mirror of
https://github.com/mlepage/heightmap.git
synced 2024-11-19 04:44:23 +00:00
67 lines
1.7 KiB
Plaintext
67 lines
1.7 KiB
Plaintext
Heightmap module
|
|
by Marc Lepage
|
|
|
|
|
|
OVERVIEW
|
|
|
|
The heightmap module uses the diamond-square algorithm to generate cloud or
|
|
plasma fractal heightmaps which can be used for terrain.
|
|
|
|
|
|
USAGE
|
|
|
|
-- import module
|
|
require "heightmap"
|
|
|
|
-- create 256x256 heightmap
|
|
map = heightmap.create(256, 256)
|
|
|
|
-- examine each height value
|
|
for x = 0, map.w do
|
|
for y = 0, map.y do
|
|
print(map[x][y])
|
|
end
|
|
end
|
|
|
|
-- define a custom height function (reusing the default but scaling it)
|
|
function f(map, x, y, d, h)
|
|
return 2 * heightmap.defaultf(map, x, y, d, h)
|
|
end
|
|
|
|
-- use it to create a new heightmap
|
|
map = heightmap.create(256, 256, f)
|
|
|
|
|
|
HOW IT WORKS
|
|
|
|
The heightmap must be a square the size of a power of two, plus one, so that
|
|
it can be evenly divided. For example, 4x4 cells will require 5x5 vertices.
|
|
|
|
First the four corners are seeded with a random value (C).
|
|
|
|
Then each square is used to set the value of its center (S) based on the
|
|
average of its four corners (plus some randomness).
|
|
|
|
Then each diamond is used to set the value of its center (D) based on the
|
|
average of its four points (plus some randomness).
|
|
|
|
The square and diamond steps continue until all values have been set:
|
|
|
|
4 S 2 D 2 S 1 D 1
|
|
C...C c...c c.D.c c.d.c cDdDc
|
|
..... ..... ..... .S.S. DsDsD
|
|
..... ..S.. D.s.D d.S.d dDsDd
|
|
..... ..... ..... .S.S. DsDsD
|
|
C...C c...c c.D.c c.d.c cDdDc
|
|
|
|
The default height function randomly displaces values by up to +/- 0.5 of the
|
|
step size. So above, the corners will be from -2 to +2, the center will be
|
|
the mean of the corners randomly displaced from -1 to +1, and so on.
|
|
|
|
|
|
RESOURCES
|
|
|
|
http://en.wikipedia.org/wiki/Diamond-square_algorithm
|
|
http://en.wikipedia.org/wiki/Heightmap
|
|
http://en.wikipedia.org/wiki/Fractal_landscape
|