail.mud.yahoo.com> Damien McKenna wrote: > Some more detail, might help explain what I want to do. > We have the following OUs: > Root > OU=Our Company > -> OU=User Accounts > -> OU=Departments > -> OU=Customer Service > -> User=Joe User > -> User=Jo User > -> OU=Operations > -> OU=Technology > -> User=Damien McKenna > -> OU=Aliases > -> alias email addresses.. > -> OU=Groups > -> OU=Distribution > -> email distribution addresses > -> OU=Security > -> OU=Departments > -> Group=Customer Service > -> Special Access > -> Group=Customer Service Supervisors > -> Group=Customer Service Managers Not sure this is how you should tree your ADS/OU, but I'm not in charge of your network. > What I'd like to do is make Group=Customer Service Managers > a member of Group=Customer Service Supervisors so that e.g. > they can read all of the supervisor's email, access their > files, etc. So what do I need? If your domain is in "mixed mode" you can_not_ nest Global or Domain Local groups. You can only place Global groups into Domain Local groups. It has to do with CIFS compatibility. If you put the domain in "native mode," you can nest Global or Domain Local groups with each other. You can also use the Universal Group (whose use is recommended only sparingly). As I mentioned, you should learn the proper use of Domain Local and Global groups regardless of mode, because they help avoid NTFS file corruption due to unavailable SIDs on ACL and other NTFS filesystem information. While Microsoft doesn't tell you that technical fact (the limitation of NTFS' design is why you should use Domain Local and Global as approprate), but they do give you sound planning on how to use both Domain Local and Global groups regardless. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Wed Aug 17 13:09:59 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] RE: Nested group security in Active Directory (Win2000)? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859293@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050817170959.53222.qmail@web34110.mail.mud.yahoo.com> Damien McKenna wrote: > I'll take a look, I might have a book for that at home. I'd stop by your local Barnes'n Noble, Books-A-Million, Borders, etc... this evening and spend 45-60 minutes reading the ~40 pages of Chapter 8 of the book. The stores _always_ have the Microsoft 70-218 book and I high recommend you undersatnd how groups work before dorking with them any more. It'll be the best 45-60 minutes you'll spend in a month. It'll also give you some basic planning information too. Also be sure to read up on delgation of authority in the ADS tree, so you can have managers/supervisors adding/removing/modifying portions of the ADS tree. It's actually very straight-forward wizard, although there are a few gotchas. BTW, the 70-218 is "to-the-point" on everything for the MCSA that the MCSE gets in 4 exams, so it doesn't always give you things that the 70-217 and 70-219 would which focus solely on ADS implementation and design -- including little "case studies." Most of the "case studies" are rather simplistic and not very "real world" anyway, hence why I don't recommend them unless you are taking the actual exams. I typically recommend people hit this and just search for "good design" or "case study" when it comes to the ADS tree, delegation, etc... ;-ppp http://www.google.com/microsoft.html -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From jasonb at edseek.com Wed Aug 17 17:18:25 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Appropriate PSU for Socket 754 and 6600GT AGP? Message-ID: <200508171718.25872.jasonb@edseek.com> I'm looking at PSUs on newegg and, like cases, you have about 200 choices. I'm focusing on the ones that actually have at least 10 reviews and immediately discarding any that mention DoA or serious reliability issues. I am curious what would be adequate ratings for the 3.3, 5, and 12 rails? Should I buy a dual +12V rail PSU? Few on newegg have any reviews at all, so they're apparently not that popular yet. The dual +12V PSUs I looked at boasted lower +5V and +3.3V ratings, around say 20A, versus the single +12V PSUs that generally have > 20A on those rails. What's a reasonable efficiency level? I'm definitely going to get one with 24-pin ATX and SATA power connectors for future expansion, so I want this to be a PSU I'll have for several years into the future. (The Socket 754 board I get probably won't have 24-pin, although I haven't chosen a final candidate yet.) Thanks. From b.j.smith at ieee.org Wed Aug 17 20:39:01 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <200508171718.25872.jasonb@edseek.com> References: <200508171718.25872.jasonb@edseek.com> Message-ID: <1124325541.4737.15.camel@bert64.mobile.smithconcepts.com> On Wed, 2005-08-17 at 17:18 -0400, Jason Boxman wrote: > I'm looking at PSUs on newegg and, like cases, you have about 200 choices. > I'm focusing on the ones that actually have at least 10 reviews and > immediately discarding any that mention DoA or serious reliability issues. > I am curious what would be adequate ratings for the 3.3, 5, and 12 rails? > Should I buy a dual +12V rail PSU? Few on newegg have any reviews at all, so > they're apparently not that popular yet. The dual +12V PSUs I looked at > boasted lower +5V and +3.3V ratings, around say 20A, versus the single +12V > PSUs that generally have > 20A on those rails. They are typically ATX 2.0. ATX 1.0 and ATX 2.0 mainboards are designed differently, with different voltage regulators based on the change in what the power supply is providing. > What's a reasonable efficiency level? > I'm definitely going to get one with 24-pin ATX and SATA power connectors for > future expansion, so I want this to be a PSU I'll have for several years into > the future. (The Socket 754 board I get probably won't have 24-pin, although > I haven't chosen a final candidate yet.) I'd go cheap, but at least UL Listed. Get a decent ATX 1.0 power supply with a single +12V rail, one that gives you quite a bit on +12V so you can handle the PS' Molex connector, while still getting quite a bit on the +3.3V and +5V for the mainboard/CPU. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Wed Aug 17 22:01:12 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <1124325541.4737.15.camel@bert64.mobile.smithconcepts.com> References: <200508171718.25872.jasonb@edseek.com> <1124325541.4737.15.camel@bert64.mobile.smithconcepts.com> Message-ID: <1124330472.4737.54.camel@bert64.mobile.smithconcepts.com> On Wed, 2005-08-17 at 19:39 -0500, Bryan J. Smith wrote: > They are typically ATX 2.0. ATX 1.0 and ATX 2.0 mainboards are designed > differently, with different voltage regulators based on the change in > what the power supply is providing. > I'd go cheap, but at least UL Listed. Get a decent ATX 1.0 power supply > with a single +12V rail, one that gives you quite a bit on +12V so you > can handle the PS' Molex connector, while still getting quite a bit on > the +3.3V and +5V for the mainboard/CPU. Couple of notable ATX 1.0 power supplies ... First off, under $35, there is the 400W (430W peak) COOLER MASTER eXtreme Power RS-430-MSR/P for $33.99: http://www.newegg.com/Product/Product.asp?Item=N82E16817171007 +3.3V@25A, +5V@32A, +12V@19A Secondly, although only at 320W, but it seems to have all the right amps with a split +12V on ATX 1.0 (although it could be an ATX 2.0 -- having trouble verifying) ENERMAX Noisetaker EG325P-VE for $38.99: http://www.newegg.com/Product/Product.asp?Item=N82E16817103449 +3.3V@24A, +5V@24A, +12V1@16A, +12V2@16A (20A V1+V2 max) Third, there is the 420W Thermaltake Silent PurePower TT-420AD(DUAL FAN) at $39.00. http://www.newegg.com/Product/Product.asp?Item=N82E16817153006 +3.3V@30A, +5V@40A, +12V@18A Although there is one ATX 2.0 for just over $40 from Cooler Master, if you don't mind rebates ($61.99 - $20 mail-in rebate): http://www.newegg.com/Product/Product.asp?Item=N82E16817171005 +3.3V@30A, +5V@35A, +12V1@18A, +12V2@16A Specs page is here: http://www.coolermaster.com/index.php?url_place=product&p_serial=RS-450-ACLX Just a sample of brands I personally like. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Wed Aug 17 22:29:56 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <1124330472.4737.54.camel@bert64.mobile.smithconcepts.com> References: <200508171718.25872.jasonb@edseek.com> <1124325541.4737.15.camel@bert64.mobile.smithconcepts.com> <1124330472.4737.54.camel@bert64.mobile.smithconcepts.com> Message-ID: <200508172229.57133.jasonb@edseek.com> On Wednesday 17 August 2005 22:01, Bryan J. Smith wrote: > On Wed, 2005-08-17 at 19:39 -0500, Bryan J. Smith wrote: > > They are typically ATX 2.0. ATX 1.0 and ATX 2.0 mainboards are designed > > differently, with different voltage regulators based on the change in > > what the power supply is providing. > > I'd go cheap, but at least UL Listed. Get a decent ATX 1.0 power supply > > with a single +12V rail, one that gives you quite a bit on +12V so you > > can handle the PS' Molex connector, while still getting quite a bit on > > the +3.3V and +5V for the mainboard/CPU. > > Couple of notable ATX 1.0 power supplies ... > > First off, under $35, there is the 400W (430W peak) COOLER MASTER > eXtreme Power RS-430-MSR/P for $33.99: > http://www.newegg.com/Product/Product.asp?Item=N82E16817171007 > +3.3V@25A, +5V@32A, +12V@19A > > Secondly, although only at 320W, but it seems to have all the right amps > with a split +12V on ATX 1.0 (although it could be an ATX 2.0 -- having > trouble verifying) ENERMAX Noisetaker EG325P-VE for $38.99: > http://www.newegg.com/Product/Product.asp?Item=N82E16817103449 > +3.3V@24A, +5V@24A, +12V1@16A, +12V2@16A (20A V1+V2 max) > > Third, there is the 420W Thermaltake Silent PurePower TT-420AD(DUAL FAN) > at $39.00. > http://www.newegg.com/Product/Product.asp?Item=N82E16817153006 > +3.3V@30A, +5V@40A, +12V@18A > > Although there is one ATX 2.0 for just over $40 from Cooler Master, if > you don't mind rebates ($61.99 - $20 mail-in rebate): > http://www.newegg.com/Product/Product.asp?Item=N82E16817171005 > +3.3V@30A, +5V@35A, +12V1@18A, +12V2@16A > > Specs page is here: > > http://www.coolermaster.com/index.php?url_place=product&p_serial=RS-450-ACL >X Yeah, I found a review of a variant. I actually have a Cooler Master right now I think. It's been working fine, although I have no way of verifying the differentials on my rails. http://www.silentpcreview.com/article205-page1.html Even though it fails the silence test, in this room I couldn't tell the difference anyway. I think I might buy another Cooler Master since my 3xxW one is fine for the past 1.5 years and deal with the rebates. Having had like $500 in outstanding rebates once, I'm pretty used to them these days. It's too bad there hasn't been opportunity to use and abuse Office Max this past year. A few years ago I ended up with so much free stuff after rebates from them. CDRW, tons of CDRs, CDR cases, $25 40GB disk, and so on... -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Thu Aug 18 21:18:39 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: AOpen F90GS 19" LCD for $240 -- Not quite the same. In-Reply-To: <20050814042120.51114.qmail@web34106.mail.mud.yahoo.com> References: <20050814042120.51114.qmail@web34106.mail.mud.yahoo.com> Message-ID: <1124414320.6072.10.camel@bert64.mobile.smithconcepts.com> On Sat, 2005-08-13 at 21:21 -0700, Bryan J. Smith wrote: > I want to say it looks like the same, ultra-portable ("fold > base back and put in box") design as the Sceptre 19" I have. > I really want a 12ms response LCD to replace my 25ms Sceptre > for video/gaming, while being the exact same resolution (and > close to same height) so it can be the primary head with the > Sceptre as the secondary head (which would be off for > gaming). Okay, first off, it's not a "fold base back" design, so it must be a newer AOpen design. But other than that, it's pretty light, portable and comes in a reusable box. But you do have to disconnect the base to do so. It's not difficult (a crapload easier than a Dell), but the plastic tab might break off eventually. I still like the Sceptre's base, which almost seconds as a "handle" for portability. The edges of the encasing have a few millimeters off from the Sceptre. I can't tell if it's a better design (slightly smaller), or maybe thinner (and less absorbent of damage). In any case, it does look "leaner." Other than that, it's basically the same -- all internal PS (3-prong out), internal speakers, DVI and VGA. It seems like the AOpen brighter, although it clearly has more backlight when black, not always ideal. I can't tell if my Sceptre is better on text, or if it's just older (almost a year now) and not as sharp. But definitely notice the difference going from the 25ms (40fps) refresh of my Sceptre X9G-Komodo II* to the 12ms (83.33fps) refresh of this AOpen F90GS in gaming. No more "ghosting," period. At 50ms (20fps), it's intolerable. At 25ms (40fps), it's noticeable. At 20ms (50fps) it's still a slight issue. At 16ms (62.5fps) it's pretty much gone. At 12ms (83.33fps), it's completely gone. I'm sure at 8ms (125fps), it's overkill. I'm just happy to be dual-head again, now all LCD. I don't lug my CRTs with me. And I was sure to get a video card that had dual-DVI outs too. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Thu Aug 18 22:23:19 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <200508172229.57133.jasonb@edseek.com> References: <200508171718.25872.jasonb@edseek.com> <1124330472.4737.54.camel@bert64.mobile.smithconcepts.com> <200508172229.57133.jasonb@edseek.com> Message-ID: <200508182223.19814.jasonb@edseek.com> On Wednesday 17 August 2005 22:29, Jason Boxman wrote: > Yeah, I found a review of a variant. I actually have a Cooler Master right > now I think. It's been working fine, although I have no way of verifying > the differentials on my rails. > > http://www.silentpcreview.com/article205-page1.html > I went with these. MB BIOSTAR NF3 SATA AGP NF325-A7 RT - Retail POWER SUPPLY CMAX 400WCX-400B BK RT - Retail CPU AMD|SEMPRON64 2800+ RTL - Retail PQI POWER 512MB DDR PC3200 - Retail x 2 CASE MXTP|MID 0W 2BAFAN CSX147K GRY - Retail For $60 and about $120 I could've gone for 939, instead of $50 and $76, but it seems silly to buy 939 with AGP and I don't want to immediately resell my brand new eVGA 6600GT AGP for a PCIe version. (And sell it at a loss.) I'm sure 754 will be fast enough for me, anyway, even if 939 is noticeably faster than 754 with a Sempron. -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Thu Aug 18 22:32:50 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <200508182223.19814.jasonb@edseek.com> References: <200508171718.25872.jasonb@edseek.com> <1124330472.4737.54.camel@bert64.mobile.smithconcepts.com> <200508172229.57133.jasonb@edseek.com> <200508182223.19814.jasonb@edseek.com> Message-ID: <1124418770.6072.77.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-18 at 22:23 -0400, Jason Boxman wrote: > MB BIOSTAR NF3 SATA AGP NF325-A7 RT - Retail Yeah, that's the one I would go for (in an ATX). > POWER SUPPLY CMAX 400WCX-400B BK RT - Retail Yep, $35 gets you there now. > CPU AMD|SEMPRON64 2800+ RTL - Retail > PQI POWER 512MB DDR PC3200 - Retail x 2 > CASE MXTP|MID 0W 2BAFAN CSX147K GRY - Retail Cheap, but it has 2 intake (one above CPU), 2 outtake, does the jobbie. > For $60 and about $120 I could've gone for 939, instead of $50 and $76, but it > seems silly to buy 939 with AGP and I don't want to immediately resell my > brand new eVGA 6600GT AGP for a PCIe version. (And sell it at a loss.) I'm > sure 754 will be fast enough for me, anyway, even if 939 is noticeably faster > than 754 with a Sempron. Yep. This will do for now. Save $200 and upgrade later. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Thu Aug 18 22:36:14 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <1124418770.6072.77.camel@bert64.mobile.smithconcepts.com> References: <200508171718.25872.jasonb@edseek.com> <1124330472.4737.54.camel@bert64.mobile.smithconcepts.com> <200508172229.57133.jasonb@edseek.com> <200508182223.19814.jasonb@edseek.com> <1124418770.6072.77.camel@bert64.mobile.smithconcepts.com> Message-ID: <1124418974.6072.82.camel@bert64.mobile.smithconcepts.com> BTW, I hope you didn't take offense to my prior posts. I just didn't want to see you go down Socket-462 which is _dead_. AMD has Socket-754 planned out until next year, it's not dying at all. Socket-462 basically died last year, and as of now, is no longer being manufactured. Yes, I have said I don't think Socket-754 is worth it versus Socket-939, but that _assumes_ you're buying a new, higher-end PCIe card and power supply already. If you already have an AGP card, and you want to save money on the PS, etc..., then Socket-754 with Sempron is a good move. Especially now that Sempron is available in x86-64 -- at least for those of us running Linux. -- Bryan P.S. Has everyone noted how many pieces of hardware and software are now saying "not supported by XP 64-bit"? He he he. ;-> -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Thu Aug 18 22:48:52 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Opteron 100 goes Socket-939 (unreg-mem), ServerWorks HT1000 to go it alone ... Message-ID: <1124419732.6072.92.camel@bert64.mobile.smithconcepts.com> To start, AMD is moving to Socket-939 for Opteron 100, which includes use of unregistered memory (both ECC and non-ECC). This will bring down the cost of entry-level servers with single processor, single and dual- core Opteron 100 series. AMD FAQ on Opteron 100 with Direct Connect Architecture: http://www.amd.com/us-en/Processors/SellAMDProducts/0,,30_177_861_8806~85257,00.html It also seems that entry-level server mainboards with PCI-X might be dropping under $200 soon. The Broadcom/ServerWorks HT1000 might start making a sole appearance on new Socket-939 mainboards for Opteron 100. This means you'll have a single PCI-X channel, possibly one (1) 133MHz slot, two (2) 100MHz slots, or maybe two (2) 66MHz slots plus on-board peripherals (like GbE). http://www.broadcom.com/collateral/pb/HT-1000-PB01-R.pdf To date I've only seen the HT1000 implemented alongside the HT2000. But the HT1000 is clearly capable of being on its own with LPC/PCI support. This is unlike the HT2000, which must rely on the HT1000 for LPC/PCI. Although the HT2000 does add PCIe, a 2nd PCI-X channel and on-chipset dual-GbE, the HT2000+HT1000 combination is >$500 in a mainboard. I'm hopeful the HT1000 on its own will be <$200. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Thu Aug 18 23:14:21 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <1124418770.6072.77.camel@bert64.mobile.smithconcepts.com> References: <200508171718.25872.jasonb@edseek.com> <200508182223.19814.jasonb@edseek.com> <1124418770.6072.77.camel@bert64.mobile.smithconcepts.com> Message-ID: <200508182314.21369.jasonb@edseek.com> On Thursday 18 August 2005 22:32, Bryan J. Smith wrote: > On Thu, 2005-08-18 at 22:23 -0400, Jason Boxman wrote: > > MB BIOSTAR NF3 SATA AGP NF325-A7 RT - Retail > > Yeah, that's the one I would go for (in an ATX). Wasn't much in the newegg posts. Sounds like an average, inexpensive mainboard without GbE (which I can't use anyway), FRAID, or much else. I just hope the on-board sound is decent. > > POWER SUPPLY CMAX 400WCX-400B BK RT - Retail > > Yep, $35 gets you there now. > > > CPU AMD|SEMPRON64 2800+ RTL - Retail > > PQI POWER 512MB DDR PC3200 - Retail x 2 > > CASE MXTP|MID 0W 2BAFAN CSX147K GRY - Retail > > Cheap, but it has 2 intake (one above CPU), 2 outtake, does the jobbie. From the comments, the photos are wrong and there's a single rear fan slot for a 8cm, 9.2cm, or 12cm fan. It comes with a 8cm, and I'll possibly replace it with a new 12cm. -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From jasonb at edseek.com Thu Aug 18 23:17:13 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Appropriate PSU for Socket 754 and 6600GT AGP? In-Reply-To: <1124418974.6072.82.camel@bert64.mobile.smithconcepts.com> References: <200508171718.25872.jasonb@edseek.com> <1124418770.6072.77.camel@bert64.mobile.smithconcepts.com> <1124418974.6072.82.camel@bert64.mobile.smithconcepts.com> Message-ID: <200508182317.13452.jasonb@edseek.com> On Thursday 18 August 2005 22:36, Bryan J. Smith wrote: > BTW, I hope you didn't take offense to my prior posts. > I just didn't want to see you go down Socket-462 which is _dead_. No problem. I posted my OP because it'd been so long I figured I was totally out of date. Obviously, I was. I'm glad someone took the time to set me straight so I didn't spend money on dead technology. > AMD has Socket-754 planned out until next year, it's not dying at all. > Socket-462 basically died last year, and as of now, is no longer being > manufactured. > > Yes, I have said I don't think Socket-754 is worth it versus Socket-939, > but that _assumes_ you're buying a new, higher-end PCIe card and power > supply already. If you already have an AGP card, and you want to save > money on the PS, etc..., then Socket-754 with Sempron is a good move. I actually jokingly looked into trying to run one of the Athlon 64 Mobile 754s, as there are a few threads out there on which boards unofficially support the mobile microcode and what HSFs will work. I finally decided against it, as it isn't worth the hassle. The only reason I can see to try to run 64 Mobile on the desktop is to overclock the heck out of it. So, I went with the Sempron so I could just 'plug and play'. Plus, it's officially supported on my mainboard. -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From tec at homemail.bjt.net Thu Aug 18 23:39:01 2005 From: tec at homemail.bjt.net (Thomas Carlson) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Case w/ good airflow and PSU with good +12V? In-Reply-To: <20050816220400.66035.qmail@web34110.mail.mud.yahoo.com> References: <20050816220400.66035.qmail@web34110.mail.mud.yahoo.com> Message-ID: <1124422741.3948.14.camel@localhost.localdomain> I saw an Aspire X-QPack series at Fry's today ... really quite nice for around 100 bucks or so. These come in red, black, silver and blue from what I could tell and are a little bit wider and deeper than the shuttle small form factor boxes. The case reminded me of the Mondarch Hornet. Also how is the btx form factor enter into this... wait until prices drop and eventually replace the micro atx? Thomas On Tue, 2005-08-16 at 15:04 -0700, Bryan J. Smith wrote: > Jason Boxman wrote: > > Since I'm going to buy a new mainboard and CPU, I think I'm > > just going to make it a system. Any thoughts on a quality > > case with lots of airflow? My current case completely > > sucks for airflow, so it's not really appropriate for my > > upgrade anyway. (It tends to run cooler with the case side > > off, a sure sign of problems.) > > You know my opinion on boxes and airflow: > http://thebs413.blogspot.com/2005/08/small-enough-form-factor-pc.html > > > Lots of MicroATX mainboards can be had for cheap (<$50) for > older Socket-462 platforms. Otherwise, I listed my favorite > MicroATX for Socket-939 ($79) -- one of the few available > right now. > > > I'm also going to snarf up a new PSU. While I wouldn't > > mind a quieter PSU, one of average noise is fine since the > > whole system will be in a noisy server room (relatively > > speaking) which I can't quiet down in any reasonable way. > > Other fans are typically more of an issue than the PS' fan. > > > From b.j.smith at ieee.org Thu Aug 18 23:59:48 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Case w/ good airflow and PSU with good +12V? In-Reply-To: <1124422741.3948.14.camel@localhost.localdomain> References: <20050816220400.66035.qmail@web34110.mail.mud.yahoo.com> <1124422741.3948.14.camel@localhost.localdomain> Message-ID: <1124423988.6072.108.camel@bert64.mobile.smithconcepts.com> On Thu, 2005-08-18 at 20:39 -0700, Thomas Carlson wrote: > I saw an Aspire X-QPack series at Fry's today ... really quite nice for > around 100 bucks or so. > These come in red, black, silver and blue from what I could tell and are > a little bit wider and deeper than the shuttle small form factor boxes. Correct, because they have a _full_ MicroATX mainboard _and_ a _full_ ATX (not even MicroATX, but _full_ ATX) power supply. As I noted in my blog, most small form-factor cases uses a proprietary power supply, and many also use a proprietary mainboards. IMHO, it's not worth it, I'd rather have standard components. Hence why I noted in my blog I think the 8-9" x 10-11" x 13-14" MicroATX mainboard, ATX power supply "box" is on the verge of popularity as the "Small Enough" Form-Factor PC: http://thebs413.blogspot.com/2005/08/small-enough-form-factor-pc.html > The case reminded me of the Mondarch Hornet. I think they are using the same. > Also how is the btx form factor enter into this... wait until prices > drop and eventually replace the micro atx? BTX is going the way of FlexATX, Tier-1 OEM-only. BTX started out as a great idea (flip the board so AGP/PCI cards are "facing up"), but then it got too "Tier-1" focused with the mid-cooling fan and other changes. Furthermore, it can't handle dual processor or other layouts. ATX and MicroATX seem to continue to have the industry behind them. There are "flipped ATX" cases with the CPU at the bottom, and the cards at the top "facing up." And now the "box" MicroATX cases have taken off -- some with proprietary power supplies, more and more coming with full ATX power supplies for only 1-2" more in height. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Fri Aug 19 17:01:28 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] 3Ware Escalade 7006-2 ($79), 8006-2 ($89) and Promise Ultra33 ($9) pulls ... Message-ID: <20050819210128.46121.qmail@web34102.mail.mud.yahoo.com> For those of you who have ATA needs, be it true hardware ATA RAID, or just need extra ATA channels for your optical drives, "Weird Stuff Warehouse" has some ATA pulls. $79 3Ware Escalade 7006-2 ATAx2 66MHz PCI32: http://www.weirdstuff.com/cgi-bin/item/14203 $89 3Ware Escalade 8006-2 SATAx2 66MHz PCI64: http://www.weirdstuff.com/cgi-bin/item/14205 These will get you RAID-0 or RAID-1. Both will work in 32-bit PCI slots, the 8006-2 with SATA is also a 64-bit capable card. Both can run at 66MHz, as well as support 33MHz. Also, if you just need Ultra33 ATA channels for optical drives (freeing up your chipset ATA channels for hard drives), the Promise Ultra33 ATA cards are $9: http://www.weirdstuff.com/cgi-bin/item/14202 If you are worried about boot-time issues, just yank the big'ole DIP package (at the end, clearly visible) off. That's what I always do -- no more boot/BIOS. Again, these are all _pulls_ (i.e., used). -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From whittake at sbaflorida.com Fri Aug 19 20:21:46 2005 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Case w/ good airflow and PSU with good +12V? Message-ID: <1124497306.4200.10.camel@two> All this talk about internal temperatures and in general discussions about leaving a linux computer up and running for sometimes a year or more (Bryan what is the highest yours have gone to?) leaves me cold (pun intended). I have four sometimes five machines running 24/7 most of the time. I never leave the case covers on and turn the airconditioning off in my office room (20 x 20) each night, so it must get pretty hot in there someimes. So this talk about power supplies heating up and etc, etc. frankly does not make sense to me. Secondly, in reference to the leaving the linux boxes on 24/7 I have neverr had one fail because of capacitors running out of alloted life times. I have several motherboards that are 4 or 5 years old and they run 24/7 or almost. What another engineering ghost just went out the window? Me thinks the engineers are in cahoots with the manufacturing people so that they can get their new notions put into play :) Homer Whittaker From ozz at ozz.is-a-geek.net Fri Aug 19 23:11:21 2005 From: ozz at ozz.is-a-geek.net (Austin Denyer (Ozz)) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Case w/ good airflow and PSU with good +12V? In-Reply-To: <1124497306.4200.10.camel@two> References: <1124497306.4200.10.camel@two> Message-ID: <20050819231121.71f81893.ozz@ozz.is-a-geek.net> On Fri, 19 Aug 2005 20:21:46 -0400, Homer Whittaker wrote: > > All this talk about internal temperatures and in general discussions > about leaving a linux computer up and running for sometimes a year or > more (Bryan what is the highest yours have gone to?) leaves me cold (pun > intended). I've had Novell NetWare 3.12 boxes that had been up for over 5 years back in the late 90s. It broke my heart to install the y2k patches as the patches required a reboot... > So this talk about power supplies heating up and etc, etc. frankly does > not make sense to me. The main issue with power supplies (correct me if I'm wrong here) is that if the PSU is unable to supply enough juice then the voltage drops. As the voltage drops, the current increases proportionally to compensate (W=VA, so A=W/V), and the increased current is what causes the increased heat. I'm pretty sure that's what fried my 1.3GHz Athlon - the CPU was running at INSANE temperatures... > Secondly, in reference to the leaving the linux boxes on 24/7 I have > neverr had one fail because of capacitors running out of alloted life > times. I have several motherboards that are 4 or 5 years old and they > run 24/7 or almost. What another engineering ghost just went out the > window? I have a Magitronic 386sx40 that's gotta be 15 years old. It's still running... > Me thinks the engineers are in cahoots with the manufacturing people so > that they can get their new notions put into play :) There are always exceptions to the rule... Regards, Ozz. From whittake at sbaflorida.com Sun Aug 21 14:02:27 2005 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <4308C10D.5010408@sbaflorida.com> References: <4308C10D.5010408@sbaflorida.com> Message-ID: <4308C1B3.9030000@sbaflorida.com> Homer Whittaker wrote: > I am having a self induced cabling problem right now. > > Several years ago I installed a computer at my wifes desk which is 115 > feet from the systems hub. The first 10 feet or so was Cat 5 > connecting to 85 feet of 25 pair telephone wire (which was already > installed) and