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: 95
Posts: 95   Pages: 10   [ Previous Page | 1 2 3 4 5 6 7 8 9 10 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 145075 times and has 94 replies Next Thread
Sekerob
Ace Cruncher
Joined: Jul 24, 2005
Post Count: 20043
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

1 Dual core Linux64 @ 2.26 GHz average 1.18 hours !!??


Looks like that matches what Jean reported, the extra kick on Linux (64) being the removal of the high PF incurrance.
----------------------------------------
WCG Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All!
[Mar 24, 2010 9:39:02 AM]   Link   Report threatening or abusive post: please login first  Go to top 
rilian
Veteran Cruncher
Ukraine - we rule!
Joined: Jun 17, 2007
Post Count: 1452
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

Finished 2 WUs on MacOS 10.6.2 (64b) @ 2.1Ghz in 1.29 and 1.34 hours

Amazing speed up!!111
----------------------------------------
----------------------------------------
[Edit 1 times, last edit by rilian at Mar 24, 2010 9:58:56 AM]
[Mar 24, 2010 9:58:07 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: HCC1 BETA Comment thread

Even at the slow end of town, these are still very quick:

1GHz PIII/Celeron: avg 4.1 hrs
2GHz P4: avg 2.8 hrs
[Mar 24, 2010 12:37:39 PM]   Link   Report threatening or abusive post: please login first  Go to top 
sk..
Master Cruncher
http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif
Joined: Mar 22, 2007
Post Count: 2324
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

I expect the calculation refinements depend on CPU architecture and CPU speed, unless they have found a RAM bottleneck, so expect a spread of different task completion times for different CPUs, and different improvement factors from the present tasks.

Some CPUs are very different to others not just in the speed, but in overall core design (number of transistors and technology). This is why most CPUs are slightly better at crunching some tasks than other tasks; they exploit different mathematical pathways of the CPU.

Below, the core speeds do not in themselves account for all of the 45min extra time on the lesser CPU. The better design of the Q6600 accounts for about 14min improvement (about 33% of the speed). There are of course much better CPU designs than the 3year old Q6600, as well as supporting architectures wink

Two Valid, one in Progress:
BETA_ X0000090850189200708162327_ 0-- Valid 23/03/10 15:11:52 24/03/10 01:05:31 3.20 41.3 / 51.8 (Native Pent Dual Core, 2GHz, Win XP x86)
BETA_ X0000076080540200610061618_ 1-- Valid 24/03/10 01:33:25 24/03/10 10:13:43 2.35 35.8 / 40.2 (Native Q6600, 2.4GHz, Windows x86)

Well done, applause
----------------------------------------
[Edit 2 times, last edit by skgiven at Mar 24, 2010 1:30:21 PM]
[Mar 24, 2010 12:42:31 PM]   Link   Report threatening or abusive post: please login first  Go to top 
TimAndHedy
Senior Cruncher
Joined: Jan 27, 2009
Post Count: 267
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

Are we going to see more variability with the new algorithm?

I just noticed these two

BETA_ X0000076081347200610061606_ 2-- htpc2 Valid 3/24/10 03:51:21 3/24/10 08:39:57 1.16 30.6 / 19.6

BETA_ X0000090851129200708162315_ 2-- htpc2 Valid 3/23/10 17:56:25 3/24/10 00:03:24 1.65 44.0 / 26.0

There is a significant difference in time. The older version only had minor differences.
[Mar 24, 2010 12:43:59 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: HCC1 BETA Comment thread

just completed 3 and are valid on my quad 9550 w7 64b
1.82.1.85 1.86
[Mar 24, 2010 1:26:46 PM]   Link   Report threatening or abusive post: please login first  Go to top 
uplinger
Former World Community Grid Tech
Joined: May 23, 2005
Post Count: 3952
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

TimAndHedy,

These two work units are from completely different batches, so the variances between the images is what causes the runtime differences.

-Uplinger
[Mar 24, 2010 2:53:46 PM]   Link   Report threatening or abusive post: please login first  Go to top 
mikaok
Senior Cruncher
Finland
Joined: Aug 8, 2006
Post Count: 489
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

Atom N270 HT enabled

BETA_ X0000090990560200708021525_ 1-- laNina Valid 23/03/10 21:32:25 24/03/10 15:05:32 16.06 71.0 / 41.6
----------------------------------------
to infinity and beyond

[Mar 24, 2010 3:10:00 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: HCC1 BETA Comment thread

In case, the new application goes in production, is there a dry out necessary? Or will there just be a switch of the application?
[Mar 24, 2010 3:15:37 PM]   Link   Report threatening or abusive post: please login first  Go to top 
uplinger
Former World Community Grid Tech
Joined: May 23, 2005
Post Count: 3952
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC1 BETA Comment thread

Just a simple switch if it goes live.

We are sending some more results to the researchers now for them to inspect.

Also, the average runtime of the beta was 2 hours, the average on the old binary was 5.5 hours. So far good indications :)

-Uplinger
[Mar 24, 2010 3:34:50 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 95   Pages: 10   [ Previous Page | 1 2 3 4 5 6 7 8 9 10 | Next Page ]
[ Jump to Last Post ]
Post new Thread