I moved to California at the end of 2003. I've always wanted to learn OpenGL
but never had the time to really get to it. Then it finally happened and the first thing that
came to my mind was to experiment with some of my old algorithms used in
Malstorm and especially World'ed.
The picture on the left is a slice of Malstorm's
universe at a huge scale. I obviously found problems with the original algorithm and things
tend to cycle - if only I could have told that to my young self back in the 1990s.
Nevertheless, it was a weird
experience to see this picture as I could only explore an extremely
small part of the original world of Malstorm back in 1992...
And then I tried to create a new version of World'ed but this time
in realtime 3D and I tried to only use algorithms from Malstorm to
create 3D landscapes. I didn't expect the result to be like this...
The first two pictures are in realtime and the next two took less than a minute to render.
I found flaws in the algorithm but I am still amazed to get this sort of
results after only a week of work. And contrarily to the first World'ed, the world presented
here is infinite ! I could just walk forever in one direction and still find the same hill
if I ever decided to walk back to my starting point.
This is very promising and I am really planing to
get this research going and hope to link it with Beagle
one day...
Later on I went a bit further and played with the algorithms a bit more.
This time I went for pre-rendered graphics using the same techniques to
see if I could generate the planets I initially wanted to
feature in the universe of Malstorm. The picture on the left shows
such attempt and I was far from expecting this kind
of results.
The pictures at the bottom show close-up views of
one of the coast lines and some details of a mountain
chain.
I guess I could carry one exploring such
algorithms and techniques. But it will certainly
take me the next ten years to get it to an
acceptable level than can maybe compete against
modern software such as the excellent Terragen.
In any case I had a blast playing again with my old algorithms
created back in 1992 & 1993, just to see what they were capable of.
It was also, in a way,
a chance to finally turn a page of my life.