then thru a patch panel via some more 25 pair cable out > to my office. When it came into my office, I then used approx 15 feet > of Cat 5 to get to the hub. > In other words a real cluge. This setup gave up the ghost on Saturday > and her machine has not been able to see the internet since ???? > whatever happened. > I am now faced with purchasing a minimum order of 250 ft of outside > Cat 5, digging a trench some 100 foot long and then laying out the > wire into her office and getting it into my office (a separate building). > Given all the da, dq, da, can you think of any device that I can hang > on the mixed bag of wire than might help my wife's machine to reach > out and see the internet? Most anything beats the physical aspects > of laying 100 feet of outside Cat 5. > Homer Whittaker > > > > > From jasonb at edseek.com Sun Aug 21 14:10:32 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <4308C1B3.9030000@sbaflorida.com> References: <4308C10D.5010408@sbaflorida.com> <4308C1B3.9030000@sbaflorida.com> Message-ID: <200508211410.33331.jasonb@edseek.com> On Sunday 21 August 2005 14:02, Homer Whittaker wrote: > Homer Whittaker wrote: > > In other words a real cluge. This setup gave up the ghost on Saturday > > and her machine has not been able to see the internet since ???? > > whatever happened. Could've been lightning. A long cable run is a lightning magnet. You could try going fiber optic. ;) But in all seriousness, I have no experience with cable runs. Maybe someone else will have a suggestion if you don't want to deploy WiFi technology for wireless access. -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From whittake at sbaflorida.com Sun Aug 21 15:34:56 2005 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <200508211410.33331.jasonb@edseek.com> References: <4308C10D.5010408@sbaflorida.com> <4308C1B3.9030000@sbaflorida.com> <200508211410.33331.jasonb@edseek.com> Message-ID: <4308D760.9030204@sbaflorida.com> Jason Boxman wrote: >On Sunday 21 August 2005 14:02, Homer Whittaker wrote: > > >>Homer Whittaker wrote: >> >> > > > >>>In other words a real cluge. This setup gave up the ghost on Saturday >>>and her machine has not been able to see the internet since ???? >>>whatever happened. >>> >>> > >Could've been lightning. A long cable run is a lightning magnet. You could >try going fiber optic. ;) > >But in all seriousness, I have no experience with cable runs. Maybe someone >else will have a suggestion if you don't want to deploy WiFi technology for >wireless access. > > Had thought of WiFi but not too sure exactly what components might be requiired. Assume some sort of hub or router that can handle both WiFi and ethernet plus some sort of transmitter for my wifes machine. But that leaves reliabilty etc. She is unhappy with me as it is and I do not want to do a double cluge on her machine :) Homer From b.j.smith at ieee.org Sun Aug 21 16:53:15 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <200508211410.33331.jasonb@edseek.com> References: <4308C10D.5010408@sbaflorida.com> <4308C1B3.9030000@sbaflorida.com> <200508211410.33331.jasonb@edseek.com> Message-ID: <1124657595.4678.3.camel@bert64.mobile.smithconcepts.com> On Sun, 2005-08-21 at 14:10 -0400, Jason Boxman wrote: > Could've been lightning. A long cable run is a lightning magnet. You could > try going fiber optic. ;) Yeah, I'd suggest fiber-optical. Simple multi-mode fibre 1000Base-SX can go 485 meters. A 250' premade SC-SC or ST-ST connector (SC is commonly used in switches) cable runs around $200. Then figure another $50-100 for the 1000Base-SX NIC and another $50-100 for the 1000Base-SX module to a switch (or a couple hundred for a switch with 1000Base-SX). -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From tec at homemail.bjt.net Mon Aug 22 02:15:34 2005 From: tec at homemail.bjt.net (Thomas Carlson) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <1124657595.4678.3.camel@bert64.mobile.smithconcepts.com> References: <4308C10D.5010408@sbaflorida.com> <4308C1B3.9030000@sbaflorida.com> <200508211410.33331.jasonb@edseek.com> <1124657595.4678.3.camel@bert64.mobile.smithconcepts.com> Message-ID: <1124691334.2740.21.camel@localhost.localdomain> On Sun, 2005-08-21 at 15:53 -0500, Bryan J. Smith wrote: > On Sun, 2005-08-21 at 14:10 -0400, Jason Boxman wrote: > > Could've been lightning. A long cable run is a lightning magnet. You could > > try going fiber optic. ;) > > Yeah, I'd suggest fiber-optical. Simple multi-mode fibre 1000Base-SX > can go 485 meters. > > A 250' premade SC-SC or ST-ST connector (SC is commonly used in > switches) cable runs around $200. I remember these by sc = stick and click and st = stick and twist. :) > > Then figure another $50-100 for the 1000Base-SX NIC and another $50-100 > for the 1000Base-SX module to a switch (or a couple hundred for a switch > with 1000Base-SX). > > From b.j.smith at ieee.org Mon Aug 22 02:39:51 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <1124691334.2740.21.camel@localhost.localdomain> References: <4308C10D.5010408@sbaflorida.com> <4308C1B3.9030000@sbaflorida.com> <200508211410.33331.jasonb@edseek.com> <1124657595.4678.3.camel@bert64.mobile.smithconcepts.com> <1124691334.2740.21.camel@localhost.localdomain> Message-ID: <1124692791.4507.51.camel@bert64.mobile.smithconcepts.com> On Sun, 2005-08-21 at 23:15 -0700, Thomas Carlson wrote: > I remember these by sc = stick and click and st = stick and twist. :) Damn, those are good ones. Never thought of that, but they make great sense! I'll definitely use those when telling others for the first time. I'll try to remember your name for proper credit when I do. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From tec at homemail.bjt.net Mon Aug 22 21:53:57 2005 From: tec at homemail.bjt.net (Thomas Carlson) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Cat 5 mixed bag In-Reply-To: <1124692791.4507.51.camel@bert64.mobile.smithconcepts.com> References: <4308C10D.5010408@sbaflorida.com> <4308C1B3.9030000@sbaflorida.com> <200508211410.33331.jasonb@edseek.com> <1124657595.4678.3.camel@bert64.mobile.smithconcepts.com> <1124691334.2740.21.camel@localhost.localdomain> <1124692791.4507.51.camel@bert64.mobile.smithconcepts.com> Message-ID: <1124762037.3014.3.camel@localhost.localdomain> On Mon, 2005-08-22 at 01:39 -0500, Bryan J. Smith wrote: > On Sun, 2005-08-21 at 23:15 -0700, Thomas Carlson wrote: > > I remember these by sc = stick and click and st = stick and twist. :) > > Damn, those are good ones. Never thought of that, but they make great > sense! I'll definitely use those when telling others for the first > time. I'll try to remember your name for proper credit when I do. > Thanks! :) > From b.j.smith at ieee.org Tue Aug 23 12:15:20 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: 'Real World' Hard-drive throughput - SATA vs Ultra160 SCSI In-Reply-To: <1124809140.7072.11.camel@suse.something.com> Message-ID: <20050823161521.58261.qmail@web34102.mail.mud.yahoo.com> [ NOTE: I should rehash this into a Blog entry. Until then, if you get Sys Admin, see my 2004 article on "Dissecting ATA RAID Options" -- which covers some of the same concepts. ] David Simmons wrote: > Guys, > Was hoping to get some 'real world' experience/numbers > (trying hard to get around the marketing text of operations) > for SATA vs Ultra 160 SCSI. Get out the specifications of _any_ hard drive, _regardless_ of interface, and you'll get the internal bit-rate of the drive. That's the maximum data transfer rate (DTR). BTW, interface has little to do with it at the _drive_ level. SCSI adds standards for sector remapping (Integrated Drive Electronics, IDE) is far less standard in this regard). SCSI host adapters add queuing not on just a drive level, but the _bus_wide_ level. Serial Attached SCSI (SAS) does the same using the proven SCSI-2 protocol as well. Only recently has the Advanced Technology Attachment (ATA) added Native Command Queuing (NCQ) at the _drive_ level, and it's still not the same. You need an "intelligent [hardware] host adapter" and you still can't get that with Advanced Host Control Interface (AHCI) which is just a 100% software interface for controlling ATA channels. > From what I understand, SATA (not SATA-II, etc) is > 150MegaBytes/sec...and Ultra 160 SCSI is 160 > MegaBytes/sec....... First off, interface speed which has *0* to do with DTR of the disk itself. Most commodity capacities (160, 200, 250, 300, 320, 400, 500GB) can't break 80MBps yet, and most enterprise capacities (36, 73, 146GB) are about 50MBps -- individually. Although it _does_ become a consideration when your bus is parallel with _multiple_ devices. I.e., an Low Voltage Differential (LVD) SCSI parallel bus like Ultra2/80 (80MBps), Ultra3/160 (160MBps) and Ultra4/320 (320MBps) has to share all that DTR across all the devices on a channel. Which is why 3GHz (300MBps using 8/10 encoding) Serial Attached SCSI (SAS) is quickly going to replace SCSI. It leverages the "point-to-point" of Serial ATA (SATA), but adds the SCSI-2 protocol. So now instead of 1-2 channels typical on a SCSI host or RAID card, you can have an intelligent microcontroller/ASIC controlling 4-8 point-to-point connections of a _full_ 300MBps (6GHz/600MBps planned). SAS enforces the use of twisted pair cabling for lengths up to 8m (~26'), while remaining compatible with SATA-I/II** (1.5GHz/150MBps) and the new SATA-IO** (3GHz/300MBps) which go 1m (~3.25'). I'm not sure if SATA with TP can go greater than 1m, but I don't think so -- it might be a differential or other signaling used (although I don't know how with only 7-conductors, 3 being ground). [ **NOTE: SATA-II is now a _marketing_ term much like USB 2.0 is. You have to have a SATA-IO drive for 300MBps, just like you have to have an EHCI controller for 480Mbps/60MBps with USB 2.0. I haven't checked to see if SATA-IO requires a twisted pair cable, but the original SATA committee expected 3 and 6GHz signaling to require it. This might be way they have created the SATA-IO spec, whereas vendors are claiming and shipping SCSI-II with only a 1.5GHz capable cable/EMF. ] BTW, SAS can also