Advanced search

Message boards : News : Please upgrade to DRIVER 334.21 or NEWER [closed]

Author Message
Profile MJH
Project administrator
Project developer
Project scientist
Send message
Joined: 12 Nov 07
Posts: 696
Credit: 27,266,655
RAC: 0
Level
Val
Scientific publications
watwat
Message 36505 - Posted: 21 Apr 2014 | 18:09:23 UTC

Hi all,

As you may have noticed, we now have a CUDA 6.0 application. This brings with it several benefits over the older 4.2 and 5.5 applications, notably:

* Support for Maxwell GPUs
* Slightly improved performance
* Much lower CPU load

The minimum driver required for this app is 334.21, but any later version is OK.

In the longer term, our software development will assume CUDA 6. features, and it would be good to have as many of you as possible able to run it.

Currently 60% of all the WUs we get back are completed on machines with this driver or later (though many of you have been running the 55 app, because of our server scheduling policy)

If you have the opportunity, please do consider upgrading your driver!

Don't worry if you can't: the older applications aren't going away. From now on, though, they will get only maintenance updates.

Matt

klepel
Send message
Joined: 23 Dec 09
Posts: 132
Credit: 1,802,807,041
RAC: 1,492,702
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36508 - Posted: 21 Apr 2014 | 20:49:36 UTC

I have recently upgraded to 332.21. I am a little bit conservative on this issue as everything runs just fin on my computers at the moment.

So I was wondering if you can confirm that the driver 334.21 will run fine on the following cards: GTX570, GTX650ti and GTX670, the last two computers are with W7 and the first one with XP, BOINC Version is 7.0.25. Or can anybody recomened an other driver version?

Jim1348
Send message
Joined: 28 Jul 12
Posts: 455
Credit: 1,130,760,908
RAC: 149,820
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36509 - Posted: 21 Apr 2014 | 21:36:03 UTC - in response to Message 36508.

Or can anybody recomened an other driver version?

I have run a GTX 660 and GTX 650 Ti under WinXP using the 335.28 drivers, and they have worked fine for me on the longs. More recently I have used a GTX 750 Ti with the same drivers, and it has worked fine on the shorts. I would not hesitate to upgrade, but I recommend a clean install, especially if you have multiple cards on the same motherboard (i.e., use Driver Cleaner or Driver Fusion).

Jacob Klein
Send message
Joined: 11 Oct 08
Posts: 1060
Credit: 1,124,685,739
RAC: 1,301,618
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36510 - Posted: 21 Apr 2014 | 21:46:55 UTC - in response to Message 36509.
Last modified: 21 Apr 2014 | 21:50:45 UTC

:) I have 3 generations of GPUs in the same PC. When I update drivers, I do this:

1) In System -> Advanced System Settings -> Hardware -> Device Installation Settings -> Select "No, let me choose what to do" and "Never install driver software from Windows Update"... that way Windows Update will never try to install a (possibly older) version
2) In Programs and Features, uninstall "NVIDIA Graphics Driver", rebooting when prompted
3) In Programs and Features, uninstall "NVIDIA PhysX" since it doesn't require a restart, and then also uninstall "NVIDIA GeForce Experience", rebooting if prompted, and then also uninstall "NVIDIA HD Audio Driver" rebooting when prompted
4) In Programs and Features, make sure no more NVIDIA driver stuff remains
5) Finally install the new (or old) version that you wanted. Choosing "Clean install" will reset all setting and profiles to their stock defaults.
6) Check for Windows Updates drivers periodically, or revert that Windows Update setting so it includes checking for newer drivers whenever it runs.

It's honestly a mess. NVIDIA needs to fix their uninstaller to do everything in one fell swoop. But, to be fair, most people should simply install the newer version on top, with no need to uninstall the old version first. I choose to do the uninstalls, because I run the BETA NVIDIA drivers, and I want them completely removed whenever I upgrade drivers.

At any rate, I have not had any major problems with NVIDIAs latest driver lately. They do GPUGrid tasks just great. Sometimes they get a bit "power conservative" and will downclock the GPU from Max Boost, but I've created a thread on how to alleviate that, if you want to. Other than that, though, latest drivers work great for me!

Hope this helps!
Jacob

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1837
Credit: 10,434,859,894
RAC: 8,761,703
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36515 - Posted: 21 Apr 2014 | 23:14:34 UTC - in response to Message 36505.

