Maxwell Render | Official public forum

JD-

I've converted a nurbs cylinder object to a mesh within Rhino...when viewed in Rhino the quad segments are pretty uniform -- they look square.

When I export that same mesh object thru the plugin and open in Studio the quads on the length of the cylinder are now elongated and have become rectangular. I'm having a material issue when I apply a material to these longer elongated segments on an object.

Is there a way to keep the quads on an object more uniform. I've tried making the mesh "denser" but that eats up my computers' resources quickly leading to a Rhino shutdown.

Suggestions much appreciated.

jvm

The plugin just exports exactly what you've told Rhino to give it, except that any quads will be subdivided into two triangles, since Maxwell only uses triangle meshes. In the case of an actual mesh object, that means it will be exported exactly how the mesh is built in Rhino. For NURBS objects, whatever you have set up for the render mesh will be exported. If you are seeing some discrepancy between the render mesh for an object, and the mesh which shows up in the MXS, then there is some cached render mesh data stuck in your Rhino session, or some meshing parameter override affecting the object's mesh. I think the command RefreshShade may be helpful to get rid of any 'old' cached version of a render mesh. Also, it may be that you have the plugin's Cache MXS Meshes option on, and you are not actually exporting the altered mesh - I am not sure if changes to an object's render mesh parameters can always be sensed by the plugin or not.

JD-

Thanks...I went back and looked again at the mesh in Studio...hidden line mode shows the triangels. I was interpreting the mesh based on the wireframe mode.

Here's my rendering...I'm trying to understand what's happening with the bottles at the back where it looks like elongated triangles are effecting how the material looks.

jvm

I'd assume that issue is due to the presence of co-planar surfaces in your liquid/bottle geometry. As mesh points are represented using floating-point numbers, two 'identical' mesh points never really occupy the exact same space. If you like, have a look at this post, and the example .3dm I've linked to. There are a few principles demonstrated in that model, and one of them is modeling an enclosed liquid volume such that it will render correctly.

JD-

Thanks for the post...looked at it. It doesn't appear to be an issue with the modeling... if I were to use a clear material...like a clear glass (as per your example)it renders fine. It's when I try to achieve a "frosted" look to the glass where I have an issue.

I've spent the day looking at material combinations trying to work various maps into materials to get the effect...no luck.

jvm

Hmm...next question then, when you are making this frosted material, what does it basically look like? I mean, what types of layers, settings, and maps are you using in it?

The bottle is a translucent plastic or glass -- the actual product has a matte finish on the exterior of the bottle. I've been trying to use a material with two bsdf layers...one for transparency, the other for the matte finish. First layer is glass type material, second layer I've tried setting the roughness with no luck.

Separately, I've tried a number of combinations using grayscale maps, scratch maps, in reflectance, roughness and bump channels...with ineffective results.

Anytime I use sss it takes forever to clear and results are unrealistic...I'm at the stage where I'm just looking for something that can pass as a fairly reasonable facsimile.

jvm


jvm

Well, I tried to see if I could also duplicate this, but no luck so far. As for the material, here's something simple which is just a dielectric made with the plugin's material wizard. Roughness is 25, and the transmittance color is around [220,220,200]:

JD-

Your suggestion works well...much thanks JD.

jvm