Subject: Re: [boost] [gsoc] Interest check for 3d geometry proposal
From: Simonson, Lucanus J (lucanus.j.simonson_at_[hidden])
Date: 2009-03-26 12:19:00
Joel Falcou wrote:
> Simonson, Lucanus J a écrit :
>> It sounds to me like such a thing would need to employ archetecture
>> specific techniques such as compiler intrinsics and inline-assembly
>> to be in any way competitive in terms of performance with what
>> probably exists in closed source development shops. Also, these
>> things are best offloaded to the GPU if possible, and that code
>> isn't written in C++.
> simd intrinsic is actually enough but last time I proposed something
> on those lines, it was dismissed as useless ....
A new implementation of polygon clipping was dismissed as useless when I proposed it. You just need to find the people who see the value in the idea. In fact, a simd upgrade to UBLAS could be very compelling, however, you won't convince people with an idea. You need to show them working code and data showing the performance benefits. I found the people who saw the value in a new polygon clipping library, implemented it for them and when those who dismissed the idea originally saw the benchmark data they acknowledged that there was value after all. Even without simd, optimized matrix operations that take things like cache-behavior into account are dramatically faster than naïve implementations. This is why people charge money for those things, it is worth it. I know I dismissed the idea of simd boost library originally, but you could easily convince me I was wrong if you show me the evidence to prove it.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk