Jaime Vives Piqueres <jai### [at] ignoranciaorg> wrote:
>    Look at the images I just posted on p.b.i. (just forget about the ugly
> compression) : the number on the file names reflect the cases I listed. Look
> at the second image and you will see artefacts on the edges between
> surfaces. These can be worse in "real" scenes with finer radiosity settings,
> appearing as horizontal lines on the shadows.

We're actually talking about two different effects here:

(1) The classic samples-taken-during-final-render artifacts, which are those you
describe as "appearing as horizontal lines on the shadows".

(2) The wrongly-colored edges seen in your second image, causing shadows or
highlights right at the edges; so far, I actually have no clear idea where they
come from, and it seems to me they're not necessarily related to an insufficient
pretrace. I currently suspect "crosstalk" between the two adjacent surfaces
meeting at the edge but haven't made much progress in eliminating them, so it's
an interesting thing for me to see that the focal blur seems to kill them very
effectively.


Maybe it's all about the strict sampling pattern of left-to-right and
top-to-bottom, which might cause edges to be "dominated" by the top and left
surfaces. As a bit of "crosstalk" cannot be avoided, this might lead to the bad
edge shading observed. Breaking up this strict pattern, as focal blur does,
might result in a better balance between the surfaces.

Post a reply to this message