|
![](/i/fill.gif) |
Greg M. Johnson wrote:
> I just purchased a new PC.
>
> My OLD PC is an Aptiva with 64MB RAM and a 200 MHz Pentium.
> My NEW PC is an Aptiva with 256 MB RAM and a 450 MHz Pentium II.
>
> I compared the render times for three different scene files. One was
> with Bryce, and involved transparent and reflective objects. The other
> two were in POV-Ray. One was a human character made out of blobs of my
> own creation. The other was the LATHE2.POV file that is found in the
> POVDEMO directory shipped with version 3.02
>
> Render times in min:sec
>
> OLD PC NEW PC
> Bryce scene 29:30 2:49
> POV, blobs 0:29 0:11
> POV, lathe2 9:01 4:16
>
> What gives? With Bryce, I get a 10X improvement in render time, but POV
> gives me less than a 3X improvement! Does this point to some other
> "hardware" solution to a faster POV render? Could this point to a
> problem with my PC: dead memory, etc.?
No, not really. POV is float-intensive, and the speed up is proportionnalto the
processor speedup
(p2 has better fpu, about 30% faster at equal
mhz, 2.5 times the mhz on the new machine 1.3 x 2.5 = 3.25, which is
consistant with the observed speedup).
As for Bryce, I'd guess it is memory-intensive, so that it stops swapping
with 256 megs of ram. It's a bit like loading netscape on a 200 mhz machine
with 16 megs of ram... and then on a 200 mhz machine with 128. It can't
be compared. Netscape takes about 5 sec to load on my machine (a 333mhz
with 128 meg of ram)... and about 30 sec on a pentium 90 with 32 megs of
ram. ... To tell the truth, I don't use Bryce, so I can't tell the reasons. May-
be it is also sensitive to the kind of video adapter you have. Going from a
mach 64 to a 3d Rage+ II _does_ make a lot of difference.
> ___________________________
> Greg M. Johnson
> http://members.xoom.com/gregjohn
Best,
S.
--
----------------------------------------------------------
Steven Pigeon Ph. D. Student.
University of Montreal.
pig### [at] iro umontreal ca Topics: data compression,
pig### [at] jsp umontreal ca signal processing,
ste### [at] research att com non stationnary signals
and wavelets.
----------------------------------------------------------
http://www.iro.umontreal.ca/~pigeon
Post a reply to this message
|
![](/i/fill.gif) |