looks like advmethods has dried up somewhat

This forum is for discussions about the Motherboards.org Folding team. What is folding? Venture on in for a look.

Moderator: The Mod Squad

looks like advmethods has dried up somewhat

Postby len444 » Sat Oct 18, 2003 3:45 pm

recieved a tinker wu, and tried to delete, and restart folding, but kept receiving tinker wu's. then i looked at the server status page, and see no more advmethods for linux. looks like they are still available for windows, and macs though- so get them while you can. looks like i'll be dropping back a bit if i only get tinkers (sucks). might be time to try some OC, as the tinkers can endure OC'g much better. have to do what i can to get max output! :twisted:

anyway, once you get a tinker core (fahcore_65), it pretty much will stick there. i'll keep my eye on the server status page, and when adv is back for linux, i'll delete the fahcore_65.exe, work, queue, ect..., and restart. this will usually download the fahcore_78, and gromacs wu's. crap, we won't be making progress as well if advmethods dries up :(

ok 48hrs for a 71point tinker (#638), and 20hrs for a 51pt gromacs (#351).
71pt/48hrs= 1.48points/hr
51pts/20hr= 2.55points/hr

48hr(2.55points/hr)= 122points in the same time frame running a gromacs instead of a tinker! there is a majors points discrepancy between gromacs, and tinker wu's!?!

2.55-1.48= 1.07

1.07/1.48(100)= 72.29% points improvement for running a gromacs wu vs a tinker (in the same time frame)! i'm sure other comparisons will vary- but there is a definite discrepancy when it comes to points value. makes me wonder what logic is used? i figured that i'd run gromacs while i can (could have- past tense relevant to linux) as the points discrepancy was not right. yeah, i'll fold for the science of it- but please adjust the points scheme for fainess. until then, if there is an exploit that i can use- then i'll use it!
:twisted:

but now, it doesn't look too good for advmethods. keep the -advmethods flag active, and get the gromacs wu's while you can. if they totally dry up, then it'll be more of a mix of tinker, and gromacs. this will slow things down a bit for sure.

seems to be about 40,000 wu's available for windows, and macs w/ -advmethods flag, and 44,000+ wu's specificly for macs, and zero for linux. oh well

:(
len444
Black Belt 3rd Degree
Black Belt 3rd Degree
 
Posts: 3987
Joined: Thu Nov 08, 2001 1:01 am

Postby Tulatin » Sat Oct 18, 2003 5:42 pm

No More Work Units For Linux? I'm sure there'll be more out soon, otherwise, send a complaint.
Tulatin
Enlightened Master
Enlightened Master
 
Posts: 14664
Joined: Wed Dec 11, 2002 2:29 pm
Location: Canada

Postby len444 » Sat Oct 18, 2003 7:10 pm

Tulatin wrote:No More Work Units For Linux? I'm sure there'll be more out soon, otherwise, send a complaint.


no there are plenty of wu's, just that adv servers have disappeared for linux. i did post something at FAH forums regarding this, and was informed that they are switching back a server to adv today. it does appear that this will not last long. i'm running a tinker wu w/ a 9% OC, and realizing an ~7% performance gain. this will take 4 hours off from the expected 48hr duration to complete it. i did mention that a specific flag for tinkers might work for those that do serious OC'g (danny). not sure where the stability point is though. some experimentation will be needed. i'll need to buy a new heatsink, as well as an add on thermal probe. the thermistor that is included w/ the new GA-7vm400m is off by about 10degC. every friggen mb i buy has some type of defect. fic's was rma material- not sure about this one. this board has fsb OC in 1mhz increments. i can potentially take a 133mhz fsb cpu up to 165mhz. currently running it at 145mhz. i'll need to reboot to check my temps (which can't be accurate). still, i'm somewhat leary (spooked) of OC'g. i figure that a t-breb B will run sufficiently cool w/ stock hsf, and a 9% OC. i won't go higher until i get the temp probe, and hsf that can keep the cpu cool. i'll need to get some measurements first to see what will fit in the space allowed by the capacitors present. mb does have the protective film for protecting mb from hsf installation slips.
len444
Black Belt 3rd Degree
Black Belt 3rd Degree
 
Posts: 3987
Joined: Thu Nov 08, 2001 1:01 am

Postby Tulatin » Sat Oct 18, 2003 7:44 pm

Danny's OC isn't so much as extreme as it is crazy. I can go to 2.66 and fold with ease at not much of a temp difference than 2.5, so don't worry about your OCs - no need to be spook'd
Tulatin
Enlightened Master
Enlightened Master
 
Posts: 14664
Joined: Wed Dec 11, 2002 2:29 pm
Location: Canada

Postby len444 » Sat Oct 18, 2003 8:11 pm

Tulatin wrote:Danny's OC isn't so much as extreme as it is crazy. I can go to 2.66 and fold with ease at not much of a temp difference than 2.5, so don't worry about your OCs - no need to be spook'd


didn't you smoke an intel mb bios? how did that turn out (what did you do?)?
len444
Black Belt 3rd Degree
Black Belt 3rd Degree
 
Posts: 3987
Joined: Thu Nov 08, 2001 1:01 am

Postby Tulatin » Sat Oct 18, 2003 8:18 pm

Yes, i did, but that was due to PSU failure. As to what i did - i did jack $%^!@#. I replaced the board. There is no chance i could have gotten the $150 USD to rma.
Tulatin
Enlightened Master
Enlightened Master
 
Posts: 14664
Joined: Wed Dec 11, 2002 2:29 pm
Location: Canada

Postby len444 » Sat Oct 18, 2003 9:33 pm

what about replacing the bios- or was there other damage (thought it was errors in reprogramming of the bios?)?
len444
Black Belt 3rd Degree
Black Belt 3rd Degree
 
Posts: 3987
Joined: Thu Nov 08, 2001 1:01 am

Postby Tulatin » Sat Oct 18, 2003 9:57 pm

Actually, i successfully hot flashed it (tri beep, the chip was recognized and al), but STILL no post...
Tulatin
Enlightened Master
Enlightened Master
 
Posts: 14664
Joined: Wed Dec 11, 2002 2:29 pm
Location: Canada

Postby len444 » Sun Oct 19, 2003 12:34 am

upped the OC to 15% for about a 13% speed boost. seems to be somewhat linear as i OC higher. i have a feeling that the linearity will deviate when the peak is reached. about a 20% OC will bring the time duration somewhat on par w/ 2- 51pt gromacs wu's (2@ 20hr each non OC), but i would need an additional 40%+ just to equal the points value.
len444
Black Belt 3rd Degree
Black Belt 3rd Degree
 
Posts: 3987
Joined: Thu Nov 08, 2001 1:01 am

Postby Tulatin » Sun Oct 19, 2003 12:21 pm

According to your theory, a 25% OC should bring about a 20% performance increase.
Tulatin
Enlightened Master
Enlightened Master
 
Posts: 14664
Joined: Wed Dec 11, 2002 2:29 pm
Location: Canada

Next

Return to Motherboards.org Folding Team

Who is online

Users browsing this forum: No registered users and 2 guests