• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

F@H Cruncher Question

T

trevinom

Golden Member
I have a PIII 800E Chip. The motherboard it's on recognizes it as a 1Gig and runs windows XP fine. My question is, is it worth running F@H on it? I have several AMD 950's and they don't put out that much...since they can't run the BIG WU (600's). Have any of you done comparison's of the output of a 1 Gig PII and the 950's with regards to output and power consumption?

Thanks in advance
 
I slipped it on my dads PIII950. He leaves his comp on all the time, so I figured I might as well put it to use. Not great output, but every little bit helps.
 
I ran F@H on a PIII 600 for a while and it was ok what it did. Just make sure, that you reduce its energy consumption as much as possible, like having no screen on it or even turning it into a diskless cruncher, w/o harddrive, CD and floppy.
 
I ran F@H on an AMD K6-2 450 for a while and it made the deadlines (barely), the PIII should do fine.
 
Originally posted by: MDE
I ran F@H on an AMD K6-2 450 for a while and it made the deadlines (barely), the PIII should do fine.
Click to expand...


But AFAIK they redid the whole deadline thing a few months ago with a new reference system and thus that K6-2 450 wouldn't make any deadline anymore right now I guess unless those special nodeadline WUs.

:beer:
 
You won't set the world on fire with it, but if you have fun doing it.... running the timeless work units would be worthwhile.

Turn off the monitor and your power consumption would be about the same as leaving the porch light on.

-Sid
 
Originally posted by: Insidious
You won't set the world on fire with it, but if you have fun doing it.... running the timeless work units would be worthwhile.

Turn off the monitor and your power consumption would be about the same as leaving the porch light on.

-Sid
Click to expand...

LOL

With my 6 rigs upstairs, and 8 rigs downstairs, I already have a ton of 'lights' on. Just wondering if it is even worth it.



 
Hey trevinom, where ya been?

I run a 1 Ghz PIII laptop with 256MB ram.

Running BigWUs it gets over 110 points per day! 🙂
 
I have some Pentium Pro systems still crunching. Lots of PIIIs in the 500MHz range. Even a couple Celerons.
 
Originally posted by: lobadobadingdong
my 933 was putting out a double point WU about every 2.5 days, not bad for an older rig.

it and the mobo are for sale too btw.
Click to expand...

Thank you all for your input. I just found a AMD 1700+ CPU and a motherboard to put it into, so I've set that up and should keep my interests up for a few days. At that point, I'll go back to seeing if I can set up that PIII. You all are a great bunch of guys.


Martin
 
Not to offend anyone, and this isn't the "official" line, but it's my opinion that Vijay Pande(F@H head honcho, or at least he seems to be) would prefer that each individual project(protein) produces meaningful results as quickly as possible. You may be doing the project a favor if you put your slower(relative term, I know) machines on projects that are less linear.

Fewer work units, but faster equals faster viable scientific conclusions IMHO.

I'm not trying to incite anyone, I just believe reading between the lines at the F@H website indicates that this assessment is true.
 
Originally posted by: jasong42
Not to offend anyone, and this isn't the "official" line, but it's my opinion that Vijay Pande(F@H head honcho, or at least he seems to be) would prefer that each individual project(protein) produces meaningful results as quickly as possible. You may be doing the project a favor if you put your slower(relative term, I know) machines on projects that are less linear.

Fewer work units, but faster equals faster viable scientific conclusions IMHO.

I'm not trying to incite anyone, I just believe reading between the lines at the F@H website indicates that this assessment is true.
Click to expand...
Then why have they made available the option for deadline-less WUs?
 
Originally posted by: jasong42
Not to offend anyone, and this isn't the "official" line, but it's my opinion that Vijay Pande(F@H head honcho, or at least he seems to be) would prefer that each individual project(protein) produces meaningful results as quickly as possible. You may be doing the project a favor if you put your slower(relative term, I know) machines on projects that are less linear.

Fewer work units, but faster equals faster viable scientific conclusions IMHO.

I'm not trying to incite anyone, I just believe reading between the lines at the F@H website indicates that this assessment is true.
Click to expand...


Welcome to Anandtech jasong42

I really don't understand how you came to believe Vijay would prefer computers be removed from F@H and put on other projects??

While I respect your opinion, I really don't agree with it.

Actually quite the opposite. It seems to me that F@H has done a very nice job of making options available to take full advantage of any resource a donator would like to apply.

-Sid
 
Welcome to Anandtech jasong42

I really don't understand how you came to believe Vijay would prefer computers be removed from F@H and put on other projects??

While I respect your opinion, I really don't agree with it.

Actually quite the opposite. It seems to me that F@H has done a very nice job of making options available to take full advantage of any resource a donator would like to apply.

-Sid
Click to expand...
I haven't been on those forums in a while, so they may have changed the software since then. The timeless units are definitely an option. I never actually downloaded any, for the short time I was there I only ran about 4-5 units total. To be perfectly honest, my opinion is based on what I considered to be the best arguments at the time(having to do with applying "fuzzy" math to the problems of dealing with linear projects).

Basically, you have a linear problem that requires an unknown number of steps to be solved. If there are 100 computers of various strengths running the projects, you will probably get an end time which is neither really excellent, nor really horrible, let's say 2 years. But what if you take the most important problems and assign them to the top 20 computers? In that case you would get fewer results, but since they came faster, new conclusions(and new experiments) could come faster, maybe meaning more knowledge and more innovation. An end in one year could mean new conclusions and new experiments sooner.

Of course, in that case, it's possible that the experiments "left behind" are actually the important ones, in which case time was wasted.

So my argument could cause harm if it's believed. Basically, it's an unknown where anybody with a brain could kick it wide open or take it down a blind alley, depending on blind luck.

 
Oh please, it wasn't that hard. Hehee. I understand what you're saying, but with the number of workunits being produced and the number of computers working on the project (as well as the relative long-term scope of each project?scientists are not, say, waiting on the results for a vaccine to be shipped tomorrow,) I think your theory would only really be a problem if there were a very limited quantity of computers.

But it wouldn't be distributed computing then, would it?
 
Originally posted by: Haberdasher
Oh please, it wasn't that hard. Hehee. I understand what you're saying, but with the number of workunits being produced and the number of computers working on the project (as well as the relative long-term scope of each project?scientists are not, say, waiting on the results for a vaccine to be shipped tomorrow,) I think your theory would only really be a problem if there were a very limited quantity of computers.

But it wouldn't be distributed computing then, would it?
Click to expand...

Not just "a very limited quantity of computers", but a very limited number of projects, too. But FAH has quite a few protein projects.

If all the FAH contributors computers were 10Ghz or 20Ghz or 100Ghz computers that would be best for science. Or if everybody ran their computers 24/7 that would also help the science too. But this is DC, with teams, which they encourage. They have to take what they get - Beggars can't be choosers! If a TeAmmate wants to run a 300Mhz computer for his TeAm, he can!

I bought a dual HT computer because it would give me and the TeAm the most points for my price. I do care about the science and hope for early results, but I am going to use HT for the TeAm (more so now for the TeAm, a while ago it was more for myself.) I will also continue to crunch with my 1Ghz laptop. 🙂
 
I run 2 333mhz P2 MMX systems and they usually get 400 point work units and they do complete them, just a little slower. They are stripped down boxes with a floppy drive and hard drive with 128mb ram.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top