I have the 332.21 driver on my WinXPx64 / GTX 780 Ti host, and it says it has CUDA 6.
I don't want to upgrade my driver on that host, because the new driver's "Much lower CPU load" deteriorates the GTX 780 Ti's performance under WinXP.
Could you issue Cuda 6 work to every host which has appropriate driver for it? The Maxwell GPUs will have at least the 334.21 driver, as the previous ones didn't support that card.

klepel
Send message
Joined: 23 Dec 09
Posts: 132
Credit: 1,802,807,041
RAC: 1,492,702
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36521 - Posted: 22 Apr 2014 | 0:07:39 UTC - in response to Message 36515.

I have the 332.21 driver on my WinXPx64 / GTX 780 Ti host, and it says it has CUDA 6.
I don't want to upgrade my driver on that host, because the new driver's "Much lower CPU load" deteriorates the GTX 780 Ti's performance under WinXP.
Could you issue Cuda 6 work to every host which has appropriate driver for it?

This is even a better solution. I would rather stay with 332.21 driver as well.

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36523 - Posted: 22 Apr 2014 | 0:12:35 UTC - in response to Message 36521.

Not quite sure what it is you are asking here. If you have driver 332.21 you shouldn't be getting the cuda60 app.

Matt

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36538 - Posted: 22 Apr 2014 | 13:19:27 UTC - in response to Message 36536.

334.21 is the minimum driver version that we are sending cuda60s to. This is the driver revision that accompanied the first pre-prelease version of the CUDA 6 toolkit that worked for us.

Matt

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36541 - Posted: 22 Apr 2014 | 15:10:18 UTC

I have read that the latest 3 drives are hampering the GTX780Ti and that card is already not super with Win7 due to WDDM. So I would like to stick with the 331.82 drivers for a while. However now I am not getting new work anymore from GPUGRID and a message in BOINC says that I need to upgrade the drivers.
That is not what Matt tells here:

Don't worry if you can't: the older applications aren't going away. From now on, though, they will get only maintenance updates.


But seeing this:
4/22/2014 5:08:33 PM | GPUGRID | Message from server: NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU
4/22/2014 5:08:33 PM | GPUGRID | No tasks sent
4/22/2014 5:08:33 PM | GPUGRID | No tasks are available for ACEMD beta version
4/22/2014 5:08:33 PM | GPUGRID | No tasks are available for Long runs (8-12 hours on fastest card)
4/22/2014 5:08:33 PM | GPUGRID | No tasks are available for CPU only app
4/22/2014 5:10:21 PM | GPUGRID | update requested by user
4/22/2014 5:10:24 PM | GPUGRID | Sending scheduler request: Requested by user.
4/22/2014 5:10:24 PM | GPUGRID | Requesting new tasks for NVIDIA and intel_gpu
4/22/2014 5:10:25 PM | GPUGRID | Scheduler request completed: got 0 new tasks
4/22/2014 5:10:25 PM | GPUGRID | Message from server: NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPU
4/22/2014 5:10:25 PM | GPUGRID | No tasks sent
4/22/2014 5:10:25 PM | GPUGRID | No tasks are available for ACEMD beta version
4/22/2014 5:10:25 PM | GPUGRID | No tasks are available for Long runs (8-12 hours on fastest card)
4/22/2014 5:10:25 PM | GPUGRID | No tasks are available for CPU only app


it seems that I am forced to upgrade the drivers, but then the mighty GTX780Ti will underperform in Win7.

Please advice Matt.

____________
Greetings from TJ

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36542 - Posted: 22 Apr 2014 | 15:46:21 UTC - in response to Message 36541.

TJ,

That's a problem with the scheduler. I'll be fixed shortly.

Matt

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36545 - Posted: 22 Apr 2014 | 17:50:50 UTC - in response to Message 36542.

TJ,

That's a problem with the scheduler. I'll be fixed shortly.

Matt

Thank you Matt!
____________
Greetings from TJ

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,827,709,814
RAC: 445,018
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36546 - Posted: 22 Apr 2014 | 17:56:01 UTC - in response to Message 36538.
Last modified: 22 Apr 2014 | 17:56:52 UTC

334.21 is the minimum driver version that we are sending cuda60s to. This is the driver revision that accompanied the first pre-prelease version of the CUDA 6 toolkit that worked for us.

