Latest drivers (NVidia and ATI)

[Sutaru Tsureku]

Which nVIDIA driver you tested with CUDA V2.3?
I guess you used only 197.13 with CUDA V2.3 and CUDA V3.0.14 in your test, right?

My test was CUDA v2.3 vs CUDA v3.0.14 on MB_6.08_CUDA_V12_VLARKill_FPLim2048.exe

I asked, because from my experiences and test of 190.x , 191.x and 195.x was 190.x the fastest nVIDIA driver (with CUDA V2.3).
196.x I didn't tested.

Now with 197.13 and CUDA V2.3 or V3.0.14 I'm sceptical that it's faster than 190.x + CUDA V2.3 .
It's look like the 197.13 + CUDA V3.0.14 is ~ 1 - 2 % slower than 190.x + CUDA V2.3 .

This sounds not much, but my OCed GTX260-216 have a RAC of ~ 15,000.
And ~ 1 - 2 % are ~ 150 - 300 Cr./day (RAC). This are two or four WUs/day less crunched.

So someone should make a test if 190.x + CUDA V2.3 or 197.13 + CUDA V3.0.14 is faster.

If someone would explain me how to make it with bench WUs, I would make this test in near future.
If someone could do this test faster/earlier as I, it would be also fine..

[glennaxl]

To test grab KWSN Knabench 1.81(Raistmer's modified Benchmark Package

The quick and dirty  method:
1) Make sure boinc is running then snooze the gpu leaving the cpu at 100%. (note: LAIM should be off)
2) Extract the KNAbench and prepare the science apps. Read the readme for further instructions
3) Modify Knabench181_nb.cmd -> look for "-verb -st" and append "-device N" where N is the gpu number  eg. "-verb -st -device 0" (note: there are 6 instances; its used for text output and parameter)
4) run Knabench181_nb.cmd
if you want to test N gpus at a time, just duplicate the knabench folder and modify the target gpu (-device N)

Long Method:
Make sure nothing is running in background; cpu is idle.
Number of CPUs = Number of KNAbench folder
Number of GPUs  = Number of KNAbench folder
so for q6600 and gtx 295 it would have 6 KNAbench folders
follow steps 2 & 3 above; step 3 for the gpu test folders
run Knabench181_nb.cmd for all folders

[Sutaru Tsureku]

O.K., here we go..
It was not so easy to understand how to do it.. but I got it..

The result in short.. I don't know what's faster.. 


The results on my QX6700 @ 3.14 GHz with manufacturer OCed GTX260-216:
[MB_6.08_CUDA_V12_VLARKill_FPLim2048.exe]
[only GPU calculation, CPU idle]
[fastest in yellow]


197.13 + CUDA 3.0.14

TaskName: PG0395.wu ------ TaskName: PG0444.wu ------ TaskName: PG1327.wu
47.578 secs Elapsed ----- 43.219 secs Elapsed ----- 31.391 secs Elapsed
16.203 secs CPU time --------- 15.609 secs CPU time --------- 14.922 secs CPU time


197.13 + CUDA 2.3

TaskName: PG0395.wu ------ TaskName: PG0444.wu ------ TaskName: PG1327.wu
47.078 secs Elapsed ----- 43.719 secs Elapsed ----- 32.375 secs Elapsed
16.250 secs CPU time --------- 15.813 secs CPU time --------- 15.547 secs CPU time


190.38 + CUDA 2.3 (1st run)

TaskName: PG0395.wu ------ TaskName: PG0444.wu ------ TaskName: PG1327.wu
47.078 secs Elapsed ----- 43.891 secs Elapsed ----- 32.453 secs Elapsed
16.469 secs CPU time --------- 16.172 secs CPU time --------- 16.078 secs CPU time


190.38 + CUDA 2.3 (2nd run)

TaskName: PG0395.wu ------ TaskName: PG0444.wu ------ TaskName: PG1327.wu
47.531 secs Elapsed ----- 43.063 secs Elapsed ----- 31.563 secs Elapsed
16.422 secs CPU time --------- 15.922 secs CPU time --------- 15.141 secs CPU time


The best would be to let run 3 or 5 runs per driver/CUDA version combis for to calculate the average. For to compensate 'runaway' results.
But, no.. not I again..
Who would like to do it?


