1)
Message boards :
Number crunching :
Insufficient amount WU on the server
(Message 64139)
Posted 2 Dec 2015 by blub Post: Does the WU shortage mean there is something wrong with the generation of WUs, or does it mean the crunchers keep up with what the scientists want to have computed? |
2)
Message boards :
Number crunching :
Mac OSX NVIDIA got absourd crunch rates
(Message 64121)
Posted 24 Nov 2015 by blub Post: according to http://milkyway.cs.rpi.edu/milkyway/apps.php the Mac OSX NVIDIA app for modified fit has an average performance of 704,877,187 GigaFLOPS or about 99.98% of the project total. Last time I checked the market share of Mac OSX was way lower than 99.98%, especially if one takes in account nowadays the mac computers come with intel or amd GPUs. |
3)
Message boards :
Number crunching :
New Benchmark Thread - times wanted for any hardware, CPU or GPU, old or new!
(Message 64026)
Posted 23 Oct 2015 by blub Post: hd 7970 @1052Mhz core, 1375MHz ram: 24.118s average |
4)
Message boards :
MilkyWay@home Science :
GPU v non-gpu tasks
(Message 64016)
Posted 20 Oct 2015 by blub Post: are there applications that can't run on GPU? Or can anything Milkyway needs to calculate be done on GPU? |
5)
Message boards :
Number crunching :
max. gpu temperature for permanent mw crunching
(Message 63667)
Posted 4 Jun 2015 by blub Post: try lowering the ram Voltage, as power consumption is proportional too V^2 |
6)
Message boards :
Number crunching :
Why so short Workunits?
(Message 62835)
Posted 13 Dec 2014 by blub Post: Why are the Workunits here so short? It must bee a huge overhead for the server with workunits running as short as 20 seconds. |
©2024 Astroinformatics Group