• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

PS3 tripling t-flops for Folding!

Nice! I've seen similar comments in some other forums. It seems like the PS3 has prompted people to fold with their PCs. :cool:

Welcome to the [H]orde!



 
if my math is right, the GPU client has a higher performance that the PS3 client, however there are fewer GPU clients running :(
 
if my math is right, the GPU client has a higher performance that the PS3 client, however there are fewer GPU clients running :(
haha, wow.. I was just thinking that and was going to post it.
If there were as many GPU clients as PS3 clients this thing would be kickin it hard. It's almost up to a PFLOP anyway. :eek:

I'm just wondering with this huge increase if they were ready to even be able to process all the information that is coming back. How in the world do you take a machine running at 800+ TFLOPS and use the data it's spitting out to get something coherent and useful?
 
The GPU client is more than twice as fast as the PS3 in terms of FLOPS.. And that ratio will increase as more powerful GPUs are released - think R600 - (PS3 will always have the same performance per machine). Stanford has indicated that soon enough there will be a big push for the GPU client (more diverse/advanced WUs, more PPD).. They're also backporting the PS3 cores to run on the GPU sinse the architecture of the GPU and the Cell is similar. Exciting times for this project.
 
In case you all have missed my other thread about this.. here's a graph I've been working on sinse november (manually recording flops levels daily.. - the dates shown at the bottom are only in-between dates for reference, the data is actually from a pretty-much daily record.):

tflops.png
 
Crazy stuff. 27,559 active ps3, 30621 total ps3 and 675 tflops in 3 days. There's only 160,266 active cpus and this project has been around for 7 yrs. Awesome for the project.



 
Obviously Sony has done a really cool thing here. With ATi's GPU's doing great in F@H, too, though (comparatively), why isn't MS doing the same thing for the 360?

If Sony & MS worked together they could literally cure cancer! lol..........

I'm not the only one wondering:

http://gizmodo.com/gadgets/home-entertainment/microsoft-bring-foldinghome-to-xbox-will-ya-246748.php

MS & Sony both said they were going to do just this back in 2005. So now Sony has done it a few months after the PS3 launch, what happened to MS doing it on the 360? Was there an unseen technical issue?

http://www.wired.com/news/technology/0,67835-0.html
 
Crazy stuff. 27,559 active ps3, 30621 total ps3 and 675 tflops in 3 days. There's only 160,266 active cpus and this project has been around for 7 yrs. Awesome for the project.

Yeah this is outstanding! I mean who would have thought that hundreds of thousands of PC's would be put to shame in less than a week from a gaming console. It's rather neat to see such a huge company like Sony to put time and effort into developing this co-operation. I've got a new view of Sony now:)



EDIT: You're going down SpoogeMonkey!! And you'll have been beaten by a couple of Mac's:D :p
 
This folding PR campaign is working ;)


Yeah you're totally right. It gives a 'human' side to the 'face' of Sony. I live in Toronto and it, the PS3 & Folding, have been on the news for the past few days. Even my parents paid attention and asked me if we should/could get a PS3 for movies and folding! I'm a quadripalegic so I can't very well play games YET (I'm doing lots of physio and have high hopes). I fold for the obvious reasons of so many people affected by alzheimers, cancer etc but I also do it HOPING that some of this data can help researchers with a "cure" for Spinal Cord Injuries and make wheelchairs an archaic thing of the past!
 
Yeah you're totally right. It gives a 'human' side to the 'face' of Sony. I live in Toronto and it, the PS3 & Folding, have been on the news for the past few days. Even my parents paid attention and asked me if we should/could get a PS3 for movies and folding! I'm a quadripalegic so I can't very well play games YET (I'm doing lots of physio and have high hopes). I fold for the obvious reasons of so many people affected by alzheimers, cancer etc but I also do it HOPING that some of this data can help researchers with a "cure" for Spinal Cord Injuries and make wheelchairs an archaic thing of the past!


Cheers man! I agree.. this research, although targetted at these specific diseases, will also forever move our knowledge of cellular/subcellular biology forward! It's the idea of moving the human race's understanding of the universe forward another step that drives me to fold, for all of the diseases being researched and for the greater understanding that will surely come from it aswell. Fold on! (PS3 aside,if your PC isnt folding yet, it should be!)
 
I was on the fence about buying a PS3, but as soon as I saw this, I went out and got one.

I see the program is right on the verge of going PetaFLOP! Woohoo!
 
EDIT: You're going down SpoogeMonkey!! And you'll have been beaten by a couple of Mac's:D :p

Heh, maybe. I'm pretty much a solo operation anymore since I got divorced. Just an opty165 (24/7) and an old mobile barton 2500+ that's only on 6-8 hrs/day. But I like a challenge.

clean_me said:
I'm a quadripalegic so I can't very well play games YET (I'm doing lots of physio and have high hopes). I fold for the obvious reasons of so many people affected by alzheimers, cancer etc but I also do it HOPING that some of this data can help researchers with a "cure" for Spinal Cord Injuries and make wheelchairs an archaic thing of the past!

I just saw a thing on G4 the other day about spinal cord injury. Apparently they've done some great research on using nasal stem cells. They have every kind of cell curing ability, and they exist right in your own nose! They just grab some tissue from your honker and drop it in the injured part of the spine. Currently (at least as of 1 yr ago) only 1 doctor was doing the procedure in Lisbon, but they were very optimistic that others would be able to perform the operation soon.

Parja said:
I was on the fence about buying a PS3, but as soon as I saw this, I went out and got one.

Good way to support the casue and get some more ppd. Plus you might actually play a game or two and watch a blu-ray movie :)
 
I just saw a thing on G4 the other day about spinal cord injury. Apparently they've done some great research on using nasal stem cells. They have every kind of cell curing ability, and they exist right in your own nose! They just grab some tissue from your honker and drop it in the injured part of the spine. Currently (at least as of 1 yr ago) only 1 doctor was doing the procedure in Lisbon, but they were very optimistic that others would be able to perform the operation soon.

Yeah it's good they're getting a little more public awareness out. To put when's being done succinctly here it goes:

a) the olfactory stem cells in your nose are good but not great, they're 'old' stem cells and not that great. But they(researchers/dr's) are still tinkering with this procedure in parts of china and lisbon etc

b) they're also tinkering with stem cells that are found in your "belly button"

c) the BEST stem cells are taken from the umbilical cord after a baby is born, but now I heard they can somehow "grow" these umbilical like cells in a lab

d) researchers in calgary just (month ago?) discovered the gene/protein/whatever it's called that temporarily prevents pregnant women from getting MS. I guess that's nature for you! What this gene/protein does is prevent (or in some cases regrow) the Myelin around the Spinal Cord. Think of Myelin as the rubber or plastic insulator around a cable, but with the spinal Cord as soon as the Myelin wears off in a part of the Spinal Cord the 'transmission' of signals gets stopped or severely confused/blocked. Soooo, maybe this gene can help both MS and Spinal Cord Injuries.

e) Ultimately a 'cure' will be a combination of a lot of different techniques/research.

So hopefully Folding can somehow pay off too!
 
The default PS3 group is bringing in about 1.1 million points a day :eek:

They will pass FLECOM within 3 weeks of starting....sucks for FLECOM.
 
How times have changed. You'll have more points in 6 months than I earned in the first 2 years. I've been folding for 4 yrs now on basically 2 boxes......glad I finally got a dual core so I can step it up a bit :D
 
Anyone else notice that PS3 just cracked the top 10 folders...over 18000 PS3s are currently turning in work under the default name.

Stanfords top 100
 
If PS3 was a team in itself, we'd have to be worried. That's a whole lot of power.

 
I haven't posted in this sub-forum for a couple of years, so bear with me. What I would like to know is how does the PS3 compare with typical dual-core offerings from AMD/Intel in overall PPD? Why is the PS3 making such an enormously impressive debut in F@H2? Is it primarily because of a large pre-established base, superior client, disproportionate point allotment, or is it simply due to brute processing power?
 
I haven't posted in this sub-forum for a couple of years, so bear with me. What I would like to know is how does the PS3 compare with typical dual-core offerings from AMD/Intel in overall PPD? Why is the PS3 making such an enormously impressive debut in F@H2? Is it primarily because of a large pre-established base, superior client, disproportionate point allotment, or is it simply due to brute processing power?

When left alone to fold, my C2D@3.2 running the Linux SMP client puts out 2k PPD. If the numbers being thrown about of 900 PPD for the PS3 are true, then not even two PS3s can match my one system. For now this only holds up due to the bonus points awarded to the SMP client. I'm sure my PPD will drop when the bonus is cut down or removed. Even then, I don't think I'll do worse than break even with a PS3. By the time that happens, I should have quad core, though.

 
I haven't posted in this sub-forum for a couple of years, so bear with me. What I would like to know is how does the PS3 compare with typical dual-core offerings from AMD/Intel in overall PPD? Why is the PS3 making such an enormously impressive debut in F@H2? Is it primarily because of a large pre-established base, superior client, disproportionate point allotment, or is it simply due to brute processing power?

I'll take a shot here.. Smokerngs is high again above, in more than one way... 2K PPD of a C2D rig? Yikes!!! :)

C2D rigs on the SMP client should do 1K-1.5k PPD or there abouts on the high side. PS3 is confirmed @ ~900 PPD.

PPD is not really a fair judgement of anything short of a e-penis anymore.

Yes to all the other questions since I'm already bored with typing...

Yes, 18K clients in a week equals a pre-established base of... well 18K of something doing something.

superior client- no, that's got nothing to do with anything really. More on that in a few.

disproportionate point allotment- Well, it all is... ask a AMD guy and it's wrong, ask a Intel guy and it's all wrong (a drunk one maybe). More on that in a few.

simply due to brute processing power- some, the thing do add 1+1 really fast.

The deal, or the latest deal nearest I can tell w/o being high to match Stanfords logic (ohh smile here) is the points are not only based on the amount of work streaming off the clients, but the type (read flexibility) of work the client can run.

CPU= most flexible, slowest 1+1 engine
GPU= least flexible, fastest 1+1 engine
PS3= in the middle for flexibility, fast 1+1 engine

1+1 engine = how much the sucker can add stuff up... :)

I can't explain what all the difference's in "can only run certain types of calculations" but it is what it is. This is where the SMP client comes in, faster and yet still as flexible original CPU client is.

Ok, my version of things anyways--- subject to reality.
 
I'll take a shot here.. Smokerngs is high again above, in more than one way... 2K PPD of a C2D rig? Yikes!!! :)

Around 16-18 hours from start to finish on a 1760 point Linux SMP work unit at 3.2 Ghz. That's what it does when the system does nothing but fold.

Around 8 hours to fold one of the 587 point Linux SMP work units.

That's what it takes for my rig. Remember, it's at 3.2Ghz from a stock 2.13. Just think of what I might be able to do once I get rid of the stock heatsink.

 
Around 16-18 hours from start to finish on a 1760 point Linux SMP work unit at 3.2 Ghz. That's what it does when the system does nothing but fold.

Around 8 hours to fold one of the 587 point Linux SMP work units.

That's what it takes for my rig. Remember, it's at 3.2Ghz from a stock 2.13. Just think of what I might be able to do once I get rid of the stock heatsink.


That really makes me want to get a E6400, but I'm holding out for the cheap quad core.

ND40oz said:
Anyone else notice that PS3 just cracked the top 10 folders...over 18000 PS3s are currently turning in work under the default name.

Stanfords top 100

We need to do some recruiting.
 
That really makes me want to get a E6400, but I'm holding out for the cheap quad core.

I snagged my E6400 at the end of December. I was sick of waiting and waiting to upgrade (due to money problems I've been waiting for about three years) and said to hell with it.

If the Q6600 drops down to $266 in the third quarter, I plan to find the money to buy one to replace my E6400. Hopefully I will have the money for some other parts so I have both CPUs running. Fun times ahead.

 
That's what it takes for my rig. Remember, it's at 3.2Ghz from a stock 2.13. Just think of what I might be able to do once I get rid of the stock heatsink.
I could imagine what a quad would do...

As of now, I only have a dual Opteron@2.36GHz. Nothing to boast about compared to a PS3 in a cost per production comparison.
 
Wouldn't it be better to just get a high end GPU in one of your boxen instead of a PS3?
 
Wouldn't it be better to just get a high end GPU in one of your boxen instead of a PS3?

But, the GPU won't play blu-ray movies or let you play motorstorm, etc...(Do not turn this into a PS3 is crap thread. We are talking about folding. I just want to say it has alot of other uses too.)
 
But, the GPU won't play blu-ray movies or let you play motorstorm, etc...(Do not turn this into a PS3 is crap thread. We are talking about folding. I just want to say it has alot of other uses too.)
Ok... but if you ALREADY had a PS3 and ALREADY had the best GPU in your main system, what would be better to get. Another PS3 or another GPU?
 
Ok... but if you ALREADY had a PS3 and ALREADY had the best GPU in your main system, what would be better to get. Another PS3 or another GPU?

Depends on if you have another TV...muahhahahaha ;)


Seriously with prices and everything else no object., a GPU if you just wanted to do folding alone.
 
OK, then dollar-for-production, GPUs, especially high-end ones, come on top. That's good to know because I wouldn't purchase a game console just to fold. They take up additional space and I don't know how they will hold up running 24/7. These issues would be a primary concern for me.
 
Found an interesting blog post here: http://www.codinghorror.com/blog/archives/000823.html

This is what raised my eyebrow to the whole PS3 folding thing.... I caught a mainstream news thing yesterday evening about it, and they said it was 6 times more powerful then an average PC. I didn't believe it at first. But this is a great thing, I even want to go out and purchase a PS3 now. I could actually churn out some good work :)

But, is it true that Intel/x86 CPU's can handle WU's the "cell" proc can't?

 
But, is it true that Intel/x86 CPU's can handle WU's the "cell" proc can't?
Yes, which is one of the reasons that the PC is less effective than the PS3 at folding: it's not as specialized which means it takes longer to do a work unit, but can handle more varied types.

Ars had a really good article about it they posted yesterday. It's a good read and explains the differences in the processors and why the PS3 is better than the PC for this specific application.
 
According to the above link: "...there's almost no point in adding general purpose CPUs to the folding network any more. It's a waste of time, effort, and electricity. For folding and other distributed computing efforts, it's the death of the general purpose CPU as we know it."

Strong statement. In effect, what they're saying is not to bother adding more CPUs for F@H because they're so much less efficient, and not worth the cost of purchase or running expense compared to the alternatives. The alternatives, though, are not that appealing from several standpoints, AFAIC. Is there even a Nvidia client or only an ATI GPU client?

In any case, I wonder what the AMD Barcelona will do for CPUs because of its supposed much improved FPU. I realize it will still pale compared to specialized processors, but I'm curious nevertheless.
 
General Purpose CPUs are still very much needed... In fact, without them, F@H would be very limited in the types of science they could do... Vector processors like the Cell and the GPU are very good at "implicit solvation" simulations, in which water is handled mathematically in a continuum fashion.. However, they are very week (and in the case of the GPU, virtually incapable) of handling "explicit solvent" simulations, where water atoms are handled atom by atom, in an explicit fashion, just like any other atom - for this type of simulation, the CPU is needed. Why do we use both types of simulations? Because implicit solvation only works in certain models, and explicit solvation only works well in certain other models.
 
General Purpose CPUs are still very much needed... In fact, without them, F@H would be very limited in the types of science they could do... Vector processors like the Cell and the GPU are very good at "implicit solvation" simulations, in which water is handled mathematically in a continuum fashion.. However, they are very week (and in the case of the GPU, virtually incapable) of handling "explicit solvent" simulations, where water atoms are handled atom by atom, in an explicit fashion, just like any other atom - for this type of simulation, the CPU is needed. Why do we use both types of simulations? Because implicit solvation only works in certain models, and explicit solvation only works well in certain other models.

And on a more simpler note:

General purpose CPUs will continue to get faster and more powerful on a regular basis. Yes, GPUs do the same thing, but not as many people utilize it. The PS3 on the other hand will stay static on what it puts out. I doubt there will be much in the way of enhanced performance with it throughout it's life concerning folding.

While the Cell is currently more powerful than general purpose CPUs right now in some things, that will not last forever. Five years is a long time to be stuck with the same performance. Just think of the CPUs that will be out during the five year lifespan of something like the PS3. I know my C2D is a hell of a lot faster than the Palamino core AthlonXP I was using five years ago. I was running a single core 1.7 Ghz (approximate) more inefficient CPU compared to two more efficient cores running at 3.2 Ghz I have now.

The quoted blog post seems to be assuming general purpose CPUs will not gain performance or functionality increases ever again. It also doesn't take into account new folding cores and ways of performing the required calculations that mass produced specialized processors won't be able to perform. In the not too distant future, modern CPUs of the time will make the Cell and other current specialized processors look like an obsolete 386 in comparison.

 
Back
Top