Sign in with
Sign up | Sign in
Your question

CPU'S getting stomped, even C2D's....

Last response: in CPUs
Share
March 25, 2007 2:43:50 PM

Folding at Home has introduced the ps3 to their cause, which is helping immensly. Im currently running both the cpu and the gpu versions and I remember all those ps3 fanboys, but hey, til you can match my gpu GO FISH heres a link : http://www.techpowerup.com/img/07-03-25/stats.png beat that, if ya can

More about : cpu stomped c2d

March 25, 2007 2:53:50 PM

Im guessing Im pushing a lil more than .1 teraflops to them, since not all gpu's are 1900's
March 25, 2007 3:02:12 PM

I want an Xbox 360 Folding option, I never really use mine (a couple hours every few days at most), so I could leave the thing folding pretty much all day. As soon as I get a new video card and my pc becomes my main comp my macbook and emac are going on 24/7 folding duty.
Related resources
March 25, 2007 3:12:30 PM

I wish more people felt that way. I like competition, and Tom's has their own folding team. We are currently ranked 62nd in the world, and anyone who wants to help... please do
March 25, 2007 4:04:51 PM

Yeah, I want an Xbox 360 folding client, too. I'm on a different folding team that's ranked 29th, ^^ Anyway, it seems that on average GPU clients can do twice as much work per unit as a PS3, but the number of people doing it is so small.

This type of work is what the Cell processor excels at. However, stream processing, for the win.
March 25, 2007 4:14:23 PM

Quote:
I wish more people felt that way. I like competition, and Tom's has their own folding team. We are currently ranked 62nd in the world, and anyone who wants to help... please do


Don't really care about rank - it's all for a good cause but since I was planning on it anyway I'm Folding using Tom's team ID. I just put my core rig online and by the end of the week I'll have 13 machines going, even my HTPC....
March 25, 2007 4:29:38 PM

I'm going to load every computer at the community college /high school I work at with F@H. Imagine the massive score... :mrgreen:
March 25, 2007 4:38:04 PM

Go for it.
its for a good cause 8)
March 25, 2007 5:10:46 PM

Quote:
I'm going to load every computer at the community college /high school I work at with F@H. Imagine the massive score... :mrgreen:


Muhaha, that would be awesome. I wish my college would have the hundreds/thousands of computers around campus folding 24/7. The power use would spike like an f'in bandit, but it would be quite an impressive increase. Especially because I'd get them to do it under my UID, :lol: 
March 25, 2007 6:14:13 PM

how do you get into toms folding group? I ran F@H for a while but my poor throroughbred can't really handle it...gonna go for a e4300 soon as pricecuts come maybe brisbane if i can get one for 50 bucks 8)
March 25, 2007 6:46:09 PM

http://fah-web.stanford.edu/cgi-bin/main.py?qtype=ossta...
What in the world is a "TFLOPS"? Is it a measure of performance or actual work being done? So at the current rate (and ignoring different work units), are they saying that more work is being done by 30K PS3's than 200K other machines?

http://fah-web.stanford.edu/psummary.html
Notice that for the few PS3 WU's being crunched, the atom count is in the hundreds, whle the atom count for PCs and others are in the thousands and tens of thousands..
March 25, 2007 6:48:03 PM

With Nvidia's CUDA SDK now available, I wonder if Standford will try to make a Folding@home client for the Geforce 8 series.
March 25, 2007 7:04:37 PM

Quote:
http://fah-web.stanford.edu/cgi-bin/main.py?qtype=ossta...
What in the world is a "TFLOPS"? Is it a measure of performance or actual work being done? So at the current rate (and ignoring different work units), are they saying that more work is being done by 30K PS3's than 200K other machines?

TFLOPS is the number to represent how many "Trillions of FLoating Operations Per Second" are being done; basically a measurement of work output.
March 25, 2007 7:14:00 PM

Quote:

http://fah-web.stanford.edu/psummary.html
Notice that for the few PS3 WU's being crunched, the atom count is in the hundreds, whle the atom count for PCs and others are in the thousands and tens of thousands..


PS3 work units are in the hundred thousands, 200,000 is avg. number i've seen people posting on the web, i'm sure they do larger ones because people are saying it takes 8-10hrs to fold. But when compared with a thousand or ten's of thousands work units most PCs get that take days to work out, you can see the CELL clearly pwns at this sort of work. I too use my x1900 to fold, so much nicer than the CPU because i can still web browse while i fold.

Side note, when you look at the CELL's 300Tflop performance, and compare the cost of the Stream Processors from nVidia and ATI, the CELL is much more efficient folder for the cost, being only a few hundred TFlops less than a GPU stream processor.

One last thing, if the PS3's are running F@H this well just imagine what the Opteron/CELL servers would do that are being built 8O

Edit: Grammar/Spelling
March 25, 2007 7:28:43 PM

Quote:

http://fah-web.stanford.edu/psummary.html
Notice that for the few PS3 WU's being crunched, the atom count is in the hundreds, whle the atom count for PCs and others are in the thousands and tens of thousands..


PS3 work units are in the hundred thousands, 200,000 is avg. number i've seen people posting on the web, i'm sure they do larger ones because people are saying it takes 8-10hrs to fold. But when compared with a thousand or ten's of thousands work units most PCs get that take days to work out, you can see the CELL clearly pwns at this sort of work. I too use my x1900 to fold, so much nicer than the CPU because i can still web browse while i fold.

Side note, when you look at the CELL's 300Tflop performance, and compare the cost of the Stream Processors from nVidia and ATI, the CELL is much more efficient folder for the cost, being only a few hundred TFlops less than a GPU stream processor.

One last thing, if the PS3's are running F@H this well just imagine what the Opteron/CELL servers would do that are being built 8O

Edit: Grammar/Spelling

I take it you don't have a dual-core processor? I used to fold on my laptop and I could still do all my normal tasks without any real noticeable degradation in performance. You should still be able to browse the Internet while folding with a typical single-core, though, :?
March 25, 2007 7:38:19 PM

i have a dual core albeit a 939 Athlon X2, it still dragged down the system even with 2gb of RAM. I didn't put any kind of restrictions on the type of WU's so i've done a fairly good amount of large units, the shortest one i ever had took 26hrs. It choked my system pretty bad, but like i said x1900 = insane folding with minimal CPU drain
March 25, 2007 7:55:45 PM

folding?
March 25, 2007 7:57:23 PM

Quote:
i have a dual core albeit a 939 Athlon X2, it still dragged down the system even with 2gb of RAM. I didn't put any kind of restrictions on the type of WU's so i've done a fairly good amount of large units, the shortest one i ever had took 26hrs. It choked my system pretty bad, but like i said x1900 = insane folding with minimal CPU drain


Interesting...I'm not about to speculate why, but I have two P4 folding machines, also without WU size or CPU usage restrictions that I can browse on. Hmmmm...I kind of find that odd. X2s aren't weak...they can dish it out.

Kind of odd sounding to me. But yeah, GPU folding is where the real folding could get done. If only I didn't have to pay for college...:( 
March 25, 2007 8:47:29 PM

Work output? What? So they're saying that 30K PS3s are crunching ~4x 160K PCs? Well for some reason, PS3 TFLOPS just taken a whacking of -100. Maybe they initially overestimated..

Quote:
PS3 work units are in the hundred thousands, 200,000 is avg. number i've seen people posting on the web, i'm sure they do larger ones because people are saying it takes 8-10hrs to fold. But when compared with a thousand or ten's of thousands work units most PCs get that take days to work out, you can see the CELL clearly pwns at this sort of work. I too use my x1900 to fold, so much nicer than the CPU because i can still web browse while i fold.

200,000 what? Atoms per WU? I just linked you a summary page showing the PS3 WU's in the hundreds.. I guess it kind of balanaces out though since the PS3 deadlines are in a ~day while PC deadlines are longer. And excuse me if I'm not familiar with the graphs, more of a WCG person :wink:
March 25, 2007 9:30:44 PM

WU's aren't always equal in their size or complexity. They're not the best way to determine how much work is being done, but the TFLOPS are.
March 25, 2007 10:08:33 PM

I just wonder how often people have their F@H running at home. Even if they do have it running 24/7 do they relegate it to 50% core utilization or 100%?