Matt

334.21 is for Linux and adds support for Maxwell's.
However I have successfully returned CUDA6 work using 331.49 on Ubuntu 13.10 64-bit from a Kepler:

234x-SANTI_MAR423cap310-69-84-RND5084_0 6452642 20 Apr 2014 | 10:24:50 UTC 20 Apr 2014 | 12:58:06 UTC Completed and validated 9,100.43 1,596.87 18,300.00 Short runs (2-3 hours on fastest card) v8.21 (cuda60)

Presumably that rig will stop getting CUDA6 work now, unless I manage to upgrade to the latest drivers, which I don't really need to do as I don't have a Maxwell (and I should still get CUDA42/55 work).
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36547 - Posted: 22 Apr 2014 | 18:02:51 UTC

I got new work again app 8.40 cuda60 and I have driver 331.82. They error out almost immediately, this are some lines:

4/22/2014 7:59:46 PM | GPUGRID | If this happens repeatedly you may need to reset the project.
4/22/2014 7:59:47 PM | GPUGRID | Computation for task e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0 finished
4/22/2014 7:59:47 PM | GPUGRID | Output file e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0_0 for task e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0 absent
4/22/2014 7:59:47 PM | GPUGRID | Output file e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0_1 for task e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0 absent
4/22/2014 7:59:47 PM | GPUGRID | Output file e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0_2 for task e16s63_e7s8f74-GIANNI_lig3-0-1-RND3589_0 absent
4/22/2014 8:00:27 PM | GPUGRID | Task A2ART4Ex03x7-GERARD_A2ART4E-2-14-RND6887_0 exited with zero status but no 'finished' file
4/22/2014 8:00:27 PM | GPUGRID | If this happens repeatedly you may need to reset the project.

So something is still not right, as the driver I am using is not good for cuda60, but my PC is getting these WU's. Can we solve this Matt, or do I really have to update the drivers? I am not keen to do for the 780Ti. Thanks
____________
Greetings from TJ

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36549 - Posted: 22 Apr 2014 | 18:36:01 UTC - in response to Message 36547.

Looking into it...

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36550 - Posted: 22 Apr 2014 | 18:44:00 UTC - in response to Message 36549.

Try cuda60 version 841

(By the way, this reintroduces SWAN_SYNC - if you set that to 1 you should find you get improved performance)

Matt

Profile Stoneageman
Avatar
Send message
Joined: 25 May 09
Posts: 211
Credit: 12,057,679,608
RAC: 11,742,576
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36551 - Posted: 22 Apr 2014 | 19:23:18 UTC
Last modified: 22 Apr 2014 | 19:24:41 UTC

app 8.21 errors out with 331.38 driver which is the default for the latest Ubuntu release 14.04

Profile Stoneageman
Avatar
Send message
Joined: 25 May 09
Posts: 211
Credit: 12,057,679,608
RAC: 11,742,576
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36552 - Posted: 22 Apr 2014 | 19:35:53 UTC

How about using the ACEMD beta channel just for the cuda 60 long tasks.

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36553 - Posted: 22 Apr 2014 | 19:45:42 UTC - in response to Message 36552.

Stoneageman - it's a scheduling issue, unfortunately. Itś affecting all apps equally.

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36554 - Posted: 22 Apr 2014 | 20:09:22 UTC - in response to Message 36550.

Try cuda60 version 841

(By the way, this reintroduces SWAN_SYNC - if you set that to 1 you should find you get improved performance)

Matt

It is still not working Matt. App and files are downloaded, then it says running, but nothing happened and then they error out.
So it seems driver 331.82 can no longer be used? But you mentioned that cuda42 tasks would still be around. Why does my system not getting those as I have a lesser driver then the one needed for cuda60? Or am I missing something or miss-read?
____________
Greetings from TJ

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36555 - Posted: 22 Apr 2014 | 20:33:26 UTC - in response to Message 36554.

Yeah - scheduling on the Cuda capability reported by the driver is insufficient - the 331s say they do, but they don't. We've reverted to giving cuda60s out only to 334+

Matt

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36556 - Posted: 22 Apr 2014 | 21:02:17 UTC - in response to Message 36555.

Yeah - scheduling on the Cuda capability reported by the driver is insufficient - the 331s say they do, but they don't. We've reverted to giving cuda60s out only to 334+

Matt

Thanks Matt. Highly appreciated especially as it is late evening.I have now two tasks running again at 90% GPU load at 1150MHz on the 780Ti.
____________
Greetings from TJ

Profile Stoneageman
Avatar
Send message
Joined: 25 May 09
Posts: 211
Credit: 12,057,679,608
RAC: 11,742,576
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36557 - Posted: 22 Apr 2014 | 21:11:59 UTC - in response to Message 36555.

