PS EDIT NOTE: _ShowRenderMesh is slow and I miss understood the issue.
Wrong assumption
Hardware is very, very slow at rendering lines
WHY!!!
I notice the wireframe view is much slower compared to rendered in all software in general.
As if the instruction line is much slower to render than two unlit triangles or quad that face to the camara.
I do not understand why a simple line is way less inefficient than a triangle.
For example, if you got a million lines is hard to rotate. It is not a problem with a million triangles with PBR, and tones of post-process effects.
Lucubration: I suspect that the line instruction is an old CPU old feature that was not well implemented and probably never improved.
I notice this problem in all my 3D apps and my software as well.
Can Rhino developers ask GPU features to improve this?
Or make the lines as triangles or use the GPU post-process effect to make the outside of the triangle simulating the wireframe.
Curve drawing(which yes is a totally different process from shading polygons) was revamped in V6 to use more advanced GPU features, and if you have the hardware to actually see it(I see it on my 1080Ti,) is orders of magnitude faster, the whole display in fact is massively faster.
Rhino used to be made to run on a potato, which people have been complaining about since Version 1 when “3D accelerators” did very little. Now it exploits high-end video cards much better…but there’s a price to pay for that!
Yeah, I’ve never used that command before, but whatever it’s doing is 100X slower than (what it seems to do,) making a copy of the geometry and extracting the render mesh.