Mesh3d lighting fix

[archmoj]

Possibly fixes #3353 & #2477 by applying an identical light positioning and shading algorithms for mesh3d & surface traces i.e. following the logic used in surface trace.
@plotly/plotly_js

Here is a demo illustrating the fix for reversed ranges.

[archmoj]

https://github.com/gl-vis/gl-mesh3d/compare/3353-light-reversed-ranges

[etpinard]

@archmoj https://github.com/gl-vis/gl-mesh3d/compare/3353-light-reversed-ranges looks good to me. I made one small comment in gl-vis/gl-mesh3d@9fae35a, but yeah this indeed looks more similar to how gl-surface3d handles lightposition.

That said, I did notice one odd diff in the gl3d_tetrahedra mock:

is that new "shaded" part in the bottom third of the tetrahedra face on purpose?

[archmoj]

That said, I did notice one odd diff in the gl3d_tetrahedra mock:
is that new "shaded" part in the bottom third of the tetrahedra face on purpose?

Good 👁️
I made few more changes in mesh3d and we have new baselines.

[etpinard]

@archmoj are you happy with the results now? Would this be a good time to take a second look at this PR?

[archmoj]

Yes. Now the number of changed baselines is at minimum. So, it is good to review.
Thanks @etpinard

[etpinard]

@archmoj could you explain why the green tetrahedra in

appears to be all shaded now? Where is the light coming from in that mock?

[archmoj]

@archmoj could you explain why the green tetrahedra in
[..]
appears to be all shaded now? Where is the light coming from in that mock?

Light position is the default and the reason is that those two faces have identical angle to the light source. But if you slightly rotate the scene they would show different shadings.

[etpinard]

by applying an identical light positioning and shading algorithms for mesh3d & surface traces i.e. following the logic used in surface trace.

This sounds like a step in the right direction. Let's get this in 💃

[etpinard]

thanks for fixing that new baseline 💃 !