Advanced search

Message boards : Graphics cards (GPUs) : How to get 2 GPUs running GPUGrid?

Author Message
shdbcamping
Send message
Joined: 2 May 12
Posts: 22
Credit: 145,756,579
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwat
Message 24901 - Posted: 11 May 2012 | 1:02:50 UTC

I have 2X evga classy Mobos with Hexcore I7's. One has GTX570 and GTX460 (768), the other GTX465 and Gtx460 (1024). Only one gpu in each rig is running. Any help to get others going would be helpfull.

Profile K1atOdessa
Send message
Joined: 25 Feb 08
Posts: 249
Credit: 370,186,977
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24902 - Posted: 11 May 2012 | 4:59:53 UTC - in response to Message 24901.

Do you have the "<use_all_gpus>1</use_all_gpus>" option set in the cc_config.xml file?

I don't know if that is a requirement nowadays, or just a holdover from earlier versions of BOINC. But, I have 1 rig with 2 cards and another with 4 (only 1 in each runs GPUGrid given most of the cards are older, but still useful for other projects) and I've had that option set for a couple years now.

One thing to check is to make sure that BOINC is seeing both cards. For example (in the Event Log):


5/11/2012 12:52:57 AM | | NVIDIA GPU 0: GeForce GTX 570 (driver version 295.73, CUDA version 4.20, compute capability 2.0, 1280MB, 1171MB available, 1530 GFLOPS peak)
5/11/2012 12:52:57 AM | | NVIDIA GPU 1: GeForce GT 240 (driver version 295.73, CUDA version 4.20, compute capability 1.2, 512MB, 440MB available, 414 GFLOPS peak)

Profile K1atOdessa
Send message
Joined: 25 Feb 08
Posts: 249
Credit: 370,186,977
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24903 - Posted: 11 May 2012 | 5:04:13 UTC - in response to Message 24901.

According to your host, BOINC is only reporting 1 card in each system. Otherwise instead of seeing:

NVIDIA GeForce GTX 460 (767MB) driver: 266.58

you would see:

[2] NVIDIA GeForce GTX 460 (767MB) driver: 266.58


BOINC doesn't differential between cards when reporting back, so even though you have two separate, different cards you will see 1 reported 2x (assuming BOINC can see them).

I would start there and see if BOINC's event log shows a gpu 0 and gpu 1.

shdbcamping
Send message
Joined: 2 May 12
Posts: 22
Credit: 145,756,579
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwat
Message 24912 - Posted: 11 May 2012 | 10:30:43 UTC - in response to Message 24903.
Last modified: 11 May 2012 | 10:33:20 UTC

According to your host, BOINC is only reporting 1 card in each system. Otherwise instead of seeing:

NVIDIA GeForce GTX 460 (767MB) driver: 266.58

you would see:

[2] NVIDIA GeForce GTX 460 (767MB) driver: 266.58


BOINC doesn't differential between cards when reporting back, so even though you have two separate, different cards you will see 1 reported 2x (assuming BOINC can see them).

I would start there and see if BOINC's event log shows a gpu 0 and gpu 1.


5/11/2012 6:02:32 AM Starting BOINC client version 6.10.58 for windows_x86_64
5/11/2012 6:02:32 AM log flags: file_xfer, sched_ops, task
5/11/2012 6:02:32 AM Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3
5/11/2012 6:02:32 AM Data directory: C:\ProgramData\BOINC
5/11/2012 6:02:32 AM Processor: 12 GenuineIntel Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz [Family 6 Model 44 Stepping 2]
5/11/2012 6:02:32 AM Processor: 256.00 KB cache
5/11/2012 6:02:32 AM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 nx lm vmx tm2 popcnt aes pbe
5/11/2012 6:02:32 AM OS: Microsoft Windows 7: x64 Edition, Service Pack 1, (06.01.7601.00)
5/11/2012 6:02:32 AM Memory: 5.99 GB physical, 11.98 GB virtual
5/11/2012 6:02:32 AM Disk: 465.66 GB total, 418.10 GB free
5/11/2012 6:02:32 AM Local time is UTC -4 hours
5/11/2012 6:02:32 AM NVIDIA GPU 0 (not used): GeForce GTX 465 (driver version 26658, CUDA version 3020, compute capability 2.0, 994MB, 855 GFLOPS peak)
5/11/2012 6:02:32 AM NVIDIA GPU 1: GeForce GTX 460 (driver version 26658, CUDA version 3020, compute capability 2.1, 993MB, 645 GFLOPS peak)

I don't have a clue why it sees gpu0 but is not using it. New at this. Is it something to do with my windows GPU setup? I'm sure that I have suspend GPU when in use unchecked For boinc GPUGrid. Even restarted the computer. My other Rig is the same with the exception of it being XP64

I have desktop extended. Should that make a difference?

