New to folding+Been silencing for years=Now folding for SPCR
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
New to folding+Been silencing for years=Now folding for SPCR
I'm new I'm new! Look at me!
Threw my 'puters into the ring about 10 days ago for SPCR, so far:
(2) PIII-866 folding 24/7
(2) PIII-1000 folding 24/7
(1) P4-2.4 folding 24/7
(2) P4-2.6 folding 24/7
(1) XP1700 folding whenever I remember to
I may be was down the list but at least I'm YELLOW... (come on GREEN)
Pretty happy to have climbed from 160 to 75 in that time, and Mr. Smartepants, I'm coming for you baby
cos
...I think my folding farm sucks tho... I mean every morning I fold my towel after drying off and how many molecules is that?
Threw my 'puters into the ring about 10 days ago for SPCR, so far:
(2) PIII-866 folding 24/7
(2) PIII-1000 folding 24/7
(1) P4-2.4 folding 24/7
(2) P4-2.6 folding 24/7
(1) XP1700 folding whenever I remember to
I may be was down the list but at least I'm YELLOW... (come on GREEN)
Pretty happy to have climbed from 160 to 75 in that time, and Mr. Smartepants, I'm coming for you baby
cos
...I think my folding farm sucks tho... I mean every morning I fold my towel after drying off and how many molecules is that?
http://folding.stanford.edu
It's a distributed computation thingy like SeTI, does protein folding to solve medical problems, or something.
I'm new to it too - just started a few days ago and whoa are some of those workunites big - took nearly three days. In sETI I used to go through a work unit in about six hours.
More here
It's a distributed computation thingy like SeTI, does protein folding to solve medical problems, or something.
I'm new to it too - just started a few days ago and whoa are some of those workunites big - took nearly three days. In sETI I used to go through a work unit in about six hours.
More here
Last edited by DonP on Sat Aug 30, 2003 2:28 pm, edited 1 time in total.
Actually, it was from project 638.
So let me get this straight..
The workunits can come from any number of projects, and then they are assigned to clients at random, right?
The workunits require different amounts of CPU work, and therefore there is this "points" score assigned, which is a function of the complexity/time required?
Interesting.. can't wait to get to my dual CPU boxes at work and boot them into Linux
So let me get this straight..
The workunits can come from any number of projects, and then they are assigned to clients at random, right?
The workunits require different amounts of CPU work, and therefore there is this "points" score assigned, which is a function of the complexity/time required?
Interesting.. can't wait to get to my dual CPU boxes at work and boot them into Linux
The points for each unit, (and the time limit for each as well) are derived by Stanford from a benchmark run on one of their machines.
As far as I know it's still a 500Mhz celeron.
Part of the reason for the oddity of point values is due to the difference in processors. The SSE2, 3Dnow, and ASM features of more modern CPU's effect the processing times in ways that aren't necessarily linear with the increase in Mhz's.
On some units having ASM on will help alot, so it will look like you're getting more points than you should. On others it doesn't help much at all, so you feel like you're being robbed of points, in comparison to the optimize units.
I hope that clears it up, at least alittle.
As far as I know it's still a 500Mhz celeron.
Part of the reason for the oddity of point values is due to the difference in processors. The SSE2, 3Dnow, and ASM features of more modern CPU's effect the processing times in ways that aren't necessarily linear with the increase in Mhz's.
On some units having ASM on will help alot, so it will look like you're getting more points than you should. On others it doesn't help much at all, so you feel like you're being robbed of points, in comparison to the optimize units.
I hope that clears it up, at least alittle.