What are the matrix values of a 16 float?

[magneto]

Hi,

I have a matrix4 parameter in a VOPSOP and I multiply point positions with this parameter. Is shear also part of this?

So right now I assume it's like this:

[1, 0, 0, 0]

[0, 1, 0, 0]

[0, 0, 1, 0]

[0, 0, 0, 1] -> translation

I have checked the help but couldn't find any info on this.

Thanks

Edited November 18, 2012 by magneto

[Macha]

Check this:

http://www.cs.brandeis.edu/~cs155/Lecture_07_6.pdf

[magneto]

Thanks Macha, that's a great reference. I just don't understand the point of the 4th component in each row. They don't seem to have any effect. I will see if I can get that bit from your pdf.

[sanostol]

it caused by rulez of matrix math, square matrix has a lot of advantages, for example a non square matrix is not invertible and has no determinant.

You can look at a 4x4 matrix like a coordinate system

[1, 0, 0, 0] x axis

[0, 1, 0, 0] y axis

[0, 0, 1, 0] z axis

[0, 0, 0, 1] offset

by multiplying a vector with this you get a transformed vector , position, scaling, rotation and shearing is possible by this

I can only recommend this book for the start:

http://www.amazon.co...=dp_ob_title_bk

Edited November 18, 2012 by sanostol

[Macha]

The 4th column helps to make the maths work out. You can then represent all affine transformations with one matrix, including projections so that everything is unified and you can do multiple transformations by just one matrix (a combination of all transformations).

I think they were originally invented to represent points at infinity without getting weird maths (dividing by zero and such things).

Thanks Macha, that's a great reference. I just don't understand the point of the 4th component in each row. They don't seem to have any effect. I will see if I can get that bit from your pdf.

[magneto]

Thanks alot guys, very helpful.