Profile MarkJ
Volunteer moderator
Project tester
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 730
Credit: 189,243,545
RAC: 228
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24913 - Posted: 11 May 2012 | 10:43:25 UTC - in response to Message 24912.

According to your host, BOINC is only reporting 1 card in each system. Otherwise instead of seeing:

NVIDIA GeForce GTX 460 (767MB) driver: 266.58

you would see:

[2] NVIDIA GeForce GTX 460 (767MB) driver: 266.58


BOINC doesn't differential between cards when reporting back, so even though you have two separate, different cards you will see 1 reported 2x (assuming BOINC can see them).

I would start there and see if BOINC's event log shows a gpu 0 and gpu 1.


5/11/2012 6:02:32 AM Starting BOINC client version 6.10.58 for windows_x86_64
5/11/2012 6:02:32 AM log flags: file_xfer, sched_ops, task
5/11/2012 6:02:32 AM Libraries: libcurl/7.19.7 OpenSSL/0.9.8l zlib/1.2.3
5/11/2012 6:02:32 AM Data directory: C:\ProgramData\BOINC
5/11/2012 6:02:32 AM Processor: 12 GenuineIntel Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz [Family 6 Model 44 Stepping 2]
5/11/2012 6:02:32 AM Processor: 256.00 KB cache
5/11/2012 6:02:32 AM Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 nx lm vmx tm2 popcnt aes pbe
5/11/2012 6:02:32 AM OS: Microsoft Windows 7: x64 Edition, Service Pack 1, (06.01.7601.00)
5/11/2012 6:02:32 AM Memory: 5.99 GB physical, 11.98 GB virtual
5/11/2012 6:02:32 AM Disk: 465.66 GB total, 418.10 GB free
5/11/2012 6:02:32 AM Local time is UTC -4 hours
5/11/2012 6:02:32 AM NVIDIA GPU 0 (not used): GeForce GTX 465 (driver version 26658, CUDA version 3020, compute capability 2.0, 994MB, 855 GFLOPS peak)
5/11/2012 6:02:32 AM NVIDIA GPU 1: GeForce GTX 460 (driver version 26658, CUDA version 3020, compute capability 2.1, 993MB, 645 GFLOPS peak)

I don't have a clue why it sees gpu0 but is not using it. New at this. Is it something to do with my windows GPU setup? I'm sure that I have suspend GPU when in use unchecked For boinc GPUGrid. Even restarted the computer. My other Rig is the same with the exception of it being XP64

I have desktop extended. Should that make a difference?


You need the use_all_gpus tag in your cc_config file. By default BOINC only uses the most capable gpu. If you need help with creating one just ask.

____________
BOINC blog

shdbcamping
Send message
Joined: 2 May 12
Posts: 22
Credit: 145,756,579
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwat
Message 24915 - Posted: 11 May 2012 | 10:57:53 UTC - in response to Message 24913.

I'm asking :)

Profile MarkJ
Volunteer moderator
Project tester
Volunteer tester
Send message
Joined: 24 Dec 08
Posts: 730
Credit: 189,243,545
RAC: 228
Level
Ile
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24918 - Posted: 11 May 2012 | 11:53:01 UTC - in response to Message 24915.

Step by step:

1. Start up Notepad
2. Paste the following code into it.

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

3. Save it into C:\ProgramData\BOINC as cc_config.xml (careful, Notepad likes to put .txt on the end of the filename)
4. Shut down your BOINC client if its running
5. Start BOINC up again

You should have a couple of messages in the startup about config: and it should use both GPUs

Thats it.
____________
BOINC blog

shdbcamping
Send message
Joined: 2 May 12
Posts: 22
Credit: 145,756,579
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwat
Message 24939 - Posted: 11 May 2012 | 18:08:56 UTC - in response to Message 24918.

