Advanced search

Message boards : Graphics cards (GPUs) : WARNING: don't use 6.6.1 - 6.6.20! (windows)

Author Message
ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9543 - Posted: 9 May 2009 | 14:10:03 UTC
Last modified: 21 May 2009 | 10:13:37 UTC

Hi folks,

seriously, I don't know why 6.6.20 became a recommended version. It has serious issues with work fetching, scheduling and debt handling. We expected problems due to this, but as it turns out another of 6.6.20s flaws is hitting us hard:

There are massive complains about WUs taking forever. This is a bug introduced a few versions prior to 6.6.20 and (supposedly) fixed in 6.6.23. So if you want to run a 6.6-series client, upgrade at least to 6.6.23! UPDATE: 6.6.28 is officially recommended by UCB now.

This version still has issues, so if you don't like these I recommend 6.5.0(*), which doesn't have any of them. The only issue I know of is that it doesn't differentiate between cpu and gpu projects regarding the ressource share, so be sure to put in some balanced numbers. For 4 cpu cores and one gpu give GPU-Grid about 20% ressource share, otherwise your cache will be messed up.

You could also use 6.4.7 (the old recommended version), but this may require using a cc_config.xml to use all cpu cores.. so why bother.

All versions can be found here.

(*) not speaking officially for the project

MrS
____________
Scanning for our furry friends since Jan 2002

SkyeHunter
Send message
Joined: 7 Mar 09
Posts: 12
Credit: 1,254,285
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 9582 - Posted: 10 May 2009 | 9:47:01 UTC - in response to Message 9543.

I recommend 6.5.0...


After running in the never endig WU issue, and consulting some fora, I migrated towards 6.5.0. However, this morning I discovered a new never ending WU. Just like the previous ones, It was a KASHIF job...

I re-installed 6.6.20 and the WU resumed again... I'll have to keep a close eye on it, but I wonder if their is an issue with KASHIF jobs (didn't have any problem with IBUCH WU's)...

I don't know if other projects experience the same issues.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9584 - Posted: 10 May 2009 | 11:04:54 UTC - in response to Message 9582.

There is a strange problem with some "KASHIF_HIVPR" and "IBUCH_KID" WUs. But so far it hasn't been linked to hanging tasks. There have also been rare cases of hanging WUs before, which could be resolved by a simple BOINC restart. In the case of 6.6.20 the restart doesn't help for long, as far as I understand.

MrS
____________
Scanning for our furry friends since Jan 2002

[AF] Profanateur
Avatar
Send message
Joined: 25 Oct 08
Posts: 42
Credit: 42,812,268
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 9604 - Posted: 10 May 2009 | 20:16:30 UTC

I update to 6.6.23 ans I have always the same failure :/

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 9621 - Posted: 10 May 2009 | 23:09:52 UTC - in response to Message 9604.

I update to 6.6.23 ans I have always the same failure :/

c:\cygwin\home\speechserver\gpumd2\src\pme\CPME_cufft.cu

I would understand from this that you are running BOINC under a linux emulator? Or are you just running a linux emulator for some other software?

At any rate, that is going to be an issue. Emulators do not allow driect access to hardware which is needed for CUDA, and secondly, when the video driver is "virtualized" it is going to blow up any running CUDA tasks just as if you used remote desktop to look at your PC from remote locations.

Stop using cygwin and the see if the tasks run to completion.

Michael Doerner
Send message
Joined: 28 Feb 09
Posts: 37
Credit: 666,889
RAC: 0
Level
Gly
Scientific publications
watwatwatwat
Message 9625 - Posted: 10 May 2009 | 23:50:24 UTC - in response to Message 9622.

I've had no issues w/ 6.6.20 on Linux 64-bit. Then again, I only have a measley 9600 GSO....:-( Not even SLI, just the one card.

Mike Doerner
____________

dyeman
Send message
Joined: 21 Mar 09
Posts: 35
Credit: 591,434,551
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9628 - Posted: 11 May 2009 | 5:52:04 UTC - in response to Message 9584.

I had some hanging tasks. Upgrading Nvidia drivers from 182.06 to 185.85 seems to have cured the problem (and the WUs run a bit faster also).

