From: Lally Singh (lally.singh_at_[hidden])
Date: 2007-07-19 18:22:20
On 7/19/07, Malte Clasen <news_at_[hidden]> wrote:
> iain.denniston_at_[hidden] wrote:
> I'm not sure whether this would be the right level for architecture
> specific optimizations. Having a layer that provides access to SIMD
> instructions might smooth out issues between SSE and AltiVec, but
> programming Cell or perhaps CUDA is afaik a bit more involved. I would
> start with larger operations such as FFT, SVD or "multiply a set of
> vectors with matrix A". Given a portable C++-only implementation, you
> could then provide architecture specific implementations of these
> building blocks.
For game engines, the one you want to do first is the Runge-Kutta.
That'll handle the big weight in physical simulations.
Then Collision Detection methods.
Later in the year I'll have some performance numbers to indicate where
a normal desktop CPU's spending it's cycles in a game engine, if
anyone wants to prioritize their work.
If you can get these parallelizable over 6 SPEs (e.g. what you find in
a Linux-loaded COTS PS3), 3 PPC cores (Xbox 360) or 2 R5900s
(PSP/PS2), you're in fantastic shape.
-- H. Lally Singh Ph.D. Candidate, Computer Science Virginia Tech
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk