New Nvidia Driver 197.13

Author	Message
Sten-Arne Volunteer tester Send message Joined: 1 Nov 08 Posts: 2688 Credit: 7,662,017 RAC: 22,666	Message 983192 - Posted: 24 Mar 2010, 19:57:38 UTC - in response to Message 983186.
New host ID, oldest Wu was sent 6 Nov 2009 8:18:54 UTC All tasks for computer 5337951 Claggy Hmm, and why did it get a new host ID? Would adding hardware automatically create a new host ID, but keep the totals? Sten-Arne
ID: 983192 ·

Cliff Harding Volunteer tester Send message Joined: 18 Aug 99 Posts: 607 Credit: 34,337,463 RAC: 55,837	Message 983585 - Posted: 25 Mar 2010, 18:38:49 UTC
The 197.13 version was releasd into production this morning!!!. ____________ I don't buy computers, I build them!!
ID: 983585 ·

[AF>france>pas-de-calais]symaski62 Volunteer tester Send message Joined: 12 Aug 05 Posts: 254 Credit: 69,988 RAC: 0	Message 984401 - Posted: 27 Mar 2010, 7:43:31 UTC Last modified: 27 Mar 2010, 7:49:46 UTC
17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak) 256MB RAM => 244MB RAM --------------- Display Devices --------------- Card name: NVIDIA GeForce 8400 GS Manufacturer: NVIDIA Chip type: GeForce 8400 GS Display Memory: 499 MB Dedicated Memory: 243 MB Shared Memory: 255 MB ____________ SETI@Home Informational message -9 result_overflow with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.
ID: 984401 ·

MarkJ Volunteer tester Send message Joined: 17 Feb 08 Posts: 912 Credit: 14,561,754 RAC: 8,832	Message 984431 - Posted: 27 Mar 2010, 11:33:54 UTC - in response to Message 984401.
17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak) 256MB RAM => 244MB RAM --------------- Display Devices --------------- Card name: NVIDIA GeForce 8400 GS Manufacturer: NVIDIA Chip type: GeForce 8400 GS Display Memory: 499 MB Dedicated Memory: 243 MB Shared Memory: 255 MB Interesting, so that means the driver is now reporting less memory available (or its using more). ____________ BOINC blog
ID: 984431 ·

Wembley Volunteer tester Send message Joined: 16 Sep 09 Posts: 415 Credit: 879,818 RAC: 0	Message 984492 - Posted: 27 Mar 2010, 17:09:58 UTC - in response to Message 984431.
My system is still reporting the proper amount of memory after upgrading the driver: NVIDIA GPU 0: GeForce 9500 GT (driver version 19713, CUDA version 3000, compute capability 1.1, 1024MB, 90 GFLOPS peak) I suspect something other than the driver is taking up more memory on your card. ____________ Donate with your searches and online buys: http://www.goodsearch.com/toolbar/university-of-california-setihome
ID: 984492 ·

Raistmer Volunteer developer Volunteer tester Send message Joined: 16 Jun 01 Posts: 2542 Credit: 25,662,021 RAC: 39,170	Message 984494 - Posted: 27 Mar 2010, 17:28:47 UTC - in response to Message 983175.
Found another CUDA-blower, this time bloating the database by 13,331: host 5337951. This is beginning to get a bit worrying. Yes, with advance of GPU computing current quota-based protection system is completely obsolete. It can't effectively block damaged host even if it does CPU+ single GPU computations, but if it has few GPUs... no way to stop database polluting at all now.
ID: 984494 ·

[AF>france>pas-de-calais]symaski62 Volunteer tester Send message Joined: 12 Aug 05 Posts: 254 Credit: 69,988 RAC: 0	Message 984510 - Posted: 27 Mar 2010, 18:25:57 UTC - in response to Message 984431. Last modified: 27 Mar 2010, 18:26:57 UTC
17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak) 256MB RAM => 244MB RAM --------------- Display Devices --------------- Card name: NVIDIA GeForce 8400 GS Manufacturer: NVIDIA Chip type: GeForce 8400 GS Display Memory: 499 MB Dedicated Memory: 243 MB Shared Memory: 255 MB Interesting, so that means the driver is now reporting less memory available (or its using more). stdoutdae.txt 27/03/2010 07:41:03 [---] NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 27/03/2010 08:36:09 [---] NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak) ____________ SETI@Home Informational message -9 result_overflow with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.
ID: 984510 ·

Claggy Volunteer tester Send message Joined: 5 Jul 99 Posts: 3363 Credit: 25,950,143 RAC: 1,098	Message 988705 - Posted: 12 Apr 2010, 20:58:53 UTC Last modified: 12 Apr 2010, 20:59:42 UTC
I've posted to Nvidia's Forum asking what has changed in the 197.xx drivers to cause Boinc and CUDA-Z to report less memory, No replies yet, feel free to add comments there: Cuda-Z and Boinc report less Memory with 197.xx drivers On my laptop, the reduction of reported memory caused Collatz to no longer send Cuda tasks, needed 126Mb, had 113Mb, reverted to an app_info for now, on the Desktop PC, (with 2 OS's) the reduction in reported memory caused downloaded Stock Seti Beta 6.09 tasks to not start because of lack of memory, needed something like 384Mb, had 382Mb. Claggy
ID: 988705 ·