Profile [AF>DoJ] supersonic
Send message
Joined: 8 Nov 08
Posts: 8
Credit: 3,032,744
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwat
Message 9636 - Posted: 11 May 2009 | 12:43:58 UTC

I had a never ending wu on a boinc 6.4.7

it's a IBUCH_KID

this wu

it took me 5 days to realise, as the computing card is far from where I live and work.

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9650 - Posted: 11 May 2009 | 20:54:05 UTC

@Paul: the cygwin appears relatively often in error messages, so I think it's related to the compiling machine.

@Michael: the bug solved with 6.6.23 was first described as WUs taking about 4 times longer, not really hanging. So one would have to watch the machine or the RAC closely to spot the problems. And I don't know how different the win and linux builds are, the problem may very well not exist under linux.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 9664 - Posted: 12 May 2009 | 5:01:21 UTC - in response to Message 9650.

@Paul: the cygwin appears relatively often in error messages, so I think it's related to the compiling machine.

@Michael: the bug solved with 6.6.23 was first described as WUs taking about 4 times longer, not really hanging. So one would have to watch the machine or the RAC closely to spot the problems. And I don't know how different the win and linux builds are, the problem may very well not exist under linux.

MrS

Yeah, I just crashed one on my other rig. I was interested in that error I had the other day and was curious if it was crashed because of the game or the task. It appears that if the game crashes the driver, and I just did it again tonight. Then, you can also crash the task.. which I did url=KX20708-SH2_US_8-1-10-SH2_US_8670000_0]in this case;/url]. Now what was interesting to me was the fact that the game is pretty stable in the sense that I played a full scenario through (about 20 some hours) and it did not have a problem.

And, yes I got the:

Cuda error: Kernel [fft_data_swizzle_out] failed in file
'c:\cygwin\home\speechserver\gpumd2\src\pme\CPME_cufft.cu' in line 61 : the launch timed out and was terminated.


Hmm, looks like I toasted a couple more tasks ...

Cuda error: Kernel [copy_mul] failed in file 'com.cu' in line 46 : the launch timed out and was terminated.


Seems to be the error if you crashed the Nvida driver kernel. Not a big deal for me ... this is the slower system so it can afford to not have as many tasks queued ...

Anyway,

p20000-RAUL_pYEpYI1205-0-10-RND6536_0
p10000-RAUL_pYEpYI1205-0-10-RND6116_0

Crashed because of the game killing the driver ... sigh ... this stuff should be easier than this by now ...

TomaszPawel
Send message
Joined: 18 Aug 08
Posts: 121
Credit: 59,836,411
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9671 - Posted: 12 May 2009 | 12:54:56 UTC - in response to Message 9664.

If you want 6.6.20 to work good, you must do clean install.

Uninstal your previous version of BOINC

Delete all files and folders of BOINC

Instal 6.6.20

Have fun.

I do this and no problem so far.
____________
POLISH NATIONAL TEAM - Join! Crunch! Win!

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 9680 - Posted: 12 May 2009 | 20:01:38 UTC - in response to Message 9671.

I do this and no problem so far.

You will ... :)

The problem is not the install. It is the debts. People that do an update install do not get the debts reset and that causes issues. A clean install fixes that. So does an install and debt reset using the cc_config file.

The problem is that the LTD calculations are not correct. Not that they listen to me when I prove a bug, even down to the lines of code ... but I have not found where the issue might be (yet) partly because there have been other pressing issues to chase.

Though it may be time to take another look...

Part of the problem is that there are two parts to the problem and I am not sure which one has the most impact. They have already told me I am wrong about part of the first problem (RR SIm has issues, several issues not the least of which is that it does not model the actual system in the way it will process the tasks, I already proved that once though they have ignored my notes)...

LTD calculations also seem to be hammered and there seems to be issues with it heading negative forever ... requires occasional debt resets, the faster the system the more often the resets...

uBronan
Avatar
Send message
Joined: 1 Feb 09
Posts: 139
Credit: 575,023
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 9705 - Posted: 13 May 2009 | 12:05:24 UTC

Well yes i reset the debts a few times a month not sure what it does screw up though but works fine for me.

Profile Valentin Kolinko
Send message
Joined: 1 Apr 09
Posts: 7
Credit: 10,165,794
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 9747 - Posted: 14 May 2009 | 11:06:19 UTC
Last modified: 14 May 2009 | 11:12:52 UTC

Hello. Sorry i bad speak english. I used boinc 6.6.20 and my work never end. Why? My system: XP x64, GPU: gf 9600GT
last version nvidea driver

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 9765 - Posted: 14 May 2009 | 16:20:35 UTC - in response to Message 9705.

