1)
Message boards :
Number crunching :
Benchmark results - times wanted for any hardware, CPU or GPU, old or new!
(Message 61408)
Posted 15 Mar 2014 by Phoenix Post: Ah wicked, thx :), was hoping 1 of those would show up. The short tasks take 189s, so the linked WU must be one of those. For the benchmark I run 1 task and deactivated all cpu work units. Now I am back to my normal setup: 2 milkyway gpu tasks in parallel and 4 WCG (clean energy project) tasks on the cpu (i5 3570). With this setup the cpu and gpu run at 99-100%. I know it isn't optimised for maximum MW crunching power but I also want to support the other project ;) I am also running games and other things on the pc so my times can vary alot. Overall I am very satisfied with my gpu. I paid about 90€ when it came out and the performance is still ok :) |
2)
Message boards :
Number crunching :
Benchmark results - times wanted for any hardware, CPU or GPU, old or new!
(Message 61404)
Posted 15 Mar 2014 by Phoenix Post: Since no HD 5830 has been reported yet: The slow ones take 282s each. http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=690429473 The gpu runs at stock values: 800 MHz gpu clock, 1000 MHz memory clock. http://www.sapphiretech.com/presentation/product/product_index.aspx?pid=1153&lid=1 Greetings Phoenix |
3)
Message boards :
Number crunching :
nbody-plum files don't run - MW stops running
(Message 52680)
Posted 27 Jan 2012 by Phoenix Post: As far as I understand the n-body WUs can only be processed by your CPU. Please check if your CPU is allowed to work for this project. There is also an option to block getting n-body WUs from the server. |
©2023 Astroinformatics Group