EDIT: Or maybe also longer WUs for 'benching'?

[glennaxl]

Nvidia 257.21 drivers released today.

I havn't done any any benchmarks yet but for multiple gpu users (vista/7), you don't need to extend your desktop (using dummy plugs or real monitors). ALL cuda gpus will be detected.

[efmer (fred)]

Nvidia 257.21 drivers released today.

I havn't done any any benchmarks yet but for multiple gpu users (vista/7), you don't need to extend your desktop (using dummy plugs or real monitors). ALL cuda gpus will be detected.

Nice, already used the beta drivers with these features, but maybe this release driver is even better. Was still a lot slower than to XP 64 version.

[Ghost0210]

Nvidia 257.21 drivers released today.

I havn't done any any benchmarks yet but for multiple gpu users (vista/7), you don't need to extend your desktop (using dummy plugs or real monitors). ALL cuda gpus will be detected.

Well blow me......seems like nVidia have finally got their act sorted with these drivers
no problems installing over the top of the 257.15 beta drivers AND I can remove some of the cables plugged into my rig as I don't have to use a KVM to get both my nv cards working

[Jason G]

Trying out 258.19 now (Developer Driver)... Seems Good, time will tell.

[Vyper]

Trying out 258.19 now (Developer Driver)... Seems Good, time will tell.

Trying too

[Jason G]

Trying too

See what you reckon.  'Feels' good, but slower Cuda (by quite a bit) I reckon.

[Vyper]

Well that's an old trick, how do you make your new generation faster when you can't? , You gradualy makes the older hardware slower by a tiny bit per driver version.. That's how you make people think that newer is so much better than the older..

Too bad that the car industry hasn't got to that point, but you might think what they could incorporate in the future too when the car is in for a service when a regular modern car has firmware upgrades as hell nowadays too

2 cents.

Kind regards Vyper

[hellsheep]

I've been using the 257.21 drivers since release day. Upgraded from 197.something i haven't noticed any major performance increases, maybe by a few seconds here and there. The one thing i do like is the multi-gpu support without any dummy plugs and things like that. At least they finally got that part right.

[IrishFBall32]

Downloading 257.21 now... wonder what the chances are that this will allow service-mode installs on Win7/Vista, seeing as how they must have decoupled CUDA from the user mode drivers... (or I suppose maybe not)

[efmer (fred)]

Downloading 257.21 now... wonder what the chances are that this will allow service-mode installs on Win7/Vista, seeing as how they must have decoupled CUDA from the user mode drivers... (or I suppose maybe not)

Not very likely.

[chelski]

Have anyone been able to run the CUDA version on a Nvidia Quadro NVS 3100m with 256M memory?

Been trying to run the optimised app (via the Unified Installer) and both Cuda 2.3 and 2.2 versions tend to error out before completion.  Machine is a Lenovo Thinkpad T410, using driver 257.21 and OS is Windows 7 - I have switched it to classic mode in order to reduce video memory consumption (now round about 43M)

If anyone have any insights please let me know.

[Gizbar]

Have anyone been able to run the CUDA version on a Nvidia Quadro NVS 3100m with 256M memory?

Been trying to run the optimised app (via the Unified Installer) and both Cuda 2.3 and 2.2 versions tend to error out before completion.  Machine is a Lenovo Thinkpad T410, using driver 257.21 and OS is Windows 7 - I have switched it to classic mode in order to reduce video memory consumption (now round about 43M)

If anyone have any insights please let me know.

I think you'll have problems. The only time I got a 256Mb GFX card to work was on Windows XP 32. It was also running at 1280x1024 as the screen I had was only a 17" CRT at the time. The screen modes for Vista and Win 7 seem to require more memory, and not leave enough for the app to use...

Giz.