Personally I have my Opty 165 crunching through F@H for about ~240 ppd and its on 24/7 at 100% but I game alot on the weekend so it is subordinate to that.

I want a GeForce folding client ><

Also there are still P2's and Original Pentium's doing folding on folding farms, to think that a handful of multi core PS3's couldn't out do 10000 old CPU's is ridiculous. 8O
March 25, 2007 10:34:07 PM

Quote:
i have a dual core albeit a 939 Athlon X2, it still dragged down the system even with 2gb of RAM. I didn't put any kind of restrictions on the type of WU's so i've done a fairly good amount of large units, the shortest one i ever had took 26hrs. It choked my system pretty bad, but like i said x1900 = insane folding with minimal CPU drain


That's bs. I've got two clients running all day at 100% each on the pc (1graphical, 1console) and performance drop is "0" when set priority to "idle". So, cut the crap
March 25, 2007 11:07:59 PM

Quote:
That's bs. I've got two clients running all day at 100% each on the pc (1graphical, 1console) and performance drop is "0" when set priority to "idle". So, cut the crap



you need to chillax, and address people in a more respectable demeanor, the reason my system hangs is because i run graphical, and i give my client top priority as to fold faster. Now i don't use the CPU client anymore, and i should have stated the conditions in which i ran the client, my bad. Now your also running a CPU which makes mine look a snail on a drag strip, so there will be a considerable difference.

so how about next time you try to address someone you be a little more polite.
March 26, 2007 6:48:14 AM

Quote:
how do you get into toms folding group?


I would like to know this as well.
March 26, 2007 8:13:48 AM

Here is a LINK
that has an updated chart. Clearly the PS3 is owning the F@H. Im just glad that ppl will use their machines for a good cause.
March 26, 2007 8:39:01 AM

Call me an idiot... no actually dont do that, but WHAT purpose does folding actually serve???

It cant just be a pi$$ing contest. Is there some international medical research lab with no supercomputer or something?

Just adding to the above... Wasnt there some news awhile back about NV/ATI/DX10 that would bring HUGE improvements to F@H gpu results, but u would need to re-write/modify the s/ware?
March 26, 2007 10:26:41 AM

Quote:
Call me an idiot... no actually dont do that, but WHAT purpose does folding actually serve???

It cant just be a pi$$ing contest. Is there some international medical research lab with no supercomputer or something?

Just adding to the above... Wasnt there some news awhile back about NV/ATI/DX10 that would bring HUGE improvements to F@H gpu results, but u would need to re-write/modify the s/ware?


http://folding.stanford.edu/

"What is protein folding and how is folding linked to disease? Proteins are biology's workhorses -- its "nanomachines." Before proteins can carry out these important functions, they assemble themselves, or "fold." The process of protein folding, while critical and fundamental to virtually all of biology, in many ways remains a mystery.

Moreover, when proteins do not fold correctly (i.e. "misfold"), there can be serious consequences, including many well known diseases, such as Alzheimer's, Mad Cow (BSE), CJD, ALS, Huntington's, Parkinson's disease, and many Cancers and cancer-related syndromes.

You can help by simply running a piece of software. Folding@Home is a distributed computing project -- people from through out the world download and run software to band together to make one of the largest supercomputers in the world. Every computer makes the project closer to our goals.

Folding@Home uses novel computational methods coupled to distributed computing, to simulate problems thousands to millions of times more challenging than previously achieved."
March 26, 2007 11:17:36 AM

Well said.

Folding@Home = saving lives
March 26, 2007 11:39:19 AM

Quote:
Well said.

Folding@Home = saving lives
There's no proof of that yet, as the scientists analyzing data may still find no real use for it; still better to try I guess. I just can't justify leaving my PC on 24/7 when CPUs run folding@home so poorly. Now if an 8800GTX client were available...
March 26, 2007 11:43:00 AM

Quote:
Well said.

Folding@Home = saving lives
There's no proof of that yet, as the scientists analyzing data may still find no real use for it; still better to try I guess. I just can't justify leaving my PC on 24/7 when CPUs run folding@home so poorly. Now if an 8800GTX client were available...

Exactly, I'd rather risk possibly saving a life than leave my computer idle with 0 chance.

CPU client isn't bad... they just don't have near the FP power that GPU's/PS3's do.
March 26, 2007 11:49:05 AM

Quote:


you need to chillax, and address people in a more respectable demeanor, the reason my system hangs is because i run graphical, and i give my client top priority as to fold faster. Now i don't use the CPU client anymore, and i should have stated the conditions in which i ran the client, my bad. Now your also running a CPU which makes mine look a snail on a drag strip, so there will be a considerable difference.

so how about next time you try to address someone you be a little more polite.



Sorry fragile flower, and what's on BOLD is bs again
March 26, 2007 5:41:41 PM

Quote:
Well said.

Folding@Home = saving lives
There's no proof of that yet, as the scientists analyzing data may still find no real use for it;

All the projects are done for a reason...
http://folding.stanford.edu/results.html
March 26, 2007 5:49:44 PM

From: http://folding.stanford.edu/FAQ-PS3.html

Q:It seems that the PS3 is more than 10X powerful as an average PC. Why doesn't it get 10X PPD as well?

A: We balance the points based on both speed and the flexibility of the client. The GPU client is still the fastest, but it is the least flexible and can only run a very, very limited set of WU's. Thus, its points are not linearly proportional to the speed increase. The PS3 takes the middle ground between GPU's (extreme speed, but at limited types of WU's) and CPU's (less speed, but more flexibility in types of WU's). We have picked the PS3 as the natural benchmark machine for PS3 calculations and set its points per day to 900 to reflect this middle ground between speed (faster than CPU, but slower than GPU) and flexibility (more flexible than GPU, less than CPU).

Q: The PS3 is outrunning all the rest of the FAH client types. Should I stop my existing PC/GPU/... FAH clients?

A: No, the other clients are valuable to us too and we have chosen a points system to try to reflect the relative merits of each different platform to our scientific research. For example, the SMP client has been producing some very exciting scientific results and continues to be very important in our work. By supporting machines with lots of different functionality, we have a very rich set of hardware on which to run calculations, allowing us to tailor calculations to the hardware to achieve maximum performance
March 26, 2007 6:23:29 PM

Quote:
Well said.

Folding@Home = saving lives


Folding@Home = burning energy, killing the environment.

Just donate to charity and write a big number on a spreadsheet every now and then to remind yourself how fast your PC could be at crunching WUs. It's cheaper, more tax efficient and greener.
March 26, 2007 6:30:31 PM

Quote:
From: http://folding.stanford.edu/FAQ-PS3.html

Q:It seems that the PS3 is more than 10X powerful as an average PC. Why doesn't it get 10X PPD as well?

A: We balance the points based on both speed and the flexibility of the client. The GPU client is still the fastest, but it is the least flexible and can only run a very, very limited set of WU's. Thus, its points are not linearly proportional to the speed increase. The PS3 takes the middle ground between GPU's (extreme speed, but at limited types of WU's) and CPU's (less speed, but more flexibility in types of WU's). We have picked the PS3 as the natural benchmark machine for PS3 calculations and set its points per day to 900 to reflect this middle ground between speed (faster than CPU, but slower than GPU) and flexibility (more flexible than GPU, less than CPU).

Q: The PS3 is outrunning all the rest of the FAH client types. Should I stop my existing PC/GPU/... FAH clients?

A: No, the other clients are valuable to us too and we have chosen a points system to try to reflect the relative merits of each different platform to our scientific research. For example, the SMP client has been producing some very exciting scientific results and continues to be very important in our work. By supporting machines with lots of different functionality, we have a very rich set of hardware on which to run calculations, allowing us to tailor calculations to the hardware to achieve maximum performance


Basically: "We want all the platforms folding, because each can do different things."
March 26, 2007 6:49:19 PM

Quote:
Well said.

Folding@Home = saving lives


Folding@Home = burning energy, killing the environment.

Just donate to charity and write a big number on a spreadsheet every now and then to remind yourself how fast your PC could be at crunching WUs. It's cheaper, more tax efficient and greener.

