well I guess it's not even necessary to surpass the quality of a rendered 
element, but rather use the flexibility offered by various AOV passes 
optimally. Quite often it's about adding subtle things like additional 
reflections etc. that I wouldn't bother to set up an extra layer for if I'm on 
a schedule...which seems to be pretty much the point of your Video anyway. If 
used correctly you can probably even keep your CGI passes energy conserving to 
some extend (apart from minor shader mismatches like snell vs complex fresnel 
implementations etc. i guess) and I could see it very handy with Packshot stuff.

Just out of curiosity - since all the stuff that Escher does seems to be 
framebuffer based, is it possible to integrate it into a deep compositing 
pipeline? If so, it could potentially even become a very handy tool for faking 
volumetric color bleeding within eg. a generated volume noise...since as far as 
i understand it a deep depth pass basically resmbles a "voxel container" with a 
set z-step-size. Like the FumeFx illumination maps (which afaik are in screen 
space as well). could be really handy for atmospheric effects on large env shots

anyway, just sharing my thoughts.
cheers



_______________________________________________
Nuke-users mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Reply via email to