Advanced search

Message boards : Graphics cards (GPUs) : ACEMD2 CUDA3.1 is now out

Author Message
Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17791 - Posted: 1 Jul 2010 | 8:46:35 UTC

I have uploaded the new CUDA3.1 application for everyone.
Remember to use SWAN_SYNC=0 for maximum performance (at the price of one CPU core).

Let us know the performance improvement.

gdf

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17792 - Posted: 1 Jul 2010 | 9:32:21 UTC - in response to Message 17791.

For 65nm GTX260 too?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17793 - Posted: 1 Jul 2010 | 10:44:24 UTC - in response to Message 17792.
Last modified: 1 Jul 2010 | 10:45:02 UTC

Anyone with a CUDA3.1 driver will receive it.
(Windows only for now)

gdf

Profile K1atOdessa
Send message
Joined: 25 Feb 08
Posts: 249
Credit: 370,186,977
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17794 - Posted: 1 Jul 2010 | 11:58:54 UTC - in response to Message 17793.

Anyone with a CUDA3.1 driver will receive it.
(Windows only for now)

gdf

Any testing with non-GTX 4X0 cards? It appears pretty good so far for these, but I haven't received any CUDA3.1 WU's on my slower / older cards. Just wondering how many of these beta's have been run on those systems.

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17795 - Posted: 1 Jul 2010 | 12:05:47 UTC

Got 6.05 cuda tasks, what to do to recieve 3.1 tasks? Beta 3.1 run nice on my GTX260.

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17796 - Posted: 1 Jul 2010 | 12:14:42 UTC - in response to Message 17795.

Got 6.05 cuda tasks, what to do to recieve 3.1 tasks? Beta 3.1 run nice on my GTX260.

The best you can do is make sure you have "Run test applications" checked on your Prefences and then hope you get one.

From what I can see this is a fairly small beta (I think it was 20 WUs), most likely because it tested really well in the lab and so far I have not seen anyone posting an error for this beta. Congrats to the GPUGrid team on this and I look forward to this version of your app's release. Hopefully the performance increases will allow you to increase overall WU size to help the project but still remain about the same average runtime we crunchers see now.

:thumbs up:
____________
Thanks - Steve

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17797 - Posted: 1 Jul 2010 | 12:53:43 UTC - in response to Message 17796.

Test 3.1 tasks I run OK http://www.gpugrid.net/forum_thread.php?id=2199&nowrap=true#17785, I have understood, that are available standard 3.1 tasks. Standard 6.05 2.2/2.3 cuda tasks crashes on my 65nm GTX260.

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17799 - Posted: 1 Jul 2010 | 15:45:48 UTC - in response to Message 17797.

Test 3.1 tasks I run OK.
I have understood, that are available standard 3.1 tasks.

I don't think a standard GPUGrid application that uses 3.1 has been released yet.

____________
Thanks - Steve

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17800 - Posted: 1 Jul 2010 | 15:57:01 UTC - in response to Message 17795.

Got 6.05 cuda tasks, what to do to recieve 3.1 tasks? Beta 3.1 run nice on my GTX260.


You just need to install cuda3.1 drivers.

gdf

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17801 - Posted: 1 Jul 2010 | 16:03:52 UTC - in response to Message 17799.

Test 3.1 tasks I run OK.
I have understood, that are available standard 3.1 tasks.

I don't think a standard GPUGrid application that uses 3.1 has been released yet.


And what about the 6.09 version in the apps? It's not among the ACEMD beta versions. Only the windows version available at the moment.

The first post says:

I have uploaded the new CUDA3.1 application for everyone.
Remember to use SWAN_SYNC=0 for maximum performance (at the price of one CPU core).

Let us know the performance improvement.

gdf


Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17802 - Posted: 1 Jul 2010 | 16:10:55 UTC - in response to Message 17800.

I have installed WHQL 257.21 since 06/15/2010 at host ID 31329 and beta 3.1 tasks run OK....but still recieving (and deleting) standard tasks as 6.05 cuda.

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17803 - Posted: 1 Jul 2010 | 16:48:35 UTC - in response to Message 17801.

Test 3.1 tasks I run OK.
I have understood, that are available standard 3.1 tasks.

I don't think a standard GPUGrid application that uses 3.1 has been released yet.


And what about the 6.09 version in the apps? It's not among the ACEMD beta versions. Only the windows version available at the moment.

The first post says:

I have uploaded the new CUDA3.1 application for everyone.
Remember to use SWAN_SYNC=0 for maximum performance (at the price of one CPU core).

Let us know the performance improvement.

gdf




My bad ... I was not reading closely.

Might I suggest that we all just crank through whatever WUs GPUGrid gives us rather than cherry picking the new ones? This way we can finish up the old ones and move everything to the new as soon as possible.

For point hunters ... make sure you don't cancel too many or the server will shut you off for a while ... possibly a long while ~ many hours!!!
____________
Thanks - Steve

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17804 - Posted: 1 Jul 2010 | 17:16:40 UTC - in response to Message 17803.

For point hunters ... make sure you don't cancel too many or the server will shut you off for a while ... possibly a long while ~ many hours!!!

I know that, but my 65nm GTX260 crashes any CUDA 2.2/2.3 task. It was usable a year ago and from that time I am crunching other project on it, but not GPUGRID. I load (and cancel) task time to time to not reach daily quota. Sometimes i recieve 3.1 beta task that rise daily quota.

GDF, are you sure that server should send standard tasks for non-Fermi class GPU with 257.21 drivers as CUDA 3.1?

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17805 - Posted: 1 Jul 2010 | 17:27:46 UTC - in response to Message 17804.

For point hunters ... make sure you don't cancel too many or the server will shut you off for a while ... possibly a long while ~ many hours!!!

I know that, but my 65nm GTX260 crashes any CUDA 2.2/2.3 task. It was usable a year ago and from that time I am crunching other project on it, but not GPUGRID. I load (and cancel) task time to time to not reach daily quota. Sometimes i recieve 3.1 beta task that rise daily quota.

My post was not directed towards you, it was just a general warning for people who do not already know the consequences if they abort to many.
____________
Thanks - Steve

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17808 - Posted: 1 Jul 2010 | 22:28:34 UTC - in response to Message 17804.

For point hunters ... make sure you don't cancel too many or the server will shut you off for a while ... possibly a long while ~ many hours!!!

I know that, but my 65nm GTX260 crashes any CUDA 2.2/2.3 task. It was usable a year ago and from that time I am crunching other project on it, but not GPUGRID. I load (and cancel) task time to time to not reach daily quota. Sometimes i recieve 3.1 beta task that rise daily quota.

GDF, are you sure that server should send standard tasks for non-Fermi class GPU with 257.21 drivers as CUDA 3.1?


It should. I will check the log files.

gdf

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17809 - Posted: 1 Jul 2010 | 22:28:56 UTC
Last modified: 1 Jul 2010 | 22:31:54 UTC

Got anybody any of these new v6.09 WUs?
I'm still getting only v6.05.
I guess we have to wait the queue to run out of v6.05 before we'll get v6.09.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17817 - Posted: 2 Jul 2010 | 8:43:08 UTC - in response to Message 17809.

My fault. The server was not giving out cuda3.1 jobs to anyone.
Now it should be fixed.

gdf

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17818 - Posted: 2 Jul 2010 | 11:18:41 UTC - in response to Message 17817.

Thanks GDF, it works now. Great.

Mark Henderson
Send message
Joined: 21 Dec 08
Posts: 51
Credit: 26,320,167
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwat
Message 17822 - Posted: 2 Jul 2010 | 13:49:49 UTC
Last modified: 2 Jul 2010 | 13:53:59 UTC

6.09 Cuda 3.1 beta is running good on my 55nm EVGA 260s. As did the previous versions also, with moderate overclock.
Just 1 WU completed so far but it seems to be 5 to 10 percent faster than 6.05 was.
Time per step seems 2x faster than I usualy see.

Profile [AF>Libristes>Jip] Elgran...
Avatar
Send message
Joined: 16 Jul 08
Posts: 45
Credit: 78,618,001
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17823 - Posted: 2 Jul 2010 | 15:58:23 UTC

For Linux, when ?

Please.

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17824 - Posted: 2 Jul 2010 | 16:00:12 UTC

My first return on a full size 3.1 is slower ... very disappointing.
No changes to system setup: GTX480 on WinXP
Examples are from the same WU type: TONI_CAPBIND*

Old version average runtime was 6550 (very little delta between runs)
# Time per step (avg over 650000 steps): 10.065 ms
# Approximate elapsed time for entire WU: 6541.937 s

New version 1 result runtime was 7768 seconds.
# Time per step (avg over 650000 steps): 11.947 ms
# Approximate elapsed time for entire WU: 7765.391 s

____________
Thanks - Steve

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17827 - Posted: 2 Jul 2010 | 16:33:39 UTC
Last modified: 2 Jul 2010 | 16:42:55 UTC

next result on the same system is for HERG* type WU.
about the same as the old app version ...

Old App average runtime 6250 sec. (very little delta between WUs)
# Time per step (avg over 625000 steps): 10.001 ms
# Approximate elapsed time for entire WU: 6250.359 s

New App 1 result
# Time per step (avg over 625000 steps): 9.815 ms
# Approximate elapsed time for entire WU: 6134.672 s
____________
Thanks - Steve

ftpd
Send message
Joined: 6 Jun 08
Posts: 152
Credit: 328,250,382
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17829 - Posted: 2 Jul 2010 | 17:38:15 UTC

Two cuda 3.1 6.09 WU windows-xp-pro gtx480 takes more than 3 hours per wu.
More 18.000 ms over 625000 steps.

So this is not good??!!
____________
Ton (ftpd) Netherlands

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17830 - Posted: 2 Jul 2010 | 17:51:15 UTC

My the first standard task succesfully finished after more than one year (65nm GTX 260, 64bit Win XP).
TONI_CAPBIND*
SWAN: Using synchronization method 0
# Time per step (avg over 325000 steps): 27.879 ms
# Approximate elapsed time for entire WU: 18121.656 s
Thanks GDF.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17834 - Posted: 2 Jul 2010 | 20:20:14 UTC - in response to Message 17823.

For Linux, when ?

Please.



Next week.

gdf

Profile [AF>Libristes>Jip] Elgran...
Avatar
Send message
Joined: 16 Jul 08
Posts: 45
Credit: 78,618,001
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17837 - Posted: 2 Jul 2010 | 21:47:22 UTC - in response to Message 17834.

Thank you.
My gpus drivers are updated to 256.35 linux version.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17839 - Posted: 2 Jul 2010 | 22:24:14 UTC
Last modified: 2 Jul 2010 | 22:33:36 UTC

I have completed four 6.09 tasks, and I'm a little disappointed.
There is a performance gain, but it's not the way I thought it would be.
Now my GPU is cooler with the new client, but that's not the important factor for me. :)
These 6.09 WUs take longer, beacuse they use less GPU.
According to GPU-Z v0.4.4 the GPU load is now around 60% to 74%.
Could you optimize the code so it would cause higher GPU load (on GTX 480)?
It was over 80% with the V6.05 client.

V6.09 m2-IBUCH_201b_pYEEI_100304-40-80-RND1404_1 . 15.422ms 9639.047s
V6.05 p41-IBUCH_501_pYEEI_100301-39-40-RND6587_1 _ 12.876ms 8047.344s
V6.09 h232f99r412-TONI_CAPBINDsp2-29-100-RND6253_0 16.624ms 10805.422s
V6.05 h232f99r558-TONI_CAPBINDsp2-22-100-RND1136_1 12.253ms 7964.406s

Also, one WU is failed approximately at 80%.

