- Joined
- Feb 6, 2003
- Location
- up state NJ
hey gang
i finally broke down and ordered parts for a new system which includes a GTX 670 FTW.
i'm going to get back into crunching SETI for our team. i also don't want to corrupt results.
being stable in 3DMark won't guarantee SETI work stability IMO.
looking into it i found GPU-Burn (Linux) which looks promising for testing CUDA cores as well as 90% graphics RAM.
would be great if we could run this from a "live" cd/thumb drive. *nudge *nudge
found a test similar to Memtest for GPU Memory called MemtestG80. how well it is at finding CUDA core problems i dunno.
[edit] another Memtest for graphics RAM CUDA GPU memtest which runs in CUDA and OpenCL (ATI/AMD) which includes "additional stress tests". [/edit]
would there be a way to run SETI work with known results, so we could MD5 them to compare what work we've done?
while preventing the upload of results to berkeley of course. how different is one WU to the next anyway?
let me know what you folks think.
i finally broke down and ordered parts for a new system which includes a GTX 670 FTW.
i'm going to get back into crunching SETI for our team. i also don't want to corrupt results.
being stable in 3DMark won't guarantee SETI work stability IMO.
looking into it i found GPU-Burn (Linux) which looks promising for testing CUDA cores as well as 90% graphics RAM.
would be great if we could run this from a "live" cd/thumb drive. *nudge *nudge
found a test similar to Memtest for GPU Memory called MemtestG80. how well it is at finding CUDA core problems i dunno.
[edit] another Memtest for graphics RAM CUDA GPU memtest which runs in CUDA and OpenCL (ATI/AMD) which includes "additional stress tests". [/edit]
would there be a way to run SETI work with known results, so we could MD5 them to compare what work we've done?
while preventing the upload of results to berkeley of course. how different is one WU to the next anyway?
let me know what you folks think.
Last edited: