Supercomputing at 1/10th the Cost. The Tesla Processor

NVIDIA TESLA GPU COMPUTING SOLUTIONS FOR WORKSTATIONS
Experience cluster level computing performance—up to 250 times faster than standard PCs and workstations—right at your desk. Convert your workstation into a NVIDIA® Tesla™ Personal Supercomputer by adding Tesla GPU computing processors. Each Tesla GPU is based on the revolutionary NVIDIA® CUDA™ massively parallel computing architecture with a rich set of developer tools (compilers, profilers, debuggers) for popular programming languages APIs like C, C++, Fortran, and driver APIs like OpenCL and DirectCompute.

www.nvidia.com...

www.youtube.com...

a little more in detail in what it can do
www.youtube.com...

want to own a supercomputer at your own home! ? in a desktop I sure do

Surprised the Government has not banned this ( example cure for cancer ) www.accelereyes.com...
in days then months/years decades as it stands now something like the Insane Computer Proteus IV did (Demon Seed)

it looks like that Hal 9000 (2001Space Odyssey) and Proteus IV from the movie (Demon Seed) Bio computer news.nationalgeographic.com... might become a reality

put this in the Honda Robot to make a real android that you see only in science fiction. MR Asimov your I robot might become so true
ASIMO imagine that asimo.honda.com...
asimo.honda.com...

before you post a comment the Tesla Processor has been in the market for almost a year ago
I would like to post it here as to imagine all the possibility's

and like to see what ATS Members think about the Tesla Processor in what it can do for the future in the hands of the public

[edit on 8-2-2010 by Wolfenz]

[edit on 8-2-2010 by Wolfenz]

Tesla is market speak for fast, slightly more opened to developers GPU running standards. We've seen GPU offloading for years in the technology sector in niche markets (FFT applications in science to name one), just moreso now it's being pushed, as people realise cpu technology has started to slow in development - when compared to ATI or Nvidias latest terraflop plus gpus. CPUs are mostly going multi core, albeit not as terrifically faster per core iteration as it used to jump in an average technology cycle.

news.softpedia.com...

Japanese supercomputer in 2008 broke into the top500 with Nvidia tesla GPU tech, a first.

My 5 year old, top of the range x800xt gpu would have been considered a supercomputer a few years before its release.. 145gflops in a can. PS3 is 2tflops peak PR speak with a unicorn crapping on it and much less sustained (around 500gflops) with 6 cores and some gpus whacked in. Latest ATI gpu is 4.6tflops in a can with 2 cores.

It's amazing how quick GPU architechture is developing in contrast to your typical desktop CPU.

sad part is that it has to be pretty before "people" do anything useful with a computer,

computers are basicly the same they where in the 70´s sure you can argue that the processing power has out done it self by miles compared to them old cans but it still wont change the matter of fact that they are still oversized calculators people rearly even know how to use the full potential of ,

sure its smoother and you can just click but its basicly the same as it where in the 70´s you just dont need to know the mechanics behind it all....

Originally posted by zerbot565

sure its smoother and you can just click but its basicly the same as it where in the 70´s you just dont need to know the mechanics behind it all....

Too true. 0s and 1s since nam, why are we not looking at electron spin in crystal lattices and getting home quantum computers.... I guess it's the same answer as when you consider wheels and petrol engines for 100 years plus...

reply to post by GhostR1der


It's not because GPU technology is advancing faster, or is generally better or faster, it's just due to the specific nature of GPUs tends to lend GPUs to performing lots and lots of highly-accurate maths problems (due to the inherent accuracy of the rendering pipelines), and the inclusion of programmable hardware (programmable shaders).

CPU technology has advanced fantastically before and since NVIDIA released their Tesla lineup, but in a very general way, that benefits every other aspect of computing. You can't run an operating system on a Tesla, which is why you'll never find a supercomputer using only these GPGPUs. There will always be normal processors in there, too, coordinating the processing, and running the actual OS itself.

It's a bit rude to say that desktop CPUs have not advanced that much in the last few years. Intel has really pulled out all the stops recently, and has come out with some fantastic chips.

Nice to see Tesla's name getting a bit of big name recognition.

Kind of ironic, though. The digital logic of computers depends on DC, and can't run directly on Tesla's AC.

Other than that, It's pretty sweet for people who need to do a lot of math-intensive work, which pretty much means simulations and rendering. I'm not really in the market.

Originally posted by zerbot565
sad part is that it has to be pretty before "people" do anything useful with a computer,

computers are basicly the same they where in the 70´s sure you can argue that the processing power has out done it self by miles compared to them old cans but it still wont change the matter of fact that they are still oversized calculators people rearly even know how to use the full potential of ,

sure its smoother and you can just click but its basicly the same as it where in the 70´s you just dont need to know the mechanics behind it all....

On basic HW layer there is no "revolutionary" development as you said. But if you take SW layer into consideration then ... I'm pretty amazed. Especially OpenSource "movement" is something unpredictable in capitalistic society.
Personally I lost my attention to "top" HW development when I switched from 3d animation on Win to network administration on Linux.

while this is sweet.
you guys may be interested in this thread
www.abovetopsecret.com...

REVERSIBLE COMPUTING: Computers 10^10 more efficient!