Yeah - scheduling on the Cuda capability reported by the driver is insufficient - the 331s say they do, but they don't. We've reverted to giving cuda60s out only to 334+

Matt

Mmmm...it's not working so far with 331.38 driver on Linux

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36560 - Posted: 22 Apr 2014 | 21:24:07 UTC - in response to Message 36557.

Hi Stoneageman,

This is a Linux-specific problem - turns out the boinc client in't reporting the driver version to the server, so the scheduler can't make the right allocation.

I am working on a patch (to the client)..

Matt

GPUGRID Role account
Send message
Joined: 15 Feb 07
Posts: 134
Credit: 1,349,535,983
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36561 - Posted: 22 Apr 2014 | 21:25:31 UTC - in response to Message 36560.

PS Stoneage - hope you don't think I'm mucking you but just to get the #1 slot off you! :-)

Profile Stoneageman
Avatar
Send message
Joined: 25 May 09
Posts: 211
Credit: 12,057,679,608
RAC: 11,742,576
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36563 - Posted: 22 Apr 2014 | 21:50:19 UTC

Ha! On that host I've changed to the 337.12 beta driver (after some hassle) so cuda60 tasks are now running OK on that. Just don't have the time to do the rest of the farm just now.
PS, don't you ever sleep?

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36617 - Posted: 24 Apr 2014 | 17:10:13 UTC

Okay I have now updated to latest driver (337.50). The 780Ti goes to boast 1158MHz, without any program like AfterBurner running. Problem: temperature goes quickly to 83°C when crunching GPUGRID. I don't like those high values so I started AfterBurner, it is still working in boast speed but temperature is 76°C with 91% fan speed. What I now notice is that the GPU load is 66-67% and that is less then it has been with 331.82 driver. I have no Swan_Sync settings yet.

No statistical data yet, but as others mentioned that with drivers after 331.82 the GTX780Ti and higher hampered little on Win7. I will let it run overnight and see what happens. However temperature is now 76°C steady and was 69-72° until this afternoon, when I updated. Will also try with stock clock (875MHz) what temperature does, later tomorrow.
____________
Greetings from TJ

Jeremy Zimmerman
Send message
Joined: 13 Apr 13
Posts: 61
Credit: 726,605,417
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwat
Message 36620 - Posted: 24 Apr 2014 | 17:55:41 UTC - in response to Message 36617.

TJ,

With some of the new WU's actually taking advantage of all those cores/SMUs/Shader units of the 780Ti, the air cooling is getting a wee bit strained especially with spring and warmer temps in the house. This is why I am using the Precision and prioritize temperature over power. It is cutting voltage and throttling GPU frequency as needed to keep the temperature desired. That is of course after the custom fan profile is at 100%.

I am enjoying the 87-90% utilization that I have seen, but those cards are sweating heavily. :) Work harder there yee poor silicon I say...work harder! Dear researchers, no easy WU's, make my silicon work hard! But do not crash them. :)

Regards,
Jeremy

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36623 - Posted: 24 Apr 2014 | 18:26:56 UTC - in response to Message 36620.

Thank you Jeremy, that give me some relieve.

It is now a Gerard WU that runs at 66% and 77°C at 1055.6MHz.
This morning same WU type ran at 88% at 70°C and 875MHz.
And with ambient temperature 28°C and warmer weather on the way, I like the last values despite the lower clock.
I will wait and see how much faster it is tomorrow and then try to throttle the card with PrecisionX or revert to 331.82 drivers.
____________
Greetings from TJ

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36625 - Posted: 24 Apr 2014 | 22:06:58 UTC

After some experimenting with the core clock I see that if I set it to 1060MHz with AfterBurner it stays at 1060 and GPU load is 66%. If I set he core clock to 875MHz, the GPU load increases to 71%. I have not seen this before but can imagine that it works like this.
Have set fan to max. 100% but stays at 75°C with clock at 1060MHz.

The first WU the 780Ti did with 337.50 driver is about 2000 seconds faster then with 331.82 but core clock was higher too. But temperature also and that bothers me the most as I know that my attic will become warmer in the coming weeks.
____________
Greetings from TJ

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36640 - Posted: 25 Apr 2014 | 12:22:32 UTC

Hello Jeremy, one more question if I may.
I have now set the temperature target at 72°C and the power target at 88. With a new WU starts the cards boast but after a few minutes the temp. rises and clock goes down. This is off course what I want. However with the fan at 90% the temperature is 74°C steady with a GPU load of 83-84%.

