From the ruby console I changed the value of Sketchup.break_edges and performed a few run time tests on big meshes. I could not oberve any time difference between true and false values of Sketchup.break_edges variable. Apparently there is a much more costly operation that is performed at the end user move completion. I wish the testers at trimble had a few large mesh cases to push the developers for a performance improvement...
That performance improvement has been in developer's mind for ages. They've improved it over the years, the question though is that it seems the real performance issue is rooted deep within sketchup itself, and improving it without breaking sketchup is a real challenge...
But this is what I've been reading here and there. I also think testers should push it harder!
All work was done in SketchUp. Steps: Import photo, trace the outlines onto a plane, trace a skeleton, displace the central parts of the skeleton, weave meshes inside the blocks. I depended largely on Tgi3D's curve capabilities.
dphases.jpg
I am also attaching the sketchup file with the 3D warehouse licesing terms (i.e free). It can be easily slapped on a plane.
enjoy
I think so, too.
I didn't notice the shade removal. It eliminates these marked problems.
Here a more difficult test. A coin. A friend posted a scanned image. His scanner has a very narrow angle for lighting. Difficult then.
Another cycles render. Just a low poly cylinder, bumps only, no displacements.
Works for distant camera, complicated scenes.
Amusing, but who would know, other than nurses and doctors? Butt, nipple, vibrator, female-to-male connection, stud aren't just words used by writers to Penthouse Forum; they are legitimate building-trades words (but, then again, the building trades industry is predominantly male....)
The two 'wireframes' of the forms overlaid shows they are not identical - but why such small differences in their forms have such dramatic affects is their smooth appearance is a mystery...Capture.PNG
