Hello, I was thinking about the complexities involved with supporting non-uniform scaling. And I thought...why do I need it? Where would a good use for it be in game programming? The only thing I can think of off the top of my head is if you had some stretchy/bouncy effect on a model and didn't want to use animation data to make it happen. For that matter, how often do you use uniform scaling? Almost everything is just translate->rotate->draw anyway. Thanks.

You use non-uniform scaling every time you set up a projection matrix that isn't square.

Also, there aren't really any "complexities"... the non-uniform scaling matrix is pretty much the simplest and easiest-to-understand matrix used in computer graphics.

I don't mean for the projection matrix. I mean for models in your scene. The complexities are when you have world to local transforms and parent to child transforms going down your scene graph heirarchy.

Right? Unless I'm the one complicating it. :D

Quote:Original post by Chris81
I don't mean for the projection matrix. I mean for models in your scene. The complexities are when you have world to local transforms and parent to child transforms going down your scene graph heirarchy.

Right? Unless I'm the one complicating it. :D

You're right, but you're also kind of complicating it: Scaling tends to be the last transformation node on a scene graph before the geometry node.