Does anyone know why the teapot is so famous for computer graphics? I've seen it countless times in demonstrations of 3d rendering software. Are there special properties about the teapot that make it impressive to render? Or is it just something that the graphics community adopted as standard?
Zyy7390
Probably because it’s a daily, simple object but contains diverse 3D structures such as smooth surface, sharp corners, etc.
bchee
@tyleryath That's a good point. I looked it up and it's called the "Utah teapot" because it was founded at University of Utah and is considered the "Hello World" model for an easy 3d scene with somewhat complex model.
https://en.wikipedia.org/wiki/Utah_teapot
MingweiSamuel
Googling about the Gumbo model - each of the colored patches is a different bezier surface, showing how they can be stitched together to make a complex shape - however, whenever the model actually needs details, bezier curves don't really work because they are too smooth, so there are just multiple tiny bezier surfaces stiched together such as on the tusks.
sandykzhang
How come we preferred triangles when rasterizing pixel colors but now use a grid mesh for computing bezier curves?
amandaawan
From my understanding, we actually do use triangles. The mesh seems to be there as it is easily drawn, but we later use them to create half triangles that we would presumably fill in when we rasterize pixel colors.
Does anyone know why the teapot is so famous for computer graphics? I've seen it countless times in demonstrations of 3d rendering software. Are there special properties about the teapot that make it impressive to render? Or is it just something that the graphics community adopted as standard?
Probably because it’s a daily, simple object but contains diverse 3D structures such as smooth surface, sharp corners, etc.
@tyleryath That's a good point. I looked it up and it's called the "Utah teapot" because it was founded at University of Utah and is considered the "Hello World" model for an easy 3d scene with somewhat complex model. https://en.wikipedia.org/wiki/Utah_teapot
Googling about the Gumbo model - each of the colored patches is a different bezier surface, showing how they can be stitched together to make a complex shape - however, whenever the model actually needs details, bezier curves don't really work because they are too smooth, so there are just multiple tiny bezier surfaces stiched together such as on the tusks.
How come we preferred triangles when rasterizing pixel colors but now use a grid mesh for computing bezier curves?
From my understanding, we actually do use triangles. The mesh seems to be there as it is easily drawn, but we later use them to create half triangles that we would presumably fill in when we rasterize pixel colors.