NVIDIA Tesla C1060


log in

Advanced search

Questions and Answers : GPU applications : NVIDIA Tesla C1060

Author Message
N0ADM
Send message
Joined: 17 Jun 11
Posts: 4
Credit: 44,440
RAC: 0
United States
Message 1123045 - Posted: 30 Jun 2011, 17:50:39 UTC

Has any one used this GPU board NVidia Tesla C1060? would be compined with 2)GeForce GTX 560 Ti Video Cards, ASUS Rampage III Black Edition ROG Motherboard, Intel Core i7-990X 3.46 GHz, 24GB of DDR3 Ram, 2) 1TB sata drives, looking for more computing power.

thanks for the advice

OzzFan
Volunteer tester
Avatar
Send message
Joined: 9 Apr 02
Posts: 13662
Credit: 31,494,196
RAC: 11,319
United States
Message 1123053 - Posted: 30 Jun 2011, 18:16:38 UTC - in response to Message 1123045.

It is estimated that the nVidia Tesla C1060 is about as powerful as a GTX 285, but consumes considerably more power.

If you're looking for more computing power, I'd suggest going with triple-SLI and get another GTX 560 Ti. Make sure your power supply can handle it first, of course.

Profile Lint trapProject donor
Send message
Joined: 30 May 03
Posts: 871
Credit: 27,837,024
RAC: 10,588
United States
Message 1123097 - Posted: 30 Jun 2011, 20:49:25 UTC - in response to Message 1123053.

If you're looking for more computing power, I'd suggest going with triple-SLI and get another GTX 560 Ti. Make sure your power supply can handle it first, of course.


I think the 560 Ti (like too many of Nvidia's later offerings) only does 2fer-SLI; there's just one connector on it.

Martin

N0ADM
Send message
Joined: 17 Jun 11
Posts: 4
Credit: 44,440
RAC: 0
United States
Message 1123111 - Posted: 30 Jun 2011, 21:53:32 UTC

the GT 560 series i have can only run in 2 way SLI configuration, one reason i was thinking of adding the tesla to the system.

OzzFan
Volunteer tester
Avatar
Send message
Joined: 9 Apr 02
Posts: 13662
Credit: 31,494,196
RAC: 11,319
United States
Message 1123117 - Posted: 30 Jun 2011, 22:23:56 UTC - in response to Message 1123111.

Then don't run it in SLI, just run all three individually.

Profile Ageless
Avatar
Send message
Joined: 9 Jun 99
Posts: 12393
Credit: 2,666,323
RAC: 958
Netherlands
Message 1123124 - Posted: 30 Jun 2011, 22:40:23 UTC - in response to Message 1123117.

Have you seen the width of those things? How do you add 3 next to each other in one system? Mustn't be a tower of sorts either, as how do you hang three of those things underneath each other without ripping your motherboard to pieces?

I see an Asus GTX560 TI DCII TOP/2DI/1G weighs in at 2.87lbs, that's 1.29 kilos. You sure you want 4 kilos of videocard in your tower? ;-)
____________
Jord

Fighting for the correct use of the apostrophe, together with Weird Al Yankovic

N0ADM
Send message
Joined: 17 Jun 11
Posts: 4
Credit: 44,440
RAC: 0
United States
Message 1123125 - Posted: 30 Jun 2011, 22:41:02 UTC - in response to Message 1123117.

bionc will see all 3 cards and utilize all 3 in that way? if so thanks for the help.

N0ADM
Send message
Joined: 17 Jun 11
Posts: 4
Credit: 44,440
RAC: 0
United States
Message 1123127 - Posted: 30 Jun 2011, 22:44:05 UTC - in response to Message 1123125.

i figure while im am still in the building of this system mind as well utilize all features that i paid for and its for a good cause. figured my budget for this build was 7K and i am about 5K currently

Profile [seti.international] Dirk Sadowski
Volunteer tester
Avatar
Send message
Joined: 6 Apr 07
Posts: 7115
Credit: 61,257,416
RAC: 6,709
Germany
Message 1123150 - Posted: 30 Jun 2011, 23:11:07 UTC
Last modified: 30 Jun 2011, 23:14:41 UTC

If you are a gamer it's an other topic..

But for CUDA calculation you need to disable SLI for max performance. Newer driver should do this automatically.

So if you would like to build a crunching machine, you could add different cards in one PC case.
But you shouldn't combine non- (8xxx to GTX2xx series)/ & Fermi (GTX4xx+) cards in one PC case, because BOINC can't send different numbers of WUs to different (non-/&Fermi) GPUs.
Because with the currently available CUDA apps you would need to let run 2+ WUs/GPU simultaneously for whole Fermi chip usage.
On non Fermi GPUs this isn't possible, there only 1 WU/GPU.


- Best regards! - Sutaru Tsureku, team seti.international founder. - Optimize your PC for higher RAC. - SETI@home needs your help. -
____________
BR

SETI@home Needs your Help ... $10 & U get a Star!

Team seti.international

Das Deutsche Cafe. The German Cafe.

OzzFan
Volunteer tester
Avatar
Send message
Joined: 9 Apr 02
Posts: 13662
Credit: 31,494,196
RAC: 11,319
United States
Message 1123183 - Posted: 1 Jul 2011, 1:06:32 UTC - in response to Message 1123125.

bionc will see all 3 cards and utilize all 3 in that way? if so thanks for the help.


Yes, BOINC will see all three cards and utilize all 3.

Greg W Jones
Send message
Joined: 28 Nov 02
Posts: 29
Credit: 25,473,210
RAC: 0
United States
Message 1139514 - Posted: 12 Aug 2011, 18:39:15 UTC - in response to Message 1123183.

So, you're saying that it is not recommended to combine a GTX260 and a GTX560Ti in the same computer for boinc crunching?

I currently have a GTX260 and was planning to add a GTX560Ti in the same computer. I am just starting to read and learn about optiziming my GPU crunching. So, I have not yet enabled multiple WUs per GPU yet.

Any advise would be appreciated.

Thanks
Greg

____________

Profile BilBg
Volunteer tester
Avatar
Send message
Joined: 27 May 07
Posts: 2879
Credit: 6,452,347
RAC: 2,884
Bulgaria
Message 1140663 - Posted: 15 Aug 2011, 1:55:43 UTC - in response to Message 1139514.


Better move the GeForce GTX 260 to another computer (Core 2 Quad CPU Q6600 @ 2.40GHz)
and put the GTX 560Ti in "this" (Core 2 Quad CPU Q9450 @ 2.66GHz) computer.
(unless you want to use the GTX 260 as PhysX card for gaming)

It is possible to use the mixed GTX 260/560Ti configuration with both standard and optimized apps (which I see you use now: "Multibeam x38g Preview, Cuda 3.20" (Lunatics_x38g_win32_cuda32.exe))
but GTX 260 is not capable of running effectively 2 tasks at once.


If you want to test it - do it now before adding the GTX 560Ti :

Suspend all tasks
Exit BOINC

Go to ....\projects\setiathome.berkeley.edu\
Make copy of app_info.xml (for easy undo of changes)
Edit app_info.xml with Notepad
Make the changes (info bellow) and save the file

Start BOINC
Resume 2 CUDA tasks

Look closely for the running times and possible errors


The changes to make:
In the following section change the all 6 occurrences of:
<count>1</count>
to:
<count>0.5</count>