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17840 - Posted: 2 Jul 2010 | 23:59:02 UTC
Last modified: 2 Jul 2010 | 23:59:52 UTC

While I am seeing higher GPU usage I am also seeing higher CPU usage which is causing more kernel thrashing which I believe is what is slowing this version of the app back. I have cut BOINC Manager down another thread to see if this helps. I was leaving one free so now I have two thread open for GPUGrid on each machine, not good for my WCG crunching. Overall I am seeing performance LOSSES of a few minutes on WinXP (480) and enormous (like at least an hour per WU) LOSSES on Win7 (a 285 and a 295). The only WU type that I am seeing an improvement on is TONI_HERG by a couple minutes on both systems. I am seriously considering rolling back.

Perhaps the betas proved the programming was stable and perhaps I was overly optimistic in think the beta runtimes / ms per step were indivcative of production performance.

Sorry if I sound harsh, I think I had my hopes up high and should crunch for aliens and esoteric math problems to regain my focus on GPUGrid which I really do think is a fantastic project. Time for a break from posting.
____________
Thanks - Steve

coldFuSion
Send message
Joined: 22 May 10
Posts: 20
Credit: 85,355,427
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17841 - Posted: 3 Jul 2010 | 0:38:12 UTC
Last modified: 3 Jul 2010 | 0:40:30 UTC

Results seem to vary drastically with my WinXP

I392-TONI_KIDln-12-100-RND6196_0
# Device 0: "GeForce GTX 470"
SWAN: Using synchronization method 0
# Time per step (avg over 1300000 steps): 4.924 ms!!!
# Approximate elapsed time for entire WU: 6401.594 s

h232f99r440-TONI_CAPBINDsp2-25-100-RND7492_0
# Device 0: "GeForce GTX 470"
SWAN: Using synchronization method 0
# Time per step (avg over 650000 steps): 16.178 ms
# Approximate elapsed time for entire WU: 10515.719 s

Not sure what the difference between these 2 applications is but if the second can be optimized like first then that would be seriously impressive

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17845 - Posted: 3 Jul 2010 | 7:04:28 UTC - in response to Message 17841.

The beta workunit has a size of the simulation box of 64x64x64. Cuda3.1 optimizes FFT in multiple of 2. So it was a bit faster and other machines (WinXP) reproduced it. As they were overclocked even improved the reference speed we have found locally.

After these tests in gpugrid, it seems that other workunits are not any better than cuda3.0. I will do some tests next week on other workunit types.

Nevertheless, CUDA3.1 works well also on holder card, and seems to fix several bug. So it is a good release.

gdf

Betting Slip
Send message
Joined: 5 Jan 09
Posts: 589
Credit: 2,039,762,925
RAC: 1,511,935
Level
Phe
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17847 - Posted: 3 Jul 2010 | 8:16:30 UTC - in response to Message 17845.

@GDF Please read this and give me an optout of 3.1 if that's possible because the result will be the probable loss of two crunchers or more for this project.


____________

Snow Crash
Send message
Joined: 4 Apr 09
Posts: 450
Credit: 539,316,349
RAC: 0
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17851 - Posted: 3 Jul 2010 | 11:52:54 UTC

Serious note ... the majority of the degraded performance on my 2xx series cards was because the last time I switched hardware around I forgot to delete the SWAN_SYNC environ var and apparently the new version of the app recognizes and uses it. This is why I had an even greater amount of kernel usage than normal. While some WUs are faster and some are slower I hope the project gets the benefits from a more stable app.
____________
Thanks - Steve

Profile MarkJ
Volunteer moderator
Project tester
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 730
Credit: 189,243,545
RAC: 0
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17852 - Posted: 3 Jul 2010 | 12:30:21 UTC - in response to Message 17847.

@GDF Please read this and give me an optout of 3.1 if that's possible because the result will be the probable loss of two crunchers or more for this project.


You could just go back to older drivers.
____________
BOINC blog

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17854 - Posted: 3 Jul 2010 | 21:18:52 UTC
Last modified: 3 Jul 2010 | 21:20:42 UTC

I've just installed v258.19 driver, and it reports itself like this in Boinc Manager:

2010.07.03. 23:09:18 NVIDIA GPU 0: GeForce GTX 480 (driver version 25819, CUDA version 3020, compute capability 2.0, 1536MB, 1345 GFLOPS peak)

It means that soon(?) tere will be CUDA 3.2

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17855 - Posted: 3 Jul 2010 | 23:10:34 UTC - in response to Message 17854.

This CUDA 3010 WU look promising!
http://www.gpugrid.net/result.php?resultid=2611239

It finished about 40% faster than the CUDA 3000 tasks. So although there are some that are about 50% slower, there are some faster WU's as well.

GDF just needs to work out what the difference is, and make sure we get the fast ones rather than the slow ones.

Good luck with that,

Details:

The fast one,
2611239 1660106 3 Jul 2010 13:39:37 UTC 3 Jul 2010 18:51:08 UTC Completed and validated 6,601.92 6,598.16 4,727.93 7,091.89 ACEMD2: GPU molecular dynamics v6.09 (cuda31)

Name I531-TONI_KIDln-13-100-RND2019_0
Workunit 1660106
Created 3 Jul 2010 13:19:57 UTC
Sent 3 Jul 2010 13:39:37 UTC
Received 3 Jul 2010 18:51:08 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 71363
Report deadline 8 Jul 2010 13:39:37 UTC
Run time 6601.921875
CPU time 6598.156
stderr out

<core_client_version>6.10.56</core_client_version>
<![CDATA[
<stderr_txt>
# Using device 0
# There is 1 device supporting CUDA
# Device 0: "GeForce GTX 470"
# Clock rate: 1.41 GHz
# Total amount of global memory: 1341718528 bytes
# Number of multiprocessors: 14
# Number of cores: 112
SWAN: Using synchronization method 0
MDIO ERROR: cannot open file "restart.coor"
# Time per step (avg over 1300000 steps): 5.077 ms
# Approximate elapsed time for entire WU: 6599.953 s
called boinc_finish

</stderr_txt>
]]>

Validate state Valid
Claimed credit 4727.92939814815
Granted credit 7091.89409722222
application version ACEMD2: GPU molecular dynamics v6.09 (cuda31)

An example of the older CUDA 3000 tasks,
2603160 1654062 2 Jul 2010 1:29:51 UTC 2 Jul 2010 6:15:33 UTC Completed and validated 8,218.06 8,105.92 4,428.01 6,642.02 ACEMD2: GPU molecular dynamics v6.05 (cuda30)

An example of the newer slow WUs,
2611961 1657969 3 Jul 2010 17:01:16 UTC 3 Jul 2010 22:26:14 UTC Completed and validated 12,871.38 12,867.41 4,535.61 6,803.41 ACEMD2: GPU molecular dynamics v6.09 (cuda31)

coldFuSion
Send message
Joined: 22 May 10
Posts: 20
Credit: 85,355,427
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17857 - Posted: 4 Jul 2010 | 0:11:59 UTC

just want to suggest that everyone take a slow deep breath....


and remind everyone we are using bleeding edge drivers with bleeding edge developer tools.

