Message boards :
Wish list :
SLI & XFire Support
Message board moderation
| Author | Message |
|---|---|
liveoncSend message Joined: 1 Jan 10 Posts: 292 Credit: 41,567,650 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]()
|
I read that GPUGRID.net wasn't interested in extending the time limit awarded to WU because it would cause delayes to the projects. I was thinking about SLI & XFire support? Why, because not only would that decrease the time that it take to crunch a WU on already powerful GPU's, but also will allow HTPC's & other PC's that use two or more less powerful GPU's.
|
GDFSend message Joined: 14 Mar 07 Posts: 1958 Credit: 629,356 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() |
SLI & XFire don't work for compute. gdf |
liveoncSend message Joined: 1 Jan 10 Posts: 292 Credit: 41,567,650 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]()
|
Can't yet, or won't ever? If SLI slips the graphic up in 2,3, or 4 parts on one screen. Can't CUDA work in such a way or maybe GPUGRIDE.net could WU split up into sets? Is there not a way to make better use of 2 GPU's or more working on the same task? The PCI-E lanes have a maximum speed, if 32 lanes were used instead of 16? Or is this just not possible?
|
GDFSend message Joined: 14 Mar 07 Posts: 1958 Credit: 629,356 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() |
I don't know for sure, but probably not possible. gdf |
©2025 Universitat Pompeu Fabra