What have you set in PrecisionX at temp. and power targets?
By the way I am now running 337.50 beta driver.
____________
Greetings from TJ

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1837
Credit: 10,434,859,894
RAC: 8,761,703
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36648 - Posted: 25 Apr 2014 | 17:39:40 UTC - in response to Message 36550.

Try cuda60 version 841

(By the way, this reintroduces SWAN_SYNC - if you set that to 1 you should find you get improved performance)

Matt

Since this version has the SWAN_SYNC, I've upgraded my drivers, and this version is working fine.
However I had a strange period of getting tasks which led to the decision of upgrading the drivers:
My hosts had the 332.21 (and the 326.80) driver, and they received and completed CUDA 5.5 tasks normally (CPU time = RUN time).
Then my hosts received a couple of CUDA 6.0 tasks which all have failed.
Then my hosts received CUDA 4.2 tasks which all have completed successfully.
Then I've upgraded my drivers to 337.50, and now my hosts are receiving and completing CUDA 6.0 tasks normally.

After that, I've checked the nvcuda32.dll and the nvcuda64.dll in both drivers (332.21 and 337.50), and all four dll's state that they are "NVIDIA CUDA 6.0.1 drivers" (right click -> properties -> version tab -> Product name field), but they have different file sizes. So the CUDA 6 included in previous drivers than 334.21 is not working (to be polite).

Yeah - scheduling on the Cuda capability reported by the driver is insufficient - the 331s say they do, but they don't. We've reverted to giving cuda60s out only to 334+

My experiences assure this.
Why haven't they increased at least the last digit of the driver's version number? (You don't have to answer this)

Jeremy Zimmerman
Send message
Joined: 13 Apr 13
Posts: 61
Credit: 726,605,417
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwat
Message 36662 - Posted: 26 Apr 2014 | 3:53:30 UTC - in response to Message 36640.
Last modified: 26 Apr 2014 | 4:11:07 UTC

TJ,

Here are my Precision Settings for the machine with the 780Ti cards.

Power Target = 105%
Temp Target = 72°
Power and Temp are NOT linked
Prioritize Temp Target (click on arrow to point down) ***EDIT***
GPU Clock Offset = +38 (not of much good now that winter is over)
MEM Clock Offset = 0
Fan Curve = Auto
Under Fan Curve
Fan Speed Update 5000 msec
Temperature hysteresis (in °C) = 2
Force Fan Speed on each Period is not checked
35% at 30°C
40% at 50°C
45% at 60°C
60% at 65°C
100% at 70°C

Since I have the EVGA ACX cooling cards which just blow the hot air inside the case, I am using the Cooler Master Half932 case which can circulate air in and out pretty quick. Built a duct to take outside 'cool' air directly to the cpu, and then 4x120mm on the side of case blowing on the gpu's, and 3x120mm exhausting at top. Those fans are all linked to cpu temp so the fan profiles in Asus AI Suite3 are set to run them roughly where I want. Could also just run them straight from 12V, 7V, or 5V, but I like the AI Suite for cpu/case fan control.

I have a smaller half case for another machine, and both the video card and cpu could not run full clock until a side fan was installed when crunching. Side fans or open cases are critical for the non exhausting gpu's. It is really surprising what 100 cfm of outside air on a side panel blowing onto a gpu can do.

Regards,
Jeremy

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36672 - Posted: 26 Apr 2014 | 15:24:50 UTC - in response to Message 36662.

Thank you very match Jeremy.

You have some settings different, that could explain my to high temperatures.
I will go for your settings and let it run for a day to see how that goes.
I have one 20cm fan in the top and 14cm at the back. No side fan.


____________
Greetings from TJ

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36693 - Posted: 27 Apr 2014 | 15:34:54 UTC - in response to Message 36662.

Hello Jeremy,

I have used your settings and indeed the card never became above 72°C, and GPU use is ~86% for a Gianni Ligand with driver 331.82

So perhaps I will update to the latest driver again and then with your settings still in place see the results. But I like few more WU's finished first to compare.
But I am very glad that I now know how I can keep the GPU at 72°. So thanks again for your help.
Kindest regards.


____________
Greetings from TJ

Jeremy Zimmerman
Send message
Joined: 13 Apr 13
Posts: 61
Credit: 726,605,417
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwat
Message 36696 - Posted: 27 Apr 2014 | 17:00:58 UTC - in response to Message 36693.

TJ,

Glad to hear it is replicated. The only issue I am facing now is the downclock of GPU speed.