Oh please. Charity is the biggest rip off on the planet.
Giving people money is always a bad idea. I'd much rather give them resources.
March 26, 2007 7:41:41 PM

http://arstechnica.com/news.ars/post/20070326-why-the-p...
Quote:
First, most Ars readers are aware that, as a general rule, the TFLOPS metric is a pretty poor way to judge performance. Different instruction set architectures (ISAs) will render the same calculation using different numbers of floating-point operations, sometimes skewing the results significantly when you compare TFLOPS ratings across ISAs. So the TFLOPS ratings given by F@H are fun, but the project itself acknowledges that there are major problems with this metric. In fact, the F@H group's initial numbers for the PS3's TFLOPS rating were wrong, and were revised downward by 50% last night.

Here's a simple analogy to illustrate the logic behind the rankings. The GRE exam has three sections: verbal reasoning, quantitative reasoning, and analytical writing. Let's say that three different students took the exam, and they were then ranked relative to each other based on the average of their three scores.

Because student G is a mathematical genius and an autistic savant who can't do anything but math, he was asked to take only the quantitative exam—an exam that he's insanely good at. So his GRE average reflects his performance on that single exam. Student P excels at both math and vocabulary, so took he took only the quantitative and the verbal reasoning exams. Thus P's GRE average reflects the average of only these two exams. Student C is good at vocabulary, average at math, and a solid writer. He took all three exams, and all three scores contributed to his GRE average.

If we were to rank the average scores of all three students, student G would outclass the other two by a wide margin, student P would come in second place, and student C would be stuck at a far, far distant third place. This is because G (the GPU in the analogy) took only the test that he was insanely good at, P (the PS3) took the two tests at which he excelled, and C (the general-purpose CPU) had to take all three tests.

Ultimately, the TFLOPS/CPU rankings given above align pretty much exactly with the degree of specialization of each type of processor. The GPU is far and away the most specialized of the three, so it sits comfortably atop the rankings. The PS3 has a lower degree of specialization than the GPU, but a significantly higher degree than the general-purpose CPU. Indeed, you could almost use each processor's TFLOPS/CPU score as a sort of "degree of hardware specialization" rating.

The final thing that's worth noting is that the pool of CPUs that make up the "Windows" portion of the client list varies widely, from older Pentium 4 models to brand new Core 2 Duos and everything in between. The GPUs are much more uniform in terms of hardware types (all ATI), and the PS3 is the most uniform of them all. So the PS3-to-PC comparison isn't just apples-to-oranges. It's more like apples-to-citrus.
March 26, 2007 7:50:28 PM

Quote:
Sorry fragile flower, and what's on BOLD is bs again



no not quite fragile, just people like you make this place seem like an unfriendly atmosphere to talk, spread there opinions and ideas, because you will belittle there attempt at an open debate. In this case you attack my experience of 100% CPU utilization on an X2 3800+ running graphical F@H, without knowing anything about my setup or my situation.

Think before you speak, or before you attack someone be sure you're clear on there meaning or understand the situation by asking questions.
March 26, 2007 8:10:47 PM

I have been involved in the Alpha and Beta testing of the GPU for about 5-6 months now. To give you an idea the Gpu cores are worth about 320 per WU. The typical CPu unit is worth about 370. As it currently stands Dual core or twin graphcs cards don't boost performance unless you hack the registry and set flags to do so. I am running twin X1950 pros running the GPU client on each one. It takes about 13 hours to finish a unit. I figure that I generate aout 1200 points worth of work units a day or about as much as 10 Core 2's or AMD duals cores.. They take between 48 and 60 hours to generate the full GROMACS unit. Substantially less for the Tinker or small Gromac. The 512mb X1950XT and XTX are about 30% faster. http://fahwiki.net/index.php/How_do_I_add_flags_using_a... . The most valuable designs are the Clearstream processors because the Clearstream is something like 100 tflops of double precision computing.

GROMACS ON CLEARSPEED

Clearspeed offers a CPU architecture which, while similar in architecture to GPUs in terms of many parallel FPUs, goes beyond GPUs by having a more flexible architecture and many other scientifically useful properties, such as double precision FPUs. Clearspeed has been working on porting GROMACS and has done some public demos. Once they have completed the port, we will start testing its feasibility to be used in FAH. http://folding.stanford.edu/FAQ-highperformance.html

The cpus except a few AMD's are confined to single precision because of in the case of the Intel, chips only have 1 complex instructiuon decoder per core vs 3 for the AMD 64 series. AMD running windows Xp is single precision by default. They are more capable of running double precison becasue they have 3 complex decoders. That is only possible currently if you are running 64 bit linux. "The SMP client supports OSX/Intel natively (which means a major points boost for OSX donors) as well as 64-bit linux (with 32-bit linux hopefully to come soon).

The limitations of the GPU design limit GPU's to Single precision. The Cell BE is going to be the most effective when IBM decides to enable double precision calculations.

"On average, Cell is eight times faster and at least eight times more power-efficient than current Opteron and Itanium processors, despite the fact that Cell's peak double-precision performance is fourteen times slower than its peak single-precision performance. If Cell were to include at least one fully usable pipelined double-precision floating-point unit, as proposed in the Cell+ implementation, these performance advantages would easily double"

"To evaluate Cell's potential, Berkeley Lab computer scientists evaluated the processor's performance in running several scientific-application kernels, and then compared this performance against other processor architectures. The results of the group's evaluation were presented at the ACM International Conference on Computing Frontiers, held May, 2006 in Ischia, Italy, in a paper by Samuel Williams, Leonid Oliker, Parry Husbands, Shoaib Kamil, and Katherine Yelick of the Future Technologies Group in Berkeley Lab's Computational Research Division, and by John Shalf from DOE's National Energy Research Scientific Computing Center, NERSC. "
Read all about it here: http://www.supercomputingonline.com/article.php?sid=118... I don't think that any one here or Jon Stokes (especailly) has the credentials of any of these people.

All the talk about Quad core and so forth is a bunch of marketing hype. The Cell BE or Clearstream or other GPU based accelerators are more energy efficient and at least an order of magnitude faster. The cpu will still be necessary to handle instructons and out of order sequencing. But they don't need quad cores to do that. The way forward will require IMC, northbridge based designs are obsolete.
March 26, 2007 8:17:28 PM

Quote:
Sorry fragile flower, and what's on BOLD is bs again



no not quite fragile, just people like you make this place seem like an unfriendly atmosphere to talk, spread there opinions and ideas, because you will belittle there attempt at an open debate. In this case you attack my experience of 100% CPU utilization on an X2 3800+ running graphical F@H, without knowing anything about my setup or my situation.

Think before you speak, or before you attack someone be sure you're clear on there meaning or understand the situation by asking questions.

And I call it BS again. So, what now?
March 26, 2007 8:24:49 PM

Oooooooh. Them's fighting words! :lol: 

Seriously, keep it chill. No hits below the belt, no references to past girlfriends, immediate family or that incident in primary/elementary school involving a bottle of glue and a eraser. :wink:

Tap gloves and fight.
March 26, 2007 8:30:55 PM

Getting in an argument on the internet is the same as getting last place in the special olympics, and apparently PMR likes getting last.

and with that comment i am done, i tried being civil, so i won't say anymore
March 26, 2007 8:44:44 PM

SUP, how many flops is a Core 2 Duo? I'm pretty sure that a core 2 is faster then a Cell proc.
March 26, 2007 9:19:52 PM

Ninja, what's the THG team #?
March 26, 2007 9:22:45 PM

40051. You can find it listed in the main page of the CPU forums. :) 
March 26, 2007 9:34:09 PM

Thx. :) 
March 26, 2007 9:42:56 PM

No problem man. :) 
March 26, 2007 9:52:13 PM

Quote:
Seriously, keep it chill. No hits below the belt, no references to past girlfriends, immediate family or that incident in primary/elementary school involving a bottle of glue and a eraser.


Your literacy level disappoints me.
March 26, 2007 9:57:05 PM

Quote:
Seriously, keep it chill. No hits below the belt, no references to past girlfriends, immediate family or that incident in primary/elementary school involving a bottle of glue and a eraser.


Your literacy level disappoints me.0noZ w3 gotz a grammar naz1 0n 0ur handz!!!!
March 26, 2007 9:57:06 PM

Damn your pedantry. :( 
!