Page 1 of 1

Shading artifacts emitter noise transparency ghost

Posted: Fri Jan 03, 2014 1:54 am
by feynman
Hi,

strange triangular artifacts and a transparency ghost afflict this very simple scene. Two emitter planes left and right as softboxes and 5 emitters inside the white shell are used, pointing upwards from the grille (displacement and transparency mapped). The white shell material has no transparency, in fact the triangular artifacts and the ghost transparency appear also with the default material assigned - that is very odd. Same result when OBJ exported from MATLAB or Rhino or Alias. What am I doing wrong here?

http://bayimg.com/JAFofaaFm

By the way, is it at all feasible with a small 5 node render farm to attempt an SSS material for the shell here - or will that require some insanely high SL or weeks of render time? I remember reading somewhere that SSS and emitters don't go so well together as far as noise/time is concerned?

Thanks for any hints!

Re: Shading artifacts emitter noise transparency ghost

Posted: Sat Jan 04, 2014 12:55 am
by tom
Could you send me the scene (MXS+textures) so, we can see what's wrong with the shell getting transparent itself? Thanks.

Re: Shading artifacts emitter noise transparency ghost

Posted: Sat Jan 04, 2014 1:16 am
by feynman
Thanks, Tom. The problem went away today - the UVs on the OBJ from Alias were bad (which is odd, since Alias/Wavefront "invented" the OBJ format decades ago). But, if you don't mind - do you have any idea as to the tessellated shading that occurs with the default MR material also? Is it to do with the two emitter planes maybe that light the object from opposing sides? Thanks!

Re: Shading artifacts emitter noise transparency ghost

Posted: Sat Jan 04, 2014 3:16 pm
by tom
It's very hard to make an accurate conclusion about faceting without checking the scene. It could also be bad direction of shading normals or something else...

Re: Shading artifacts emitter noise transparency ghost

Posted: Sat Jan 04, 2014 5:17 pm
by feynman
tom wrote:It's very hard to make an accurate conclusion about faceting without checking the scene. It could also be bad direction of shading normals or something else...
Thanks for replyin'... I try uploading it to a Dropbox folder later today.