To keep the 3d view port fast and responsive, I keep models as low poly as possible. (duhh) However, when i import a model from say, sketchup to blender, all quads are converted/cut diagonally into tris. This essentially adds an edge for every face, and begs the question:
does an edge add “weight” to a model like a vertex does? Or am i waisting my time converting tris to quads?
Alt+J (tris to quads) often misses some tris that could easily be converted to quads, leaving me with a notable amount of manual work. But if an edge does not bog things down like a vertex does, i wont worry about it.
If someone could shed some light on the subject in general, i would appreciate it. I am here to learn! Thanks all
Why don’t you answer your own question.
Have a high poly model made with quads. Convert the quads to tris. Can you see any appreciable difference in the speed of the 3d view. I would say it is more the practicality of using quads. Subsuf a triangular mesh and see the horrible result. Controlled loop cuts in a triangulated mesh?, forget it. These should be considered more to how to model rather than possibly undetectable differences.
Yeaaaaah… i suppose this was a pretty easy one to test. I guess i just thought someone might know off hand.
The results were far from undetectable though:
a mesh of 1,280,666 quad faces was then converted to tris faces
tris made the 3d view a bit slower
tris took about 60% longer to render
tris ran an down animation fps about 60% as well
tris increased the file size of the .blend by 67%
Conclusion:
tris do add “weight” to a model/scene.