1)
Message boards :
Number crunching :
Testing ATI Application Availability
(Message 35659)
Posted 14 Jan 2010 by ableeker Post: I've been using MW 0.20 on a 4850 running XP32 with Catalyst 8.12 for quite a while now because anything else caused problems, but I've been running 0.21 on Catalyst 9.12 for the last couple of days, and it seems to be working great! |
2)
Message boards :
Number crunching :
ATI Stream SDK v2.0 Beta with OpenCL™ 1.0 Support
(Message 32386)
Posted 15 Oct 2009 by ableeker Post: I don't really know anything about this stuff, and the people who do know probably already heard about this, but AMD has announced a Stream SDK with OpenCL support. Could this be a first step towards a combined ATI/Nvidia program? Or wouldn't this affect MW at all? |
3)
Message boards :
Number crunching :
BOINC 6.10.12 fixes Ati issues? (6.10.13 has been released as well)
(Message 32072)
Posted 7 Oct 2009 by ableeker Post: According to the thread "Development BOINC 6.10.13 released" in the Number crunching forum of Seti@Home there are a number of Ati related issues fixed in BOINC 6.10.12. The unofficial change log mentions the following: - client: if downloaded project list file is garbage, ignore it. Nice, Gipsel is in there! By the way, the 6.10.13 change log has the following: - client: fix crashing bug introduced in [18605] Do the fixes in 6.10.12 mean we don't have to rename the DLL's anymore? And could this possibly mean we can now use the latest Catalyst drivers with Milkyway on XP? Uhm, actually, what does it all mean? Does anybody know if this will have any effect on the way we run the Ati GPU version of Milkyway@Home? |
4)
Message boards :
Number crunching :
ATI GPU app 0.19f fixes the ps_sgr_208_3s errors
(Message 26079)
Posted 20 Jun 2009 by ableeker Post: Sorry to bother you again, would 6.6.28 for Windows x86 be all right? Or should I go for 6.6.20? Thanks! |
5)
Message boards :
Number crunching :
ATI GPU app 0.19f fixes the ps_sgr_208_3s errors
(Message 26078)
Posted 20 Jun 2009 by ableeker Post: Thanks for the tip! I'll try 6.6.2x and see what I get. The GPU version is still incredible, going from hours to minutes per unit! Amazing job! |
6)
Message boards :
Number crunching :
ATI GPU app 0.19f fixes the ps_sgr_208_3s errors
(Message 26056)
Posted 19 Jun 2009 by ableeker Post: I've tried 9.2 first, but this one came up with a load of errors, something about search parameters not found, so I went back to 8.12, which seems to be working correctly. At the moment it seems to do 3 units in 3 minutes, so 1 unit in about 1 minute. Not the 9 seconds I have seen, but still much better than 2 hours. How fast is not easy to say, as it's running a number of WU's at the same time, and SaH as well. Can someone tell me if there's a way to see how long one unit takes? |
7)
Message boards :
Number crunching :
ATI GPU app 0.19f fixes the ps_sgr_208_3s errors
(Message 25968)
Posted 18 Jun 2009 by ableeker Post: Thanks, I'll try that. |
8)
Message boards :
Number crunching :
ATI GPU app 0.19f fixes the ps_sgr_208_3s errors
(Message 25937)
Posted 18 Jun 2009 by ableeker Post: I use a Radeon 4850 and Catalyst 9.5 on Windows XP SP3. I've copied the three files to the Milkyway folder, I've copied and renamed the CAL DDL's, and I've shut down and restarted BOINC. When I look at the messages it says near the top Found app_info.xml; using anonymous platform, and according to the Task Manager I have a dozen of so astronomy_0.19_ATI_SSE2f.exe running. However, the dozen or so tasks I have running according to BOINC Manager take about two hours to complete. I don't see tasks completed in seconds. I also get VPU recoveries. Can anybody tell me if that's normal? |
©2024 Astroinformatics Group