Using terrain layers to change the simulation behaviour

[alin256]

One cool idea for procedural terrain generation would be different geologic layers that affect the steering bias. So when you're in a "hard" layer, maybe it doesn't turn as sharply (and so on). It could be a color lookup like the collision detection.

Originally posted by @gradyh in https://github.com/CodingTrain/Directional-Boring/issues/2#issuecomment-1172479254

[alin256]

There are two things that I can think of that can be affected by the layers

• [ ] When steering from a softer layer to a harder layer there is resistance to steering. Which can be on top or instead of current randomness.
• [ ] When you calculate reflection response from the boulders the speed of sound in different layers is different. Ignoring the fact that there will reflection form any contrast we can account for time that sound-wave spends in each layer but use the constant of the top layer when estimating the seismic respnce.

[Djinn-Indigo]

Wait, so in that second bullet, are you talking about implementing seismic tomography? (Beyond just outlining the tops of the boulders.) I feel like that's way beyond the simplistic scope of the game... but is also totally awesome. You could even have a sensor deployment phase and a monitor; make it a whole mini game.

[alin256]

It is already implemented (1st reflection) for the boulders.

How hard can it be?

Disclaimer: my seismic guy is on vacation, so I have no idea what I am doing.

[Djinn-Indigo]

I'm actually starting school for geophysics in a few months, so it probably wouldn't kill me to take a crack at it. I'll see what I can come up with in a few days. Disclaimer: I also have no idea what I'm doing.

[alin256]

Keep me updated :)

[alin256]

@Djinn-Indigo, I have experimented with setting randomness constant to a different value depending on the dirt level.

Here is the code update FYI: https://github.com/alin256/Directional-Boring/commit/dcbaa6dc6a0866116189eadc0dffb3a4ea081867