Vec/Pnt speed/implementation (was: Re: Expanded use of Boost.Function)

	Hi All,

this thread is moving far away from the subject, so to help distinguish 
the two threads going on I'd like to adjust the subject for this one.

Gerrit Voss wrote:
> 
> Just a quick update, I ran some more test, same compiler options,
> VS2005:
> 
> original opensg:
> Ratio: 1.32355
> 
> hand coded Vec3f class (with __forceinline):
> Ratio: 1.29539
> 
> gmtl (even more templates) :
> 
> Ratio: 0.992 - 1.0003
> 
> So more templates are better, I knew it . Ok, I guess we have to do
> some more tests and have a closer look 

So score one for GMTL. Have you tried it with a more complicated 
expression (a realistic one might be 'p = p + s * v1 + (1-s) * v2' or 
something like that). In a case like that I would assume GMTL has even 
more advantage.

	Dirk

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/