Alex Perry writes:
>
>Can we patch the sgdPointInTriangle back to PointInTriangle
>_and_ keep the improvements from Norman in the tree ?

Alex

I believe all of my improvements are in the tree.

Although I have yet to add the IntersectCachedLeaf() routine 
which can give yet another order of magnitude (or two) of speedup 
with out much penalty for misses for most cases where one is doing 
multiple queries in close proxsition.

FWIW
I believe the sgdPointInTriangle() is currently faster for our use then 
the commented out 'testbed' PointInTriangle() that I 'inadvertenly' 
left in the code, in that it often takes the early out faster :-)
 < a bounding box test though ugly to see in code is hard to beat > 

FYI
the sgdPointinTriangle() was specifically written for use with these 
FGFS routines and was moved to PLib so others could take advantage
of it recently.

However, as you observed, my test routine is 'elegant' and 'quite' fast
it was gleaned from a recent discussion on the games-algorithms list
and is very close to what was being advocated there as an 'optimium' 
solution and IMHO deserves study until understood :-)

Cheers

Norman


_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel

Reply via email to