Folding@home?

[Impaler]

Does it suck? It seems more "serious" then Boinc..What's the difference?
Pardon my incompetance..:book1:

[NeoGen]

I'll be glad to answer that question for you Impaler... just pull a chair and make yourself confortable... this is gonna be long. :p

Folding@Home is an older project that runs on its own platform, built from the ground up way before boinc was even thought.
It is a very advanced protein folding study project, backed by the Stanford University, that has already published numerous papers to the scientific community and helped advance the study of proteins and their relation to cancers and diseases in many ways.

One can say that it is indeed one of the most "serious" distributed computing projects out there.

Why has it not got "boinc'ed" like many have been lately? I do not know this but I have a feeling that their platform and mechanism are not very compatible with boinc. Don't forget that it is a platform built way before boinc existed.

Nevertheless, there was an attempt to create a boinc client for Folding@Home a year ago or so, but they failed and never tried again.

Most people who are into Distributed Computing do it more by the competition, by the stats, by the fellowship and the people...etc. But there are people who strongly believe that they should only crunch for projects that benefit mankind, and they choose the medical and biology projects, of which in my oppinion the best ones are really Folding@Home (non-boinc) and Rosetta@Home (for the Boinc devotees)

But enough with the speech or else I won't stop writing tonight. LOL

What drives you to crunch for DC projects Impaler? The stats, the fellowship, the wish to do something better for mankind,... or all of them together? :icon_smile:

[drezha]

No it doesn't suck!

In fact In my opinion, it's better than ANY BOINC project out there. ;) But then I'm biased because every system I can get hold of runs it ;)

Folding is studying protiens. Proteins have shapes caused by the stuff they are made out of. Of these there are 4 main ways that the it's put together. Primary and secondary structures are to do with the order amino acids are put together, but the tertiary and quarternary structures are due to folding. When a protien folds wrong, it screws up and doesn't work properly. They're studying the effects of this etc. Various diseases are thought to be caused by mis folding such as alzehiemer's etc.

Project wise, it doesn't use the BOINC platform. It uses it's own client software. This means it can't be run with other projects. (BOINC can let you help various projects from one PC splitting the time each has to run, Folding only folds) so it's folding or the highway ;)
AMD's usually outstrip much of what Intel processor's can do and perform better than average per Ghz than the benchmark machine. (All WU's are run on a P4 3Ghz with no SSE on and given 110 points per day, thus a 220 point unit took the benchmark machine 48 hours to complete) The credit system, IMHO is a lot lot fairer than BOINC's.
Unit's can take between 12 hours and I currently have one on my linux box thats takes about a week (though the long units are optional)
The project has many work servers so if one is down, the project still contiues. Most, if not all BOINC projects have one server. (ie if that dies, the project is down and you waste time waiting for new work) Folding will have little time when it's not working (ie when it uploads data and downloads data)

Unfortunatly, it can not cache work units as the scientists want the data back ASAP to work on it. (one project complete leads onto the next one etc) so getting work back as fast as possible is best.

If you want any more questions, feel free.
I think I'm fairly clued up on Folding as I'm pretty much the folding guy on this forum ;)

Hope that answers some questions, if not arising more and being rather long :p


EDIT:Beaten by Neogen :lol:

Anyhow...Folding still plan to bring a BOINC version out in the future but they are working on 3 new clients at the minute. Thats Version 6 of the current one, a GPU client (now in beta) and a client for the PS3. Boinc is on the to do list but on the back burner a bit. They're also having a problem with the programming team.

[NeoGen]

I think I'm fairly clued up on Folding as I'm pretty much the folding guy on this forum ;)
I testify on that, you are indeed the folding guy on these parts!

We should be ashamed... we had to hire a fellow cruncher from Team Phoenix Rising to come here and help us with folding@home. LOL :lol:

EDIT: I'm just kidding about the hiring part.... he works for free. :lol: :P

[gatekeeper53]

Can you enlighten me on weather or not Tanpaku is going for the same thing but they are trying to get it to work on Boinc and no science is being done at this point. But what they are doing is getting them closer to having a workable Boinc platform to do folding research. Is this close? I've been to Tokyo U. and that is no clown college.

[Brucifer]

The University of Tokyo is a (if not the) premier university in Japan. VERY hard to get in to, extremely competitive.

[drezha]

I couldn't tell you to be honest. I've no idea what Tanpaku is doing.

From a brief read it seems possible it may be doing similiar work, but there's no mentioning of it focussing on disease. It looks like it's just folding to fiund the structures. I'm not sure if F@H use Brownian Dynamics method either.


Ah but Neogen, you cant run every project. ;) And besides, I crunch maths projects for AMD every now and then and you guys help me get that going...No one at TPR is intrested in the non BOINC or folding projects really...

[Nflight]

From what I have read they are similar in focus of medical projects. Tanpaku and Folding@home both work on similar angles to the same equation. Each one is different. The complexity of the focus means that each one is really resolving issues what the original design of the project was meant for.

Stanford project is more in depth with these recent updates; http://folding.stanford.edu/results.html
"Folding@Home - Our goal: to understand protein folding, misfolding, and related diseases"

The Tanpaku project on the other hand has this; http://issofty17.is.noda.tus.ac.jp/doc/first_E.html
"TANPAKU" is the project that is aiming at attacking "protein structure prediction"

Since my focus is doing investigative research, I seem to be able to resolve almost anything sent my direction quickly and effectively.

[Impaler]

Thanks for all the answers :-)

[Strongbow]

Completely off topic (sorry) but just following on from Impaler's quote.

My favorite quote around the same subject has to be from Bob Monkhouse...

" I want to die like my father, peacefully in his sleep, not screaming and terrified, like his passengers."

Now back on the thread again...

I like what they are doing in Folding@home, they seem to be pushing the boundry a bit with their approach and it's looking very impressive. However, as a cruncher then BOINC is far more interesting and popular so Tanpaku is the one for me for the time being!