GPU on MB + GPU discrete in PCI Express x16 slot


log in

Advanced search

Message boards : Number crunching : GPU on MB + GPU discrete in PCI Express x16 slot

Author Message
Neil L. Carter
Volunteer tester
Send message
Joined: 6 Dec 99
Posts: 53
Credit: 3,367,296
RAC: 6,009
United States
Message 1360410 - Posted: 23 Apr 2013, 22:31:24 UTC

Greetings:

I've been running ATI GPU units for some time now. A friend recently pointed out that my mothernoard has a GPU installed in it as well, an ATI HD3300.

Is there a way to get both units to crunch?

Thanks!

Neil
____________

Profile HAL9000
Volunteer tester
Avatar
Send message
Joined: 11 Sep 99
Posts: 3581
Credit: 98,389,964
RAC: 77,866
United States
Message 1360643 - Posted: 24 Apr 2013, 13:04:37 UTC

The HD 3000 series is not OpenCL capable, but the hybrid AP app could be used with it. If you really wanted to use it.
However IIRC there is not currently a way to tell BOINC to only run certain apps on specific GPU's. It would require running 2 instances of BOINC. Your current one and a 2nd configured only to use the HD 3300 & the hybrid application.
____________
SETI@home classic workunits: 93,865 CPU time: 863,447 hours

Join the BP6/VP6 User Group today!

Profile ignorance is no excuse
Avatar
Send message
Joined: 4 Oct 00
Posts: 9529
Credit: 44,432,515
RAC: 121
Korea, North
Message 1360723 - Posted: 24 Apr 2013, 16:50:49 UTC

Also note that the 3300 GPU using the hybrid app would require a good deal of CPU time. This and the slow speed of the 3300 would deter most from bothering to run apps on it. You'd probably be better off just disabling that GPU
____________
In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

End terrorism by building a school

Neil L. Carter
Volunteer tester
Send message
Joined: 6 Dec 99
Posts: 53
Credit: 3,367,296
RAC: 6,009
United States
Message 1360857 - Posted: 24 Apr 2013, 23:51:47 UTC

OK, scratch that! I thought it might be a good idea.

Thanks, guys!!

Neil

Message boards : Number crunching : GPU on MB + GPU discrete in PCI Express x16 slot

Copyright © 2014 University of California