The Mother Board

Motherboards.org forums. Free tech support, motherboard ID, and more.
It is currently Mon Oct 15, 2018 3:14 am

All times are UTC - 8 hours




Post new topic Reply to topic  [ 13 posts ]  Go to page Previous  1, 2
Author Message
 Post subject:
PostPosted: Sun Oct 19, 2003 1:48 pm 
Offline
Black Belt 3rd Degree
Black Belt 3rd Degree

Joined: Thu Nov 08, 2001 1:01 am
Posts: 3987
Tulatin wrote:
According to your theory, a 25% OC should bring about a 20% performance increase.


according to my math, and a linear assumption- 20% OC will bring the 48hr 71pt tinker down to 40hrs. at 15%OC at present, the time is about 42-43 hrs to complete it. i'm using 40 hrs from a ball park average time to complete 2- 51pt gromacs w/ o an OC (@~20hr ea.). then the 2 gromacs are worth 102pts, and the tinker is worth 71pts, so it's obvious what puts more points up in the same time frame. we still contribute to the science of folding. i'll delete the fahcore_65.exe, work, config, queue.dat fahlog, ect, when this stinker is finished, and uploaded to stanford. then i'll redo the config w/ -config -advmethods -forceasm -verbosity 9 to get some adv wu's while there are still some available.

run -advmethods -forceasm while you can get a higher concetration of adv/ gromacs wu's, as this supply will more than likely dry up.
    tinker frame rate w/ o OC: 7.3min/ frame(400 frames)(1hr/60min)= 48.67hrs
    10%OC: 6.7min/ frame(400frames)(1hr/60min)= 44.67hrs
    15%OC: 6.35min/ frame(400 frames)(1hr/60min)= 42.33hrs


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Mon Oct 20, 2003 1:14 am 
Offline
Mobo-fu Master
Mobo-fu Master

Joined: Mon Nov 22, 1999 1:01 am
Posts: 10244
Location: Mackay, Queensland, Australia
In my experiments with my Barton 3200+ I have seen the time taken to generate one point vary between 12 minutes and 22 minutes depending on the work unit with the average being 17 minutes per point (85 points per day).

It is hard to comprehend why there is such a large variation in the points scoring rate between work units on the same machine when the same baseline machine is supposedly used to establish each work units points value.

Obviously Stanford can't test every work unit on their baseline machine so I guess most are just estimated from past experience with similar work units or something. If their points allocation methods were accurate the points generation rate of any given machine should be very similar regardless of the work unit surely?

There are certainly more tinkers about now than there has been in the past that's for sure.


Aussie

_________________
Our lives begin to end the day we become silent about things that matter - Martin Luther King


Top
 Profile  
Reply with quote  
 Post subject:
PostPosted: Tue Oct 21, 2003 12:01 am 
Offline
Black Belt 3rd Degree
Black Belt 3rd Degree

Joined: Thu Nov 08, 2001 1:01 am
Posts: 3987
finished the stinker, and deleted all fah files except for the fah linux executable. redid the config w/ -forceasm -advmethods -verbosity 9 flags, and fetched a gromacs wu. get them while i can, i will.

currently overclocked 20%. seems stable so far.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 13 posts ]  Go to page Previous  1, 2

All times are UTC - 8 hours


Who is online

Users browsing this forum: trendictionbot [Bot] and 5 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group