On Apr 17, 2004, at 4:54 PM, Dan Clemmensen wrote:
Another interesting point: From the point of view of raw computational
power, the sum of all the 3D graphics cards in the world probably
exceeds the sum of all CPUs in the world. Sure, the graphics cards are
special-purpose, but what if you had the resources to carefully create
algorithms that are well suited to graphics cards? basically, this
requires that you map your problem onto the space that is easily
addressed by graphics cards. The extremely obvious problem is graphics
rendering (DUH!) This has actually been done: you can use graphics
cards to run the POV-RAY backend. When you do this, a few (10?)
machines with graphics cards can render scenes that would otherwise
take hundreds of high-end CPUs.
Essentially all modern motherboard "chipsets" have embedded 3D
graphics engines. This resource is basically unused most of the time
in most machines.
Quite a few people are expressing keen interest in using GPUs for
other things besides graphics. Here are a few links:
http://www.gpgpu.org - General Purpose Computation Using Graphics
Hardware
Arial0000,8080,0000wwwcg.in.tum.de/Research/data/Publications/sig03.pdf
- paper on implementing various linear algebra operations on GPUs
http://www.sintef.no/static/AM/gpu/summary.html
http://www.eyetap.org/about_us/people/fungja/research/ - cool paper on
using GPUs for computer vision projects