* Only happening to GPU1 and not GPU2 (identical cards/bios).
* Downclocks to 548Mhz.
* Shutdown/Restart of Boinc will not reset the speed.
* Must shutdown and restart the computer to get speed resent and may go 0-3 days before downclocking again.
* Is not a thermal issue since the card is staying 72°C (actually ticks to 73 on occasion).
* Happens with both 335.23 and 337.50 drivers.
* Was not happening on the 331.82 drivers.
* Does not happen to the GTX680/GTX460 XP machines with the 331 or 335 drivers.
* I have not tried Jacob Klein's force Max Boost speed yet because I liked the temperature control of Precision. The 780Ti will sit max boost at 68C on the low utilization WU's and climbs to the upper 70's on high utilization WU's (without temp control set in Precision). So max boost would be a little rough with my current cooling setup.

So, I set for Max Boost, and it works wonderfully, even when the drivers would otherwise stupidly downclock due to supposed low utilization. Forcing Max Boost works wonders.
http://www.gpugrid.net/forum_thread.php?id=3647&nowrap=true#36320


So I will be going back to 331.82 today since I will not be able to watch the systems close. I think the new app 8.41 is now scheduling WU's correctly so I should be ok.

Regards,
Jeremy

TJ
Send message
Joined: 26 Jun 09
Posts: 815
Credit: 1,469,215,105
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36699 - Posted: 27 Apr 2014 | 17:58:12 UTC - in response to Message 36696.
Last modified: 27 Apr 2014 | 18:03:02 UTC

Indeed Jeremy, with 331.82 drivers you will get app 8.41 and cuda42. I have not had any errors since I revered back to those drivers, but that is only about 30 hours. My 780Ti runs now at 888MHz with 72°C. I am happy with that.
Good luck with your system downgrading the driver!

Edit: with those "old" drivers we can not use SWAN_SYNC what should give our big cards some extra performance.
____________
Greetings from TJ

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 783
Credit: 1,393,022,995
RAC: 1,224,664
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36793 - Posted: 4 May 2014 | 19:23:14 UTC - in response to Message 36550.

Try cuda60 version 841

(By the way, this reintroduces SWAN_SYNC - if you set that to 1 you should find you get improved performance)

Matt

I've had a small problem twice with cuda60 v8.41:

Task 9763720
Task 9784050

Both times, I noticed that the task had 'stalled' - had counted up an unusually large elapsed time, and was making no progress.

No message on screen, no crash. I simply suspended the task for a few seconds, then resumed it, and it started from the last checkpoint without any fuss.

Stderr has this error logged:

<stderr_txt>
# GPU [GeForce GTX 670] Platform [Windows] Rev [3301M] VERSION [60]
# SWAN Device 1 :
# Name : GeForce GTX 670
# ECC : Disabled
# Global mem : 2048MB
# Capability : 3.0
# PCI ID : 0000:08:00.0
# Device clock : 1084MHz
# Memory clock : 3054MHz
# Memory width : 256bit
# Driver version : r334_89 : 33523
# GPU 0 : 78C
# GPU 1 : 57C
# GPU 1 : 58C
# GPU 1 : 59C
# GPU 1 : 60C
SWAN : FATAL : Cuda driver error 719 in file 'swanlibnv2.cpp' in line 1965.
# SWAN swan_assert 0
# GPU [GeForce GTX 670] Platform [Windows] Rev [3301M] VERSION [60]
# SWAN Device 1 :
# Name : GeForce GTX 670
# ECC : Disabled
# Global mem : 2048MB
# Capability : 3.0
# PCI ID : 0000:08:00.0
# Device clock : 1084MHz
# Memory clock : 3054MHz
# Memory width : 256bit
# Driver version : r334_89 : 33523
# GPU 0 : 69C
# GPU 1 : 38C

(same both times)

Note how far the temperature of GPU 1 has fallen - the tasks were probably stalled for several hours before I noticed.

Also, that 'SWAN swan_assert 0' on restart is new - I don't have SWAN_SYNC set.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,827,709,814
RAC: 445,018
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36812 - Posted: 7 May 2014 | 21:28:59 UTC - in response to Message 36793.
Last modified: 7 May 2014 | 21:30:46 UTC

I would predominantly be worried about the message,

    SWAN : FATAL : Cuda driver error 719 in file 'swanlibnv2.cpp' in line 1965.

Apparent in both failures.
However, I would also be inclined to put the temps down to the buggy 335.23 driver; which not only fails to boost but on occasion downclocks, at least in my experience.
FWIW I suggest trying a 337.x driver...
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Richard Haselgrove
Send message
Joined: 11 Jul 09
Posts: 783
Credit: 1,393,022,995
RAC: 1,224,664
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36814 - Posted: 8 May 2014 | 10:28:51 UTC - in response to Message 36812.