Let's not prematurely flee like rats from a sinking ship just because every single GPU transistor is not working at 100% optimization and productivity.

Remember the main underlying goal is to help crunch protein simulations for medical research.

If some work units are slower now than before...so what? GPU crunching is still exponentially faster than crunching the same thing on CPU based applications.

If we keep our wits about us and work together we can help improve application productivity by providing constructive feedback.

Profile liveonc
Avatar
Send message
Joined: 1 Jan 10
Posts: 292
Credit: 41,567,650
RAC: 0
Level
Val
Scientific publications
watwatwatwatwatwat
Message 17858 - Posted: 4 Jul 2010 | 2:13:36 UTC - in response to Message 17857.

I second that!

I guess that I could run Prime95 or Linpack 24/7 in all I wanted to do was produce heat & waste electricity, especially in Summer. If I like getting points, I'm sure i could do that on Facebook for free w/o needing more then a netbook.

But I like GPUGRID. They do work, they have a friendly forum where stupid questions aren't answered back with a verbal gutting or impalement, & they're quick to adopt to all the new things which are still raw.

That one day, there would be other things then research in mental disorders, is something I can but hope. I'm more interested in cancer research, cures for STD's, & illness' that normally don't get enough funding, this is none of the above, but if GPUGRID shows the way, maybe the rest of the pack will follow one day...
____________

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17860 - Posted: 4 Jul 2010 | 9:15:18 UTC - in response to Message 17855.

I got one of those fast work units:
# Time per step (avg over 1300000 steps): 4.617 ms
# Approximate elapsed time for entire WU: 6002.672 s
Claimed credit 4728
Granted credit 7092
This WU generated 94% CPU load, wich makes me contented.
But I know the other WUs did run almost at this 'efficiency level' before, and I'd like to have it back, or more :)

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17862 - Posted: 4 Jul 2010 | 11:20:04 UTC - in response to Message 17860.
Last modified: 4 Jul 2010 | 11:44:54 UTC

These CUDA 3.1 I531-TONI_KIDln WU’s are clearly much faster than the CUDA 3000 WUs, and the 3.1 TONI_HERG WU's are just the same as 3.0, but the TONI_CAPBIND, KASHIF_HIVPR are much slower.

TONI_KID fast, and again

Before last weekend’s downtime my Fermi on XP had a steady RAC of 65K. It has now dropped to 55K and is showing no signs of increasing.
If I just crunched the TONI_CAPBIND it would drop to 45K!
If I just crunched KASHIF_HIVPR it would drop to about 53K
If I could just crunch TONI_KIDln it would rise to 92K. Yes, these are twice as fast as TONI_CAPBIND under CUDA 3.1. So, as you can see there is a massive performance difference between tasks!
Before you (GPUGrid team) sit down to work out why this is, could you build the Work Units according to this, so the Fermi’s just get TONI_KIDln Work Units? The slower tasks could just be compiled for 2.2, as they are faster that way. This would immediately speed up the research on Fermi cards (almost double it).
Thanks,

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17866 - Posted: 4 Jul 2010 | 12:00:33 UTC - in response to Message 17862.

We can reproduce the problem so we can probably fix it.
http://www.gpugrid.net/forum_thread.php?id=2209&nowrap=true#17865

gdf

Bedrich Hajek
Send message
Joined: 28 Mar 09
Posts: 340
Credit: 3,821,471,309
RAC: 941,032
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17867 - Posted: 4 Jul 2010 | 12:29:25 UTC

The IBUCH_ _pYEEI units are also slower on my computer.

Werkstatt
Send message
Joined: 23 May 09
Posts: 116
Credit: 122,267,748
RAC: 29,512
Level
Cys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17871 - Posted: 4 Jul 2010 | 19:24:38 UTC - in response to Message 17809.

Got anybody any of these new v6.09 WUs?

Hi,
yes, today I updated to cuda 3.1, resetted the project. A couple of minutes later the download of all necessary files started. For abaout half an hour I received messages, that I have a wrong system, that a fermi-GPU is required aso.
But then - surprise, surprice - a 6.09WU started. Right now my old GTX260 is crunching it. Usually it faild after ~16%, now it reached 18% .. and it is still working!
I will give an update, whether it finished or failed.

Regards,
Alexander

Werkstatt
Send message
Joined: 23 May 09
Posts: 116
Credit: 122,267,748
RAC: 29,512
Level
Cys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17872 - Posted: 4 Jul 2010 | 20:05:39 UTC - in response to Message 17871.


I will give an update, whether it finished or failed.

No reason to be happy, wu failed after ~2 hours.
So CUDA3.1 + old 260 : Sorry, the problem is still there!

Regards
Alexander

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17877 - Posted: 4 Jul 2010 | 23:15:53 UTC - in response to Message 17866.

We can reproduce the problem so we can probably fix it.
http://www.gpugrid.net/forum_thread.php?id=2209&nowrap=true#17865

gdf


We know where is the problem. It is due to a change of data structure in CUDA3.1 which seems to be very bad in most cases.
A fix is on its way, but it might require a couple of days.
Be patient in the meanwhile.

gdf

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17886 - Posted: 5 Jul 2010 | 14:25:54 UTC - in response to Message 17877.

It's a problem of a routine in cuda3.1. We have patched it for now.
New application release coming up.

gdf

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17892 - Posted: 5 Jul 2010 | 17:17:12 UTC - in response to Message 17886.