be interconnected with hubs and other concentrators, trunking 4 or even 8 x 3GHz lines for 1.2-2.4GBps. This is much, much cheaper to do than parallel SCSI. SAS is purposely designed to not only replace parallel SCSI, but to bring a more cost effective option than FC-AL, and a far lower-overhead than iSCSI. But it's limitation is the length, so it's good as long as you don't go outside of the server room. > say the SATA is from onboard controller and SCSI Ultra 160 > will be PCI add-on card. You're comparing apples and oranges. ATA (including SATA) is a _dumb_ bus arbitrator between PCI[-X|e] and the Integrated Drive Electronics (IDE). ATA is dead _dumb_ and other than some registers for bus timing/configuration, it's the system memory/CPU talking to the drive. SCSI has its own _hardware_ host adapter with intelligent management and queuing, plus a full command set. SCSI host adapters are already half-way to a full, intelligent hardware RAID design. That's why ATAPI (ATA Peripheral Interface) was required for non-simple block transfers like most optical drives require. But even then, ATAPI is done in software, between the system memory/CPU and the end-drive. > Q: Would the drives be comparable in performance? For what? ATA provides dead _dumb_ block I/O without any blocking. That's great for 1 drive at 1 operation, such as typical desktop usage. The second you start queuing a lot of operations, SCSI wins. ATA can't service requests at all, it relies on the system CPU/OS. ATA with NCQ now adds it, but _only_ for one drive. That means it's great for a desktop or even a workstation with 1 drive, but once you start adding drives, then NCQ loses it's benefits. Because the SCSI host adapter is queuing for _all_ drives, not just 1. Straight Just a Bunch of Disks (JBoD) really depends on the application, and ATA is typically all you need today. Things change once you start talking about an intelligent ATA RAID controller. Now you have ATA with intelligence, queuing, SRAM (non-blocking) or DRAM (buffering). For RAID-0, 1 and 10 (simultaneous RAID-0 and 1 in hardware), ATA with a non-blocking ASIC and SRAM is most ideal. That's 3Ware's legacy Escalade design (pre-9000 series), using the direct I/O of ATA. For RAID-3, a non-blocking ASIC and SRAM, plus a little DRAM for extra XOR buffer, is also ideal -- especially when the width of the bus matches the data channel (not including parity). That' the NetCell SR3x00 (32-bit -- 2 drive + partiy) and SR5x00 (64-bit -- 4 drive + parity). ATA is still ideal because it's direct I/O. For RAID-4 or RAID-5, you're now going blocks of (typically) 32KB striped, with dedicated parity (RAID-4) or striped (RAID-5). Now you want a microcontroller with lots of buffer (DRAM). ATA or SCSI doesn't matter -- the I/O isn't direct, so non-block is useless. Furthermore, SCSI can have lots of benefits with its higher spindles for response time (especially for RAID-5), let alone other features (like sector remapping standard**). [ **NOTE: Many intelligent ATA RAID controllers, like 3Ware, reserve parts of the ATA disk for sector remapping when they use a block volume -- i.e. RAID-0, RAID-10, RAID-5. RAID-1 mirroring on 3Ware is the only time it uses the "raw" (non-volume) disk. ] In the end, the future is Serial Attached SCSI (SAS). Almost _all_ new intelligent RAID controllers being designed are SAS because they also do SATA. SAS is basically an intelligent host SATA with SCSI-2 adopt. In fact, LSI's new SAS host adapters actually have RAID-0, 1 and 1e (1 "enhanced" ~ RAID-10?) in the hardware for the local 8 channels. They then sell software RAID add-ons for multiple cards (RAID-0, 5, 6, etc...). There is also the Broadcom BCM8506 IC (I believe that's the P/N) that is available for $60 in quantity. It does 8-channel SAS with on-board hardware RAID (a little SRAM on the IC, plus RAID-5 logic) and can glue upto 768MB of SDRAM for buffer. I can also arbitrary to either PCIe or PCI-X (and bridge in between for embedded designs). It should launch a new set of sub-$300 SAS/SATA RAID cards sometime late this year or early next year. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dsimmons at powersmiths.com Tue Aug 23 12:19:00 2005 From: dsimmons at powersmiths.com (David Simmons) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: 'Real World' Hard-drive throughput - SATA vs Ultra160 SCSI In-Reply-To: <20050823161521.58261.qmail@web34102.mail.mud.yahoo.com> References: <20050823161521.58261.qmail@web34102.mail.mud.yahoo.com> Message-ID: <1124813940.7072.22.camel@suse.something.com> Bryan, I've always enjoyed your postings because they are thorough and I learn more than the original question....but, all I really wanted to know was: On Tue, 2005-08-23 at 09:15 -0700, Bryan J. Smith wrote: > Q: Would the drives be comparable in performance? Do you really get 150MegaBytes per second on SATA (of course we both know that the answer is 'No'...and like you stated (I think) it's more around 80MegaBytes/sec). Not having experience with SCSI, I was wanting to know if that same principle held true for Ultra160...because (according to the marketing) the Ultra40, Ultra80 would be inferior...while Ultra320 would be 'twice' as fast? I find it hard to believe that it's that simple...BUT, on a side-by-side comparison...and just looking for a ball park answer, how would they compare? Thanks - Dave From b.j.smith at ieee.org Tue Aug 23 13:04:13 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: 'Real World' Hard-drive throughput - SATA vs Ultra160 SCSI In-Reply-To: <1124813940.7072.22.camel@suse.something.com> Message-ID: <20050823170413.34312.qmail@web34110.mail.mud.yahoo.com> David Simmons wrote: > Not having experience with SCSI, I was wanting to know if > that same principle held true for Ultra160...because > (according to the marketing) the Ultra40, Ultra80 would be > inferior...while Ultra320 would be 'twice' as fast? Ultra320 just means that there is enough DTR that several Ultra320 drives can burst and get probably over 150MBps "real world performance" (after arbitration disconnects, protocol overhead, etc...). Real world application of SCSI typically means that you don't put more than 3-4 fixed disks on a bus if they are going to be used simultaneously. But serial is taking over parallel, and SAS is going to replace SCSI as the preferred local/closet storage solution. In the new age of SAS host adapters, you're basically going to get "hardware RAID-0/1 for free." > I find it hard to believe that it's that simple...BUT, on a > side-by-side comparison...and just looking for a ball park > answer, how would they compare? Ultra320 just means that, assuming you have Ultra320 drives, you can get twice the DTR burst on the bus than Ultra160. Including all the overhead, with 3-4 disks, on a RAID controller I can get around 150MBps "real-world" per Ultra320 channel than about 100MBps per Ultra160 channel. If I'm using an i960/IOP30x RAID controller, then it doesn't matter, I will get _nothing_ close to that. I have to be running with a StrongARM or, more recently, XScale to break 100MBps. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From damien at mc-kenna.com Tue Aug 23 20:26:49 2005 From: damien at mc-kenna.com (Damien McKenna) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Thoughts on Intel's next generation plans? Message-ID: <430BBEC9.2080203@mc-kenna.com> Any thoughts on Intel's next generation plans? At an overview it sounded just like what anyone would have guessed - continue the Pentium-M (aka Pentium III with some improvements) with a few bits out of the Pentium-IV that were worth keeping. I honestly kinda thought they'd push further, but we've got a year+ to see what the final product is like. -- Damien McKenna, husband, father, geek. damien@mc-kenna.com - http://www.mc-kenna.com/ From b.j.smith at ieee.org Tue Aug 23 21:18:13 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Thoughts on Intel's next generation plans? In-Reply-To: <430BBEC9.2080203@mc-kenna.com> References: <430BBEC9.2080203@mc-kenna.com> Message-ID: <1124846293.4515.24.camel@bert64.mobile.smithconcepts.com> On Tue, 2005-08-23 at 20:26 -0400, Damien McKenna wrote: > Any thoughts on Intel's next generation plans? At an overview it > sounded just like what anyone would have guessed - continue the > Pentium-M (aka Pentium III with some improvements) with a few bits out > of the Pentium-IV that were worth keeping. I honestly kinda thought > they'd push further, but we've got a year+ to see what the final product > is like. I've been working my @$$ off at my current client in preparation for an August 31st departure, so I haven't kept up. Do you have some good links on Intel's announcement (IDF is this week, right?)? -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Tue Aug 23 21:24:25 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Video card recommendations -- LCD details and pricing ... In-Reply-To: <20050823230712.44841.qmail@web34108.mail.mud.yahoo.com> References: <20050823230712.44841.qmail@web34108.mail.mud.yahoo.com> Message-ID: <1124846665.4515.29.camel@bert64.mobile.smithconcepts.com> Since CRT and LCD came up on our lists recently, I wanted to pass on something from another list. If you're contemplating a new CRT purchase, know that LCDs are really coming down in price and are very competitive. But know that unless you have DVI (which is now commonplace on new video cards), LCD doesn't look nearly as good over analog (and can even bend bad on cheap LCD designs). So consider LCD over a new CRT purchase, but only for DVI unless you just want something portable (like a cheap 15"). Bryan J. Smith wrote with regards to color on a LCD: > Yes, CRT currently have better guns than the latest TFT LCD > pixels can do. But the gap is closing. It's "good enough" > AFAIK, especially for desktop usage. > 98% of the LCD problems are due to people using DB15/analog. > There is _no_ reason to go LCD if you're not using DVI. > It's _self-defeating_. Bryan J. Smith wrote with regards to LCD pricing: > LCDs ... > - 15" (~16" CRT) LCD no DVI: $125+ (sometimes sub-$100) > - 17" (~18" CRT) LCD no DVI: $175+ > - 17" (~18" CRT) LCD w/DVI: $200+ > - 19" (~20" CRT) LCD w/DVI: $250+ > - 20" (~21" CRT) LCD w/DVI: $400+ > - 24" (~25" CRT) LCD w/DVI: $800+ > > 17" LCDs (equivalent to 18" CRTs because LCDs are _true_ > display size) are under $200, sometimes with DVI for under > $200. Sometimes this is not the rebate price if you can > catch a deal. NewEgg.COM has several just over $200 before > rebate, under after, including a 12ms DVI one. > > At this point, with DVI coming standard and 12ms DVI becoming > more commonplace, I can't see spending $150 on 19" being > worth it. And if you are looking at new 21" CRTs, they are > over $300, typically $400, for 20" or less viewable. That's > a 20.1" LCD which can be had for under $500 now. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Tue Aug 23 21:32:53 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: XFX DoA Issues In-Reply-To: <200508161319.57628.jasonb@edseek.com> References: <20050816170430.96236.qmail@web34105.mail.mud.yahoo.com> <200508161319.57628.jasonb@edseek.com> Message-ID: <200508232132.53312.jasonb@edseek.com> On Tuesday 16 August 2005 13:19, Jason Boxman wrote: > I just ordered the eVGA 6600GT. The BFG was quite a bit more. This time > I'm going to take the free shipping and just wait a week. I am no longer > in any hurry. I've gotten bored playing with video cards and lost interest > in playing SH3 and BF2. eVGA works like a champ. Using it right now. Worked out of the box. Even the documentation was well written, although I only flipped through it for amusement. Only downside is their 6600GT AGP has the temp sensor disabled, although one is included. Have to hack your BIOS and reflash for access to it. Might try it once I get my 754 box up. -- Jason Boxman Perl Programmer / *NIX Systems Administrator Shimberg Center for Affordable Housing | University of Florida http://edseek.com/ - Linux and FOSS stuff From b.j.smith at ieee.org Wed Aug 24 01:52:38 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] More intelligent PCIe x4/x8 storage controllers ... Message-ID: <1124862758.4515.203.camel@bert64.mobile.smithconcepts.com> Well, it appears the Intel IOP33x has be going into more products now. Unfortunately, with the exception of the Promise (sub-$400), none of them are under $500. But the Promise clearly notable for its PCIe x4 design, which is not-so-uncommon on new nForce4 Ultra/SLI mainboards. http://thebs413.blogspot.com/2005/08/new-intelligent-pcie-storage.html Still, a sub-$300 design is sorely needed, possibly a sub-$150 2-channel design like the $125 3Ware Escalade 8006-2 for 64-bit PCI/PCI-X. So far, everything under $150 has been "FRAID" (fake RAID, 100% software driver, limited Linux compatibility). Also, I revised my previous comments and added a new blog entry on Serial Attached SCSI (SAS) here: http://thebs413.blogspot.com/2005/08/serial-storage-is-future.html -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From dmckenna at thelimucompany.com Wed Aug 24 09:35:09 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Thoughts on Intel's next generation plans? Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D185943B@tlc001.tlcusa.thelimucompany.com> > Do you have some good links on Intel's announcement (IDF is this week, > right?)? Anand has some details http://www.anandtech.com/tradeshows/showdoc.aspx?i=2503 http://www.anandtech.com/tradeshows/showdoc.aspx?i=2504 http://www.anandtech.com/tradeshows/showdoc.aspx?i=2505 Also: http://techreport.com/onearticle.x/8694 http://techreport.com/onearticle.x/8695 -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From dmckenna at thelimucompany.com Wed Aug 24 09:36:46 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Video card recommendations -- LCD details andpricing ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D185943C@tlc001.tlcusa.thelimucompany.com> > If you're contemplating a new CRT purchase, know that LCDs are really > coming down in price and are very competitive. But know that > unless you have DVI (which is now commonplace on new video cards), LCD > doesn't look nearly as good over analog (and can even bend bad on > cheap LCD designs). I wish I'd done a little extra checking when we got our 17" Samsung LCD a few years ago, I forgot to check it had DVI and ended up with a VGA-only one. -- 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 Aug 24 10:07:35 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Video card recommendations -- LCD details andpricing ... In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D185943C@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D185943C@tlc001.tlcusa.thelimucompany.com> Message-ID: <1124892455.4772.23.camel@bert64.mobile.smithconcepts.com> On Wed, 2005-08-24 at 09:36 -0400, Damien McKenna wrote: > I wish I'd done a little extra checking when we got our 17" Samsung LCD > a few years ago, I forgot to check it had DVI and ended up with a > VGA-only one. Well, it wasn't until last year that you could get one with DVI for less than $500. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From dmckenna at thelimucompany.com Wed Aug 24 10:18:25 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:13 2006 Subject: [Pc_Support] Re: Video card recommendations -- LCD detailsandpricing ... Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D1859443@tlc001.tlcusa.thelimucompany.com> > On Wed, 2005-08-24 at 09:36 -0400, Damien McKenna wrote: > > I wish I'd done a little extra checking when we got our 17" > > Samsung LCD a few years ago, I forgot to check it had DVI > > and ended up with a VGA-only one. > > Well, it wasn't until last year that you could get one with > DVI for less than $500. We got ours in 2002, spent $650 on it, and there were ones at similar prices that had DVI, I just didn't think of checking for 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 Wed Aug 24 10:40:14 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] RE: Thoughts on Intel's next generation plans? In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D185943B@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D185943B@tlc001.tlcusa.thelimucompany.com> Message-ID: <1124894414.4772.56.camel@bert64.mobile.smithconcepts.com> On Wed, 2005-08-24 at 09:35 -0400, Damien McKenna wrote: > Anand has some details > http://www.anandtech.com/tradeshows/showdoc.aspx?i=2503 > http://www.anandtech.com/tradeshows/showdoc.aspx?i=2504 > http://www.anandtech.com/tradeshows/showdoc.aspx?i=2505 > Also: > http://techreport.com/onearticle.x/8694 > http://techreport.com/onearticle.x/8695 So they are doing a partial refit of Pentium Pro, just like they did Pentium 4. But instead of massively retrofitting long pipes for clock, they are just making it more efficient. This is basically what AMD did from Athlon to Athlon64/Opteron. They are designing the control so 4 pipes can complete duties simultaneously. Maybe a little more than a typical 18-month refit, but much less than a 36-48 month redesign. A lot of people are getting confused on the threading aspect. They think in terms of HyperThreading which is a _marketing_ hack to add good out-of-order (OO) execution. This new design is more efficient so HyperThreading is likely to be self-defeating, just like it would be on the Athlon designs. HyperThreading was barely a bonus on the P4, where lots of stages were doing _nothing_. So what will happen is that once Intel adds more and more cores, they will add threading control logic to make better utilization of those cores. This is not unlike what anyone else has planned. Let me rephrase that, threading will _not_ be used to make more efficient use of a single core, it will be used to make more efficient use of multiple cores. This is because the concept of "HyperThreading" to make more efficient use of a single core is pretty much a "57 pipes and nothing's on" (play that against Bruce Springsteen's "57 channels and nothing's on") hack for Pentium 4 that _dies_ with it. So, in conclusion, I hope to God this isn't it for Intel. Because if AMD is headed into virtualization, God help Intel. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From jasonb at edseek.com Wed Aug 24 10:35:13 2005 From: jasonb at edseek.com (Jason Boxman) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Re: Video card recommendations -- LCD detailsandpricing ... In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D1859443@tlc001.tlcusa.thelimucompany.com> References: <5C9DC445A45FEC4185D272DAF6AF37D1859443@tlc001.tlcusa.thelimucompany.com> Message-ID: <200508241035.13785.jasonb@edseek.com> On Wednesday 24 August 2005 10:18, Damien McKenna wrote: > > On Wed, 2005-08-24 at 09:36 -0400, Damien McKenna wrote: > > > I wish I'd done a little extra checking when we got our 17" > > > Samsung LCD a few years ago, I forgot to check it had DVI > > > and ended up with a VGA-only one. > > > > Well, it wasn't until last year that you could get one with > > DVI for less than $500. > > We got ours in 2002, spent $650 on it, and there were ones at similar > prices that had DVI, I just didn't think of checking for it. And until today I was wondering what the importance was. I'm glad this threaded showed up. My only problem is the cost of DVI KVM switches. I suppose I can keep a DVI LCD on a direct connection to my desktop and make the secondary head my KVM head. That'd work, unless I went DVI on my second monitor and really wanted to not convert from DVI to VGA back to DVI through the KVM. Sigh. From b.j.smith at ieee.org Wed Aug 24 10:45:08 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] RE: Thoughts on Intel's next generation plans? In-Reply-To: <1124894414.4772.56.camel@bert64.mobile.smithconcepts.com> References: <5C9DC445A45FEC4185D272DAF6AF37D185943B@tlc001.tlcusa.thelimucompany.com> <1124894414.4772.56.camel@bert64.mobile.smithconcepts.com> Message-ID: <1124894708.4772.60.camel@bert64.mobile.smithconcepts.com> On Wed, 2005-08-24 at 09:40 -0500, Bryan J. Smith wrote: > Let me rephrase that, threading will _not_ be used to make more efficient > use of a single core, it will be used to make