I would predominantly be worried about the message,
    SWAN : FATAL : Cuda driver error 719 in file 'swanlibnv2.cpp' in line 1965.

Apparent in both failures.
However, I would also be inclined to put the temps down to the buggy 335.23 driver; which not only fails to boost but on occasion downclocks, at least in my experience.
FWIW I suggest trying a 337.x driver...


I had another one yesterday:

Task 10212213

Only seems to happen on Gerard's tasks, though this is a slightly different sub-type.

Yes, I'm primarily concerned about the SWAN FATAL - that should trigger a boinc_temporary_exit, but doesn't.

No, a 20 degree drop in temperature isn't the result of a drop in boost - it's a complete cessation of processing, for several hours. I caught yesterday's much sooner, and it only had time to cool down by 2 degrees.

337.50 is still in Beta - I'll leave that to the rest of you, thanks.

Jim1348
Send message
Joined: 28 Jul 12
Posts: 455
Credit: 1,130,760,908
RAC: 149,820
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36817 - Posted: 8 May 2014 | 13:39:23 UTC - in response to Message 36814.
Last modified: 8 May 2014 | 13:44:25 UTC

Only seems to happen on Gerard's tasks, though this is a slightly different sub-type.

I had a BSOD a few days ago (very unusual), which I put down to performing a Windows update. Something similar happened on another PC, which I thought was connected to the most recent security update to Internet Explorer. However, on looking though BoincTasks, it seems that I was running Gerards on both of the GTX 660s at the time (WinXP, 335.28 driver).

There is nothing in the Stderr output to indicate anything other than a shutdown to install the updates, but now I am beginning to wonder.
http://www.gpugrid.net/result.php?resultid=9785157
http://www.gpugrid.net/result.php?resultid=9786033

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,827,709,814
RAC: 445,018
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36818 - Posted: 9 May 2014 | 20:07:36 UTC - in response to Message 36817.

If the updates are automatic, that would do it!
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Jim1348
Send message
Joined: 28 Jul 12
Posts: 455
Credit: 1,130,760,908
RAC: 149,820
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwat
Message 36819 - Posted: 9 May 2014 | 23:21:16 UTC - in response to Message 36818.

If the updates are automatic, that would do it!

That was on a dedicated PC, and I have automatic updates turned off (it is WinXP after all). I leave it running 24/7, and so don't have to worry if the work unit has some sort of restart problem. But it looks like some of the Gerards do, since the BSOD occurred apparently right after the reboot. I say "apparently", because I don't normally have a monitor connected to it, but manage it over the LAN.

As I recall, it was only after a couple of hours that I noticed no progress, and no LAN connection, and then found the BSOD after connecting a monitor. It looks like bad luck, and probably won't happen again on that machine.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,827,709,814
RAC: 445,018
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 36821 - Posted: 10 May 2014 | 16:04:30 UTC - in response to Message 36819.

These things occasionally happen, for unknown reasons. You could run for several months without issue, get a blue screen, shut down, start up and not get another blue screen for months... It's good to post any concerns, just in case there is a bad batch, but it's best to wait and see if it was a one off before hitting the panic button and rebuilding your system. I also had a Gerard WU fail 6days ago on an otherwise good GPU and good system.
____________
FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

SJC_Steve
Send message
Joined: 31 Oct 12
Posts: 17
Credit: 167,541,702
RAC: 0
Level
Ile
Scientific publications
watwatwatwatwatwatwatwat
Message 38721 - Posted: 28 Oct 2014 | 0:11:54 UTC
Last modified: 28 Oct 2014 | 0:14:41 UTC

I upgraded my Nvidia drivers using xorg-edgers PPA to version 340.46 and now Boinc doesn't recognize my GPU.

I upgraded to Ubuntu 14.04.1 LTS with no change -- no GPU.

I then updated my Boinc-client to the latest version 7.4.23 with no change -- no GPU.

I reboot the system and restarted the boinc-client with no change -- no GPU.