[AF>france>pas-de-calais]symaski62 Volunteer tester Send message Joined: 12 Aug 05 Posts: 254 Credit: 69,988 RAC: 0	Message 990101 - Posted: 18 Apr 2010, 13:30:42 UTC - in response to Message 988705.
http://cuda-z.sourceforge.net/ memory 8400 GS => 243.562 MB total Global :) ____________ SETI@Home Informational message -9 result_overflow with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.
ID: 990101 ·

whoDunIt Send message Joined: 18 Sep 05 Posts: 2 Credit: 1,573,415 RAC: 0	Message 990585 - Posted: 20 Apr 2010, 3:32:43 UTC - in response to Message 990101. Last modified: 20 Apr 2010, 3:34:00 UTC
I had this same problem when switching to the 197.13 driver. It didnt really matter for the MSI 9500 GT 512MB card but for my BFG 8600 GT 256MB card i couldnt get any new astropulse tasks, at the time thats all that was sending out, as it reported my memory at 243MB so i went back to the older driver 196.21. Same with the 197.25 driver. Claggy please fill us in if you get any reply. NVIDIA GeForce 9500 GT (499MB) driver: 19713 NVIDIA GeForce 8600 GT (256MB) driver: 19621 Windows 7 x64 an both machines, BOINC 6.10.43 on both as well. Im also wondering if i should switch back to 196.21 on the 9500GT. Could this failure in reporting total ram affect performance of the card? Thanks for any information.
ID: 990585 ·

[AF>france>pas-de-calais]symaski62 Volunteer tester Send message Joined: 12 Aug 05 Posts: 254 Credit: 69,988 RAC: 0	Message 1004580 - Posted: 16 Jun 2010, 9:06:29 UTC
17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak) 27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak) 16/06/2010 10:28:29 NVIDIA GPU 0: GeForce 8400 GS (driver version 25721, CUDA version 3010, compute capability 1.1, 244MB, 43 GFLOPS peak) voila ! :) ____________ SETI@Home Informational message -9 result_overflow with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.
ID: 1004580 ·

soft^spirit Send message Joined: 18 May 99 Posts: 6231 Credit: 27,514,663 RAC: 9,266	Message 1004628 - Posted: 16 Jun 2010, 12:25:02 UTC
I may get more life out of that 9600 GT yet!! Seriously.. it is either time to buy new fans for it, or retire it for something faster. But I am a multi-tasker.
ID: 1004628 ·

Author

Message

Sten-Arne
Volunteer tester
Avatar

Send message
Joined: 1 Nov 08
Posts: 2688
Credit: 7,662,017
RAC: 22,666

New host ID, oldest Wu was sent 6 Nov 2009 8:18:54 UTC

All tasks for computer 5337951

Claggy

Hmm, and why did it get a new host ID? Would adding hardware automatically create a new host ID, but keep the totals?

Sten-Arne

ID: 983192 ·

Cliff Harding

Volunteer tester
Avatar

Send message
Joined: 18 Aug 99
Posts: 607
Credit: 34,337,463
RAC: 55,837

The 197.13 version was releasd into production this morning!!!.
____________

I don't buy computers, I build them!!

ID: 983585 ·

[AF>france>pas-de-calais]symaski62
Volunteer tester
Send message
Joined: 12 Aug 05
Posts: 254
Credit: 69,988
RAC: 0

---------------
Display Devices
---------------
Card name: NVIDIA GeForce 8400 GS
Manufacturer: NVIDIA
Chip type: GeForce 8400 GS
Display Memory: 499 MB
Dedicated Memory: 243 MB
Shared Memory: 255 MB

____________
SETI@Home Informational message -9 result_overflow
with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.

ID: 984401 ·

MarkJ

Volunteer tester
Avatar

Send message
Joined: 17 Feb 08
Posts: 912
Credit: 14,561,754
RAC: 8,832

17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)
17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)

27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak)

256MB RAM => 244MB RAM

---------------
Display Devices
---------------
Card name: NVIDIA GeForce 8400 GS
Manufacturer: NVIDIA
Chip type: GeForce 8400 GS
Display Memory: 499 MB
Dedicated Memory: 243 MB
Shared Memory: 255 MB

Interesting, so that means the driver is now reporting less memory available (or its using more).

