Darren New wrote:
> That's what I essentially wound up doing. I have my own macro preprocessor
> (instead of using SDL).  I was hoping to be able to (for example) describe
> the shape of the outside walls once and reuse it on every floor, but POV's
> bounding boxes aren't magically clever.  Then I realized if I have a
> five-story building with 50 light bulbs on each floor, I'll be dead before
> it renders, so I essentially put in my own version of "light groups" to
> simply not expand the lights in the macro expansions unless the camera was
> in "the same room", but it's all manual at the moment. I was hoping I
> could have basically one expansion of the macro work for all the different
> traces. (I take dozens of traces of each room and make a walk-thru. I had
> to put closed doors between the rooms so I could turn off lights in a way
> that kept you from seeing they're off.)
> 
> I was just wondering if it was worth modifying my macros to allow lights
> in lightgroups, or whether leaving it as "the camera and light are in the
> same room so include the light" bit is the best I can do. (Which implies
> re-expanding the macros for each room, which takes several minutes because
> the macro expander is not written with efficiency in mind.)

I don't know if this is at all relevant for what you're doing, but I'll say
it anyway :)

A generic way to see if a light affects a scene:
Render a very small image for each light; each image having only one light
turned on. Take the images that aren't completely black. Those are the
relevant lights :)

I thought about it for re-renders of long animations. Render the whole
animation at a tiny size, compare the tiny images with the ones I did
before the last change, take the differing ones, re-render only those at
full size.

Post a reply to this message