We have found a very bad bug in CUDA3.1. I have removed the CUDA3.1 application until we can see the extent of the problem.

gdf

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17894 - Posted: 5 Jul 2010 | 21:58:24 UTC - in response to Message 17892.

Maybe NV will fix it in CUDA3.2 :)
I hope they will release it in a month.

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17952 - Posted: 14 Jul 2010 | 12:08:33 UTC - in response to Message 17892.

Are there any news about that very bad bug?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17976 - Posted: 15 Jul 2010 | 13:47:41 UTC - in response to Message 17952.

On a machine with a mixed Fermi/g200 installation of cards, results on g200 for the FFT are simply wrong. This is a really bad bug.

The only solution is to deliver the CUDA3.1 application to Fermi cards only.

gdf

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17978 - Posted: 15 Jul 2010 | 14:14:06 UTC - in response to Message 17976.
Last modified: 15 Jul 2010 | 14:14:47 UTC

New application is out for test in beta. Only fermi cards can get the cuda31 app due to the cuda3.1 bug.
This version maintains the level of performance of cuda3.1 for all workunits.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17980 - Posted: 15 Jul 2010 | 14:42:58 UTC - in response to Message 17978.
Last modified: 15 Jul 2010 | 15:21:14 UTC

Picked up two, both failed in about 1 sec!

2599077 1651913 15 Jul 2010 14:35:06 UTC 15 Jul 2010 14:36:31 UTC Error while computing 1.45 0.00 0.00 --- ACEMD beta version v6.30 (cuda31)
2598716 1651698 15 Jul 2010 14:36:31 UTC 15 Jul 2010 14:40:05 UTC Error while computing 0.00 0.00 0.00 --- ACEMD beta version v6.30 (cuda31)

Name 100-GIANNI_TESTDHFR7-1-5-RND4717_0
Workunit 1651913
Created 1 Jul 2010 8:39:18 UTC
Sent 15 Jul 2010 14:35:06 UTC
Received 15 Jul 2010 14:36:31 UTC
Server state Over
Outcome Client error
Client state Compute error
Exit status -40 (0xffffffffffffffd8)
Computer ID 71363
Report deadline 20 Jul 2010 14:35:06 UTC
Run time 1.453125
CPU time 0
stderr out

<core_client_version>6.10.56</core_client_version>
<![CDATA[
<message>
- exit code -40 (0xffffffd8)
</message>
<stderr_txt>
# Using device 0
# There is 1 device supporting CUDA
# Device 0: "GeForce GTX 470"
# Clock rate: 1.41 GHz
# Total amount of global memory: 1341718528 bytes
# Number of multiprocessors: 14
# Number of cores: 112
SWAN: Using synchronization method 0
SWAN : Module load result [.nonbonded.cu.] [200]
SWAN: FATAL : Module load failed


</stderr_txt>
]]>

Validate state Invalid
Claimed credit 0
Granted credit 0
application version ACEMD beta version v6.30 (cuda31)

NVIDIA GeForce GTX 470 (1279MB) driver: 25721
Operating System Microsoft Windows XP

coldFuSion
Send message
Joined: 22 May 10
Posts: 20
Credit: 85,355,427
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 17982 - Posted: 15 Jul 2010 | 15:06:50 UTC - in response to Message 17980.
Last modified: 15 Jul 2010 | 15:07:24 UTC

Same results on my GTX 470...immediate failure

Sorry i can't post the error details ATM but I am fairly sure they identical to skgiven's

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17986 - Posted: 15 Jul 2010 | 16:29:51 UTC - in response to Message 17982.

This one also failed,
2598612 1651618 15 Jul 2010 14:40:05 UTC 15 Jul 2010 16:10:07 UTC Error while computing 0.00 0.00 0.00 --- ACEMD beta version v6.30 (cuda31)

They are not even starting,
SWAN: FATAL : Module load failed

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17993 - Posted: 15 Jul 2010 | 21:43:58 UTC - in response to Message 17986.

Removed, tomorrow another one.
gdf

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 17997 - Posted: 16 Jul 2010 | 13:29:28 UTC - in response to Message 17993.

Trying now with acemdbeta 6.32 for windows.

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 17999 - Posted: 16 Jul 2010 | 14:12:56 UTC - in response to Message 17997.
Last modified: 16 Jul 2010 | 14:36:18 UTC

Ran one Beta task. It worked on my GTX470 on XP x86, Boinc ver. 6.10.56
Cant find it online yet!
Here are the Boinc details:

16/07/2010 15:00:29 GPUGRID Starting 63-GIANNI_TESTDHFR7-1-5-RND3683_0
16/07/2010 15:00:30 GPUGRID Starting task 63-GIANNI_TESTDHFR7-1-5-RND3683_0 using acemdbeta version 632
16/07/2010 15:05:22 GPUGRID Computation for task 63-GIANNI_TESTDHFR7-1-5-RND3683_0 finished
16/07/2010 15:05:23 GPUGRID Started upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_0
16/07/2010 15:05:23 GPUGRID Started upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_1
16/07/2010 15:05:31 GPUGRID Finished upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_0
16/07/2010 15:05:31 GPUGRID Started upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_2
16/07/2010 15:05:36 GPUGRID Finished upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_1
16/07/2010 15:05:36 GPUGRID Started upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_3
16/07/2010 15:05:47 GPUGRID Finished upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_2
16/07/2010 15:05:47 GPUGRID Started upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_7
16/07/2010 15:05:51 GPUGRID Finished upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_7
16/07/2010 15:05:52 GPUGRID Finished upload of 63-GIANNI_TESTDHFR7-1-5-RND3683_0_3
16/07/2010 15:05:52 GPUGRID Sending scheduler request: To report completed tasks.
16/07/2010 15:05:52 GPUGRID Reporting 1 completed tasks, not requesting new tasks
16/07/2010 15:05:53 GPUGRID Scheduler request completed


GDF, see here too - http://www.gpugrid.net/forum_thread.php?id=2227&nowrap=true#17998

Bedrich Hajek
Send message
Joined: 28 Mar 09
Posts: 340
Credit: 3,821,471,309
RAC: 941,032
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18009 - Posted: 16 Jul 2010 | 22:16:59 UTC

I just finished 3 of these beta units, they all came out completed and validated.

