> are you guys picking what normals index is supposed to represent what angle manually, using your eyes, or are you letting your algorithms do it for you automatically?

> i mean, are you using your programs to identify all instances of your 26 angles on the official voxels and seeing what indexes are actually used most frequently in those instances.

I'm the one that uses the manual method. I'll put my research online when I have the time, but here it is in short :
make a test slab that has only 1 base color and the whole range of normal indexes on it, set it up in the game, make the 8 facing direction and screencap it.
Check out the different RGB values on the test voxel and put them in a table for each of the directions.
Next is to take a TS style object and give it the same base color as your reference.
For a given direction you look at the possible RGB matches. Then you pick another direction and do the same , etc... If you have more then 3 or 4 matches, that's your normal index you've been looking for...

It works much faster then I first anticipated.

The real problem is to find existing voxels from which you can obtain conclusive information. I'll continue the research today, hoping to find all the 26 angles you guys need by then.

vipr, as for doing double work : realize that it can't hurt to double check data by using a different methode. That is a proven and valid scientific method...