Software Rendering School: Part I

Software Rendering School: Part I
Authors: Mihail Ivanchev and Hans Thörnquist
Description: This article introduces the series and talks about the mathematical background necessary for the upcoming software-rendering series.

Thanks for these articles on software rendering. It's nice to have people like you on this planet.

Anyway, here's what I'm up to. I'm porting the source code provided in your articles, to the BREW platform. BREW is for mobile phones. Initially I'll convert the entire set of articles as is and will then replace the floating point math part by fixed point math.

Later, maybe, I'll optimize the code for performance, on the BREW platform, as my need directs me. My prime motive, however, is to learn Software Rendering. Not that I have anything against the OpenGL people or the DirectX guys and all the rest of the good people. I know MESA exists. It's too tough for me right now. It's just my interest to know how things function under the hood.

And yeah, I've already ported the code provided in "Software Rendering School: Part I".

10-4.

Hi! congratulations for this tutorial, it's great! but i've got a doubt about the size of the matrix, you said usually matrix is 4x4 size but i don't understand why 4x4 and not 3x3 :P

Thanks for your time

oyzzo, a 3x3 matrix would only allow us to represent strictly linear transformations - rotation, scaling, orthogonal projection, but not translation or perspective projection. Adding the extra dimension lets us represent affine transformations like translation and also projective transformations like perspective projection, which are very important in computer graphics.

Reedbeta thanks for your answer now i understand!