Thanks a buncho! Now all 4 GPU's are crunching for GPUGrid :)

Profile K1atOdessa
Send message
Joined: 25 Feb 08
Posts: 249
Credit: 370,186,977
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24952 - Posted: 11 May 2012 | 20:38:22 UTC - in response to Message 24939.

Very good. Good to hear it was something simple like that cc_config setting. That RAC will be blowing up in no time.

TheFiend
Send message
Joined: 26 Aug 11
Posts: 97
Credit: 1,475,298,554
RAC: 651,315
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24963 - Posted: 11 May 2012 | 22:03:26 UTC

Will I be able to use a GTX 460 and GTX550Ti together by using the <use_all_gpus>1</use_all_gpus> command?

Profile skgiven
Volunteer moderator
Project tester
Volunteer tester
Avatar
Send message
Joined: 23 Apr 09
Posts: 3967
Credit: 1,803,940,389
RAC: 502,063
Level
His
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 24965 - Posted: 11 May 2012 | 22:18:24 UTC - in response to Message 24963.

Yes, if you put them in the same machine. You will also need to install the driver for the card you move.
____________
FAQ's

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

shdbcamping
Send message
Joined: 2 May 12
Posts: 22
Credit: 145,756,579
RAC: 0
Level
Cys
Scientific publications
watwatwatwatwatwatwat
Message 24986 - Posted: 12 May 2012 | 10:59:22 UTC

Yes you can. the driver for the 560 will work fine for the 460. I've got the GTX570 and GTX460 768 meg running fine together with the 266.58 driver and XP64. I also have a GTX465 and GTX460 1024 meg running the 266.58 Vista/W7 version in a wt premium 64bit machine.

TheFiend
Send message
Joined: 26 Aug 11
Posts: 97
Credit: 1,475,298,554
RAC: 651,315
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 25029 - Posted: 13 May 2012 | 10:30:33 UTC
Last modified: 13 May 2012 | 10:30:52 UTC

GTX460 and GTX550TI now running and the load is well within the capability of the PSU..... :-)

Clay
Send message
Joined: 21 Feb 12
Posts: 2
Credit: 242,939,219
RAC: 226,270
Level
Leu
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwat
Message 46016 - Posted: 31 Dec 2016 | 7:46:50 UTC - in response to Message 24918.

I tried this and it did not work for me.
Windows 10
GTX1060
GTX970
32 GB Ram

The log file recognizes two cards it says the 970 is not used.

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

Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1814
Credit: 9,953,056,994
RAC: 6,299,531
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 46017 - Posted: 31 Dec 2016 | 9:27:30 UTC - in response to Message 46016.

You should not mix Pascal cards with older cards in the same system.
Changing GPUs from/to older to/from Pascal mid-run will result in immediate computation error.
The default location of cc_config.xml is c:\ProgramData\BOINC\
You should re-read the configuration files in BOINC manager (Options menu)

BelgianEnthousiast
Send message
Joined: 7 Apr 15
Posts: 20
Credit: 240,322,825
RAC: 735,023
Level
Leu
Scientific publications
wat
Message 46221 - Posted: 18 Jan 2017 | 8:27:46 UTC

Hi,

I'm having two GPU's and would like to run GPUGRID on both GPU's at the same time.

I've been reading up as much as I could on the different fora how to get to it and created the cc_config.xml with the option <use_all_gpus>1</use_all_gpus>.

I put the cc_config file in E:\BOINC\DATA as indicated.
It worked just once where both GPU's received a WU and were crunching simultaneously. Afterwards... No more dual GPU usage, just using the GTX 1070... :-(

BOINC Mgr version is 7.6.33 (x64)
Running Windows 10 Pro
NVIDIA Driver : 8.1.9600.0