Well yes i reset the debts a few times a month not sure what it does screw up though but works fine for me.

Because you reset the LTD it starts fresh all the time as far as allocating time to projects. Which is not a HUGE deal, but means that if a project has been off line for a bit, you lose the debt you "owe" them and don't balance things properly IAW your shares.

I am pretty sure there are more issues with the calculations though so not sure how well it is tracking teh LTD as it is ... I can't prove it yet and am chasing other issues so, don't want to take on another set of problems yet.

I think there are also major problems with RR Sim but cannot prove them yet... again, have not really started to look at that problem either ... though my preliminary look told me it was hammered ...

Profile Valentin Kolinko
Send message
Joined: 1 Apr 09
Posts: 7
Credit: 10,165,794
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 9777 - Posted: 14 May 2009 | 19:27:33 UTC

somebody will prompt that to me to do? upgrade to the version 6.6.23 ?
How many on time occupy calculations on 9600 GT ?

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9785 - Posted: 14 May 2009 | 20:58:33 UTC - in response to Message 9777.

upgrade to the version 6.6.23 ?


That's part of the first post in this thread.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Valentin Kolinko
Send message
Joined: 1 Apr 09
Posts: 7
Credit: 10,165,794
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 9801 - Posted: 15 May 2009 | 14:24:19 UTC

Greetings! I upgrade BOINC to version 6.6.23, GPUGRID do not work! Writes "Waits for the turn" and calculations do not go! ((
Please write what version of the client the working and what version of the driver approaches is better?
P.S. XP x64, GeForce 9600GT, driver: 182.50
Please help

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 9806 - Posted: 15 May 2009 | 19:55:49 UTC - in response to Message 9801.

Greetings! I upgrade BOINC to version 6.6.23, GPUGRID do not work! Writes "Waits for the turn" and calculations do not go! ((
Please write what version of the client the working and what version of the driver approaches is better?
P.S. XP x64, GeForce 9600GT, driver: 182.50
Please help


I am not absolutely sure about the 64-bit version of the 182.50 driver, but I do know that is one of the preferred versions on XP 32-bit.

As far as BOINC goes, 6.5.0, 6.6.23, or now 6.6.28 are the versions that I use and suggest.

I am not familiar with the message you describe. are you sure you do not have the task or project suspended?

can you take a picture of the BM window and post a link to it?

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9809 - Posted: 15 May 2009 | 20:21:32 UTC

I also don't understand what you mean, but I guess this is the answer.

MrS
____________
Scanning for our furry friends since Jan 2002

Profile Valentin Kolinko
Send message
Joined: 1 Apr 09
Posts: 7
Credit: 10,165,794
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwat
Message 9839 - Posted: 16 May 2009 | 11:29:17 UTC - in response to Message 9809.

I also don't understand what you mean, but I guess this is the answer.

MrS



thankx! it's work!
somebody speak at russian?

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 9853 - Posted: 16 May 2009 | 12:56:59 UTC - in response to Message 9839.

Net ;)

MrS
____________
Scanning for our furry friends since Jan 2002

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10258 - Posted: 28 May 2009 | 15:57:50 UTC - in response to Message 9543.

I see there is Boinc 6.6.28 now available - I am going to give that a go.
I have had nothing but problems with 6.6.20.
Usually I get some way into a project run, then it just finishes for no reason - it says something like uploading. After 84% and two or three days into a run, this is not what you want, 4 or 5h short of the finishing mark!

After several successive failures I gave up on it in the hope that things would get sorted out. So hopefully they have been.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 484
Credit: 554,932,448
RAC: 10,334
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10279 - Posted: 29 May 2009 | 1:35:32 UTC

I thought I'd prepared to use more than one CUDA - BOINC 6.2.28, and created cc_config.xml. However, only one CUDA is recognized. I tried upgrading the driver for my Nvidia 9800 GT using the Microsoft option to do this; however this actually downgraded the driver from 7.15.11.8250 to 7.15.11.7578 instead and stopped any BOINC use of the CUDA at all until I reversed this downgrade.

Where can I find the proper site to look for new drivers and is there any other one recommended for this card?

Where can I find some Nvidia software for counting the number of CUDAs present to see if the expected second one was actually installed?

Does an Nvidia card CUDA need a monitor attached before GPUGRID can use it? I don't have space to add a second monitor.

In case it makes any difference, I also installed the 64-bit Vista SP2 update on my machine about this time.

MarkJ
Volunteer moderator
Project tester
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 735
Credit: 198,776,069
RAC: 10,556
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10297 - Posted: 29 May 2009 | 12:58:16 UTC - in response to Message 10279.

I thought I'd prepared to use more than one CUDA - BOINC 6.2.28, and created cc_config.xml. However, only one CUDA is recognized. I tried upgrading the driver for my Nvidia 9800 GT using the Microsoft option to do this; however this actually downgraded the driver from 7.15.11.8250 to 7.15.11.7578 instead and stopped any BOINC use of the CUDA at all until I reversed this downgrade.

Where can I find the proper site to look for new drivers and is there any other one recommended for this card?

Where can I find some Nvidia software for counting the number of CUDAs present to see if the expected second one was actually installed?

Does an Nvidia card CUDA need a monitor attached before GPUGRID can use it? I don't have space to add a second monitor.

In case it makes any difference, I also installed the 64-bit Vista SP2 update on my machine about this time.


What did you put in your cc_config file. Was it like this?


<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>


You can download nvidia drivers directly from the nvidia web site here. It would seem the 182.50 drivers are more reliable with GPUgrid work, even though they aren't the latest any more.

I understand that you may need to extend your desktop onto the other card to get it to see it under Vista. I use XP which behaves differently.
____________
BOINC blog

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 484
Credit: 554,932,448
RAC: 10,334
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10302 - Posted: 29 May 2009 | 15:13:01 UTC - in response to Message 10297.


What did you put in your cc_config file. Was it like this?


<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>


You can download nvidia drivers directly from the nvidia web site here. It would seem the 182.50 drivers are more reliable with GPUgrid work, even though they aren't the latest any more.

I understand that you may need to extend your desktop onto the other card to get it to see it under Vista. I use XP which behaves differently.


Your cc_config.xml file appears to start with a blank line; mine doesn't, but the rest is identical. Is the blank line required?

The driver download site says I'm already using 182.50 and offers only 185.85 for download.

How do I extend the desktop onto the other card without using a second monitor? How do I check if I even have a second card without opening the case?

anthonmg
Send message
Joined: 11 Apr 09
Posts: 17
Credit: 11,086,149
RAC: 0
Level
Pro
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 10307 - Posted: 29 May 2009 | 17:58:24 UTC - in response to Message 9621.

I have to use remote desktop into my computer with the CUDA card for business purposes and it keeps crashing the runs. Strangely I've been doing this since I started with the project months ago and it only started causing problems recently. Since I can't get around it (if I log in and try to pause the job it will crash the unit before I can even get to it). Thus, I'm detaching from the project since it's taking too much time to try and figure out the issues and if remote desktop is going to act like that I have no way forward. It's been a pleasure and best of luck to the project.

pharrg
Send message
Joined: 12 Jan 09
Posts: 36
Credit: 1,075,543
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 10378 - Posted: 2 Jun 2009 | 15:35:11 UTC
Last modified: 2 Jun 2009 | 15:42:12 UTC

I'm running Vista 64bit with SP2 and things are running fine. I didn't extend the desktop or any other changes, other than to disable SLI. You have to do that to let BOINC see each card as a seperate device since CUDA doesn't support SLI yet. Also, check after updating drivers since I've seen that installing a new driver turns SLI back on sometimes. The SLI limitation is an issue with nVidia since they develop the CUDA language. You'll see that after you turn it off, in the messages you see when you start BOINC back up each card now shows as a different CUDA device (exit out of BOINC completely before making SLI changes so it can redect). The other thing to watch is heat on the video cards. I have to set my machine to crank up the fans on the video cards to about 95% speed to keep them cool while doing CUDA processing. If your cards overheat, you'll get errors and crashes. Otherwise, latest drivers and turn off SLI and mine is humming along on every project I run.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 484
Credit: 554,932,448
RAC: 10,334
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10383 - Posted: 2 Jun 2009 | 16:20:53 UTC - in response to Message 10378.

Where do you find the software to disable SLI under 64-bit Vista SP2?

By the latest drivers, do you mean 182 series or 185 series? For what Nvidia card?

How do you exit BOINC completely? I've found that since the Vista SP2 update, at least for the 32-bit version, you need to exit BOINC completely to allow full backups to complete without fatal errors, and I'm looking for a way to do this that doesn't have errors in all workunits already in progress as a side effect.

Profile Aardvark
Avatar
Send message
Joined: 27 Nov 08
Posts: 28
Credit: 82,362,324
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 10384 - Posted: 2 Jun 2009 | 17:20:27 UTC
Last modified: 2 Jun 2009 | 17:33:35 UTC

The latest driver available from Nvidia is 185.85. To disable SLI right click anywhere on your screen & open NVIDIA control panel. From here left click on "set multi-GPU and PhysX configuration". Then sellect "Do not use multi-GPU mode" then click on the "Apply" button at bottom of window, this will turn off SLI. To switch on SLI just reverse process (more or less). All that is required to see both GPU's on this card is to disable SLI. No need to extend desktop. Try with SLI disabled first, should be OK.
I am running 64 bit Vista SP2 with 185.85 driver and BOINC 6.6.31 on an Nvidia 9800 GX2 (dual GPU) with no problems. I normally exit BOINC by right clicking on the BOINC icon in the task bar & firstly set "snooze" and then sellect "exit". I found that if I just exited BOINC then the tasks would remain running. The 9800GX2 is a hot card, which seems to like running at 80-85 degrees C. I have left the fans on automatic and have noticed no issues to date.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,995,359,260
RAC: 0
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10388 - Posted: 2 Jun 2009 | 19:57:01 UTC - in response to Message 10384.

GPUZ

http://www.techpowerup.com/downloads/1379o/GPU-Z.0.3.4.exe

This will tell you the info about your cards and if SLI is enabled, I think!

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 484
Credit: 554,932,448
RAC: 10,334
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10400 - Posted: 3 Jun 2009 | 0:22:17 UTC - in response to Message 10384.
Last modified: 3 Jun 2009 | 0:42:50 UTC

The latest driver available from Nvidia is 185.85. To disable SLI right click anywhere on your screen & open NVIDIA control panel. From here left click on "set multi-GPU and PhysX configuration". Then sellect "Do not use multi-GPU mode" then click on the "Apply" button at bottom of window, this will turn off SLI. To switch on SLI just reverse process (more or less). All that is required to see both GPU's on this card is to disable SLI. No need to extend desktop. Try with SLI disabled first, should be OK.


I opened the NVIDIA control panel and found a section labelled Set PhysX configuration, but nothing referring to multi-GPU mode.

I'm preparing to try 185.85 again tomorrow, though - don't want to switch drivers in the middle of a workunit.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 484
Credit: 554,932,448
RAC: 10,334
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10401 - Posted: 3 Jun 2009 | 0:28:14 UTC - in response to Message 10388.
Last modified: 3 Jun 2009 | 0:56:19 UTC

GPUZ

http://www.techpowerup.com/downloads/1379o/GPU-Z.0.3.4.exe

This will tell you the info about your cards and if SLI is enabled, I think!


When I try to run it on my 64-bit Vista SP2 machine, I get the error message:

C:\Users\Robert\Downloads\GPU-Z.0.3.4.exe is not a valid Win32 application.

I haven't bothered trying it on my 32-bit Vista SP2 machine, where I already know that the Nvidia card is not one GPUGRID can use.

pharrg
Send message
Joined: 12 Jan 09
Posts: 36
Credit: 1,075,543
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwatwatwatwatwatwatwat
Message 10402 - Posted: 3 Jun 2009 | 1:09:35 UTC
Last modified: 3 Jun 2009 | 1:10:27 UTC

My nVidia control panel appears a little different from what others above described, but it may be hardware related since I'm running multiple GTX260's. In my control panel, go to the section for 3D Settings, then under that Set PhysX and SLI Configuration. On that page, section 1 is for PhysX, choose enable. In section 2 for SLI, click Disable. Then click the apply button. It will take a few seconds and you may see your screen blink off for a second.

To exit BOINC, simply rick click the icon in the system tray and choose exit. If you have the manager window open, go to the file menu and choose exit. If the project software is written well, it should save the work and pick up where it left off next time BOINC runs.

Oh, and I am running nVidia driver version 185.85 and BOINC version 6.6.28.

