With a little luck he may answer in this post. In general it would not hurt to try the latest Indigo version

Fair enough. Indigo has seen some improvements on this front but they are older than that.bioman666 wrote:Almost. 3.2.13, but I haven't read anything on this bump issue on changelogs to date.Your Indigo is up to date, is it ?
Thanks I'll try this later, I don't have enough time right now and using some photoshop i can get a mixed image of both bump shade frequency without all the trigangles thing, so it's ok to me so far.You can also try to offset the outer profile Edges if you're working in SketchUp. You can do that with a plugin called ToolsOnSurface by Fredo6 on the SketchUcation forums. Or you can stamp a rectangle into the surface and hide the Edges, then export. That should make the triangles smaller.
In fact when you change the frequency the height should be changed inversely (twice the frequency gets half the height).bioman666 wrote:Ok here's what i got to date. Obviously, frequency of the noise shader matters when it comes to this bump issue.
Mmm that explains a lot. Thanks for the heads up !In fact when you change the frequency the height should be changed inversely (twice the frequency gets half the height).
Do i get this right (see my translation into graphics) ?From the render I can now say that the high frequency noise features are higher than their caracteristic width, if that's making sense. It's stretched really.
You mean Oren-Nayar is a useless waste of CPU when roughness is set this high ? I've used this blended Oren-Nayar/phong material as found in the "rippled plastic" material in the Indigo material library, and I've not tested it extensively.edit: I'll add this: you formerly recreated an Oren-Nayar-like surface, with a pretty high roughness. Probably beyond the sigma limits.
Absolutely right. I'm proceeding visually myself at this stage, just ensuring that the bump mapping is not producing harsh shadows before I augment the frequency and scale the height down accoedingly.bioman666 wrote:Do i get this right (see my translation into graphics) ?From the render I can now say that the high frequency noise features are higher than their caracteristic width, if that's making sense. It's stretched really.
No, using an Oren-Nayar material would be faster in all cases I assume. I meant that there is no way to have such a rough material in Indigo withot using ISL as you did first.bioman666 wrote:You mean Oren-Nayar is a useless waste of CPU when roughness is set this high ?edit: I'll add this: you formerly recreated an Oren-Nayar-like surface, with a pretty high roughness. Probably beyond the sigma limits.
Users browsing this forum: No registered users and 11 guests