gemelli david wrote:
> 
> Another "Arnold like building".
> as you can see, there is a problem on the stairs. I increased the count
> but the problem is the same !

As far as I know this is a bug, or perhaps rather a flaw, in the
radiosity algorithm; a low error_bound does produce sharper shadows than
the default setting, but the shadows are subject to splotchiness. A low
pretrace_end helps, but even so, values below .01 don't seem to have any
significant effect compared to .01.

Probably, the code needs to be rewritten to handle things like this
better. Maybe if it worked more like photons? Dunno.

Anyway, rendering at a larger size should reduce the artifacts, but it's
a nasty workaround.

-Xplo

Post a reply to this message