Modern 3D animation softwares like blender and maya use quads instead of triangles in contexts of modeling. Is there a computational reason to that, or is it just because it’s easier for human intuitions?
sberkun
It's for humans. Quads are generally easier to work with for common operations, such as applying subserf modifiers. One notable property of quads is that they often form edge loops and face loops, which make it much easier to deform a model uniformly (for example, if you're modelling a person, you can make their arm longer by selecting a face loop, stretching it, then adding a bunch of edge loops to make the geometry uniform again).
I found this article really helpful for explaining triangles vs quads.
Staffjamesfong1
@sberkun That's right! In fact, later on in this course we will talk about different subdivision surface methods. One such method is Catmull-Clark, which is specifically designed to work well on quad meshes.
Most software will internally convert quads (and other N-gons) into triangles before drawing them to the screen. This makes the drawing code simpler, since it only needs to deal with triangles as inputs.
Staffmcallisterdavid
@ZiqiShi-HMD think about what happens when you draw a line between opposite vertices in a quad!
Modern 3D animation softwares like blender and maya use quads instead of triangles in contexts of modeling. Is there a computational reason to that, or is it just because it’s easier for human intuitions?
It's for humans. Quads are generally easier to work with for common operations, such as applying subserf modifiers. One notable property of quads is that they often form edge loops and face loops, which make it much easier to deform a model uniformly (for example, if you're modelling a person, you can make their arm longer by selecting a face loop, stretching it, then adding a bunch of edge loops to make the geometry uniform again).
I found this article really helpful for explaining triangles vs quads.
@sberkun That's right! In fact, later on in this course we will talk about different subdivision surface methods. One such method is Catmull-Clark, which is specifically designed to work well on quad meshes.
Most software will internally convert quads (and other N-gons) into triangles before drawing them to the screen. This makes the drawing code simpler, since it only needs to deal with triangles as inputs.
@ZiqiShi-HMD think about what happens when you draw a line between opposite vertices in a quad!