 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I have to model scene with a lot of parallelepiped. Some of them can be boxes
and some not. Perhaps there is any person with done comparision about rendering:
mesh, prism, boxes, polygons, clipped planes (perhaps something other). The rule
is that objects are sometimes multicopied - I know that copied mesh saves memory
but what about rendering ? Is rendering of 12 triangles faster than rendering of
one box? Of course I can make test but under my WinNT I never sure if there
wasn't other process. I have not dedicated comp for POV. The short schema of my
scenefile is something like this :
#macro CreateParallelepiped(Args...)
box{} or mesh{} or prism{} or polygon{} something else
#end
#macro ObjectOne()
#local Object=CreateParallelepiped(...)
union{
object{Object transform{somewhere1}}
object{Object transform{somewhere2}}
object{Object transform{somewhere3}}
object{Object transform{somewhere4}}
}
#end
#macro ObjectTwo()
#local Object=ObjectOne()
union{
object{Object transform{somewhereA}}
object{Object transform{somewhereB}}
object{Object transform{somewhereC}}
}
#end
ObjectTwo()
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Wlodzimierz ABX Skiba wrote:
>
> I have to model scene with a lot of parallelepiped. Some of them can be boxes
> and some not. Perhaps there is any person with done comparision about rendering:
> mesh, prism, boxes, polygons, clipped planes (perhaps something other). The rule
> is that objects are sometimes multicopied - I know that copied mesh saves memory
> but what about rendering ? Is rendering of 12 triangles faster than rendering of
> one box?
box and mesh are probably the fastest. You could also try making several
'objects' with one mesh, but i don't know if that will improve things
much.
> Of course I can make test but under my WinNT I never sure if there
> wasn't other process. I have not dedicated comp for
If you use the CPU time in the Task manager, values should be comparable
even if other processes are running.
Christoph
--
Christoph Hormann <chr### [at] gmx de>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christoph Hormann wrote in message <3AA75DF8.B40888BC@gmx.de>...
> Wlodzimierz ABX Skiba wrote:
> > Of course I can make test but under my WinNT I never sure if there
> > wasn't other process. I have not dedicated comp for
>
> If you use the CPU time in the Task manager, values should be comparable
> even if other processes are running.
even if memory swaping is used ?
unfortunatly without POV I'm using 113 of 130 MB
when I run POV and prepare large enough scene - I think swaping will be used :-(
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Wlodzimierz ABX Skiba wrote:
>
> even if memory swaping is used ?
> unfortunatly without POV I'm using 113 of 130 MB
> when I run POV and prepare large enough scene - I think swaping will be used :-(
>
Swaping is done by the system process and therefore the time for it is
added there. If you do a long render with a lot of swapping you will
notice that the system process time will increase a lot.
Christoph
--
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other
things on: http://www.schunter.etc.tu-bs.de/~chris/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3aa76427@news.povray.org>, "Wlodzimierz ABX Skiba"
<abx### [at] abxartpl> wrote:
> even if memory swaping is used ?
> unfortunatly without POV I'm using 113 of 130 MB
> when I run POV and prepare large enough scene - I think swaping will be
> used :-(
In that case, using meshes might be the fastest, if you #declare a mesh
and use many copies of it. You might be able to save enough memory to
avoid swapping, and get a noticeably faster render...
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Christoph Hormann <chr### [at] gmxde> wrote:
: You could also try making several
: 'objects' with one mesh, but i don't know if that will improve things
: much.
AFAIK that can improve the rendering speed a lot. This is because when
all the triangles are inside one mesh, povray can make octree optimizations
when making intersection tests (that is, it can discard most of the triangles
at each step). If all the objects are separate, then povray can't do this
kind of optimization.
(I don't know if povray can optimize a union of several meshes this way.
Does someone know?)
--
char*i="b[7FK@`3NB6>B:b3O6>:B:b3O6><`3:;8:6f733:>::b?7B>:>^B>C73;S1";
main(_,c,m){for(m=32;c=*i++-49;c&m?puts(""):m)for(_=(
c/4)&7;putchar(m),_--?m:(_=(1<<(c&3))-1,(m^=3)&3););} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Chris Huff wrote in message ...
> In that case, using meshes might be the fastest, if you #declare a mesh
> and use many copies of it. You might be able to save enough memory to
> avoid swapping, and get a noticeably faster render...
I knew that but the problem is that I have not environment for test but want
prepare object for publish and therefore want optimal solution and therefore I
ask if someone have done tests :-)
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote in message <3aa76cce@news.povray.org>...
> AFAIK that can improve the rendering speed a lot
hmm, I can simple reorganize my macros to return only sets of traingles and join
it as one mesh at end - but what about texturing ? I have to use 12 textures in
over million triangles. It is imposible to split it into 12 meshes becouse they
are mixed in groups. How it increase memory if I apply texture to every triangle
?
> (I don't know if povray can optimize a union of several meshes this way.
> Does someone know?)
Does someone know?
ABX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
