Drystones or Whetstones - which takes care for faster crunching?

Message boards : Number crunching : Drystones or Whetstones - which takes care for faster crunching?
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile ChipCrusher

Send message
Joined: 21 Mar 05
Posts: 61
Credit: 14,499
RAC: 0
Message 6183 - Posted: 24 Jun 2007, 18:01:27 UTC

Hi,

What is more \"responsible\" for fast crunching, Dry- or Whetstones? Why is the CPU capability in Boinc reported anyway in two values?

Kind regards,
CC
<img src="http://www.boincstats.com/stats/banner.php?cpid=95ef1305806d9b867f1f28e110dbbfe5">
ID: 6183 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mikie Tim T

Send message
Joined: 21 Apr 05
Posts: 16
Credit: 57,866
RAC: 0
Message 6185 - Posted: 24 Jun 2007, 18:22:01 UTC - in response to Message 6183.  

Hi,

What is more \"responsible\" for fast crunching, Dry- or Whetstones? Why is the CPU capability in Boinc reported anyway in two values?

Kind regards,
CC


A Whetstone measures floating-point performance, which is primarily what is used by most all of the BOINC projects. The Dhrystone benchmark measures integer and string performance, which is primarily what you use out of your computer for office applications, web browsing, etc. So, for crunching, the Whetstone benchmark gives an indication of relative performance.

ID: 6185 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile ChipCrusher

Send message
Joined: 21 Mar 05
Posts: 61
Credit: 14,499
RAC: 0
Message 6186 - Posted: 24 Jun 2007, 19:05:21 UTC

Okay, sounds good!
But then my 3GHz Xeon is 20-25% slower as my 2GHz 146 Opteron... That\'s strange... :(

Thanks Mikie!

Regards,
CC
ID: 6186 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Mikie Tim T

Send message
Joined: 21 Apr 05
Posts: 16
Credit: 57,866
RAC: 0
Message 6193 - Posted: 25 Jun 2007, 4:12:07 UTC
Last modified: 25 Jun 2007, 4:14:34 UTC

You\'re welcome. Opterons used to be much stronger in floating point performance than Intel chips, but since the C2D came out, I think that it might be the other way around or at least much closer. Also, I believe that the benchmark results are per core, not for the processor overall, so that might cause part of the perceived discrepency. You might want to check to make sure that nothing else is taking up CPU on the Xeon, especially when it\'s running the benchmark.

ID: 6193 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mpan3

Send message
Joined: 24 Apr 06
Posts: 64
Credit: 29,899
RAC: 0
Message 6358 - Posted: 6 Jul 2007, 17:21:45 UTC

I believe you have one of the older generation Xeons (aka Pentium D xeon), which is considerably slower than the newer (core 2 Duo xeons) which would explain why your Opteron is beating it so badly.
Contribution to BURP total: 0.5% (manually updated)
ID: 6358 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Number crunching : Drystones or Whetstones - which takes care for faster crunching?