Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 4
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 3209 times and has 3 replies Next Thread
duanebong
Advanced Cruncher
Singapore
Joined: Apr 25, 2009
Post Count: 134
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
i5 2500K or i7 2600K?

I'm building a new PC for WCG, and need advise choosing between the Sandybridge i5 2500K and i7 2600K. The wild card seems to be hyperthreading (4C/4T vs 4C/8T).

I've searched for performance differences of having HT on/off. However, what I've found gives no clear picture whether HT makes a difference. Some claim huge improvements for WCG and folding:
http://www.realredraider.com/vbulletin/showthread.php?t=28679
http://www.legitreviews.com/article/1501/19/

However, other sources claim more heat and performance decreases:
http://ixbtlabs.com/articles3/cpu/ci7-turbo-ht-p1.html

What have you experienced? Does HT make a difference for WCG?
----------------------------------------

[Jan 4, 2011 8:17:44 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: i5 2500K or i7 2600K?

Hello duanebong,
Our experience indicates up to 25% greater throughput using HT. Of course, this was measured on the i7, which has the largest caches in Intel's i3/i5/i7 lineup. Performance is best using a variety of projects, which reduces contention for the same resources. Clean Energy Phase 2 has some potential I/O problems which could slow things down dramatically. Either avoid it or carefully limit it. If your memory or cache is limited, look over Sekerob's estimates of memory used per project and choose projects with a small footprint if you want maximum throughput.

Lawrence
[Jan 4, 2011 2:14:55 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: i5 2500K or i7 2600K?

What have you experienced? Does HT make a difference for WCG?

Yes, just don't expect 2x the performance/results/points by using 8 HT logical cores (i7) vs. 4 physical cores (i5). Of course, you will get 2x the "CPU Time" (good for badges).

Take a look at this article about Sandy Bride Efficiency: http://www.tomshardware.com/reviews/sandy-bridge-efficienct-32-nm,2831.html

Power consumption on both CPUs is very similar, but the i7 will do more work.
[Jan 4, 2011 7:23:57 PM]   Link   Report threatening or abusive post: please login first  Go to top 
duanebong
Advanced Cruncher
Singapore
Joined: Apr 25, 2009
Post Count: 134
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: i5 2500K or i7 2600K?

Thanks very much for the quick responses. A 25% improvement for WGC sounds great, and the badge hunting is a bonus.

From some benchmarks floating on the web of other applications (nobody seems to publish WCG benchmarks, though occassionally I can find some for folding@home), I was fearing it would just generate more heat with zero or even negative performance gains. I suppose it's very application dependant.

Does memory speed become a bottleneck when running 8 threads? On my Q6600 quad, scaling was poor when running the 3rd and 4th work units. Does it help to get DDR2000 or faster memory?
----------------------------------------

[Jan 5, 2011 9:21:30 AM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread