|
|
Unix and Linux fellows,
I've just published online the first results of a rather extensive
benchmark study conducted on 12 different x86 platforms (running GNU/Linux)
and using the current version of POV-Ray (3.6.1). The article is the first
of a series of two dedicated at analyzing how to obtain a fast POV-Ray binary
(as measured by running the official benchmark.pov). The study compares the
speed of the official POV-Ray 3.6.1 for Linux binary versus ~50 unofficial
binaries built on the test platforms (using several versions & optimization
options of the GCC and ICC compilers).
http://pov4grasp.free.fr/articles/fastpov1/
The article shows the overall benefit in recompiling POV-Ray from
sources, particularly on the most recent platforms: namely the Intel
Pentium 4 / Xeon, Pentium-M, and the AMD k8 familly (Athlon64 / Opteron).
All of the unofficial binaries built for this study take advantage of
platform-specific optimizations, while the official POV-Ray for Linux
binary does not (for compatibility reasons).
Note that the article is relatively long and somewhat technical.
Comments are welcome :-)
- NC
Post a reply to this message
|
|