PDA

View Full Version : LW and their Anisotropic Shader



Malorie
12-30-2005, 05:37 AM
First of all let me say that I LOVE my LW. So this is by no means just a hack on NT or LW. It just happens to be a pet peeve of mine because I like to model mechanical parts and it keeps coming up.

When is NT going to get their anisotropic shader right?

I was all excited when I ordered my latest upgrade (I haven't upgraded from 7.5c, so it's been a while) when I saw in their list of new features "improved anisotropic shader". 'cmon, they just added layers to it from what I saw on my quick run through it. It STILL doesn't have any way to orient the effect to the model it is applied to. I mean 'cmon! Anisotropy is caused by micro stucture on the surface of the OBJECT, caused from the machining process. Why on earth would you orient the effect to the LIGHTSCOURCE?

Here. I'll offer the information on how they could fix the shader. Here's the math. It took me all of two seconds to find with a google search.

Anisotropic Info (http://www.cs.ubc.ca/~heidrich/Papers/IMDSP.98.pdf)

OK, I've vented and feel much better now. :D
Malorie

Red_Oddity
12-30-2005, 05:52 AM
Have you tried TBs Shadertree? I'n not sure it will fix anything...
(What would be nice would be the ability to use a UV map to orient the structure of the anisotropic effect, not sure if there's a plugin that does this though)

Malorie
12-30-2005, 11:37 AM
I guess I'll have to write my own. Worley had one, but not sure if they still do. But I'm not paying $200-$400 for an Anisotropic shader.

I'll just get to work on it today. I have a few ideas on how to impliment it in LScript.

Later,
Malorie

Chuck
12-30-2005, 01:52 PM
The paper at the link is for simulating anistropic shading in real time in OpenGL and the method described fails for situations with more than two light sources. Not sure it applies to the changes that would be needed to address the concerns you have with the LightWave's anistropic shader, but I've passed your comments and the link along to the team.

Malorie
12-30-2005, 08:14 PM
As I read further into that paper, I saw that also. I'm glad there is an ear out there and I appologise if I came off a bit hard nosed about it. The anisotropic sader in LW is one of those things that just keeps biting me in the butt. And when it does I go off on a tangent. :)

BTW, since my rant, I've been working on a couple of ways to create an anisotropic shader in LScript. One of my hold-ups is that I need to have LW evaluate a pixel more than once. Is there a way? Or is there someone who knows a way? Or would I need to have my script do all that math?

Could I somehow have my script check to see which pass the render is on and have the user required to use anti-aliasing. Just kind of thinking out loud here. ;)

I really do love my LightWave.
Malorie