| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 35
|
|
| Author |
|
|
Kjetil-Hestvik
Advanced Cruncher Norway Joined: Oct 1, 2009 Post Count: 116 Status: Offline Project Badges:
|
Here is a couple of HCC results on my 8120, stock:
----------------------------------------X0000130230994201110251010_ 0-- AMD X8 Valid 25.11.11 23:36:42 26.11.11 10:56:22 2.02 27.5 / 25.2 X0000130230590201110251015_ 1-- AMD X8 Valid 25.11.11 23:29:13 26.11.11 08:46:07 2.00 27.3 / 24.7 X0000128831113201108251513_ 1-- AMD X8 Valid 25.11.11 23:10:53 26.11.11 08:46:07 2.04 27.8 / 24.7 OC from 3.1 - 3.5 Ghz X0000129830016201109220946_ 0-- AMD X8 Pending Validation 26.11.11 17:21:26 27.11.11 04:07:26 1.54 23.7 / 0.0 X0000130261246201110131059_ 1-- AMD X8 Pending Validation 26.11.11 19:19:40 27.11.11 04:25:06 1.46 22.4 / 0.0
Crunching in memory of the love of my life
----------------------------------------![]() [Edit 1 times, last edit by Kjetil-Hestvik at Nov 27, 2011 9:20:50 AM] |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges:
|
An average HCC task on my Quad core AMD Opteron 1352 takes 2.06h
This suggests that an Opteron at 2.1GHz (DDR2 and old hardware) is near enough as fast per core as an 8120 at 3.1GHz! For reference, an i7-2600 takes about 1h10min per task (8 at a time), and i7-920 @2.8GHz takes about 1.66h. |
||
|
|
kateiacy
Veteran Cruncher USA Joined: Jan 23, 2010 Post Count: 1027 Status: Offline Project Badges:
|
Here is a couple of HCC results on my 8120, stock: X0000130230994201110251010_ 0-- AMD X8 Valid 25.11.11 23:36:42 26.11.11 10:56:22 2.02 27.5 / 25.2 X0000130230590201110251015_ 1-- AMD X8 Valid 25.11.11 23:29:13 26.11.11 08:46:07 2.00 27.3 / 24.7 X0000128831113201108251513_ 1-- AMD X8 Valid 25.11.11 23:10:53 26.11.11 08:46:07 2.04 27.8 / 24.7 OC from 3.1 - 3.5 Ghz X0000129830016201109220946_ 0-- AMD X8 Pending Validation 26.11.11 17:21:26 27.11.11 04:07:26 1.54 23.7 / 0.0 X0000130261246201110131059_ 1-- AMD X8 Pending Validation 26.11.11 19:19:40 27.11.11 04:25:06 1.46 22.4 / 0.0 I've never gotten into overclocking, so sorry if this is a dumb question. Is it usual for an overclock of this size (3.1 to 3.5 GHz is just under 13% increase) to produce such a big reduction in run time (over 25%)? ![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
kateiacy, I'd not conclude anything from different results, even close by. Seen HCC from one to the next add 15-20 CPU minutes to the runtime. (and someone today talking about eating heart out), my last series of HCC runs under W7-64 on Q6600 stock ran faster and on Linux-64 they still do 1:15-1:25.
----------------------------------------Happy crunching. --//-- edit: Sorry, AMD does not have that feature. [Edit 1 times, last edit by Former Member at Nov 27, 2011 2:11:41 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Is it usual for an overclock of this size (3.1 to 3.5 GHz is just under 13% increase) to produce such a big reduction in run time (over 25%)? No, you would normally see a linear improvement at best. I.e. no more than X% faster for an X% increase in clock speed, with diminishing returns the higher you go.I wonder what the picture would be using a comparable Intel chip with all others held the same. Don't know about comparable, but here are some numbers for i7 2600 @ stock running eight threads with Windows 7 and Ubuntu 11.10:POEM barrelmutations average time: Win 7 6,446 sec (1.79 hours)Windows takes 44% longer. (Vs. 68% longer on kateiacy's E-350) HCC average time: Win 7 1.21 hours (5 tasks, 1.16 - 1.26 hours)Windows takes ~7% longer. (Or thereabouts, there is enough variation from task to task that I would need a larger sample to say for sure.) So benchmarks aside, there are real differences as well, but the Windows/Linux performance difference seems to vary with the project. |
||
|
|
|