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1073
Credit: 4,497,101,754
RAC: 415,156
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18011 - Posted: 16 Jul 2010 | 22:49:21 UTC
Last modified: 16 Jul 2010 | 22:51:04 UTC

These beta WUs are failing on the GTX 460:

<core_client_version>6.10.57</core_client_version>
<![CDATA[
<message>
- exit code -40 (0xffffffd8)
</message>
<stderr_txt>
# Using device 0
# There is 1 device supporting CUDA
# Device 0: "GeForce GTX 460"
# Clock rate: 0.81 GHz
# Total amount of global memory: 774307840 bytes
# Number of multiprocessors: 7
# Number of cores: 56
SWAN : Module load result [.fastfill.cu.] [200]
SWAN: FATAL : Module load failed

</stderr_txt>

While the clock speed is also incorrect above, it's reported correctly by MSI Afterburner v1.61.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18020 - Posted: 17 Jul 2010 | 9:08:41 UTC - in response to Message 18011.

The beta's are working well for GF100 cards, just not for GF104 cards (GTX460). The scientists are now aware of the problem because Beyond reported a difference in Compute Capability between the cards; CC2.0 for GF100 cards, CC2.1 for GF104 cards. GDF said they will try to create a new Beta for the GTX460 on Monday. So check back Monday evening. Apparently the GTX460 works on Collatz. Those wanting to test their card & tune the fans for a nice low temps, you could figure it out there.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18023 - Posted: 17 Jul 2010 | 11:25:25 UTC - in response to Message 18020.

GTX470 on XP running a Beta:
2678845 1704781 17 Jul 2010 8:32:57 UTC 17 Jul 2010 11:05:25 UTC
Completed and validated 280.06 276.80 187.28 280.92 ACEMD beta version v6.32 (cuda31)

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18024 - Posted: 17 Jul 2010 | 11:34:47 UTC - in response to Message 18023.

GTX470 on XP running a Beta:
2678845 1704781 17 Jul 2010 8:32:57 UTC 17 Jul 2010 11:05:25 UTC
Completed and validated 280.06 276.80 187.28 280.92 ACEMD beta version v6.32 (cuda31)


This is really good speed. A standard clocked GTX480 is 5ms/step under Linux. So only 10% slower for an overclocked GTX470. As good as it can get.

gdf

coldFuSion
Send message
Joined: 22 May 10
Posts: 20
Credit: 85,355,427
RAC: 0
Level
Thr
Scientific publications
watwatwatwatwatwatwatwatwatwat
Message 18029 - Posted: 17 Jul 2010 | 18:18:48 UTC - in response to Message 18023.

GTX470 on XP running a Beta:
2678845 1704781 17 Jul 2010 8:32:57 UTC 17 Jul 2010 11:05:25 UTC
Completed and validated 280.06 276.80 187.28 280.92 ACEMD beta version v6.32 (cuda31)


Nearly identical results on my OCed GTX470 on XP
2681352 1706384 17 Jul 2010 15:17:24 UTC 17 Jul 2010 15:36:53 UTC
Completed and validated 269.16 266.86 187.28 280.92 ACEMD beta version v6.32 (cuda31)

Time per step: 5.344 ms

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18031 - Posted: 17 Jul 2010 | 19:20:34 UTC
Last modified: 17 Jul 2010 | 19:22:55 UTC

Standard clocked GTX480 is 5ms/step under Windows XP too:
2679396
2679141
2679030

And they ran with 93% GPU usage, that's great!

I can't hardly wait for the real workunits. :)
I hope they will run this fast and efficient too.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18039 - Posted: 18 Jul 2010 | 18:09:24 UTC - in response to Message 18031.

I ran a Folding@home task today while running this Beta. It ran very slow compaired to the other Betas, but at least it ran. You cant do this with non-Fermi cards; the GPUGrid tasks would just fail!

2686592 1708583 18 Jul 2010 Completed and validated 277.58 275.69 187.28 280.92 ACEMD beta version v6.32 (cuda31)
2686539 1707219 18 Jul 2010 Completed and validated 1,442.30 1,438.45 187.28 280.92 ACEMD beta version v6.32 (cuda31)
2686534 1707190 18 Jul 2010 Completed and validated 277.84 276.08 187.28 280.92 ACEMD beta version v6.32 (cuda31)

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18040 - Posted: 18 Jul 2010 | 18:13:33 UTC

Five more 6.32 test WUs on standard clocked GTX480 on Win XP x86:
2688266
2686600
2686535
2688266
2683261
The last one failed, during a BSOD, I'm not sure what caused this (I was browsing, but nothing serious). One of my rosetta WU failed too.
Every test WU are slightly under 5ms/step.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18043 - Posted: 19 Jul 2010 | 9:14:58 UTC - in response to Message 18040.

New beta 6.34 is now out. Hopefully it should work with GTX460 cards, but we could not test it.

gdf

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1844
Credit: 10,645,982,644
RAC: 9,980,171
Level
Trp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18044 - Posted: 19 Jul 2010 | 9:29:14 UTC - in response to Message 18043.
Last modified: 19 Jul 2010 | 9:30:02 UTC

The beta 6.34 is immediately errored out on my GTX480.
Task 2691577

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18045 - Posted: 19 Jul 2010 | 9:52:28 UTC - in response to Message 18044.

Any more cases of this problem?

gdf

JG4KEZ(Koichi Soraku)
Send message
Joined: 3 Aug 07
Posts: 5
Credit: 243,198,915
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18046 - Posted: 19 Jul 2010 | 10:11:14 UTC - in response to Message 18045.

