New subject line, old problem. An update to my illum/glow learning experience has been posted:
http://www.rumblestrip.org/rad/rad_01.html
-RPG
Fitzsimmons, Rob wrote:
Hey Rob,
I'm amazed you're getting those quality renders in only 12 minutes?!
(aside from the coffer bug) they look great.
Is the speed increase due to using mkillum?
Hi Rob,
It's true!! Even less than 12 minutes. Yes, it's because of mkillum. If you think of mkillum as a tool to generate an "ies file" for a complex bit of geometry, it makes a lot of sense (to me). You move the coffer's contribution to the room illuminance from the indirect calculation to the direct one, which is much faster.
So now I have a great looking room, but a bad-looking coffer. The illum bug that Georg & Greg have been discussing also seems to have shown up for a visit in my current renderings. =8-\
Before I can use Greg's tip about excluding the coffer materials, I need to learn a bit about coming up with a valid -av number. But, to be honest, this already gives a good impression of the space. The whole point of the rendering was to see how bad the glare would be from the coffers (answer: bad), which can be judged just fine from the direct component of the glows.
That said, I would still like to get the coffers looking better. I will have to revisit this problem when I have time...
OK, this is the last of it here. Thanks again everyone!
http://www.rumblestrip.org/rad/rad_01.html#break4
-Rob