[linux-audio-dev] Traps in floating point code

[Jens M Andreasen]

On tor, 2004-07-01 at 23:40, Erik de Castro Lopo wrote:
> On Thu, 01 Jul 2004 18:18:41 +0200
> Benno Senoner <sbenno at gardena.net> wrote:
> 
> > Eric what do you think ? can something like that be coded efficiently 
> > using SSE/SSE2 ?
> 
> Probably not. There are some algorithms which simply can't be vectorized.
> 

In Bennos case (outputting several streams in paralel) there is the
obvious case of doing 4 streams in one go. 

The downside is that (for polyphony beyond 4 including voice stealing)
assignment algorithms are sickenly hard to optimize and very likely to
contain bugs that only shows up once in while.

> Erik