Well I didn't say my code was *well* written. Apparently there's a lot of trickery with copying global memory to cached memory to speed up operations. Cached memory takes (IIRC) one clock cycle to read or write, and global GPU memory takes six hundred cycles. And there's all this whatnot and nonsense about aligning your threads with memory locations that I don't even bother with.
The Tesla 1060 is a video card with no video output (strictly for processing) that has something like 240 processor cores and 4 GB of DDR3 RAM. Just doing math on large arrays (1k x 1k) I get a performance boost of about a factor of forty over a dual core 3.0 GHz Xeon.
The CUDA extension set has FFT functionality built in as well, so it's excellent for signal processing. The SDK and programming paradigm is super easy to learn. I only know C (and not C++) and I can't even make a proper GUI, but I can make my array functions run massively in parallel.
The trick is to minimize memory moving between the CPU and the GPU because that kills performance. Only the brand newest cards support functionality for "simultaneous copy and execute" where one thread can be reading new data to the card, another can be processing, and the third can be moving the results off the card.
One way that the video people can maybe speed up their processing (disclaimer: I don't know anything about this) is to do a quick sweep for keyframes, and then send the video streams between keyframes to individual processor cores. So instead of each core gets a piece of the frame, maybe each core gets a piece of the movie.
The days of the math coprocessor card have returned!
Wow, you are correct. When I downloaded the paper and scanned the references, I somehow didn't see the reference listed.
My mistake.
I guess I have to change my point to "how is this new now?" But I guess a nice experiment in any case.
sorry.
n = (index of refraction)
h = (well, hbar, the planck constant)
k = (photon wavenumber)
http://arxiv.org/abs/cond-mat/0502014
This paper from MIT showed conclusively through experiment (almost 4 years ago) that in a refractive material the medium temporarily gives up its momentum to the photon, so that the momentum of the photon in the medium is nhk.
It's too bad that this new experiment didn't cite the prior art.
The rule on staying alive as a program manager is to give 'em a number or give 'em a date, but never give 'em both at once.