____________
BOINC blog

ID: 984431 ·

Wembley
Volunteer tester

Send message
Joined: 16 Sep 09
Posts: 415
Credit: 879,818
RAC: 0

My system is still reporting the proper amount of memory after upgrading the driver:

NVIDIA GPU 0: GeForce 9500 GT (driver version 19713, CUDA version 3000, compute capability 1.1, 1024MB, 90 GFLOPS peak)

I suspect something other than the driver is taking up more memory on your card.

____________

Donate with your searches and online buys:
http://www.goodsearch.com/toolbar/university-of-california-setihome

ID: 984492 ·

Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 2542
Credit: 25,662,021
RAC: 39,170

Found another CUDA-blower, this time bloating the database by 13,331: host 5337951. This is beginning to get a bit worrying.

Yes, with advance of GPU computing current quota-based protection system is completely obsolete.
It can't effectively block damaged host even if it does CPU+ single GPU computations, but if it has few GPUs... no way to stop database polluting at all now.

ID: 984494 ·

[AF>france>pas-de-calais]symaski62
Volunteer tester
Send message
Joined: 12 Aug 05
Posts: 254
Credit: 69,988
RAC: 0

17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)
17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)

27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak)

256MB RAM => 244MB RAM

---------------
Display Devices
---------------
Card name: NVIDIA GeForce 8400 GS
Manufacturer: NVIDIA
Chip type: GeForce 8400 GS
Display Memory: 499 MB
Dedicated Memory: 243 MB
Shared Memory: 255 MB

Interesting, so that means the driver is now reporting less memory available (or its using more).

stdoutdae.txt

27/03/2010 07:41:03 [---] NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)

27/03/2010 08:36:09 [---] NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak)

____________
SETI@Home Informational message -9 result_overflow
with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.

ID: 984510 ·

Claggy

Volunteer tester
Send message
Joined: 5 Jul 99
Posts: 3363
Credit: 25,950,143
RAC: 1,098

I've posted to Nvidia's Forum asking what has changed in the 197.xx drivers to cause Boinc and CUDA-Z to report less memory, No replies yet, feel free to add comments there:

Cuda-Z and Boinc report less Memory with 197.xx drivers

On my laptop, the reduction of reported memory caused Collatz to no longer send Cuda tasks, needed 126Mb, had 113Mb, reverted to an app_info for now,
on the Desktop PC, (with 2 OS's) the reduction in reported memory caused downloaded Stock Seti Beta 6.09 tasks to not start because of lack of memory, needed something like 384Mb, had 382Mb.

Claggy

ID: 988705 ·

[AF>france>pas-de-calais]symaski62
Volunteer tester
Send message
Joined: 12 Aug 05
Posts: 254
Credit: 69,988
RAC: 0

http://cuda-z.sourceforge.net/

memory
8400 GS => 243.562 MB total Global

:)
____________
SETI@Home Informational message -9 result_overflow
with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.

ID: 990101 ·

whoDunIt
Send message
Joined: 18 Sep 05
Posts: 2
Credit: 1,573,415
RAC: 0

I had this same problem when switching to the 197.13 driver. It didnt really matter for the MSI 9500 GT 512MB card but for my BFG 8600 GT 256MB card i couldnt get any new astropulse tasks, at the time thats all that was sending out, as it reported my memory at 243MB so i went back to the older driver 196.21. Same with the 197.25 driver.

Claggy please fill us in if you get any reply.

NVIDIA GeForce 9500 GT (499MB) driver: 19713
NVIDIA GeForce 8600 GT (256MB) driver: 19621

Windows 7 x64 an both machines, BOINC 6.10.43 on both as well.

Im also wondering if i should switch back to 196.21 on the 9500GT. Could this failure in reporting total ram affect performance of the card? Thanks for any information.

ID: 990585 ·

[AF>france>pas-de-calais]symaski62
Volunteer tester
Send message
Joined: 12 Aug 05
Posts: 254
Credit: 69,988
RAC: 0

17/02/2010 18:19:28 NVIDIA GPU 0: GeForce 8400 GS (driver version 19562, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)
17/02/2010 19:15:06 NVIDIA GPU 0: GeForce 8400 GS (driver version 19621, CUDA version 3000, compute capability 1.1, 256MB, 43 GFLOPS peak)

27/03/2010 08:36:09 NVIDIA GPU 0: GeForce 8400 GS (driver version 19713, CUDA version 3000, compute capability 1.1, 244MB, 43 GFLOPS peak)

16/06/2010 10:28:29 NVIDIA GPU 0: GeForce 8400 GS (driver version 25721, CUDA version 3010, compute capability 1.1, 244MB, 43 GFLOPS peak)

voila ! :)
____________
SETI@Home Informational message -9 result_overflow
with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.

ID: 1004580 ·

soft^spirit
Avatar

Send message
Joined: 18 May 99
Posts: 6231
Credit: 27,514,663
RAC: 9,266

I may get more life out of that 9600 GT yet!!

Seriously.. it is either time to buy new fans for it, or retire it for something faster. But I am a multi-tasker.

ID: 1004628 ·