This is fantastic for kids or new coders. Its good coding practice, makes apparent the power of a simple, elegant algorithm, and has a big visual/fun payoff. You could spend weeks extending the concept.

And for a 3d example: https://goo.gl/images/JpSXTW

edit: a friend of mine wrote a game called dragon strike with terrain gen and I can't remember any real interactive 3d being done on a PC before that: https://extralives.files.wordpress.com/2013/09/dragonstrike_...

I'm sure there was prior art in some other pc game, maybe someone remembers...

There are tons of early examples of interactive 3D depending on what techniques and technologies you want to consider.

DragonStrike is most like a flight sim, of which the Microsoft Flight Simulator series dates back to 1976-77(for 8080 and Altair!), with FS 3 and 4, from 1988-1989 being most like DragonStrike in terms of tech. Contemporary examples within other genres include: Arctic Fox. M1 Tank Platoon, F-15 Strike Eagle II, Midwinter, Carrier Command, Stunts, and Test Drive III. "Flight sim" style 3D, with wireframes, flat polys and some sprites, was considered the main way to do interactive 3D for a long time, with each renderer having to pick and choose which details to focus on - the most common compromise being to impose heavy distance fog in order to convey detail in screenshots.

In the arcade, flat-shaded, color 3D was done in Atari's I, Robot in 1983! One year later, a simplified raytrace model was used to prerender terrain lighting for Marble Madness - a subtle solution to a problem that otherwise would consume a lot of artist time. Most of the 80's in arcades was dominated by sprite-scaling games with pseudo-3D road engine hardware like Outrun and Space Harrier, though. Wolfenstein 3D and Doom are in many ways more in the pseudo-3D genre than the flight sims, since they exploit a very specific hack with raycasting that gives decent quality and flexibility.

Edit: Wolf3D style raycasting was purportedly used in Alternate Reality: The Dungeon in 1986. You wouldn't know it at a glance, though, since it's a slow-paced RPG where your view is glued to 90-degree angles and snapped to the horizontal center of the grid cells.

great recap!

iRobot was awesome, but to be fair it would have never been possible with only a general purpose cpu.

so many of those games showed great skills in optimization and design.

Now add missiles! Scorched Earth ftw.

If this is new to you, you should also investigate Perlin noise used to generate terrain. There are many examples online. It uses many of the same concepts although it's not quite as intuitive. I suspect it is more efficient, though, although that strongly depends on how good the PRNG you're using is.

It seems like the old Moon Patrol (game) mountains, but better.

Did he just re-invent fractals? That's roughly how most terrain generators work.

I think that midpoint displacement is a well known technique. If that is a fractal or not depends on the definition of fractal that you prefer.

Anyway, it's a nice article. I like the graphs with the different values of the displacement decay powers. It's probably also well known, but the comparison is nice.

Better terrain generators will use higher-order curves rather than straight lines (i.e., instead of picking an increment for the midpoint and adding a multiple of 1 - |x|, add a multiple of 1 + cos(x)) in order to avoid having visible "corners". And of course terrain generators usually work in two dimensions, which makes this slightly more complex.

But yes, the "pick your dimensionality, randomly adjust the height in the middle, then recurse" approach is very standard.

Diamond square is often swapped out for Perlin noise in terrain generators. They give very similar effects, but Perlin's method is a little more flexible along certain parameters(e.g. arbitrary offsets, zoom-in) while diamond square has more control over the nature of features at different scales, which can be used to make terrain more or less noisy.

Yes, he basically rediscovered a generalization of Brownian motion (for r=sqrt2), which is self similar.

In fact his algorithm is one way to get a Brownian bridge.

Yes, I think you could call it a fractal. The Wikipedia page linked to from the article says that this technique is also called a "random midpoint displacement fractal".

Midpoint displacement is mentioned in The Science of Fractal Images (and the Beauty of Fractal Images), and also in Mandelbrot's book Fractal Geometry of Nature.

From a phone, sorry about upside down: http://m.imgur.com/JGCpnIu

A snippet from "Science of Fractal Images"