fireemblem555

get a sense of humor and learn to not read into shit so much. the only IDIOT, is you. thanks for trying to imply that im an idiot, because you have no sense of humor. i was making a joke about PS3's, not about the program. but thanks for having no comprehension skills.

and i guess having a picture of a pixelated mario makes you a fanboy? i dont recall saying "OMG NINTENDO TOWTALLY RAWKZ, PS3 SUCKZORS". please be a little less moronic and think before you speak.

AND ftr, i dont play consoles, so you trying to imply im a fanboy of anything is absurd.

please pull the stick out of your ass and lighten up. thanks.

love,
idiotism

idiotism

Amazing!

*cough cough* looks like sony just got my idea for burn-in. . .

threepointone

And somehow I get this feeling Sony's contributing to most of this effort. Heck, I wouldn't be surprised if they were "burning in" PS3s at the factory with F@H!

threepointone

Why? Consider the fact that the PC base on F@H encompasses pretty much any CPU capable of completing a work unit in time. I run 10 to 12 machines 24/7 for F@H and two of them are 200 MHz Pentiums. The rest are made up of AMD and Intel processors (Durons, Semprons, Celerons, P4s) at anything from 900 MHz to 2.8 GHz.

Comparing nearly 14,000 PS3s all running the same Cell processor against all other types of CPUs running F@H gives an unbalanced picture.

What if all PC machines were exclusively made up of a particular high-performance CPU or dual/quad CPU architecture?

I'm not into math (maybe my naive statements here demonstrate that fact all too clearly), but I think the results depicting PCs as being less powerfull overall (which is patently true next to the PS3) is comparing apples with oranges in this particular case.

Bottom line, I think any machine capable of contributing to F@H is good, so don't take my comments here as being negative. I'm just puzzled by how this has turned into a (kind of), "My machine can lick your machine" type of thing.

If I had a Cray, it would be contributing as well but it shouldn't be trivialized by pointing out how it does compared to some generalized concept of what an "average PC" might be. If I had a PS3, it would be contributing also and it would be just as important as any other machine on the project no matter the power.

It's distributed computing and the F@H project makes all contributing CPUs part of one, unified machine dedicated to protein folding. Let' not forget that in all of this.

uraphake

Part of the reason the Intel-based macs are so much better on average is probably because an Intel Mac is, at most, 2 years old, and based on the Core architecture, which blows the old Pentium 4's away. PC's include pretty much anything, ranging from an old Pentium 2 that folds all day because it has nothing else to do, to the state of the art quad core server. Most PCs are older, because not everyone can afford to buy a new computer all the time.

It'd be interesting to see how the G5 Macs and earlier look without the Intel Macs averaged in. I have a feeling they'd be comparable to PC. I think this is a fair comparison, because there was a surge in Mac purchases when Intel came out, but there wasn't one on PC's with the Core processors (Or at least not as big of one).

TVarmy

Slash3

ishould

The PS3 is not hard to find on store shelves at all. There are few electronics or game stores (online or brick-and-mortar) which do not have them readily available. Most stores in my area have stacks of them.

macserv

Mac OS X (All):
11417 active / 14 TFLOPS = approx 1.23 GFLOPS per client

Mac OS X (Intel Only):
2704 active / 7 TFLOPS = approx 2.59 GFLOPS per client

Analysis
It would seem that, all-in-all, Mac OS X machines contribute approximately 30% more computing power per active client. I certainly wouldn't have guessed that.

Also impresive: Intel-based Mac OS X machines blow away the Windows average, offering 164% more computing power.

Conclusion:
More Mac users need to be running F@H, especially if they've got an Intel box.

macserv

Did you guys know that there are "teams" for Folding@Home ?? I belong to Team #1971 "Team Mac OSX" and am currently ranked 5th on our team and 256th in the the whole project.

Please seriously consider using all your computational resources (Mac, PC, Linux, GPU and PS3) for the Folding@Home Project.

Stop on by our forums at www.teammacosx.com and check us out. We have our own installer scripts, support, and some of the best folders on the planet.

If you would be interested in folding for Team MacOS X, just come up with a username at Stanford and enter "1971" as your team designation on your computers or PS3's. The teams make it fun, but we fold for the benefit of the science.

Thanks for your time, and I hope you will at least stop by the forums at www.teammacosx.com, all are welcome anytime.

J.R.

jrrodeffer

Any info on what kind of wear and tear F@H causes on your CPU running full blast all the time?

edmicman

Anyways good numbers none the less. I wonder how many people will still be running F@H application when the newness and novelty wears on the the realization of the electric bill sets in.

Kristofff

It's not what most people buy a game console for.

dalziel86

Aleung

My name is Frank Pages (aka C4Diver) and I'm the Clan Leader of AFK (305 BOYZ / AFK) for the the PS3. My clan is composed of over 120 members from accross the US as well as a few from Mexico and Canada. Check out our website for more information: www.305boyz.tk.

The folding@home program was recently brought up to my attention by one of my clan leaders. I researched it and thought this was a great thing for the AFK clan to become involved with as well as an opportunity for us to give back to the worldwide community as a whole. As soon as the Download became available for PS3 I created a folding team account, #55059.

Last night emails were sent out to all clan members registered in our website asking for their support in this cause(as well as that from their friends and family). As of last night we had 8 members sign up via their PS3s and after a recent check 15 work units have been completed thus far. I would like to take this opportunity to extend my invitation to anyone else out their who would like to work with Team AFK. Just follow the directions as indicated (http://folding.stanford.edu/FAQ-PS3.html) for your PS3/PC and just enter 55095 when asked if you want to join a team.

Everyone has something to offer, so join AFK in this great cause.

Sincerly,

Frank Pages,
305 BOYZ / AFK Clan Leader
aka: C4Diver

C4Diver

But this is cool. I knew the cell proc was awesome, I just think Sony's implementation sucks.

trekkie

120.1 volts
1.77 amps
60 hertz

Of course these values fluctuate +/- 1% or 2% based on powerline quality... and in Florida, that ain't so good :-)

When the PS3 is idle and sitting at the XMB the power usage is 191 watts. Sittin on a static browser page (google.com) the power usage is 182 watts.

chelsel

drizzt

I find it amusing that there was a big fuss about how the PS3 had 2 Tflops (compared to 360's 1 Tflops) right before it launched in the US. Amazing how real-world performance is only 1% of claimed...

Hao

Aren't you a proud little fanboy, IDIOT ism? We're glad to see your priorities are in the right place.

I'm glad to see the PS3 being utilized in the folding@home project. The performance increase it offers to the network is astounding. Hopefully, more and more PS3 owners will participate in the program.

kingofgrills

Yeebles

while on the other hand, you have the PS3 owners, with no good games, probably pissed off they paid $600 for it, with the software using up all of the processing power of the PS3, because the owners dont want to play it.

idiotism

urban_ninjya

We all know that the Cell has 9 cores, which immediately gives a 9x multiplier over a PC. Further, each core is very fast, such that it's not unreasonable to say that a Cell core is 2x faster than a PC's. That gets you to 18x. Factor in more idle time for PS3s and low output by older PC boxes and the 25x figure is believable.

Now, as a Computer Scientist, I have to point out that the Cell is not a magic box - programming for that architecture is immensely difficult. Dr. Dobb's wrote an article about it, and the bottom line is that a 60-line program required 1200 lines to adapt to the Cell. Lines of code isn't a perfect measure of complexity, but 60x the code for 9x the processing power is pretty painful. It's also important to keep in mind that not all applications are parallelizable.

Regardless, Sony obviously has taken the time to write an optimized folding program, and the benefits are clear. For highly parallelizable problems that require intense computational horsepower, the investment in code overhead can be worth it.

http://www.ddj.com/dept/64bit/197801624

CSU_Maestro

PS3 = 0.024480 est.
GPU = 0.058908 est

atomriot

w00zzy

I like simpler numbers though:

2700 active IntelMacs vs 14000 PS3's -- say there are 5x as many PS3's.

7 TFLOPS Total for the Intel Macs -- multiply that by 5 to get the result for if there were the same number of Intel Macs as there are PS3's -- that gives you 35 TFLOPS if there were 14000 Intel Macs.

So, if with the same number of Intel Macs and PS3's, you'd get 35 TFLOPS total for the Macs, and nearly 350 TFLOPS total for the PS3's -- a factor of 10 different.

EQC

Intel Macs - 0.002588
PS3 - 0.024480

This is 9.45 X faster.

foghornleghorn

OniMirage

bigboy6998

yes 7/2704 = .00259

but 338/13807 = .0245

It is indeed running faster by a factor of 10.

Jeff_McAwesome

The PS3 was definitely running warm, but not too bad...in vertical position the front vents were cold, top and back pretty warm. My gaming laptop runs MUCH hotter.

Pixelantes Anonymous

PS3 0.0027572872864769106953221359214054

Only seems to be about 6% faster than the the macs per processor. I guess the power is just having more cores.

Another thing that comes to mind is.. how the hell are there so many PS3 being wasted for this application? Aren't they suppose to be fairly hard to find on the shelves?

urban_ninjya

Christopher

Harry_Greek

P.S. How bout that all you PS3 naysayers.

w00zzy

PS3 still looks to be about a factor of 10 faster per machine...

EQC

Whammo

TheCapt

To give an idea of recent performance gains, at the same clock speed, my Core 2 Duo crunches data 3.5x faster than my old P4.

I'd like to see the PS3 compared head-head with a Core 2 Duo (pick the clock speed) to get a more realistic idea of how fast it is.

Phineas J. Whoopie

Leo--wisconsin--USA

mopar_man

Brian Lam

Marty200

Brian Lam

Having a muppet day.

TFIF.

nzruss

Turing_Machine

Is this just a cover for Vijay calculating how building a black hole from common household materials? hey cool, where do I buy one?

nzruss

DragTimes

ripfire

scrag

whoops, petabyte* is what i meant to say.

But that's probably all the fact checking you are going to get out of this (extremely lazy) CS major.

bignaz2k

Seriously, this is some really really good sh1t going on. Makes me think about adding a PS3 to my xbox 360 and wii collection (that, and "resistance").

hypereric

matsayz

bignaz2k