nah Craig, games (simulations specifically) are what cell excels at. The GPGPU which ATI videocards can dish out are on par with the capabilities of the RSX in the PS3. The RSX and Xenos are capable GPUs at about 1.8 Teraflops each. The Cell coming in at around 240Gigaflops is more flexible and can do more types of custom calculations than the GPUs but all that is def changing. Future chips look like they will be having the GPU and CPU on the one chip. PS3 is king... for now. I expect it to be surpassed very soon by the dominant PC platform once the next gen of GPUs (NVidia 8800 GTX anyone?) start to get more mainstream with appropriate applications developed for them.
Disclaimer: I do not work for the F@H team, so this is not official by any means - its tricky science folks!
Basically we have the proteins sequences of all the human protiens (pretty much) but we do not know what they do because it is the 3D structure that matters.
The F@H effort shows great scientific promise (and they have publications in the most prestigious journhals to prove that) because simulating the process of protein folding, that happens quandrillions of times inside our cells each second is almost an impossible problem to solve computationally. If we could do it we might be able to eventually not have to do experiments but similate the functions of each cell solely on the computer. Also determining the structure of a single protein can take months or years of work and for some of the most important proteins this is impossible (membrane proteins that are some of the most important drug targets out there). Furhermore you usually get a static snapshot and not the dynamic "live" picture of what the protein does.
Alzheimer's disease is important because it literally is a disease of protein folding. A protein which I think is called alpha-amylase has a low probability of assuming a misfolded state that causes is to stick to other similar proteis. This forms like a crystallization nucleus that causes more and more proteins to stick together and eventually form such a big clump that it kills surrounding brain cells and so your get a brain that is full of holes. You spend often years as a vegetable until you eventually die.
F@H looks at the dynamics of alzheimer protein folding trying to eventually devise a drug that would prevent the formation of the aberrantly folded alpha-amylase proteins (aberrant conformations). This would prevent the formation of plaques in indicviduals that were predetermined to be at risk either based on family history or genetic testing. It might even dissolve already formed plaques, although this woukd be a lot more difficult. Patient might eventually even regain the use of some brain functions they have lost (which is common in people with brain damage).
Do F@H is high quality basic science (perhaps highest quality of all the public computing projects). And it might eventually save lives.
Well, Vijay is probably right when it comes to the CPU's - however, it wouldn't be the CPU that would be used for folding on an Xbox360, it would be the GPU. The GPU is an enhanced R580 - which is what folding@home uses for its GPU client, in beta, and gives more FLOPS than the PS3. Here's a link to their FLOPS page so you can see what does what: http://fah-web.stanford.edu/cgi-bin/...?qtype=osstats and then here's a link to the GPU FAQ page which would apply to the Xbox360: http://folding.stanford.edu/FAQ-ATI.html
The reason a client for the 360 hasn't happened at the moment is that Sony provided the push and the resources to create the PS3 client and Microsoft has shown no interest in doing so for the Xbox360. The university that runs all this stuff doesn't have the resources to do it on the 360, itself, at the moment - it would have to be deeply integrated with MS's online environment and would need their help.
"If the PS3 had any GAMES you wouldn't have to claim a folding program as the systems killer app! LOL!!!!"
*****************************************
LOL!!
I tried playing MotorStorm while sleeping in bed but found not being able to see the screen from under the covers proved difficult. I then tried playing Resistance from work, but my blue-tooth controller lost it's signal about halfway to the office.
Unfortunately, it seems that playing my PS3 24hrs a day 7 days a week is just not possible. Well, either I can just turn the darn thing off when I am not around it or I guess I could just leave it on and run Folding@Home and contribute my super computer ... er ... Playstation 3 to a good cause.
some people are just plain ignorant and are a waste of valuable skin. PS3 owners, don't listen. Keep helping the good cause.
"Obviously the guy is a PS3 fanboy, let me get this straight, triple core 3.2 Ghz processor (which is more raw power then the typical PC) and the fact that there are more 360s in use then PS3s ."
3.2 x 3.0 = 9.6 GHz
1PPU + 6SPUs = 3.0 + (6.0 x 3.2) = 22.2GHz.
Where do you get 1PPU = 6. GHz? So your telling me the 2.8 GHz cell is actually a 6Ghz processor and IBM didn't make a big deal about breaking 4GHz? Your on crack. Stop being stupid.
and the folding@home GPU trial is specifically for ATI cards at the moment but they have used Nvidia cards in the past. The new batch concentrate on stream processors. The Cell is a general purpose fancy stream processor itself, as is the 8800GTX. Vijay already stated that the Xenon/Xenos combination is not as useful in this case. No harm no foul. Still, the April 27 release of more optimised Cell code has tripled the output of the F@H client on the PS3. thats amazing!!! :O
As a scientist, I think some of the earlier comments about people not caring about this project are quite short-sighted. Games have, for the better part of the last decade helped pushed the boundaries of what hardware can do to such a point that we now have 2 incredibly powerful consoles with enough computer power to help solve problems (how proteins fold) that have plagued biochemists since the 70s. Scientists have been trying to calculate how proteins fold with very little success due to the sheer complexity of the problem. I think pekkoh covered most of the potential benefits of having such knowledge. Its not just about diseases directly related to misfolding however (e.g. alzheimers), but could also be of great use in drug design as a number of drugs in development at present are protein based and some are being used in bioprocesses to develop drugs. If we knew more about the folding of these molecules it would be easier to make these proteins more efficient and potentially reduce the concentration of drugs needed meaning more people could be treated with the same volume of material.