<app_info> <app> <name>setiathome_enhanced</name> </app> <file_info> <name>Lunatics_x38g_win32_cuda32.exe</name> <executable/> </file_info> <file_info> <name>cudart32_32_16.dll</name> <executable/> </file_info> <file_info> <name>cufft32_32_16.dll</name> <executable/> </file_info> <app_version> <app_name>setiathome_enhanced</app_name> <version_num>610</version_num> <platform>windows_intelx86</platform> <plan_class>cuda_fermi</plan_class> <avg_ncpus>0.040000</avg_ncpus> <max_ncpus>0.040000</max_ncpus> <coproc> <type>CUDA</type> <count>1</count> </coproc> <file_ref> <file_name>Lunatics_x38g_win32_cuda32.exe</file_name> <main_program/> </file_ref> <file_ref> <file_name>cudart32_32_16.dll</file_name> </file_ref> <file_ref> <file_name>cufft32_32_16.dll</file_name> </file_ref> </app_version> <app_version> <app_name>setiathome_enhanced</app_name> <version_num>609</version_num> <platform>windows_intelx86</platform> <plan_class>cuda23</plan_class> <avg_ncpus>0.040000</avg_ncpus> <max_ncpus>0.040000</max_ncpus> <coproc> <type>CUDA</type> <count>1</count> </coproc> <file_ref> <file_name>Lunatics_x38g_win32_cuda32.exe</file_name> <main_program/> </file_ref> <file_ref> <file_name>cudart32_32_16.dll</file_name> </file_ref> <file_ref> <file_name>cufft32_32_16.dll</file_name> </file_ref> </app_version> <app_version> <app_name>setiathome_enhanced</app_name> <version_num>608</version_num> <platform>windows_intelx86</platform> <plan_class>cuda</plan_class> <avg_ncpus>0.040000</avg_ncpus> <max_ncpus>0.040000</max_ncpus> <coproc> <type>CUDA</type> <count>1</count> </coproc> <file_ref> <file_name>Lunatics_x38g_win32_cuda32.exe</file_name> <main_program/> </file_ref> <file_ref> <file_name>cudart32_32_16.dll</file_name> </file_ref> <file_ref> <file_name>cufft32_32_16.dll</file_name> </file_ref> </app_version> <app_version> <app_name>setiathome_enhanced</app_name> <version_num>610</version_num> <platform>windows_x86_64</platform> <plan_class>cuda_fermi</plan_class> <avg_ncpus>0.040000</avg_ncpus> <max_ncpus>0.040000</max_ncpus> <coproc> <type>CUDA</type> <count>1</count> </coproc> <file_ref> <file_name>Lunatics_x38g_win32_cuda32.exe</file_name> <main_program/> </file_ref> <file_ref> <file_name>cudart32_32_16.dll</file_name> </file_ref> <file_ref> <file_name>cufft32_32_16.dll</file_name> </file_ref> </app_version> <app_version> <app_name>setiathome_enhanced</app_name> <version_num>609</version_num> <platform>windows_x86_64</platform> <plan_class>cuda23</plan_class> <avg_ncpus>0.040000</avg_ncpus> <max_ncpus>0.040000</max_ncpus> <coproc> <type>CUDA</type> <count>1</count> </coproc> <file_ref> <file_name>Lunatics_x38g_win32_cuda32.exe</file_name> <main_program/> </file_ref> <file_ref> <file_name>cudart32_32_16.dll</file_name> </file_ref> <file_ref> <file_name>cufft32_32_16.dll</file_name> </file_ref> </app_version> <app_version> <app_name>setiathome_enhanced</app_name> <version_num>608</version_num> <platform>windows_x86_64</platform> <plan_class>cuda</plan_class> <avg_ncpus>0.040000</avg_ncpus> <max_ncpus>0.040000</max_ncpus> <coproc> <type>CUDA</type> <count>1</count> </coproc> <file_ref> <file_name>Lunatics_x38g_win32_cuda32.exe</file_name> <main_program/> </file_ref> <file_ref> <file_name>cudart32_32_16.dll</file_name> </file_ref> <file_ref> <file_name>cufft32_32_16.dll</file_name> </file_ref> </app_version> </app_info>


____________



- ALF - "Find out what you don't do well ..... then don't do it!" :)

Questions and Answers : GPU applications : NVIDIA Tesla C1060

Copyright © 2014 University of California