Even GTX460 does not work by the same error.
http://www.gpugrid.net/result.php?resultid=2691593
http://www.gpugrid.net/result.php?resultid=2691594

<core_client_version>6.11.1</core_client_version>
<![CDATA[
<message>
- exit code -40 (0xffffffd8)
</message>
<stderr_txt>
# Using device 0
# There are 2 devices supporting CUDA
# Device 0: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073283072 bytes
# Number of multiprocessors: 7
# Number of cores: 56
# Device 1: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073414144 bytes
# Number of multiprocessors: 7
# Number of cores: 56
SWAN: Using synchronization method 0
SWAN: FATAL : swanMallocHost failed


</stderr_txt>
]]>

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18047 - Posted: 19 Jul 2010 | 10:48:39 UTC - in response to Message 18046.

Removed.
Another coming out now.
gdf

JG4KEZ(Koichi Soraku)
Send message
Joined: 3 Aug 07
Posts: 5
Credit: 243,198,915
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18049 - Posted: 19 Jul 2010 | 12:55:29 UTC - in response to Message 18047.

It still becomes the error on my GTX460.

http://www.gpugrid.net/result.php?resultid=2691594
http://www.gpugrid.net/result.php?resultid=2692279

<core_client_version>6.11.1</core_client_version>
<![CDATA[
<message>
- exit code -40 (0xffffffd8)
</message>
<stderr_txt>
# Using device 0
# There are 2 devices supporting CUDA
# Device 0: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073283072 bytes
# Number of multiprocessors: 7
# Number of cores: 56
# Device 1: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073414144 bytes
# Number of multiprocessors: 7
# Number of cores: 56
SWAN : Module load result [.fastfill.cu.] [200]
SWAN: FATAL : Module load failed


</stderr_txt>
]]>


Specifications
M/B: ASUS P5E
CPU: Inel Xeon X3360
RAM: PC2-6400 2GBx2
GPU: Kuroutoshikou GF-GTX460-E1GHD (made by Sparkle Computer) x2
Driver: FW258.96
OS: Windows XP Pro SP3 x86

--
JG4KEZ

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18050 - Posted: 19 Jul 2010 | 13:27:09 UTC - in response to Message 18049.

application 6.36 is out

gdf

Profile Bikermatt
Send message
Joined: 8 Apr 10
Posts: 36
Credit: 1,809,027,204
RAC: 158,961
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18051 - Posted: 19 Jul 2010 | 13:57:09 UTC - in response to Message 18050.

Looks good!


http://www.gpugrid.net/result.php?resultid=2691597

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18052 - Posted: 19 Jul 2010 | 14:06:23 UTC - in response to Message 18051.
Last modified: 19 Jul 2010 | 14:06:53 UTC

Well,
ok it runs, we can start tests.
it is very slow at least on Windows 7.
Anyone with XP and SWAN_SYNC=0?

gdf

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18054 - Posted: 19 Jul 2010 | 14:40:08 UTC - in response to Message 18052.
Last modified: 19 Jul 2010 | 14:56:05 UTC

JG4KEZ(Koichi Soraku) has...

JG4KEZ(Koichi Soraku)
Send message
Joined: 3 Aug 07
Posts: 5
Credit: 243,198,915
RAC: 0
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18055 - Posted: 19 Jul 2010 | 14:43:40 UTC - in response to Message 18052.

6.36 worked in my GTX460 normally.
GPU load is around 95%.

http://www.gpugrid.net/result.php?resultid=2691958
http://www.gpugrid.net/result.php?resultid=2691640

<core_client_version>6.11.1</core_client_version>
<![CDATA[
<stderr_txt>
# Using device 1
# There are 2 devices supporting CUDA
# Device 0: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073283072 bytes
# Number of multiprocessors: 7
# Number of cores: 56
# Device 1: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073414144 bytes
# Number of multiprocessors: 7
# Number of cores: 56
SWAN: Using synchronization method 0
MDIO ERROR: cannot open file "restart.coor"
# Time per step (avg over 50000 steps): 11.216 ms
# Approximate elapsed time for entire WU: 560.783 s
called boinc_finish

</stderr_txt>
]]>

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18056 - Posted: 19 Jul 2010 | 14:45:18 UTC - in response to Message 18055.
Last modified: 19 Jul 2010 | 14:45:48 UTC

half the speed of a GTX480.

gdf

6.36 worked in my GTX460 normally.
GPU load is around 95%.

http://www.gpugrid.net/result.php?resultid=2691958
http://www.gpugrid.net/result.php?resultid=2691640
<core_client_version>6.11.1</core_client_version>
<![CDATA[
<stderr_txt>
# Using device 1
# There are 2 devices supporting CUDA
# Device 0: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073283072 bytes
# Number of multiprocessors: 7
# Number of cores: 56
# Device 1: "GeForce GTX 460"
# Clock rate: 1.40 GHz
# Total amount of global memory: 1073414144 bytes
# Number of multiprocessors: 7
# Number of cores: 56
SWAN: Using synchronization method 0
MDIO ERROR: cannot open file "restart.coor"
# Time per step (avg over 50000 steps): 11.216 ms
# Approximate elapsed time for entire WU: 560.783 s
called boinc_finish

</stderr_txt>
]]>

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18057 - Posted: 19 Jul 2010 | 14:57:44 UTC - in response to Message 18056.
Last modified: 19 Jul 2010 | 15:17:03 UTC

Perhaps the GTX460's are not using all of their shaders?
Anyway, half the speed of a GTX480 is not bad; a GTX460 costs less than half the price of a GTX480!

It would seem that under Win7 the 1GB version is 10% faster than the 786MB version:

http://www.gpugrid.net/result.php?resultid=2691759
http://www.gpugrid.net/result.php?resultid=2691597

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18058 - Posted: 19 Jul 2010 | 15:43:54 UTC - in response to Message 18057.
Last modified: 19 Jul 2010 | 15:44:18 UTC

Maybe we can order one and try to optimize for it, it should have done much better.

gdf

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1073
Credit: 4,497,101,754
RAC: 415,156
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18059 - Posted: 19 Jul 2010 | 15:45:00 UTC - in response to Message 18052.
Last modified: 19 Jul 2010 | 15:48:54 UTC

Well,
ok it runs, we can start tests.
it is very slow at least on Windows 7.
Anyone with XP and SWAN_SYNC=0?

gdf

Works here too but it's so slow that the 460 will be staying on Collatz for now. In Collatz the GPU usage is 99% and temp 56C. In GPUGRID the GPU usage is 46% and the temp 44C. Win7-32. Why can't we get an app that runs at 95+ percent when every other GPU project does that easily?

http://www.gpugrid.net/result.php?resultid=2691759
http://www.gpugrid.net/result.php?resultid=2691643
http://www.gpugrid.net/result.php?resultid=2691630

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18061 - Posted: 19 Jul 2010 | 16:48:52 UTC - in response to Message 18059.