more efficient use of > multiple cores. > This is because the concept of "HyperThreading" to make more > efficient use of a single core is pretty much a "57 pipes and nothing's > on" (play that against Bruce Springsteen's "57 channels and nothing's > on") hack for Pentium 4 that _dies_ with it. Er, I meant "57 stages and nothing's on." -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Sat Aug 27 18:15:34 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Jetway 939GT4-SLI comes up with a better idea for only $129 ... Message-ID: <20050827221534.6629.qmail@web34104.mail.mud.yahoo.com> For those considering a full-size nForce4 SLI mainboard, AnandTech found a lot of good in the new Jetway 939GT4-SLI mainboard. Far from cutting corners, the board seems to be solid and well designed overall, even for overclocking (not my bag, but it does bode well for non-overclocked stability). But probably the most intersting aspect is the three (3) physical PCIe x16 slots. Instead of dealing with a set of jumpers or a card to change the configuration, you simply use the middle slot for a full PCIe x16 slot, or the two outter PCIe x8 slots for SLI. Not only straight-forward, but possibly no more cost for Jetway overall than using jumpers or a card. [ SIDE NOTE: I could not assertain if the outter PCIe slots are at least usable a PCIe x1 slots when the middle slot is used at the full PCIe x16. I could not find the manual on Jetway's Tawainese site either. So it seems only the dedicated PCIe x1 may work in any configuration for anything other than video. ] Just a consideration if you're looking for a full ATX mainboard for Socket-939. AnandTech Review: http://www.anandtech.com/mb/showdoc.aspx?i=2506 NetEgg.COM Product Page: http://www.newegg.com/product/Product.asp?Item=N82E16813153030 Jetway's Tawainese Product Page: http://www.jetway.com.tw/evisn/product/amd-k8/939gt4-sli/939gt4-sli.htm -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Sat Aug 27 19:45:13 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Intel's Continued Marketing Evolution Message-ID: <20050827234513.19794.qmail@web34111.mail.mud.yahoo.com> My latest blog entry: http://thebs413.blogspot.com/2005/08/intels-continued-marketing-evolution.html -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From dmckenna at thelimucompany.com Mon Aug 29 11:18:30 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Intel's Continued Marketing Evolution Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D18594D8@tlc001.tlcusa.thelimucompany.com> > My latest blog entry: > http://thebs413.blogspot.com/2005/08/intels-continued-marketin > g-evolution.html The I/O processor is a huge leap in the right direction. My first real computer, an Amiga A500, had an entire 3-or-4 chip chipset designed to take the load off the CPU, something that the big-nobs still haven't matched, some twenty years later. As you've mentioned, if you've got support chips, even low-performance ones, its amazing how much real performance is left in your CPU when it isn't constantly doing IRQs and other junk. -- Damien McKenna - Web Developer - Damien.McKenna@thelimucompany.com The Limu Company - http://www.thelimucompany.com/ - 407-804-1014 #include From dmckenna at thelimucompany.com Mon Aug 29 11:53:49 2005 From: dmckenna at thelimucompany.com (Damien McKenna) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Pioneer shows of Blu-ray DVD burner Message-ID: <5C9DC445A45FEC4185D272DAF6AF37D18594DF@tlc001.tlcusa.thelimucompany.com> http://www.engadget.com/entry/1234000500056429/ Pioneer is working on their BRD-101A drive which adds blu-ray reading/burning support to their existing DVD-+R/RW/DL laser set. Getting the cut, though is support for ye olde CDs, I guess it was too difficult for them to squeeze in yet another laser head; alternatively it could be an attempt to push people away from the older technology... Time will tell. -- 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 Aug 29 12:48:47 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Pioneer shows of Blu-ray DVD burner In-Reply-To: <5C9DC445A45FEC4185D272DAF6AF37D18594DF@tlc001.tlcusa.thelimucompany.com> Message-ID: <20050829164847.77491.qmail@web34107.mail.mud.yahoo.com> Damien McKenna wrote: > http://www.engadget.com/entry/1234000500056429/ > Pioneer is working on their BRD-101A drive which adds > blu-ray reading/burning support to their existing > DVD-+R/RW/DL laser set. > Getting the cut, though is support for ye olde CDs, I guess > it was too difficult for them to squeeze in yet another > laser head; alternatively > it could be an attempt to push people away from the older > technology... Time will tell. Interesting that Pioneer is supported Sony/Philips Blu-Ray. If figured them to be in the DVD-HD camp. Sadly enough, Pioneer's support of Sony/Philips DVD+R/+RW _sucks_ in their current DVD-R/RW base recorders/rewriters. So I wonder how good their Blu-Ray support will be? Or maybe Pioneer is trying to get a jump on the gun. Sony's PS3 will, obviously, be Blu-Ray and there is no DVD-HD consumer set-top planned (XBox 360 is DVD-ROM). So if you can't beat'em, join'em? I hope the Blu-Ray support is better than their DVD+R/+RW (not that I think DVD+R/+RW was worth supporting, but Blu-Ray probably will be). -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Mon Aug 29 17:34:19 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Re: Awesome deal on Tyan mobo... -- Tyan mainboard guide, why PCI-X, DL145 In-Reply-To: <431379E2.4060109@carter.cc> Message-ID: <20050829213419.27546.qmail@web34102.mail.mud.yahoo.com> Carter Manucy wrote: > Ah, you mean more along the lines of the Tyan S2895? > I was trying to decide between that one and the S2882. HyperTransport Tunnels/Bridges: nForce Pro 2200: 20 PCIe, 2xGbE, 4xSATA, PCI/LPC nForce Pro 2050: 20 PCIe, 2xGbE, 4xSATA nForce4 Ultra/SLI: 20 PCIe, 2xGbE, 4xSATA, PCI/LPC AMD8151: AGP 3.0 AMD8131: (2) PCI-X 1.0 AMD8111: 2xATA, PCI/LPC Tyan Product Configurations -- 4-way: S4881: nFPro2200 (CPU#0) + AMD8131 (CPU#2) S4880/2: AMD8131 + AMD111 (CPU#0) Tyan Product Configurations -- 2-way: S2895: nFPro2200 + AMD8131 (CPU#0) + nFPro2050 (CPU#1) S2891/2: nFPro2200 + AMD8131 (CPU#0) S2885: AMD8151 + AMD8131 + AMD8111 (CPU#0) S2881/2: AMD8131 + AMD8111 (CPU#0) S2875: AMD8151 + AMD8111 (CPU#0) -- *NO* AMD8131 HP/Sun Product Configurations -- 4-way DL585: AMD8131 + AMD8111 (CPU#0) + AMD8131 (CPU#2) SFV40z: (2) AMD8131 + AMD8111 (CPU#0) + AMD8131 (CPU#1) > Only about $30 difference on Newegg... but seems like a > world of difference in the options there. For a desktop, the S2882 doesn't even have AGP. You'd want at least the S2885 for AGP. Or the S2895 for PCIe. On a server, then you'd probably like the S2882/S2892. Or the S2881/S2892 which fits in 1U. Depends if you want PCIe slots for I/O (the 289x models). I've seen some of the S288x models for under $400. > This box doesn't have a lot of need for PCI-X stuff, Huh? What storage are you going to use? You said this was a server, correct? BTW, do _not_ confuse PCI-X with PCIe (PCI-Express). PCI-X is the legacy, 64-bit (at 66-266MHz) slot. PCIe doesn't have much in the way of I/O cards yet. So you definitely want PCI-X on a server! > tho - just going to be a 1U box, but I want to buy it to > last for a while. Unless you are going to do software RAID (not recommended IMHO), you want PCI-X. The PCIe cards are now just coming out, and there's not much selection (let alone only 1 of them is under $400). > I realized after I sent that e-mail that it was RDRAM on > that riser... > oof! Nevermind on that deal! Thanks for the heads-up on > the other issues, tho. No problem. > We just bought 3 DL145's - I really like them, although I'm > pissed that the Gen-2 box doesn't have working drivers for > Windows 2000... spent 3 days with tech support going over > that and finally just said forget it and installed W2K3 on > it. Yep. Microsoft and Tier-1 OEMs, as well as "superstore" vendors, are all in bed with each other. Upgrade every 2-3 years. That's how they get you. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From b.j.smith at ieee.org Tue Aug 30 00:39:48 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] Re: Awesome deal on Tyan mobo... -- Tyan mainboard guide, why PCI-X, DL145 In-Reply-To: <4313BA1F.40909@carter.cc> References: <20050829213419.27546.qmail@web34102.mail.mud.yahoo.com> <4313BA1F.40909@carter.cc> Message-ID: <1125376788.4504.35.camel@bert64.mobile.smithconcepts.com> On Mon, 2005-08-29 at 21:45 -0400, Carter Manucy wrote: > I don't need a whole lot of storage on this box - it just OCR's > documents all day long. But you don't want your storage tying up a legacy, shared PCI bus, let alone strangled by it. > The case that it's going into is a 1U Supermicro chassis If you're going to purchase a Tyan board, you might as well go with a Tyan barebones solution in 1U. Here are the B2881 (S2881) and B2891 (S2891), respectively: http://www.tyan.com/products/html/gx28b2881.html http://www.tyan.com/products/html/gt24b2891.html > with a SCSI backplane (2 3.5" bays) - Both of these Tyan solutions have (4) 3.5" hot swap bays -- your choice, U320 or SATA -- a slim CD-ROM and optional slim floppy. They have (2) riser cards, plus whatever the mainboard on its ATX area. BTW, Any reason you're not considering SATA instead? If you're scanning and serving large files, then you need data transfer rate (DTR) more than latency. Commodity 300, 320, 400GB drives are going to give you much, much higher DTR. Vendors now have commodity disks rated for 24x7, 1Mh MTBF -- e.g., Western Digital Caviar RE series are available in 120, 160, 200, 250, 300 and 320GB (WD1200SD through WD3200SD) -- all under $200. But even if you want enterprise 10Krpm drives, you can get those in SATA too. The sub-$200 73GB Western Digital Raptor WD740GD is the same exact hard drive as the Hitachi Ultrastar 53GB U320 SCSI, only with SATA logic. Then you can drive them with a 2-channel 3Ware Escalade 8006-2 (~$125) or a 4-channel 3Ware Escalade 8506-2 (~$275) in RAID-1 or RAID-10, respectively. > so I planned on getting a Tyan w/SCSI on-board But what does that SCSI connect to? The legacy PCI bus? You're going to overload your system with disk load. Why build a monster dual-Opteron when your disk is going to be the bottleneck? > to handle the single SCSI drive it's going to use. Then why build a monster dual-Opteron system when disk is going to be so limited? You'd be better off going with a $225 mainboard like this: http://www.foxconnchannel.com/products_motherboard_2.cfm?pName=NFPIK8AA-8EKRS And a single (or up to 4) SATA drives using a single Socket-940 Opteron 1xx -- either the 73GB Raptor 10Krpm if you need latency or a 320GB Caviar RE if you need DTR. And there are forthcoming Socket-939 Opteron 1xx mainboards using the ServerWorks HT1000 chipset with a single PCI-X channel. > But now that I think about it, I have to make sure that > the power supply can handle one of these boards!!! That's why you should go with the Tyan 1U barebones solutions. > My needs for the machine are mostly just something that can pass info > off of the wire, push it thru memory quickly, then dump it back out the > wire (to the other server that holds all of the files once they've been > OCR'd). In reality, a dual-proc box is really overkill, as the > application isn't even multi-threaded. Then just go for the Foxconn Socket-940 Opteron 1xx mainboard and use the on-board SATA channels. > Sooo that's why I was thinking about the S2895. But the 2895 doesn't > have video out the back, which is a real problem. Why? If you're going to use it as a workstation, you're going to find the on-board 8MB Rage XL PCI is going to be a major PITA in capability. In the Tyan 1U barebone configurations, they have 2 slots for risers. One the S2891/S2892, you can use one for a PCIe x16 slot so you can put in a better video card (like a GeForce 6200TC or 6600). > Which takes me back to the S2882. But there are _also_ the S2891/2892 _too_! I don't think you realized what I posted before ... Tyan Product Configurations -- 2-way: S2895: nFPro2200 + AMD8131 (CPU#0) + nFPro2050 (CPU#1) S2891/2: nFPro2200 + AMD8131 (CPU#0) S2885: AMD8151 + AMD8131 + AMD8111 (CPU#0) S2881/2: AMD8131 + AMD8111 (CPU#0) The S28x1/S28x2 are 1U capable mainboards, the S28x5 are for more workstation configurations. The S288x series is the older AMD8000 series, the S289x is the newer nForce Pro series (with an AMD8131 for PCI-X). > I've also got to build another server (or buy) ... this other one is > going to be more of a storage facility, tho... so I'm looking for a nice > case with a SATA backplane on it - anything on top of your list that you > like? I prefer to keep units down to 2U if I can help it due to space > issues. I like Chembro cases, but they're a real pain in the ass to > track down thru the major suppliers that I normally use. Again, see the Tyan barebone solutions: http://www.tyan.com/products/html/barebone.html I'd normally recommend Monarch Computer for 1-3U rackmount solutions using Tyan products, but they seem to be getting bad service reviews as of late. > And oh, I wanted to let you know - a while back I'd asked if you had any > suggestions for good places to buy systems from - you suggested > Monarch... well, I ended up buying two units from them. And recently > one of the Tyan boards crapped out in one of them, so we sent it off for > an RMA... over TWO MONTHS later we finally got the unit back. Did you have a service level agreement (SLA) with them? If you did, I'd get your procurement/legal involved. I'd also let AMD know regardless. You're not the only one I've heard having issues as of late. Many people on various Red Hat Enterprise / CentOS lists are too. > They were highly disorganized, couldn't tell us what was going on with the > unit, claimed that the chassis was damaged and had to be replaced - on and on > and on. So I don't think I'll be buying from them again anytime soon - > just wanted to let you know FYI. Yeah, I've heard similar from people lately. I'd recommend the HP DL145, DL385 or DL585 if you're looking for Opteron solutions from now on. And if you're looking at the DL585, you might as well take the Sun Sunfire V40z for a spin. > Thanks (as always) for the help/suggestions/info! No problem. -- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com ---------------------------------------------------------------------- The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman From b.j.smith at ieee.org Wed Aug 31 12:12:30 2005 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] WTB: Good printer stand for 75lbs. printer (with 16"x16" footprint)? Message-ID: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> I'm looking for a good printer stand for my new Dell 3100cn color laser printer. It weights about 75lbs. and has a 16"x16" footprint, so I want something that is not pressed wood or cheap plastic. So I'm looking for something solid plastic or more solid wood, possibly metal (but no glass). I don't know if I should bother with an office supply store, and maybe just hit KMart, Target or Walmart to see what they have in a very hard, thick plastic or metal. Up here in the midwest we have "Weekends Only" stores that regularly get in overstock stuff -- like really nice, old-fashioned wood tables (you know, the screw in post-type). But they didn't have any and I'm heading back today. So, if anyone has any suggestions, please pass them on. It doesn't have to raise too far off of the floor -- in fact, I probably won't want it to for center-of-gravity/top-heavy considerations. Just something so I can put paper and other things underneath, while not having it sit directly on the floor. -- Bryan J. Smith | Sent from Yahoo Mail mailto:b.j.smith@ieee.org | (please excuse any http://thebs413.blogspot.com/ | missing headers) From whittake at sbaflorida.com Wed Aug 31 12:51:23 2005 From: whittake at sbaflorida.com (Homer Whittaker) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] WTB: Good printer stand for 75lbs. printer (with 16"x16" footprint)? In-Reply-To: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> References: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> Message-ID: <4315E00B.7000901@sbaflorida.com> Bryan: You should be able to build one out of wood. The advantages are that you can be quite specific as to the dimensions, height, and even thickness of wood (weight considerations). Using my head a little bit, I realize you do not have the tools so why not go to a Home Depot or such wherever you are and see if someone in the lumber department would take the job on for $X? Draw out what you want (even an EE should be able to do that) :) Homer Whittaker Bryan J. Smith wrote: >I'm looking for a good printer stand for my new Dell 3100cn >color laser printer. It weights about 75lbs. and has a >16"x16" footprint, so I want something that is not pressed >wood or cheap plastic. So I'm looking for something solid >plastic or more solid wood, possibly metal (but no glass). > >I don't know if I should bother with an office supply store, >and maybe just hit KMart, Target or Walmart to see what they >have in a very hard, thick plastic or metal. Up here in the >midwest we have "Weekends Only" stores that regularly get in >overstock stuff -- like really nice, old-fashioned wood >tables (you know, the screw in post-type). But they didn't >have any and I'm heading back today. > >So, if anyone has any suggestions, please pass them on. It >doesn't have to raise too far off of the floor -- in fact, I >probably won't want it to for center-of-gravity/top-heavy >considerations. Just something so I can put paper and other >things underneath, while not having it sit directly on the >floor. > > > > From philb at philb.us Wed Aug 31 19:41:12 2005 From: philb at philb.us (Phil Barnett) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] WTB: Good printer stand for 75lbs. printer (with 16"x16" footprint)? In-Reply-To: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> References: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> Message-ID: <200508311941.12303.philb@philb.us> On Wednesday 31 August 2005 12:12 pm, Bryan J. Smith wrote: > I'm looking for a good printer stand for my new Dell 3100cn > color laser printer. It weights about 75lbs. and has a > 16"x16" footprint, so I want something that is not pressed > wood or cheap plastic. So I'm looking for something solid > plastic or more solid wood, possibly metal (but no glass). I have a table saw and all related tools. Get some plywood and come out here. Alternately, I have a truck and there is a Home Depot about 2 miles from here. We can get the supplies when you arrive. Advantages: Strong and exact dimensions. Get to be with a really nice guy. Disadvantages: Requires driving and thinking. -- Don't think that a small group of dedicated individuals can't change the world. it's the only thing that ever has. From pberry2 at cfl.rr.com Wed Aug 31 20:04:38 2005 From: pberry2 at cfl.rr.com (Wise Linux User Patrick) Date: Tue Oct 31 13:17:14 2006 Subject: [Pc_Support] WTB: Good printer stand for 75lbs. printer (=?iso-8859-1?q?with=0916?="x16" footprint)? In-Reply-To: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> References: <20050831161230.72786.qmail@web34106.mail.mud.yahoo.com> Message-ID: <200508312004.39200.pberry2@cfl.rr.com> On Wednesday 31 August 2005 12:12 pm, Bryan J. Smith wrote: > I'm looking for a good printer stand for my new Dell 3100cn > color laser printer. It weights about 75lbs. and has a > 16"x16" footprint, so I want something that is not pressed > wood or cheap plastic. So I'm looking for something solid > plastic or more solid wood, possibly metal (but no glass). I saw a tool cart at Harbor Freight Tools, here, http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=6650 but, bought the one of chromed steel at Costco, for $49. Don't see it on the web page, with the other Apex shelving. Probably also in most supply houses. I like the openness for heat dissipation, and to run wires through... -- Check these out: http://knopper.net/knoppix http://livecdlist.com http://distrowatch.com http://yolinux.com http://safeharbordome.com