PDA

View Full Version : Screen savers saving the world?



StickMan
2007-08-06, 11:16 PM
I know that in the past I've seen a number of different groups that have had a Screen saver that when ran helped to break down the human genome, cancer and some other various things that will take hundreds of computers to do. Does anyone know of any such programs?

I figure if i can have my computer looking for a cure for cancer when I'm not using it, that would be cool or some other good cause it would be well worth it.

The Great Skenardo
2007-08-06, 11:19 PM
Well, there are a lot of statistics showing that such and such many computer hours have been logged...but I wonder what most of that time is actually spent doing? The time spent is only relevant if your approach to trying to find a solution is in any way effective, and if you'd know a solution if you saw one. I'm not convinced that's the case with these.

Rawhide
2007-08-06, 11:19 PM
[email protected] (http://folding.stanford.edu/)

adanedhel9
2007-08-06, 11:44 PM
I know of two such projects off the top of my head:

[email protected] (http://setiathome.berkeley.edu/) watches the skies for extra-terrestrial radio signals.

[email protected] (http://folding.stanford.edu/) analyses protein folding, which could lead to medical breakthroughs.


Well, there are a lot of statistics showing that such and such many computer hours have been logged...but I wonder what most of that time is actually spent doing? The time spent is only relevant if your approach to trying to find a solution is in any way effective, and if you'd know a solution if you saw one. I'm not convinced that's the case with these.

I would assume that the parent organizations gain something from these projects, else they would stop supporting them. I don't think either project is necessarily trying to find the answer to its stated problem; rather, it's just trying to narrow it down.

For example, [email protected] doesn't try to find extra-terrestrial signals. It just tries to find something that might be indicative of an unnatural and uexpected radio signal. These finding are then shipped off to experts who take a closer look for themselves.

StickMan
2007-08-06, 11:47 PM
[email protected] has a team thing and like a point system if anyone else on the board downloads the program and wants to make a Giant in the Playground team That would be fun.

Rawhide
2007-08-06, 11:50 PM
[email protected] is having some results (http://folding.stanford.edu/results.html).

Rawhide
2007-08-07, 10:28 AM
A GitP team sounds like a really good idea. I've gone and created a team called "Giant in the Playground" for everyone who would like to join [email protected], Team ID: 82447 (http://fah-web.stanford.edu/cgi-bin/main.py?qtype=teampage&teamnum=82447)

Thes Hunter
2007-08-07, 10:38 AM
[email protected] is having some results (http://folding.stanford.edu/results.html).

Wow, [email protected] did p53! I didn't know that!

I will have to see if I can install it on my computer then. Then I might actually make a contribution to SCIENCE!

Rawhide
2007-08-07, 10:46 AM
Oh, one great thing about [email protected], is that it is completely non profit and all results are publicly available.

http://folding.stanford.edu/faq.html#project.own

StickMan
2007-08-07, 10:57 AM
A GitP team sounds like a really good idea. I've gone and created a team called "Giant in the Playground" for everyone who would like to join [email protected], Team ID: 82447 (http://fah-web.stanford.edu/cgi-bin/main.py?qtype=teampage&teamnum=82447)

woo who I had a good idea!!!

sikyon
2007-08-07, 11:12 AM
Well, there are a lot of statistics showing that such and such many computer hours have been logged...but I wonder what most of that time is actually spent doing? The time spent is only relevant if your approach to trying to find a solution is in any way effective, and if you'd know a solution if you saw one. I'm not convinced that's the case with these.

As far as I know, the system doesn't actually solve anything but mathematical formulas. It basically makes supercomputers via parallel computing open to researchers (normal supercomputers cost ALOT). I'm doing research right now, and on occasion I use a similar network (though it's not publically parallel). Basically, a request for a calculation is put in, the calculation is broken down into parallel parts and a PC does a part while otherwise unoccupied.

It'll probably slow your computer down abit though, especially in terms of RAM (I'm not familiar with the particular system, but if it does things like analyze protein folding then it might do it using quantum-mechanical calculations, which would take alot of ram).

Thes Hunter
2007-08-07, 11:22 AM
Well, there are a lot of statistics showing that such and such many computer hours have been logged...but I wonder what most of that time is actually spent doing? The time spent is only relevant if your approach to trying to find a solution is in any way effective, and if you'd know a solution if you saw one. I'm not convinced that's the case with these.

The end point for protein folding is finding the lowest energy configuration. This is done by calculating the known energies of bond lengths and configurations from other similar known confirmation's energy values.

It is theorized that one of the best places to begin attempting to solve a protein folding structure is to look at the lowest energy confirmation, and then adjust for hydrogen bonds, di-sulfide bridges, salt bridges and etc.

So all this said, yes, with protein folding they have a defined end point.

But then again, I fell asleep in my lecture on protein folding modeling and protein crystalization. The guy had a monotone, and I was up late the night before cramming amino acid catabolism pathways into my brain. :smallbiggrin:

Atreyu the Masked LLama
2007-08-07, 11:42 AM
[email protected] is fun. My roomie and I both have folding on our computers. He's a bit of a folding addict and even has a little computer that does nothing more than fold.

StickMan
2007-08-07, 11:42 AM
It'll probably slow your computer down abit though, especially in terms of RAM (I'm not familiar with the particular system, but if it does things like analyze protein folding then it might do it using quantum-mechanical calculations, which would take alot of ram).

Well yes but you can always turn the program off when you need the ram. So if you want to game then you should still be fine if you turn it off when gaming. Right?

adanedhel9
2007-08-07, 12:26 PM
Well yes but you can always turn the program off when you need the ram. So if you want to game then you should still be fine if you turn it off when gaming. Right?

I can't say for sure on [email protected], but typically these types of programs are set up as your screen saver. So they aren't active and aren't consuming any processor time or RAM when you are using the computer. You probably can set it up to run continuously, but that's probably not a good idea on a gaming box.

StickMan
2007-08-07, 12:39 PM
See now I can't get it to run as my screen saver that is under the menu but I can't get it to work. I have the program running right now though. Its kinda fun watching proteins fold.

magicwalker
2007-08-07, 01:49 PM
The University of Washington has a similar program, but I can't remember for the life of me what it is. They had timers that tracked how many hours you logged, and offered benefits to people in the dorms that met certain hour marks.

Pretty neat way to get a few thousand extra processors running for you.

sikyon
2007-08-07, 02:34 PM
I can't say for sure on [email protected], but typically these types of programs are set up as your screen saver. So they aren't active and aren't consuming any processor time or RAM when you are using the computer. You probably can set it up to run continuously, but that's probably not a good idea on a gaming box.

No, what I meant about taking up RAM wasn't related to that. Basically, the way it should work is that while your computer is in screensaver, your processor would indeed be working like crazy and your RAM would be used. However, as soon as your screensaver goes off, it would store the data at whatever point it was last at, and maybe send it out. The thing about RAM is, however, that old programs tend to keep their data in the RAM unless it's being replaced by something new. An example is... say... when you have a RAM intensive game going, and then you close it by accident and start it up again it'll start up much faster because alot of the data will still be there. I think.


The end point for protein folding is finding the lowest energy configuration. This is done by calculating the known energies of bond lengths and configurations from other similar known confirmation's energy values.

It is theorized that one of the best places to begin attempting to solve a protein folding structure is to look at the lowest energy confirmation, and then adjust for hydrogen bonds, di-sulfide bridges, salt bridges and etc.

So all this said, yes, with protein folding they have a defined end point.

But then again, I fell asleep in my lecture on protein folding modeling and protein crystalization. The guy had a monotone, and I was up late the night before cramming amino acid catabolism pathways into my brain.

Hehehe

No, that's the way most of it is done. I'm running lowest energy configuration simulations for silicon nanowires as I type this. I would assume because protein systems are much larger in terms of atoms and have many different types, the calcuations would be done with emperical forcefields (ie. classical ones). Those might actually not be that intensive on your computer, especially when broken down. The kind I'm doing right now involve solving huge matricies for numerical approximations to systems, and the upper limit is like 200 atoms on supercomputers anyways.

adanedhel9
2007-08-07, 05:39 PM
No, what I meant about taking up RAM wasn't related to that. Basically, the way it should work is that while your computer is in screensaver, your processor would indeed be working like crazy and your RAM would be used. However, as soon as your screensaver goes off, it would store the data at whatever point it was last at, and maybe send it out. The thing about RAM is, however, that old programs tend to keep their data in the RAM unless it's being replaced by something new. An example is... say... when you have a RAM intensive game going, and then you close it by accident and start it up again it'll start up much faster because alot of the data will still be there. I think.

Well, yes. But no.

When the [email protected] process ends, it and whatever data it used are almost guaranteed to remain in active memory for some non-trivial time. But, the OS knows those regions of memory are open for use, so when you start your memory-intensive game, the OS gives the game access to those regions. The game doesn't have any difficulty overwriting the [email protected] data, any more than it would have overwriting empty memory, or memory used by a dozen smaller, terminated processes.

Yes, there are circumstances where loading the same data/application as previously loaded could, theoretically, result in shorter load times. But there's no guarantee that it would, and even if it did, once the application was loaded, there'd be no change in performance.

Rawhide
2007-08-08, 06:54 AM
Congratulations to the folder known as "marcus" who has just become the first folder to complete a work unit for Giant in the Playground!

StickMan
2007-08-08, 11:17 AM
Ok never mind her computer just had to catch up to its self I guess its only going to take a day after all.

Rawhide
2007-08-11, 10:10 PM
I've set up a thread for the Folding at home team: http://www.giantitp.com/forums/showthread.php?t=53501

Khaldan
2007-08-11, 10:36 PM
Great, Rawhide made me go download something.. wonder if i can get it to run on a PSP.. *goes to see more*