CUDA and openCL Benchmarks

MAGIC Quantum Mechanic
MAGIC Quantum M...
Joined: 18 Jan 05
Posts: 1886
Credit: 1405747946
RAC: 1167734

I was just wondering if

I was just wondering if anyone had tried the newest GeForce drivers since I was checking an email from them.

I haven't tried the new 310.90 yet

http://www.geforce.com/drivers/results/55217

(I have been using 306.97 since they first came out)

Nobody316
Nobody316
Joined: 14 Jan 13
Posts: 141
Credit: 2008126
RAC: 0

RE: I was just wondering if

Quote:

I was just wondering if anyone had tried the newest GeForce drivers since I was checking an email from them.

I haven't tried the new 310.90 yet

http://www.geforce.com/drivers/results/55217

(I have been using 306.97 since they first came out)

2 post above your post tho not for xp... in windows 7 2 work units at same time has about a min longer added to the time but works really good for my testing of it... different cards have had different results.. like many was using the 310.70 driver and it would stop and crash my pc and not theirs... give it a shot but keep the older driver handy just in case you need to go back to it...

PC setup MSI-970A-G46 AMD FX-8350 8 core OC'd 4.45GHz 16GB ram PC3-10700 Geforce GTX 650Ti Windows 7 x64 Einstein@Home

Christoph
Christoph
Joined: 16 May 07
Posts: 10
Credit: 4994563
RAC: 0

RE: I was just wondering if

Quote:
I was just wondering if anyone had tried the newest GeForce drivers since I was checking an email from them.


I use the new drivers and have not encountered a problem.

Gamboleer
Gamboleer
Joined: 5 Dec 10
Posts: 173
Credit: 168389195
RAC: 0

Edit: Deleted. Re-testing

Edit: Deleted. Re-testing with one card only for proper results.

Gamboleer
Gamboleer
Joined: 5 Dec 10
Posts: 173
Credit: 168389195
RAC: 0

NEW DATA: 7750, 2 tasks =

NEW DATA:

7750, 2 tasks = 4680 seconds.

Jeroen
Jeroen
Joined: 25 Nov 05
Posts: 379
Credit: 740030628
RAC: 0

Back in early November I put

Back in early November I put another request in with NVIDIA to re-enable PCI-E 3.0 support in their latest Linux drivers. Any of the drivers post 295.33 had 3.0 support disabled at least for LGA2011 processors. NVIDIA recently added an option to enable PCI-E 3.0 with their 310.32 drivers.

http://www.geforce.com/drivers/results/56283

PCI-E 3.0 can be enabled via a kernel module parameter.

modprobe nvidia NVreg_EnablePCIeGen3=1

This should give a nice performance boost with BRP4 when using Linux, Kepler, and LGA2011 processors. I am still seeing a 10-15% increase in performance with BRP4 when switching from x16 2.0 to x16 3.0.

Denis Puhar, dr. med.
Denis Puhar, dr...
Joined: 5 Nov 09
Posts: 36
Credit: 7006583
RAC: 0

Hi! I've been using the

Hi!

I've been using the 310.90 drivers on my GA NVIDIA GeForce GTX 550Ti Windforce edition (factory OC to 970 Mhz) almost from the time they became available.

No problems with them at all including E@H GPU tasks (0% error rate on hundreds of WUs).

Yesterday I decided to try out the 313.96 beta drivers - till now I've encountered NO problems at all (maybe a very slight performance drop regarding the CPUt - but nothing more than max of 5s, but that could be also to other factors).

D

“A little knowledge is a dangerous thing. So is a lot.” - Albert EINSTEIN

Jeroen
Jeroen
Joined: 25 Nov 05
Posts: 379
Credit: 740030628
RAC: 0

My hard drive running Windows

My hard drive running Windows for my dual 580 system crashed a few days ago unfortunately. Since then I switched to a PXE boot of Linux until I can get a replacement disk which will probably be a SSD this time around.

Running two tasks per GPU and driver 310.32, I am seeing run time of slightly over 1500 seconds per GPU, which is a bit better than what I saw in Windows. In Windows, a single task would take 820-840 seconds to complete per GPU. Running two tasks per GPU in Windows did not give any gains over running one task last I tried. Both cards are in x16 2.0 slots.

Neil Newell
Neil Newell
Joined: 20 Nov 12
Posts: 176
Credit: 169699457
RAC: 0

RE: Thx

I think the HD5830 entry needs correcting; it was given as 2,916s x1 higher up in this thread, however in the PDF 2,916s appears in the x2 column.

Only noticed it because I thought I'd try an AMD GPU, and bought an HD5830 on eBay as it appeared to have amazing performance for the price :) When I got 2,850s x1, that was slower than expected so I checked back and spotted the typo.

Neil Newell
Neil Newell
Joined: 20 Nov 12
Posts: 176
Credit: 169699457
RAC: 0

RE: Back in early November

Quote:
Back in early November I put another request in with NVIDIA to re-enable PCI-E 3.0 support in their latest Linux drivers. Any of the drivers post 295.33 had 3.0 support disabled at least for LGA2011 processors. NVIDIA recently added an option to enable PCI-E 3.0 with their 310.32 drivers.

I wonder if that has any effect on PCIe 2.0 platforms that NVIDIA only let run at PCIe 1.0 speeds (e.g. Intel X38). Certainly the 295.xx drivers only run at PCIe 1.0 speed on some Intel PCIe 2.0 platforms (you can check by running the settings program and looking under the 'PowerMizer' link - 2.5GT/s for PCIe 1.0, 5.0GT/s for PCIe 2.0 and presumably 10GT/s for PCIe 3.0).

Thanks for pointing it out, I'll give it a go!

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.