| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: GO Fight Against Malaria
Thread: Results on AMD-FX Inconclusive? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 62
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
"Let the speculation begin..."
Thought you'd never ask. Speculation#1: The stock Bulldozer 8150 can't beat a stock Intel i7-2600K crunching any current WCG projects WUs in time-to-completion. Speculation#2: The stock Bulldozer 8150 can't beat a 4.0GHz overclocked Phenom-II 1100T crunching any current WCG projects WUs in time-to-completion. I would have asked for FLOPS, benchmarks, and the likes but WCG WUs time-to-complete is the benchmark for me (not that WCG gives me more points for doing WUs faster than my wingWU, and that's a different story too). ; |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Thank you for your input, but I was more interested in which processes each science uses and which ones are likely to see a bottleneck at the FPU.
----------------------------------------We all agree that they're not as fast as an overclocked 1100T. I have one of those running at stock speed and the 1100T is slightly faster on HCC and C4CW but it can't match the times in HCMD2 (at least from the limited data I have so far). My goal here is to use the FX on projects where it performs the best.  Distributed computing volunteer since September 27, 2000 |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Back to the focus of this thread, a second wingperson has reported in and it's again inconclusive:
----------------------------------------Project Name: GO Fight Against Malaria Created: 01/04/2012 17:14:23 Name: GFAM_x1j3iWt_NDP_2WATs_0002541_0445 Minimum Quorum: 2 Replication: 3 Result Name App Version Number Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit GFAM_ x1j3iWt_ NDP_ 2WATs_ 0002541_ 0445_ 2-- - In Progress 1/6/12 22:02:08 1/10/12 22:02:08 0.00 0.0 / 0.0 GFAM_ x1j3iWt_ NDP_ 2WATs_ 0002541_ 0445_ 1-- 611 Inconclusive 1/6/12 06:32:07 1/6/12 13:43:36 5.56 156.5 / 0.0 GFAM_ x1j3iWt_ NDP_ 2WATs_ 0002541_ 0445_ 0-- 611 Inconclusive 1/6/12 06:31:54 1/6/12 21:58:57 4.67 124.3 / 0.0 Guess whatever Kevin did with the HR didn't address this issue. (And yes, these were downloaded hours after the fix was announced.) Distributed computing volunteer since September 27, 2000 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
He'll probably come back with a "typo" issue :) Frankly, my head is still spinning at finding that the FX is not producing matching results to other AMD makes, but being fine with Intel.
On my wish list of none-privacy invasive enhancements to the Result Log is the printing of the actual OS/SP and the CPU. More DIY self support of members making out bugs AND this being another member diagnosed finding, lots more hard supported were this playing up again. --//-- |
||
|
|
davyboy
Cruncher Joined: Apr 28, 2007 Post Count: 9 Status: Offline Project Badges: 
|
Still seeing the same issue with GFAM and DSFL on AMD 6274 (Interlagos).
----------------------------------------[Edit 1 times, last edit by davyboy at Jan 7, 2012 4:39:54 PM] |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
The repair units are back and all have gone invalid for the FX8150.
----------------------------------------Distributed computing volunteer since September 27, 2000 |
||
|
|
PMH_UK
Veteran Cruncher UK Joined: Apr 26, 2007 Post Count: 766 Status: Offline Project Badges: 
|
My AMD's work unit went invalid, as all did before:
----------------------------------------Project Name: GO Fight Against Malaria Created: 01/05/2012 05:18:53 Name: GFAM_x1j3iWt_NDP_2WATs_0002567_0206 Minimum Quorum: 2 Replication: 3 Result Name App Version Number Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit GFAM_ x1j3iWt_ NDP_ 2WATs_ 0002567_ 0206_ 2-- 611 Valid 08/01/12 02:39:01 08/01/12 22:30:46 10.95 105.4 / 106.0 GFAM_ x1j3iWt_ NDP_ 2WATs_ 0002567_ 0206_ 0-- 611 Valid 06/01/12 22:33:30 08/01/12 02:34:20 7.89 106.6 / 106.0 GFAM_ x1j3iWt_ NDP_ 2WATs_ 0002567_ 0206_ 1-- 611 Invalid 06/01/12 22:32:44 07/01/12 23:31:57 13.82 121.8 / 53.0 <--- Mine Paul.
Paul.
|
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
With regards to AMD_FX processors, apparently only on Windows machines are they matching with Intel processors. On Linux they are matching other AMD processors. We have updated the hr assignment code to reflect that. This should resolve KWSN - A Shrubbery's invalids.
|
||
|
|
Chris Holvenstot
Cruncher USA Joined: Aug 26, 2011 Post Count: 19 Status: Offline Project Badges: 
|
@knreed: First, thank you for the information. But I am a little confused. One of the benefits of getting old I guess.
----------------------------------------I am running Linux on an FX8120 and a bunch of Phenom II systems. The Phenom II systems validate OK - the FX8120 does not. So it would appear that there is more going on here than the FX being misidentified as an Intel processor when running Windows. [Edit 1 times, last edit by Chris Holvenstot at Jan 10, 2012 4:11:03 AM] |
||
|
|
kateiacy
Veteran Cruncher USA Joined: Jan 23, 2010 Post Count: 1027 Status: Offline Project Badges: 
|
With regards to AMD_FX processors, apparently only on Windows machines are they matching with Intel processors. On Linux they are matching other AMD processors. Wow. Do you have any idea what mechanism would cause that?  |
||
|
|
|