|
 |
Hi all. I'm doing my first "real" image in POV-Ray, and part of it involves
a
set of nested #whiles to pave a 22x15 area of the XZ plane (with 0.28 unit
spacing on both axes). As the floor is reasonably complex, I decided to
#declare it as floor_final and #declared a simple plane as floor_dev. This
allows me to alternate between the two "versions" by replacing
object {floor_final} with object {floor_dev} to speed up rendering when
testing/modelling other parts of the scene. As predicted, the rendering time
improved visibly, but the parsing times are pretty much the same. Given that
only a small part of the scene is done, and that the floor detail level is
my
target for all other features, I'm afraid test renders will become quite
SLOW.
From the syntax "#declare" and "#while" etc etc I have to assume these work
as pre-processor directives, which would mean the code would be fully
expanded
without prior testing for actual use. This is also consistent with the
unchanging
parse time. Is there a way to make code expansion/evaluation lazy? Or
perhaps
a workaround for this matter?
Post a reply to this message
|
|
|
|
disregard this post. I figured it out by myself that #macroing the floor
instead of #declaring it solved my "problem"
"Doppelganger" <ped### [at] netcabo pt> wrote in message
news:411d6d60$1@news.povray.org...
> Hi all. I'm doing my first "real" image in POV-Ray, and part of it
involves
> a
> set of nested #whiles to pave a 22x15 area of the XZ plane (with 0.28 unit
> spacing on both axes). As the floor is reasonably complex, I decided to
> #declare it as floor_final and #declared a simple plane as floor_dev. This
> allows me to alternate between the two "versions" by replacing
> object {floor_final} with object {floor_dev} to speed up rendering when
> testing/modelling other parts of the scene. As predicted, the rendering
time
> improved visibly, but the parsing times are pretty much the same. Given
that
> only a small part of the scene is done, and that the floor detail level is
> my
> target for all other features, I'm afraid test renders will become quite
> SLOW.
>
> From the syntax "#declare" and "#while" etc etc I have to assume these
work
> as pre-processor directives, which would mean the code would be fully
> expanded
> without prior testing for actual use. This is also consistent with the
> unchanging
> parse time. Is there a way to make code expansion/evaluation lazy? Or
> perhaps
> a workaround for this matter?
>
>
Post a reply to this message
|
|
