I understand the basics of matrix transformations, how to multiply them and and how they work in relation to viewing and projection transformations, but am having a hard time “getting under the hood”, so to speak.

For instance, the idea of a local vs. global transformations, depending on whether or not you think of transformation matricies in a pre or post-multiplication sense; makes little sense to me.

Also, I can kind of see, but not really follow all that well, the notion that you can extract from a transformation matrix the rotational and translational information contained within it.

I guess I’m looking for references which go into these finer points, because while I understand the basics involved, I know I’m missing a lot of subtleties, and it’s frustrating when you can’t follow someone’s code because the author is using row-major form instead of column-major, for instance. Thanks for pointing me in the right direction