So, I decided to try to make computer games. Really? How unique. Why?

Simulations of reality have always fascinated me. With a small amount of math and a computer you can make things wobble and wiggle and oscillate in a realistic way. This was probably the reason I became a research scientis in atmospheric physics. But it turned out that solving radiative transfer in the atmosphere wasn't very exciting at all. You needed to stick to reality.

For me, a good game should be an accurate simulation of reality, just that you need to take away all the boring parts of it and emphasis the interesting parts.

  • View media
  • View media
  • View media
  • View media
  • View media
  • View media
RSS My Blogs

TBN

BergeLab Blog

Ah, the glory of normalmaping for lighting calculations!

I have previously not used normapping for any other surfaces than completely flat ones (sea-surface, buttons). The problem has been to calculate the tangent frame for each vertices. Now, I found a very nice description here

Terathon.com

on how to do it. It requires som understanding of linear algebra, but otherwise it's a quite good (and accurate) description of how to get a transformation matrix to get the lighting in coordinates that follow the texture surface of a vertice/triangle.

This really makes a difference for how surfaces look and feel. Only at very very close range is the illusion of a rough and detailed surface broke.

This would have been easier if I had not been so stubborn as to write all the code and shaders from scratch without using a framework or higher-level program. But now I know a lot more of how it works than if I just had to specify "phong" or "normalmap".

Now it's just a matter of adding normalmaps to all surface textures...

Post a comment

Your comment will be anonymous unless you join the community. Or sign in with your social account:

X