PDA

View Full Version : Maxwell v1.1 Textured Hard Shadows Jagged?

Sphere♥
07-13-2006, 04:37 PM
I just downloaded the Maxwell v1.1 Demo with the Maya plugin and Arroway sample textures. My first very simple test with the Arroway sample textures and sunlight resulted in jagged shadows:
http://img77.imageshack.us/img77/4425/jaggies19zl.jpg

I tried with a light emitter, and if the light is large, no jaggies, but if the light is small, then I get the same jaggies:
http://img109.imageshack.us/img109/1672/jaggies22bx.jpg

Now, the poly sphere is very high-res, 80x80, and all normals are soft and point out. I tried triangulating the polygons but it doesn't help.

Pretty bad... is it just me? Is it the Maya-Maxwell combo? Is it just the demo or perhaps is it just a problem with the textures (I tried many of the Arroway samples, same problem)?
denis3
07-14-2006, 12:38 AM
A known bug that's been there since the beta (2005). It's been flagged to the developers Next Limit on many occasions but they haven't fixed it yet. It might be a limitation of MLT - I'm not sure. A bit of a show stopper.

It's happens from the maya and 3dsmax plugins for certain - not sure about other plugin apps.
Jozvex
07-14-2006, 02:47 AM
Yes, this in my opinion is Maxwell's biggest problem right now. It's not plugin related, it's just a global Maxwell issue. It wants things to be real and so smoothing tricks don't work very well in it. The only 'solution' (problem minimizer) is to use highpoly geometry.

Believe me this comes up as an issue fairly often in the ATeam discussions and they're definitely working on it, it's just not easy.

:shrug:

I've heard from FinalRender users that FR has this issue too? (using a directional light)
Sphere♥
07-14-2006, 04:06 AM
Ok, thanks for the replies and clarifications. As bad as the sharp self-shadowing of curved objects is, the good news is that playing a little more with the demo I'm appreciating its strength with large emitters, what would be area lights in MR. With large emitters the shadows are uniformly soft, the lighting is great as are the bounces and it may just be an impression but under these conditions and with DOF it seems much faster than MR (although I guess you could avoid area lights and achieve much with FG in MR). Or perhaps it's the way Maxwell gives me the entire image gradually rather than in blocks as MR does? Oh well, it's just strange how my very first test with Maxwell produced such flagrant results. It seems Maxwell users are getting around this issue given the results I've seen.
CGTalk Moderation
07-14-2006, 04:06 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.