Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jeoshua]

... now that you mention it, higher dimensional perlin noise might be an interesting way to handle the ore deposits.  If done correctly, it would very adequately simulate the "bubbles of magma" that make up a planet, and slowly bubble to the surface.  Veins could be added by algorithm later, and after the tectonic deformations.

So yeah, not a bad idea.  The only reason that it's difficult to create a 3d stucture of ore veins, at this point, is that we're starting what is fundimentally many 2d maps governing one parameter apiece (height, temperature, vulcanism, etc).  If the map generation, instead, started as a 3d process and STAYED as a 3d process, the whole endeavor of a 3d volumetric map would be a much more manageable thing.

But how to give a Perlin Noise function some kind of "top" is the real question there.  It could easilly be made into a general density map, but it's fractal nature doesn't lend itself easilly to being cut off at the top, as a volumetric height map would necessitate.

[NW_Kohaku]

I'm unfortunately not terribly familiar with Perlin Noise, so pardon me if I just sort of nod along on that part. 

As for how to handle these sorts of folds and intrusions, however, it wouldn't be terribly hard to put into the current game.

We already have a measure for the area's volcanism (which is basically created out of its own noise map), and we can then say that, for a given level of volcanism, there is a given chance of any one spot you randomly picked having a volcanic intrusion formation in the area, or else there is a certain density of intrusions.  Highly volcanic areas would be more likely to have a batholith overwriting the stone layers.

As for folding, you just need to put them along the edges of changes in the basement stone.  You just have to make some sort of way to associate certain types of stone and certain geologic pressures with specific kinds of folding, and put that kind of folding all along the interface of those two types of basement stone.

[Uristocrat]

I wonder how Toady is planning to do the 3D veins?  Ideally, they'd be worked into all this folding and such so that it would seem realistic, but I suppose that one could also simply do a 3D random walk through the appropriate layer and expand a little bit outwards from that.

[Silverionmox]

I wonder how Toady is planning to do the 3D veins?  Ideally, they'd be worked into all this folding and such so that it would seem realistic, but I suppose that one could also simply do a 3D random walk through the appropriate layer and expand a little bit outwards from that.

One of the constraints he has to work with is that they should be able to be generated from a random seed.

[Uristocrat]

I wonder how Toady is planning to do the 3D veins?  Ideally, they'd be worked into all this folding and such so that it would seem realistic, but I suppose that one could also simply do a 3D random walk through the appropriate layer and expand a little bit outwards from that.

One of the constraints he has to work with is that they should be able to be generated from a random seed.

So long as it's a deterministic process, there's no need to worry.  He just has to save the RNG state and restore it properly during generation, like he already does for all the other "random" stuff.

[tps12]

Simple enough.  Perlin noise first, write the data as a height map, THEN do tectonic deformation on it.  I've seen it done with map generators for other games and general game-creation software.

Link? I've tried looking for that kind of thing before (simulating plate tectonics), and this was the only thing I found (it actually is pretty neat, but sort of broken in some ways and with no source, despite being on SourceForge).