It's odd that your clocks are really low.

gdf

Well,
ok it runs, we can start tests.
it is very slow at least on Windows 7.
Anyone with XP and SWAN_SYNC=0?

gdf

Works here too but it's so slow that the 460 will be staying on Collatz for now. In Collatz the GPU usage is 99% and temp 56C. In GPUGRID the GPU usage is 46% and the temp 44C. Win7-32. Why can't we get an app that runs at 95+ percent when every other GPU project does that easily?

http://www.gpugrid.net/result.php?resultid=2691759
http://www.gpugrid.net/result.php?resultid=2691643
http://www.gpugrid.net/result.php?resultid=2691630

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18064 - Posted: 19 Jul 2010 | 18:12:22 UTC - in response to Message 18061.
Last modified: 19 Jul 2010 | 18:59:02 UTC

Perhaps not just a reporting artefact, either from the driver or the Boinc Version:

Beyond (W7 x86)
NVIDIA GeForce GTX 460 (738MB) driver: 25856
Boinc 6.10.57
Clock rate: 0.81 GHz

JG4KEZ(Koichi Soraku) (XP x86)
[2] NVIDIA GeForce GTX 460 (1023MB) driver: 25896
Clock rate: 1.40 GHz

Bikermatt (W7 x64)
NVIDIA GeForce GTX 460 (993MB) driver: 25896
Clock rate: 1.40 GHz

W7 still reports less RAM than the card actually has.

As we can see from Bikermatt’s results, the Boinc/Driver upgrade would only get it to report the correct speed, and would not improve performance, however it does suggest something:

The difference in performance between an XP system and a W7 system is approximately 0.81/1.4 (to within a few percent) – an XP system takes about 58% of the time to do a similar task, ie 0.81/1.4 of the time.
This is also true of the older GF100 Fermi cards!

This could be a coincidence, but it makes me suspect the update (whatever it is; driver or Boinc version) that allows GPUGrid to display the correct frequency, is actually just displaying what the clock should be, rather than what they actually are, which is still 0.81 under W7 & Vista. This would also explain why the GF100 cards are also slower to about the same ratio.

Profile Beyond
Avatar
Send message
Joined: 23 Nov 08
Posts: 1073
Credit: 4,497,101,754
RAC: 415,156
Level
Arg
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18067 - Posted: 19 Jul 2010 | 20:57:58 UTC - in response to Message 18064.
Last modified: 19 Jul 2010 | 21:00:04 UTC

Perhaps not just a reporting artefact, either from the driver or the Boinc Version:

Beyond (W7 x86)
NVIDIA GeForce GTX 460 (738MB) driver: 25856
Boinc 6.10.57
Clock rate: 0.81 GHz

JG4KEZ(Koichi Soraku) (XP x86)
[2] NVIDIA GeForce GTX 460 (1023MB) driver: 25896
Clock rate: 1.40 GHz

Bikermatt (W7 x64)
NVIDIA GeForce GTX 460 (993MB) driver: 25896
Clock rate: 1.40 GHz

W7 still reports less RAM than the card actually has.

Just updated the Win7-32 system from BOINC 6.10.57 to 6.10.58 and the NVidia drivers from 25856 to 25896. Still reports .81. Now runs at 49% and 45C, 200 watts system draw when running GPUGRID, 240+ when running Collatz (99% GPU). When I get a chance I'll try it with XP. Until then it's on Collatz or DNETC. We really shouldn't have to be downgrading OSes to run a project IMO though.

http://www.gpugrid.net/result.php?resultid=2692961

BTW, as stated before MSI Afterburner shows the card running correctly at: 800 core / 1600 shader / 1900 memory.

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18069 - Posted: 19 Jul 2010 | 21:35:31 UTC - in response to Message 18067.

It looks like Boinc/GPUGrid/Drivers think it is a GF100 card.
Did you do a restart?

steinrar
Send message
Joined: 27 Sep 08
Posts: 1
Credit: 342,522,187
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18071 - Posted: 19 Jul 2010 | 22:10:39 UTC - in response to Message 18069.
Last modified: 19 Jul 2010 | 22:19:35 UTC

Hi

Use 2x460 GPU's on this host, run "almost" fine
http://www.gpugrid.net/show_host_detail.php?hostid=75798

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3968
Credit: 1,834,906,524
RAC: 276,970
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18073 - Posted: 19 Jul 2010 | 23:07:09 UTC - in response to Message 18071.

Yes, they work, so that is a big +ve.
The downside is that the GTX460's are a little bit slower than expected, but these are only Betas, and that might change with improved applications. It was a good idea to put them on XP - the first XP x64 I have seen in a while ;)

I guess you are not using the swan_sync=0 variable? Try leaving a core free, to see if it improves performance (I expect it will).

Good Luck,

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18076 - Posted: 20 Jul 2010 | 7:24:00 UTC - in response to Message 18073.

I will be moving the beta to production today.

gdf

Profile nenym
Send message
Joined: 31 Mar 09
Posts: 136
Credit: 589,130,728
RAC: 217,516
Level
Lys
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 18077 - Posted: 20 Jul 2010 | 8:35:04 UTC - in response to Message 18076.

For the Fermi class GPU only?

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1895
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 18078 - Posted: 20 Jul 2010 | 9:09:28 UTC - in response to Message 18077.

Yes, due to the bug in CUDA3.1.

gdf

Post to thread

Message boards : Graphics cards (GPUs) : ACEMD2 CUDA3.1 is now out