Profile Aardvark
Avatar
Send message
Joined: 27 Nov 08
Posts: 28
Credit: 82,362,324
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 10413 - Posted: 3 Jun 2009 | 13:46:51 UTC - in response to Message 10400.
Last modified: 3 Jun 2009 | 13:50:02 UTC

[quote]
I opened the NVIDIA control panel and found a section labelled Set PhysX configuration, but nothing referring to multi-GPU mode.
[quote]

What type of graphics card are you using ? If it is a single card with a single GPU then SLI will noy apply, and can be ignored. SLI is only used to link the 3D rendering functions of multiple display cards or cards with more than one GPU.

Profile robertmiles
Send message
Joined: 16 Apr 09
Posts: 484
Credit: 554,932,448
RAC: 10,334
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10414 - Posted: 3 Jun 2009 | 15:54:03 UTC - in response to Message 10413.
Last modified: 3 Jun 2009 | 15:58:34 UTC

I have a Geforce 9800GT. The specs that came with my computer weren't very clear on whether there's more than one. I'm no longer able to lift the computer and check inside. The specs did suggest that it should be able to control two monitors, though.

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 10417 - Posted: 3 Jun 2009 | 21:06:07 UTC - in response to Message 10414.

I have a Geforce 9800GT. The specs that came with my computer weren't very clear on whether there's more than one. I'm no longer able to lift the computer and check inside. The specs did suggest that it should be able to control two monitors, though.

Unless you have two cards in the system that should be a single GPU card ...

popandbob
Send message
Joined: 18 Jul 07
Posts: 67
Credit: 39,955,241
RAC: 2,990
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10482 - Posted: 12 Jun 2009 | 0:00:02 UTC
Last modified: 12 Jun 2009 | 0:01:11 UTC

I just had a long running task running 6.6.23
It took 17hours elapsed

Task: 452-GIANNI_BIND

Bob

Profile Paul D. Buck
Send message
Joined: 9 Jun 08
Posts: 1050
Credit: 37,321,185
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 10491 - Posted: 12 Jun 2009 | 17:05:26 UTC - in response to Message 10482.

I just had a long running task running 6.6.23
It took 17hours elapsed

Task: 452-GIANNI_BIND

Bob

If you see one of these in the future, where it seems to be running long, try stopping and restarting BOINC. Make sure you have the Science apps stopped. Also, try to see (before you do that) if the clocks on the GPU have been lowered.

We are trying to track down this bug ... so far to no effect ... it does seem to be something BOINC is doing in that version changes affect how often it happens ...

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10519 - Posted: 12 Jun 2009 | 22:25:08 UTC - in response to Message 10482.

Your card went into 2D mode, 800 MHz shader clock.

MrS
____________
Scanning for our furry friends since Jan 2002

popandbob
Send message
Joined: 18 Jul 07
Posts: 67
Credit: 39,955,241
RAC: 2,990
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10523 - Posted: 13 Jun 2009 | 4:41:29 UTC - in response to Message 10519.

Ok so 2D clocks would indeed cause the slowdown...
I've also had an issue with S@H tasks taking longer than they should (~20 min each versus about 7.5min) and eventually one errors out and locks up the computer.
I dunno whether its cuda or the apps but this is indeed a problem.
Bob

zoom314
Avatar
Send message
Joined: 13 Jun 09
Posts: 26
Credit: 2,309,584
RAC: 0
Level
Ala
Scientific publications
watwatwatwatwat
Message 10540 - Posted: 13 Jun 2009 | 21:13:00 UTC - in response to Message 9543.

I don't know, I'm using 6.6.20 with no problems, But then like another poster I run XP x64, I tried 6.6.28 and the WU's that I have were put into permanent waiting mode and My video driver is 185.85(NV and WHQL approved) and so I went back to 6.6.20, 6.6.31 to 6.6.36 seem to be having problems, Some of It gpu related.
____________

ExtraTerrestrial Apes
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 17 Aug 08
Posts: 2690
Credit: 1,265,819,765
RAC: 394,214
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 10548 - Posted: 14 Jun 2009 | 9:21:40 UTC - in response to Message 10540.

Depending on your crunching speed and configuration (possibly ressource share) it may take a while for the error to appear. And depending on how closely you monitor your systems you may not notice them.

MrS
____________
Scanning for our furry friends since Jan 2002

Post to thread

Message boards : Graphics cards (GPUs) : WARNING: don't use 6.6.1 - 6.6.20! (windows)