Vista (on the Amiga, of course; the PC was pretty awful for graphics in the early 90s). Vista was a big inspiration for me, as well as other landscape generation experiments I saw in the early 90s. I don't know how much they were influenced by Musgrave and Mandelbrot, but I imagine they must have been, directly or indirectly. Later I saw Mandelbrot's 1982 book:
http://www.amazon.com/Fractal-Geometry-Nature-Benoit-Mandelbrot/dp/0716711869I'd already got the bug before I found this book, but if I remember rightly it contained some great fractal mountain renders, probably by Ken Musgrave, which further inspired me. There was a picture that had steep fractal mountains surrounding a valley filled with white haze. I don't remember much more than that, but I am sure I must have learned a few things from Mandelbrot's book. But I don't think I actually read any of Ken's research until much later, except maybe what was in that book. If there were any rendering aspects to learn, I think they may have gone over my head or I was too lazy to try them, but I was developing methodologies for rendering terrains and atmospheres based on ideas I'd seen in Vista and other bits and pieces I picked up along the way. But I am sure Ken Musgrave's work made a big difference to what was happening all around. And since then I've read some of his work.
In these early days, I actually have just as strong a memory of thinking about global illumination - I didn't know what to call it back then, but I thought of it as "environment lighting" - but I didn't know about academic graphics research, and I didn't see it being done in 3D programs at the time. I understood why, because it would be computationally very expensive, but I dreamed that one day I could write my own renderer that could do this stuff. And now I have
But of course there's no end to the ways it could be improved.
At some point I got fascinated with the idea of rendering skies that were way beyond what Vista was capable of. Fractals were central to the rendering of clouds, so Mandelbrot and Musgrave were important as they were with terrains, but it was the
lighting of atmospheres that really fascinated me, and that's where I thought the biggest leaps forward could be made. It almost looked like other programs didn't care about lighting and I couldn't understand why, but it was exciting to have ideas that I knew could make them better.
I'd always been interested in rendering planets in other 3D programs and it was only a matter of time before I tried to do the same with Terragen. About 2001 was when I decided to start. At that point of course I'd seen Ken's work on planets, and, you know, knowing that something is possible is half the battle - once you know it's possible, you just dive in and figure out how to do it. But the ideas for these things build over years, and I often don't remember where they come from. It's hard to really quantify how much influence someone's had, or even remember how much you've read, when looking back.
I suppose that if Mandelbrot hadn't defined fractals and Musgrave hadn't pioneered the rendering of them, maybe I would have been writing a general purpose renderer, fueled by those earlier dreams of "environment lighting". Maybe there isn't a huge difference, but general rendering is a hard space to be influential in because you're competing with many, many other similar products.
Matt