Nvidia-smi shows;
+------------------------------------------------------+
| NVIDIA-SMI 340.46 Driver Version: 340.46 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 650... On | 0000:01:00.0 N/A | N/A |
| 37% 53C P0 N/A / N/A | 7MiB / 1023MiB | N/A Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Compute processes: GPU Memory |
| GPU PID Process name Usage |
|=============================================================================|
| 0 Not Supported |
+-----------------------------------------------------------------------------+

Messages from Boinc start-up;

1: 27-Oct-2014 17:46:13 (low) [] Starting BOINC client version 7.4.23 for x86_64-pc-linux-gnu
2: 27-Oct-2014 17:46:13 (low) [] log flags: file_xfer, sched_ops, task, coproc_debug
3: 27-Oct-2014 17:46:13 (low) [] Libraries: libcurl/7.35.0 OpenSSL/1.0.1f zlib/1.2.8 libidn/1.28 librtmp/2.3
4: 27-Oct-2014 17:46:13 (low) [] Data directory: /var/lib/boinc-client
5: 27-Oct-2014 17:46:13 (low) [] [coproc] launching child process at /usr/bin/boinc
6: 27-Oct-2014 17:46:13 (low) [] [coproc] relative to directory /var/lib/boinc-client
7: 27-Oct-2014 17:46:13 (low) [] [coproc] with data directory /var/lib/boinc-client
8: 27-Oct-2014 17:46:13 (low) [] NVIDIA drivers present but no GPUs found
9: 27-Oct-2014 17:46:13 (low) [] No ATI library found
10: 27-Oct-2014 17:46:13 (low) [] clGetPlatformIDs() failed to return any OpenCL platforms
11: 27-Oct-2014 17:46:13 (low) [] No usable GPUs found

Any advise on how to get my cruncher back crunching?
Thanks,
Steve

Vagelis Giannadakis
Send message
Joined: 5 May 13
Posts: 187
Credit: 349,254,454
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwat
Message 38723 - Posted: 28 Oct 2014 | 6:04:37 UTC - in response to Message 38721.

Maybe it is something with the NVIDIA driver, can you try uninstalling the xorg-edgers driver and install the original NVIDIA driver?
____________

RaymondFO*
Send message
Joined: 22 Nov 12
Posts: 72
Credit: 10,188,083,039
RAC: 8,735,925
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwat
Message 38725 - Posted: 28 Oct 2014 | 12:41:08 UTC - in response to Message 38721.

I upgraded my Nvidia drivers using xorg-edgers PPA to version 340.46 and now Boinc doesn't recognize my GPU.

I upgraded to Ubuntu 14.04.1 LTS with no change -- no GPU.

I then updated my Boinc-client to the latest version 7.4.23 with no change -- no GPU.

I reboot the system and restarted the boinc-client with no change -- no GPU.



This has been discussed elsewhere here. I believe this is a BOINC error that causes this issue and it would be best to uninstall all NVIDIA drivers and download the video driver from NVIDIA and re-install them manually (http://www.gpugrid.net/forum_thread.php?id=3736&nowrap=true#38456)

John C MacAlister
Send message
Joined: 17 Feb 13
Posts: 178
Credit: 132,357,411
RAC: 19,912
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 38896 - Posted: 13 Nov 2014 | 10:15:30 UTC

I am using driver 334.89. NVIDIA tells me this is the latest driver for my two GTX 650Ti GPUs.

Not being particularly tech savvy, I can't do any more. Not sure what effect, if any, this will have on my ability to process after 1 January 2015.

John

eXaPower
Send message
Joined: 25 Sep 13
Posts: 263
Credit: 1,006,020,567
RAC: 2,059,354
Level
Met
Scientific publications
watwatwatwatwatwat
Message 38899 - Posted: 13 Nov 2014 | 13:35:57 UTC - in response to Message 38896.

John: 344.65 driver is available for you're OS with (650ti) Kelper cards.

http://www.nvidia.com/download/driverResults.aspx/79527/en-us

John C MacAlister
Send message
Joined: 17 Feb 13
Posts: 178
Credit: 132,357,411
RAC: 19,912
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 38901 - Posted: 13 Nov 2014 | 17:27:33 UTC - in response to Message 38899.

many thanks: I will try to install that driver.


John: 344.65 driver is available for you're OS with (650ti) Kelper cards.

http://www.nvidia.com/download/driverResults.aspx/79527/en-us

John C MacAlister
Send message
Joined: 17 Feb 13
Posts: 178
Credit: 132,357,411
RAC: 19,912
Level
Cys
Scientific publications
watwatwatwatwatwatwatwat
Message 39510 - Posted: 17 Jan 2015 | 17:51:43 UTC

Driver 347.09 installed and I am downloading CUDA 6.5 task.

Message boards : News : Please upgrade to DRIVER 334.21 or NEWER [closed]