Despite the additional config, BOINC Mgr does not take into account my second GPU :
18/01/2017 7:08:38 | | Starting BOINC client version 7.6.33 for windows_x86_64
18/01/2017 7:08:38 | | log flags: file_xfer, sched_ops, task, coproc_debug
18/01/2017 7:08:38 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
18/01/2017 7:08:38 | | Data directory: E:\BOINC DATA
18/01/2017 7:08:38 | | [coproc] launching child process at H:\BOINC\boinc.exe
18/01/2017 7:08:38 | | [coproc] relative to directory E:\BOINC DATA
18/01/2017 7:08:38 | | [coproc] with data directory "E:\BOINC DATA"
18/01/2017 7:08:39 | | CUDA: NVIDIA GPU 0: GeForce GTX 1070 (driver version 376.19, CUDA version 8.0, compute capability 6.1, 4096MB, 3560MB available, 7046 GFLOPS peak)
18/01/2017 7:08:39 | | CUDA: NVIDIA GPU 1 (not used): GeForce GTX TITAN Black (driver version 376.19, CUDA version 8.0, compute capability 3.5, 4096MB, 4010MB available, 6396 GFLOPS peak)
18/01/2017 7:08:39 | | OpenCL: NVIDIA GPU 0: GeForce GTX 1070 (driver version 376.19, device version OpenCL 1.2 CUDA, 8192MB, 3560MB available, 7046 GFLOPS peak)
18/01/2017 7:08:39 | | OpenCL: NVIDIA GPU 1 (ignored by config): GeForce GTX TITAN Black (driver version 376.19, device version OpenCL 1.2 CUDA, 6144MB, 4010MB available, 6396 GFLOPS peak)
18/01/2017 7:08:39 | | [coproc] NVIDIA library reports 2 GPUs
18/01/2017 7:08:39 | | [coproc] No ATI library found.
18/01/2017 7:08:40 | | Host name: PRIV-PC-001
18/01/2017 7:08:40 | | Processor: 12 GenuineIntel Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz [Family 6 Model 63 Stepping 2]
18/01/2017 7:08:40 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 dca pbe fsgsbase bmi1 smep bmi2
18/01/2017 7:08:40 | | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.14393.00)
18/01/2017 7:08:40 | | Memory: 31.91 GB physical, 78.88 GB virtual
18/01/2017 7:08:40 | | Disk: 3.64 TB total, 1.79 TB free
18/01/2017 7:08:40 | | Local time is UTC +1 hours
18/01/2017 7:08:40 | | VirtualBox version: 5.1.12

Question : Why does it say : "NVIDIA GPU 1 (ignored by config): GeForce GTX TITAN Black" ?

For completeness, further down the log, I get this kind of info regarding the GPUGrid work distribution :
18/01/2017 8:28:49 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:49 | ATLAS@home | Starting task 5fyKDmNorhpn7jp7oou28CBqABFKDmABFKDmO5GKDmy3NKDmBndU0m_1
18/01/2017 8:28:49 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:51 | LHC@home | Finished download of Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__8_10__6__30_1_sixvf_boinc6472.zip
18/01/2017 8:28:51 | LHC@home | Finished download of Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__8_10__6__37.5_1_sixvf_boinc6473.zip
18/01/2017 8:28:51 | LHC@home | Started download of Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__8_10__6__45_1_sixvf_boinc6474.zip
18/01/2017 8:28:51 | LHC@home | Started download of Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__6_8__6__45_1_sixvf_boinc6463.zip
18/01/2017 8:28:51 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:51 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:52 | LHC@home | Finished download of Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__8_10__6__45_1_sixvf_boinc6474.zip
18/01/2017 8:28:52 | LHC@home | Finished download of Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__6_8__6__45_1_sixvf_boinc6463.zip
18/01/2017 8:28:52 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:52 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:54 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:55 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:57 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:58 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:58 | ATLAS@home | task 5fyKDmNorhpn7jp7oou28CBqABFKDmABFKDmO5GKDmy3NKDmBndU0m_1 suspended by user
18/01/2017 8:28:59 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:28:59 | ATLAS@home | Starting task NvnMDmdkrhpnDDn7oo6G73TpABFKDmABFKDmPeGKDmABFKDmmL5a4m_1
18/01/2017 8:28:59 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:00 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:01 | ATLAS@home | task NvnMDmdkrhpnDDn7oo6G73TpABFKDmABFKDmPeGKDmABFKDmmL5a4m_1 suspended by user
18/01/2017 8:29:02 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:02 | ATLAS@home | Starting task dhGNDmerumpnDDn7oo6G73TpABFKDmABFKDmU0IKDmABFKDmQUvmbm_1
18/01/2017 8:29:02 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:03 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:05 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:06 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:06 | ATLAS@home | task dhGNDmerumpnDDn7oo6G73TpABFKDmABFKDmU0IKDmABFKDmQUvmbm_1 suspended by user
18/01/2017 8:29:07 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:07 | ATLAS@home | Starting task CYdLDmA52mpn7jp7oou28CBqABFKDmABFKDm27FKDmQdHKDmGcRG9m_0
18/01/2017 8:29:07 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:08 | LHC@home | task Qtwo_nWP_st3m1_bb28__4__s__62.2868_60.2968__8_10__6__45_1_sixvf_boinc6474_0 suspended by user
18/01/2017 8:29:08 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:09 | GPUGRID | Starting task GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:20 | ATLAS@home | task CYdLDmA52mpn7jp7oou28CBqABFKDmABFKDm27FKDmQdHKDmGcRG9m_0 suspended by user
18/01/2017 8:29:20 | ATLAS@home | task JAYLDmbs4mpnDDn7oo6G73TpABFKDmABFKDmJCIKDmABFKDmQwjSnm_1 suspended by user
18/01/2017 8:29:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:26 | World Community Grid | task OET1_0004393_x3MX2_rig_37759_0 suspended by user
18/01/2017 8:29:26 | World Community Grid | task OET1_0004393_x3MX2_rig_3796_0 suspended by user
18/01/2017 8:29:26 | World Community Grid | task OET1_0004393_x3MX2_rig_31293_0 suspended by user
18/01/2017 8:29:26 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:29 | World Community Grid | task OET1_0004393_x3MX2_rig_33111_0 suspended by user
18/01/2017 8:29:30 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:41 | GPUGRID | task WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0 resumed by user
18/01/2017 8:29:42 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:29:42 | GPUGRID | [coproc] Insufficient NVIDIA for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0; need 1, available 0
18/01/2017 8:29:47 | GPUGRID | task GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1 suspended by user
18/01/2017 8:29:48 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:30:13 | rosetta@home | Task rb_01_15_71580_115416__t000__0_C1_SAVE_ALL_OUT_IGNORE_THE_REST_462542_241_0 exited with zero status but no 'finished' file
18/01/2017 8:30:13 | rosetta@home | If this happens repeatedly you may need to reset the project.
18/01/2017 8:30:13 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:30:13 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:30:14 | GPUGRID | task GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1 resumed by user
18/01/2017 8:30:15 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:30:15 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:30:38 | World Community Grid | task OET1_0004393_x3MX2_rig_33111_0 resumed by user
18/01/2017 8:30:38 | World Community Grid | task OET1_0004393_x3MX2_rig_37759_0 resumed by user
18/01/2017 8:30:38 | World Community Grid | task OET1_0004393_x3MX2_rig_3796_0 resumed by user
18/01/2017 8:30:38 | World Community Grid | task OET1_0004393_x3MX2_rig_31293_0 resumed by user
18/01/2017 8:30:38 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:30:38 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:31:14 | GPUGRID | Sending scheduler request: To fetch work.
18/01/2017 8:31:14 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
18/01/2017 8:31:15 | GPUGRID | Scheduler request completed: got 0 new tasks
18/01/2017 8:31:15 | GPUGRID | No tasks sent
18/01/2017 8:31:15 | GPUGRID | This computer has reached a limit on tasks in progress
18/01/2017 8:31:38 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:31:38 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:32:38 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:32:38 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:33:39 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:33:39 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:34:39 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:34:39 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:35:39 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:35:39 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:36:39 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:36:39 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:37:39 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:37:39 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:38:39 | GPUGRID | [coproc] NVIDIA instance 0; 1.000000 pending for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:38:39 | GPUGRID | [coproc] NVIDIA instance 0: confirming 1.000000 instance for WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0
18/01/2017 8:39:18 | GPUGRID | Computation for task WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0 finished
