1)
Message boards :
Number crunching :
Milestones II
(Message 25964)
Posted 18 Jun 2009 by [B^S] Beremat Post: Approaching 1 million this evening. |
2)
Message boards :
Number crunching :
ps_sgr_235_3s errors remain
(Message 24627)
Posted 9 Jun 2009 by [B^S] Beremat Post: All my CPU machines and the GPU one seem to have _1s_4_ and _2s_4_ WUs only. It's weird because with every 3s WU i've seen, the application reports 0 iterations. |
3)
Message boards :
Number crunching :
ps_sgr_235_3s errors remain
(Message 24605)
Posted 8 Jun 2009 by [B^S] Beremat Post: Mine seems to have run fine: <core_client_version>6.6.20</core_client_version> <![CDATA[ <stderr_txt> Running Milkyway@home ATI GPU application version 0.19f by Gipsel CPU: Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz (2 cores/threads) 3.00007 GHz (241ms) CAL Runtime: 1.3.145 Found 1 CAL device Device 0: ATI Radeon HD 4800 (RV770) 512 MB local RAM (remote 28 MB cached + 256 MB uncached) GPU core clock: 625 MHz, memory clock: 250 MHz 800 shader units organized in 10 SIMDs with 16 VLIW units (5-issue), wavefront size 64 threads supporting double precision 0 WUs already running on GPU 0 Starting WU on GPU 0 main integral, 160 iterations predicted runtime per iteration is 119 ms (33.3333 ms are allowed), dividing each iteration in 4 parts borders of the domains at 0 400 800 1200 1600 Calculated about 2.45253e+012 floatingpoint ops on GPU, 6.34181e+007 on FPU. Approximate GPU time 20.9531 seconds. probability calculation (stars) Calculated about 8.83997e+008 floatingpoint ops on FPU. WU completed. CPU time: 1.32813 seconds, GPU time: 20.9531 seconds, wall clock time: 21.618 seconds, CPU frequency: 3.00008 GHz </stderr_txt> ]]> Task ID 75504556 Name ps_sgr_235_1s_4_42292_1244497065_0 |
4)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 24169)
Posted 4 Jun 2009 by [B^S] Beremat Post: Today I signed up to the MilkyWay_GPU Project. It would help to read the news before bashing the project. |
5)
Message boards :
Number crunching :
ps_sgr_test_* and ps_sgr_10_1s_1
(Message 24111)
Posted 4 Jun 2009 by [B^S] Beremat Post: Hello MilkyWay@Home, That's for your work with the project, it is really appreciated. |
6)
Message boards :
Number crunching :
Milestones II
(Message 23714)
Posted 30 May 2009 by [B^S] Beremat Post: 750k |
7)
Message boards :
Application Code Discussion :
CUDA Milkyway Application v0.05
(Message 23568)
Posted 28 May 2009 by [B^S] Beremat Post: There's a bunch of parameters in the test_files folder that have numbers after them that probably go with the stars.txt files. |
8)
Message boards :
Application Code Discussion :
CUDA Milkyway Application v0.05
(Message 22970)
Posted 22 May 2009 by [B^S] Beremat Post: Awesome news! Good luck with the todo list. |
9)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 19632)
Posted 20 Apr 2009 by [B^S] Beremat Post: Something to consider: a 4870 at 30s/WU and ~100W needs 3kJ to crunch a WU. My Phenom 9850 with optimized app needs about 20mins/WU. At ~100W that's 120kJ, fourty times as much energy. Do you still think it's a good idea to run Mw on CPUs? Edit: Nevermind, not getting into this |
10)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 19270)
Posted 18 Apr 2009 by [B^S] Beremat Post: The website is ridiculously fast on the MW_GPU project without any WUs flowing. Page loads are instantaneous. |
11)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 19094)
Posted 16 Apr 2009 by [B^S] Beremat Post: it says to me that acc is already there with my e-mail..... must be copying User DB over to gpu side to make it easier???? Your account page is here: http://milkyway.cs.rpi.edu/milkyway_gpu/show_user.php?userid=21 The DB isn't being copied from what I can tell, only 22 accounts so far. Edit: Beremat Action News 5 Live Update! The whole website just got its style on. Font size is now bigger and there's a background. Logging in now works, and you can edit your profile, I think. The typo was fixed, and creating a team is possible. It's all coming together, all that's left is the replacement of "INSERT PROJECT NAME HERE" and creation of Wus. Two teams are created, and as the population count grows past 25, citizens of Milkyway_GPU are worried about overpopulation. |
12)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 19088)
Posted 16 Apr 2009 by [B^S] Beremat Post: Dave, Travis, or anyone else, if you're reading this, I found the reason why the links to the user pages from the search page aren't working. There's a typo. For example, searching my name gives this result: http://milkyway.cs.rpi.edu/milkway_gpu/show_user.php?userid=13 However, if I fix the typo and make it: http://milkyway.cs.rpi.edu/milkyway_gpu/show_user.php?userid=13 The link works. |
13)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 19083)
Posted 16 Apr 2009 by [B^S] Beremat Post: I attached :P User ID 13: http://milkyway.cs.rpi.edu/milkyway_gpu/user_search.php?search_string=[B^S]&search_type=name_prefix&country=any&profile=either&team=either&action=Search The user pages are still a 404 though. |
14)
Message boards :
Number crunching :
MilkyWay_GPU
(Message 18383)
Posted 12 Apr 2009 by [B^S] Beremat Post: Thanks for doing this, Dave. I can't wait to see a dedicated project for GPUs from MW ^_^ |
15)
Message boards :
Application Code Discussion :
ATI-optimized app 0.19e crashing on XP32
(Message 16229)
Posted 20 Mar 2009 by [B^S] Beremat Post: The versions 8.12 and 9.1 are supposed to work (on some systems WinXP systems only 8.12 appears to work, don't know why) without any modification. That are the versions you may try. The reason is because with versions 9.2 and up, AMD/ATI renamed some of the .dlls required for the MW app to work. Previous versions have the old filenames, so they will work. |
16)
Message boards :
Number crunching :
Milestones
(Message 15864)
Posted 17 Mar 2009 by [B^S] Beremat Post: Approaching 500,000. |
17)
Message boards :
Number crunching :
1 Credit per Second per GPU?
(Message 15863)
Posted 17 Mar 2009 by [B^S] Beremat Post: When the GPU client first came out, I think I hit some lucky WUs. I was averaging 1.21 C/S. Now, It's around 0.96. This is on a non-overclocked 4850. |
18)
Message boards :
Number crunching :
new workunit queue size (6)
(Message 13397)
Posted 28 Feb 2009 by [B^S] Beremat Post: Should be even better now that the workunits should take around twice as long to crunch. Just because they take longer to crunch doesn't mean that they don't give more credits. :D |
19)
Message boards :
Number crunching :
No work
(Message 13363)
Posted 28 Feb 2009 by [B^S] Beremat Post: I've been getting steady work since somewhere around 9AM EST. |
20)
Message boards :
Number crunching :
Why...
(Message 12872)
Posted 25 Feb 2009 by [B^S] Beremat Post: This really gets in the way when I'm not at the computer :\ It ends up not getting work like 5 times in a row and deferring comms for an hour. |
©2024 Astroinformatics Group