I would like to see an X-box 360 version because at present it has a larger user base than the PS3. That said, for the particular calculations needed for protein folding the Cell proccesor is more efficient then the 360 CPU. We've already seen the Cell processor more then triple the projects outputs (I wish I could have a Cell CPU instead of my AMD 3800 X2 running my PC :P). More then anything however, I'd like to see them using the GPU's of both consoles, although they're developing a program to make use of ATI based chips which would likely have a greater effect then solely programming for CPU's.
As a gamer, we obviously haven't seen the power of the PS3 utilised appropriately for games. I think we may have to wait and see MGS4 and FF13 (PS3 exclusives?) before we start to see its full gaming potential. As for me, I'm gonna stick with my PC for the time being as I just bought a GeForce 8800GTS :P
I hardly read as many idiotic and childish comments to any topic than in this thread.
The title is : Xbox 360 not as useful as PS3 for Folding@home
People a flaming each other and bashing at one console or the other in a way that has not even the slightest relevance to the topic!
The best imho was the comment from a user claiming Vijay Pande doesn't know what he's talking about.
After almost falling off my chair due to the 'laughing flash' I suffered, I wondered if that comment reflects the IQ of some users and came to the conclusion that it obviously does.
VP is the head at Stanford and he and his crew surely know best WHAT a system is capable of and WHAT not. Do you think they work(ed) with names like ATI, Nvidia, Microsoft (just to name a few) cause they don't know 'sh*t from shine-all' ?
To users not seing the potential and benefit of helping a cause like curing diseases: Go back to your console and splash some blood but leave US who DO CARE alone with your unqualified and abstruse comments.
Thank you!
Carsten
You know Vijay Pande has to be a sony lover. The latest sales releases on the 360 and the PS3 show that there are 3x as many 360's as there are PS3 that have been sold. The Xbox 360 sales= 10.5 mill. The PS3 sales= 3.5. Also seeing that there have been around 6 mill plus PS3 units shipped, one can't use the excuse of the 360 having an early start. I'm pretty sure that combined efforts from Xbox 360 users could triple the output of those PS3 suppliers. Luck on the bright side, all you unfortunate PS3 buyers that have no games to play have not wasted your 600 dollars. Your systems are being used for a greater cause.
You don't seem to get the point or are not willing to do so!
It's not about HOW MANY consoles are on the market but WHAT you can do with it and HOW MUCH work it would cost. There are folding clients for numerous platforms (Windows, Mac, Linux...) so VP can't be branded 'any brand's lover'.
Haven't you thought about the possibility that Microsoft is NOT eagerly keen to develop the program for their X-Box and the WHY SO's ?
Could it be that THEY (MS) realized that their console is NOT CAPABLE enough to stand up to a direct comparism to the PS3 when it comes to folding ?
Microsoft has been involved in the development of the clients for almost all the pc's from the start.
Now if THEY saw any chance to get a marketing freebie as Sony did, they would have jumped on board asap.
As they didn't, what does that imply in your opinion ?
A fact is that the X-Box has a serious temperature problem which would exponentially increase when using CPU heavy applications as folding (the scientific core maxes out the cpu) and I can see the THOUSANDS of users sueing MS for their melted boxes. To avoid that MS would have to reduce the workload on the cpu and thus it would become inefficient.
MS has nothing to gain but all to loose in this case, so.....
You people want to inform yourself BEFORE stating crap in the public and thus making a fool out of yourselves.
» Go to 's original post
» Go to 's original post
» Go to 's original post
Basically we have the proteins sequences of all the human protiens (pretty much) but we do not know what they do because it is the 3D structure that matters.
The F@H effort shows great scientific promise (and they have publications in the most prestigious journhals to prove that) because simulating the process of protein folding, that happens quandrillions of times inside our cells each second is almost an impossible problem to solve computationally. If we could do it we might be able to eventually not have to do experiments but similate the functions of each cell solely on the computer. Also determining the structure of a single protein can take months or years of work and for some of the most important proteins this is impossible (membrane proteins that are some of the most important drug targets out there). Furhermore you usually get a static snapshot and not the dynamic "live" picture of what the protein does.
Alzheimer's disease is important because it literally is a disease of protein folding. A protein which I think is called alpha-amylase has a low probability of assuming a misfolded state that causes is to stick to other similar proteis. This forms like a crystallization nucleus that causes more and more proteins to stick together and eventually form such a big clump that it kills surrounding brain cells and so your get a brain that is full of holes. You spend often years as a vegetable until you eventually die.
