[Pc_Support] Serious performance boost by going to SATA on nForce4 [standard] chipset ...

[Bryan J. Smith]

All --

I just wanted to point out that I just upgraded from a Seagate 200GB
7200rpm ATA to a Western Digital 320GB 7200rpm SATA and I'm seeing a
_massive_ performance boost.  I haven't done any benchmarks, and I'm not
sure it's actually throughput related, but the access time/response
"feel" is clearly much, much faster.  I noticed it is indexing my local
Evolution mbox files much faster than before -- 3x over!  Why, I'm not
sure.

Being that the WD is clearly a newer generation drive and density, it's
not really fair to the Seagate in comparison.  I know the interface
really doesn't make any difference, as neither drive approaches the
interface limitations -- unless, of course, maybe the nVidia ATA design
has some serious bottlenecks.  And it's not likely the use of the SCSI
subsystem (via nv_sata), because it's still a "dumb block" ATA data
stream -- no NCQ or anything.

So the only other thing I think it could be (besides the newer gen
drive, which is definitely part of it) is the fact that the SATA
channels are on their own, dedicated PCIe x1 channel.  The SATA isn't
contending for I/O bandwidth with any other devices on the legacy PCI
bus, and I have to assume this is a major reason.  It could also be the
bi-directional features of PCIe x1 as well, commands/writing and reading
also do not contend for the I/O channel.

Whatever it is, I really need to get two drives of the same model to
test it out further, and throw a few apps at it.  Otherwise, I am
extremely and pleasantly surprised to see the result.  Again, I can only
assume a good part of the reason is because the 4-port SATA of the
nForce (all versions -- Standard, Ultra and SLI -- the latter two being
SATA-2/300, mine is only SATA-1/150, not that it's much difference
today) is on its own PCIe x1, whereas all the other ATA is on the
legacy, shared PCI bus.


-- 
Bryan J. Smith                                     b.j.smith at ieee.org 
--------------------------------------------------------------------- 
It is mathematically impossible for someone who makes more than you
to be anything but richer than you.  Any tax rate that penalizes them
will also penalize you similarly (to those below you, and then below
them).  Linear algebra, let alone differential calculus or even ele-
mentary concepts of limits, is mutually exclusive with US journalism.
So forget even attempting to explain how tax cuts work.  ;->