Yes, but in this case the windows version shoots reflected rays where the linux
version does not,
and the results look completely different - the problem disappears if you change the
term

reflection { 0,1 fresnel }

to

reflection { 0,1 falloff 2 }

for example. In this case both versions produce absolute identical results...

Btw, in my test-scene the 'difference' produces coincident surfaces by accident, but
this is not related to the problem described, linux and windows versions output will
still look different if this is fixed...

-sascha

ABX wrote:

> On Wed, 12 Nov 2003 16:34:21 +0100, sascha <sas### [at] userssourceforgenet>
> wrote:
> 
> 
>>reflections ... produce different output!
>>
>>POV-Ray for Windows: Version 3.5.icl.win32
>>POV-Ray for Linux: Version 3.5 Unix (.Linux.gcc)
> 
> 
> Well, icl is not gcc. Intel compiler could utilize dedicated CPU/FPU commands
> related to single floats used as color components which determine
> reflection/refraction.
> 
> ABX

Post a reply to this message