18/01/2017 8:39:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:39:22 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_0
18/01/2017 8:39:22 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_1
18/01/2017 8:39:25 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_0
18/01/2017 8:39:25 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_2
18/01/2017 8:39:29 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_1
18/01/2017 8:39:29 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_3
18/01/2017 8:39:30 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_2
18/01/2017 8:39:30 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_7
18/01/2017 8:39:31 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_3
18/01/2017 8:39:31 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_7
18/01/2017 8:39:31 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_9
18/01/2017 8:39:31 | GPUGRID | Started upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_10
18/01/2017 8:39:32 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_10
18/01/2017 8:39:34 | World Community Grid | Computation for task OET1_0004393_x3MX2_rig_33111_0 finished
18/01/2017 8:39:34 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:39:36 | World Community Grid | Started upload of OET1_0004393_x3MX2_rig_33111_0_r1724252269_0
18/01/2017 8:39:38 | World Community Grid | Finished upload of OET1_0004393_x3MX2_rig_33111_0_r1724252269_0
18/01/2017 8:39:42 | World Community Grid | Computation for task OET1_0004393_x3MX2_rig_3796_0 finished
18/01/2017 8:39:42 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:39:44 | World Community Grid | Started upload of OET1_0004393_x3MX2_rig_3796_0_r43697123_0
18/01/2017 8:39:46 | World Community Grid | Finished upload of OET1_0004393_x3MX2_rig_3796_0_r43697123_0
18/01/2017 8:40:43 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:41:43 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:42:43 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:42:48 | GPUGRID | Finished upload of WT_S3F2_C25-SDOERR_BNBS2-3-4-RND1359_0_9
18/01/2017 8:42:48 | GPUGRID | Sending scheduler request: To report completed tasks.
18/01/2017 8:42:48 | GPUGRID | Reporting 1 completed tasks
18/01/2017 8:42:48 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
18/01/2017 8:42:50 | GPUGRID | Scheduler request completed: got 1 new tasks
18/01/2017 8:42:52 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-LICENSE
18/01/2017 8:42:52 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-COPYRIGHT
18/01/2017 8:42:53 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-LICENSE
18/01/2017 8:42:53 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-COPYRIGHT
18/01/2017 8:42:53 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_1
18/01/2017 8:42:53 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_2
18/01/2017 8:42:56 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_1
18/01/2017 8:42:56 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_3
18/01/2017 8:42:58 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_2
18/01/2017 8:42:58 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_3
18/01/2017 8:42:58 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-pdb_file
18/01/2017 8:42:58 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-psf_file
18/01/2017 8:42:59 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-psf_file
18/01/2017 8:42:59 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-par_file
18/01/2017 8:43:07 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-par_file
18/01/2017 8:43:07 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-conf_file_enc
18/01/2017 8:43:08 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-conf_file_enc
18/01/2017 8:43:08 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-metainp_file
18/01/2017 8:43:09 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-metainp_file
18/01/2017 8:43:09 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_7
18/01/2017 8:43:10 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_7
18/01/2017 8:43:10 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_10
18/01/2017 8:43:12 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-pdb_file
18/01/2017 8:43:12 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-GLU73ALA_S14F16_C1-SDOERR_BNBS2-2-4-RND9486_10
18/01/2017 8:43:12 | GPUGRID | Started download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-prmtop_file
18/01/2017 8:43:16 | GPUGRID | Finished download of GLU73ALA_S14F16_C1-SDOERR_BNBS2-3-prmtop_file
18/01/2017 8:43:17 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:43:22 | GPUGRID | Sending scheduler request: To fetch work.
18/01/2017 8:43:22 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
18/01/2017 8:43:23 | GPUGRID | Scheduler request completed: got 0 new tasks
18/01/2017 8:43:23 | GPUGRID | No tasks sent
18/01/2017 8:43:23 | GPUGRID | This computer has reached a limit on tasks in progress
18/01/2017 8:44:18 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:45:18 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:46:19 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:47:19 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:48:20 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:49:01 | GPUGRID | Sending scheduler request: To fetch work.
18/01/2017 8:49:01 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
18/01/2017 8:49:02 | GPUGRID | Scheduler request completed: got 0 new tasks
18/01/2017 8:49:02 | GPUGRID | No tasks sent
18/01/2017 8:49:02 | GPUGRID | This computer has reached a limit on tasks in progress
18/01/2017 8:49:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:50:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:51:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:52:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:53:21 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:54:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:55:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:56:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:57:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:58:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 8:59:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:00:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:01:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:02:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:03:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:04:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:05:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:05:39 | GPUGRID | Sending scheduler request: To fetch work.
18/01/2017 9:05:39 | GPUGRID | Requesting new tasks for CPU and NVIDIA GPU
18/01/2017 9:05:41 | GPUGRID | Scheduler request completed: got 0 new tasks
18/01/2017 9:05:41 | GPUGRID | No tasks sent
18/01/2017 9:05:41 | GPUGRID | This computer has reached a limit on tasks in progress
18/01/2017 9:06:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:07:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1
18/01/2017 9:08:22 | GPUGRID | [coproc] Assigning NVIDIA instance 0 to GLU73ALA_S13F4_C2-SDOERR_BNBS2-3-4-RND0369_1

Would be great to have a solution to this problem... !

Many thanks in advance for any help !

BelgianEnthousiast.

Seba
Send message
Joined: 30 Oct 16
Posts: 6
Credit: 27,935,274
RAC: 0
Level
Val
Scientific publications
watwat
Message 46223 - Posted: 18 Jan 2017 | 12:05:18 UTC - in response to Message 46221.
Last modified: 18 Jan 2017 | 12:24:12 UTC

Hi,

I'm having two GPU's and would like to run GPUGRID on both GPU's at the same time.

I've been reading up as much as I could on the different fora how to get to it and created the cc_config.xml with the option <use_all_gpus>1</use_all_gpus>.

I put the cc_config file in E:\BOINC\DATA as indicated.
It worked just once where both GPU's received a WU and were crunching simultaneously. Afterwards... No more dual GPU usage, just using the GTX 1070... :-(

Hi!
You have probably put cc_config.xml file to wrong place, put this file to main boinc directory E:\BOINC\ . It should resolve your problem.
My cc_config file contains:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>

It works fine with 2 different gpu.

Edit:
I see one of your cards is not pascal architecture so it will not work. Both cards must be pascal (gtx 1050,1060,1070,1080,titan x) or both cards with older architecture, otherwise boinc application will cause "Error while computing".

BelgianEnthousiast
Send message
Joined: 7 Apr 15
Posts: 20
Credit: 240,322,825
RAC: 735,023
Level
Leu
Scientific publications
wat
Message 46389 - Posted: 29 Jan 2017 | 16:27:28 UTC - in response to Message 46223.

Hi Seba,

Thanks for your quick response !

You seem to be right. When booting the PC, my system starts two WU's in parallel, one on the Titan, one on the 1070.

And as you predicted : I get calculation error on one WU (probably on the Titan card) and it happily continues to crunch on the 1070.

Is there a way around this Pascal/Non-Pascal limitation ?
It's a real shame to have a Titan card sitting there doing nothing but drawing power uselessly...

(where does this limitation stem from actually, can't the developers abstract
the hardware and use both technologies maybe with a small loss in processing power ?)

Thanks !

BelgianEnthousiast
Send message
Joined: 7 Apr 15
Posts: 20
Credit: 240,322,825
RAC: 735,023
Level
Leu
Scientific publications
wat
Message 46390 - Posted: 29 Jan 2017 | 16:32:22 UTC - in response to Message 46223.

Seba, update :

As there were no WU's available on GPUGrid, I enabled MilkyWay again this afternoon and I just noticed that it's crunching a WU on each card now, with good results !

GPUGrid seems to be the one having issues with this.

And, I observed also that while crunching on GPUGrid, it's not possible to have a MilkyWay on my second card in parallel.

Is GPUGrid blocking this somehow ?


Profile Retvari Zoltan
Avatar
Send message
Joined: 20 Jan 09
Posts: 1814
Credit: 9,953,056,994
RAC: 6,299,531
Level
Tyr
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 46393 - Posted: 29 Jan 2017 | 18:00:29 UTC - in response to Message 46390.

I've opened a new thread about this:
DO NOT MIX GTX 10X0 GPU with older GPUs in the same system

Post to thread

Message boards : Graphics cards (GPUs) : How to get 2 GPUs running GPUGrid?