> A fairly good introduction can be found on:
>
> http://freespace.virgin.net/hugo.elias/models/m_perlin.htm

Thanks. This was helpful.

So, essentially, the Perlin noise function is built on a normal random
function where a point (in whatever dimensionallity of space) is the seed,
yielding the randomness between distant points. Then interpolation is used
to make points nearby have similar values. Oh, and the fractal bit - several
different interpolated random functions of different wavelengths and
amplitudes are added together for extra "natralness". Is that about how it
works?

Andrew.

Post a reply to this message