|
 |
gRRosminet wrote:
> Hi everybody !
>
> I just want to inform you that I have tried Intel C++ Compiler v8.0 on
> povray ... the result is that the rendering time for the benchmark scene
> is now 33 minutes with these options :
> CFLAGS = -O3 -mcpu=pentium4 -march=pentium4 -ip
>
> before this, I used a povray compiled with gcc and "-O3 -mcpu=pentium4
> -march=pentium4 -f...." options and the benchmark was rendered in 38
> minutes
Recent tests we made with the 3.6 beta showed that the best performance
with gcc is achieved with plain '-O3' and without additional '-f...'
(the best -mcpu/-march flags probably depend on the processor).
Christoph
--
POV-Ray tutorials, include files, Sim-POV,
HCR-Edit and more: http://www.tu-bs.de/~y0013390/
Last updated 11 Jan. 2004 _____./\/^>_*_<^\/\.______
Post a reply to this message
|
|
|
|
Christoph Hormann <chr### [at] gmx de> wrote:
> Recent tests we made with the 3.6 beta showed that the best performance
> with gcc is achieved with plain '-O3' and without additional '-f...'
That makes lots of sense.
If there's some optimization which is not turned on with -O3 (which
is the maximum -O flag) there's probably a reason. If there would be
some additional optimization which would increase the performance
of the code, wouldn't they include it in -O3?
I suppose those optimizations not included are experimental, theoretical
(ie. they *may* speed up, but also slow down) or just buggy.
--
#macro M(A,N,D,L)plane{-z,-9pigment{mandel L*9translate N color_map{[0rgb x]
[1rgb 9]}scale<D,D*3D>*1e3}rotate y*A*8}#end M(-3<1.206434.28623>70,7)M(
-1<.7438.1795>1,20)M(1<.77595.13699>30,20)M(3<.75923.07145>80,99)// - Warp -
Post a reply to this message
|
|