F@H looks at the dynamics of alzheimer protein folding trying to eventually devise a drug that would prevent the formation of the aberrantly folded alpha-amylase proteins (aberrant conformations). This would prevent the formation of plaques in indicviduals that were predetermined to be at risk either based on family history or genetic testing. It might even dissolve already formed plaques, although this woukd be a lot more difficult. Patient might eventually even regain the use of some brain functions they have lost (which is common in people with brain damage).
Do F@H is high quality basic science (perhaps highest quality of all the public computing projects). And it might eventually save lives.
» Go to 's original post
The reason a client for the 360 hasn't happened at the moment is that Sony provided the push and the resources to create the PS3 client and Microsoft has shown no interest in doing so for the Xbox360. The university that runs all this stuff doesn't have the resources to do it on the 360, itself, at the moment - it would have to be deeply integrated with MS's online environment and would need their help.
» Go to 's original post
*****************************************
LOL!!
I tried playing MotorStorm while sleeping in bed but found not being able to see the screen from under the covers proved difficult. I then tried playing Resistance from work, but my blue-tooth controller lost it's signal about halfway to the office.
Unfortunately, it seems that playing my PS3 24hrs a day 7 days a week is just not possible. Well, either I can just turn the darn thing off when I am not around it or I guess I could just leave it on and run Folding@Home and contribute my super computer ... er ... Playstation 3 to a good cause.
some people are just plain ignorant and are a waste of valuable skin. PS3 owners, don't listen. Keep helping the good cause.
» Go to 's original post
3.2 x 3.0 = 9.6 GHz
1PPU + 6SPUs = 3.0 + (6.0 x 3.2) = 22.2GHz.
So why is he a fanboy?
» Go to 's original post
» Go to 's original post
and the folding@home GPU trial is specifically for ATI cards at the moment but they have used Nvidia cards in the past. The new batch concentrate on stream processors. The Cell is a general purpose fancy stream processor itself, as is the 8800GTX. Vijay already stated that the Xenon/Xenos combination is not as useful in this case. No harm no foul. Still, the April 27 release of more optimised Cell code has tripled the output of the F@H client on the PS3. thats amazing!!! :O
» Go to 's original post
» Go to 's original post
» Go to 's original post
the 360 is more suited to running game code than the ps3. ps3 is more suited to running apps like this.
just depends on which type of app you prefer your £400 quid box to run!
» Go to 's original post
I would like to see an X-box 360 version because at present it has a larger user base than the PS3. That said, for the particular calculations needed for protein folding the Cell proccesor is more efficient then the 360 CPU. We've already seen the Cell processor more then triple the projects outputs (I wish I could have a Cell CPU instead of my AMD 3800 X2 running my PC :P). More then anything however, I'd like to see them using the GPU's of both consoles, although they're developing a program to make use of ATI based chips which would likely have a greater effect then solely programming for CPU's.
As a gamer, we obviously haven't seen the power of the PS3 utilised appropriately for games. I think we may have to wait and see MGS4 and FF13 (PS3 exclusives?) before we start to see its full gaming potential. As for me, I'm gonna stick with my PC for the time being as I just bought a GeForce 8800GTS :P
» Go to 's original post
The title is : Xbox 360 not as useful as PS3 for Folding@home
People a flaming each other and bashing at one console or the other in a way that has not even the slightest relevance to the topic!
The best imho was the comment from a user claiming Vijay Pande doesn't know what he's talking about.
After almost falling off my chair due to the 'laughing flash' I suffered, I wondered if that comment reflects the IQ of some users and came to the conclusion that it obviously does.
VP is the head at Stanford and he and his crew surely know best WHAT a system is capable of and WHAT not. Do you think they work(ed) with names like ATI, Nvidia, Microsoft (just to name a few) cause they don't know 'sh*t from shine-all' ?
To users not seing the potential and benefit of helping a cause like curing diseases: Go back to your console and splash some blood but leave US who DO CARE alone with your unqualified and abstruse comments.
Thank you!
Carsten
» Go to 's original post
» Go to 's original post
It's not about HOW MANY consoles are on the market but WHAT you can do with it and HOW MUCH work it would cost. There are folding clients for numerous platforms (Windows, Mac, Linux...) so VP can't be branded 'any brand's lover'.
Haven't you thought about the possibility that Microsoft is NOT eagerly keen to develop the program for their X-Box and the WHY SO's ?
Could it be that THEY (MS) realized that their console is NOT CAPABLE enough to stand up to a direct comparism to the PS3 when it comes to folding ?
Microsoft has been involved in the development of the clients for almost all the pc's from the start.
Now if THEY saw any chance to get a marketing freebie as Sony did, they would have jumped on board asap.
As they didn't, what does that imply in your opinion ?
A fact is that the X-Box has a serious temperature problem which would exponentially increase when using CPU heavy applications as folding (the scientific core maxes out the cpu) and I can see the THOUSANDS of users sueing MS for their melted boxes. To avoid that MS would have to reduce the workload on the cpu and thus it would become inefficient.
MS has nothing to gain but all to loose in this case, so.....
You people want to inform yourself BEFORE stating crap in the public and thus making a fool out of yourselves.
CarstenLast edited on Sat 26 May 2007 by ceemosp
Post Comment
Login or register to reply to this topic
Create a new account or login to take part in this topic discussion.