Message boards :
Number crunching :
New GTX 1070 and SOG vs CUDA 42....
Message board moderation
Author | Message |
---|---|
jdzukley Send message Joined: 6 Apr 11 Posts: 19 Credit: 26,357,809 RAC: 74 |
Help... A short time ago, I upgraded my main crunching unit from 2 GTX 570's to 2 GTX 1070's. All new work from SETI was being crunched 4 times faster, RAC was going up... All was good. Tasks were all SOG tasks. Then a few days ago, all new work received was, and still is CUDA 42 tasks. Both GPU's are now only at 30% while with SOG they were 90%+. The CUDA 42 tasks seam to take even longer than they did than with my GTX 570 cards... With SOG tasks, it appeared to me that SETI did pick the best type of work for my GPU's and all was well. However, in the case of CUDA 42, this is not even close. So now what should I do to regain performance? Do I need to configure - cc_config.xml to run multiple jobs on the 1070's or do I sit tight and wait for SOG type work to be made available again? And or do I need 2 different config files based on what SETI is sending me, and need to swap them out based on the type of work being received???? I would appreciate your feedback. |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
You should upgrade to the new beta lunatics version of the apps. http://mikesworld.eu/download.html You will need a 7-zip to unzip the files, selection of SoG is the best option for those 1070s. Cuda are much slower than SoG when crunching the new work units. Good luck Zalster |
petri33 Send message Joined: 6 Jun 02 Posts: 1668 Credit: 623,086,772 RAC: 156 |
Hi, If you are running a dedicated cruncher you may consider Linux. You'd see an enormous RAC increase with the right app. Petri To overcome Heisenbergs: "You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.