| 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: 68
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It is fast though! I got the 2.2GHz i7 2720qm (4c/8t) ... and even going full blast with 8 WCG tasks it is turbo-ing up to 2.7GHz sustained! (maybe because discrete graphics are on so it can applying the power savings from idle integrated graphics to the cores?). It also seems to blow out a good bit less heat than my i7 720qm...
-j |
||
|
|
duanebong
Advanced Cruncher Singapore Joined: Apr 25, 2009 Post Count: 134 Status: Offline Project Badges:
|
Have a 2600K up and crunching for a week now. If you intend to overclock, for sure the stock heatsink is not enough. I'm running at 4.5GHz and using a Scythe Mugen2 heatsink. I see temps hit 73 degrees when running 8 hyperthreads. On prime95 stress test, hits 78 degrees after 15mins.
----------------------------------------As for the SATA ports recall, I'm not sure there's anything we can do since all boards are affected. If you return a board, you'll just have a CPU sitting there idle until replacement boards come in April. The 3gbps SATA ports are supposed to fail after 2-3Y, and only with a 5-15% probability. Solution is simply to use the other ports. My Asus has extra ports from a Marvell controller - in the worst case still supports a total of 4 unaffected drives. Instead of replacing boards, a far cheaper solution would be to just give every affected user a free SATA card. ![]() [Edit 2 times, last edit by duanebong at Feb 4, 2011 1:50:54 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
A 13 Line Patch That Boosts Intel Sandy Bridge Performance
http://www.phoronix.com/scan.php?page=article...tel_snb_13lines&num=1 Think I want one :D |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Watts draw at 100% load [for] a highly overclocked 4500+MHz is 212W.. At default 3500MHz the draw at 100% load is 193W. Just to say that I'm getting quite a bit less power draw for a 2600K system: * 107 W at stock 3.5 GHz (at 100% load with eight WCGC threads) * 130 W at 4.1 GHz (eight WCG threads) * ~145 W at 4.1 GHz (eight Prime95 threads) System details here, but the primary difference to Movieman's system seems to be a more efficient power supply and a low-power video card. With 4.1 GHz and eight threads I get ~38,000 WCG PPD, or 54 GFLOPS (using the 700 WGC PPD = 1 GFLOPS conversion rate), compared to 24,000 PPD for an i7 920 @ stock. An i7 980/990 Extreme or dual Xeon is still more powerful, but I could buy two 2600K systems for the cost of one i7 Extreme or entry-level dual Xeon. [Edit 1 times, last edit by Former Member at Mar 6, 2011 12:04:00 PM] |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1326 Status: Offline Project Badges:
|
. Just to say that I'm getting quite a bit less power draw for a 2600K system: * 107 W at stock 3.5 GHz (at 100% load with eight WCGC threads) * 130 W at 4.1 GHz (eight WCG threads) * ~145 W at 4.1 GHz (eight Prime95 threads) System details here, but the primary difference to Movieman's system seems to be a more efficient power supply and a low-power video card. With 4.1 GHz and eight threads I get ~38,000 WCG PPD, or 54 GFLOPS (using the 700 WGC PPD = 1 GFLOPS conversion rate), compared to 24,000 PPD for an i7 920 @ stock. An i7 980/990 Extreme or dual Xeon is still more powerful, but I could buy two 2600K systems for the cost of one i7 Extreme or entry-level dual Xeon. I like the sound of figures above. What the avarage turn around time & points per project? How long dose it take your system to finish a CEP2 task? My 980X & stock takes between 3.5 & 7 hours have had 1 or 2 take the full 12 hours on CEP2 ![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I don't have full stats per project, just notes for the WUs returned in two 24-hour periods, both with the processor at 4.1 GHz.
The first 24h period used eight threads and a mix of all projects. (~38k WCG points, 47 WUs in 24h): Project #WUs CPU hours The second period used four threads and crunched C4CW and CEP2 exclusively (~29k WCG points, 38 WUs in 24h): Project #WUs CPU hours |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1326 Status: Offline Project Badges:
|
Can you say if it will out perform and 930 or 950? I think so and also it will use MUCH less electrical. Will it out preform a 980X? Thanks for the times. I'd be running round 12 hours a day if I build one ![]() [Edit 1 times, last edit by Speedy51 at Mar 22, 2011 5:39:06 AM] |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges:
|
No, it does not outperform a 980X, but can get to about 75%
That said, it is about 1/4 the purchase price and cheaper to run. They make for the best crunching systems right now. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Will it out preform a 980X? No. Compared to the previous generation Sandy Bridge has higher performance per clock, per core and per watt, but 980X has 50% more cores to begin with. I don't have a 980X, so I don't have accurate data, but I'd guesstimate that a stock 980X would give 25-30% more WCG PPD than a stock 2600K if both are running WCG on all cores, similar to this POV-ray benchmark . |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1326 Status: Offline Project Badges:
|
this POV-ray benchmark was an interesting read thanks
----------------------------------------![]() |
||
|
|
|