PDA

View Full Version : Rank these for implementation in Lightwave-Open subdiv,micropolydisplacement,surface



prometheus
10-17-2014, 07:26 PM
Hi there!

Im throwing out some bones and see if some of you can grow some flesh on it..

What type of displacement enhancement or subdivision enhancement would you guys prefer to see in Lightwave, I reckon all..but in what order, which might be the most valuable and important one?

1.Pixar opensubdiv.
2.Surface (not geometry) displacement ala octane.
3.Micropolygon displacement.

Michael

jwiede
10-17-2014, 08:34 PM
For clarification, would the "surface displacement" feature or the "micropoly displacement" feature implicitly include bucket rendering?

We already have (bucket-less) micropoly rendering thanks to Denis, but the lack of buckets unfortunately means its RAM requirements are grossly impractical for mose use cases. Neither of the latter two features are likely to be practically useful unless they can rely on the presence of bucket rendering infrastructure as well.

djwaterman
10-17-2014, 09:02 PM
Surface displacement, the type that happens on render, like Octane.

m.d.
10-17-2014, 09:31 PM
micro polygon displacement happens at render time as well.....

both types tessellate the geometry as needed on rendertime

the only difference between the 2 is microploygon displacement can displace any arbitrary vector.... micropolygon displacement can use an RGB vector displacement map, where surface displacement can only displace along the normals... using a greyscale heightmap.

You could make a mountain for instance, but not a cave with an overhang within....surface displacement will only displace up and down the normal

Micropoly displacement makes surface displacement redundant

jasonwestmas
10-17-2014, 09:32 PM
Yeah, we would probably need buckets to render any high poly scene. The way lightwave freezes meshes currently is grossly inefficient. Of course the APS that LW has does help a little.

hrgiger
10-18-2014, 06:18 AM
open sub div. But micropoly would be a close second.

tburbage
10-18-2014, 10:48 AM
I would vote for Pixar opensubdiv -- either impl or at least compatibility, including compatible edge weighting/crease set support. Then to the extent that the FBX format provides a way to transfer that data, update the LW FBX support to match. I think this is going to become increasingly important over the next couple of years so better to set the foundation now.