and thanks Blackley1, that is the only thing I would consider is missing from the tut.

I look forward to reading more!

x += 0.005f;
and
y += 0.005f;

and a good starter data is

8 Octaves
2.85 Lacunarity
.45 gain
1.0 offset

and thats all you need to get the RidgeMF working . . . Hope this helped i needed to downlaod the entire project just to get those points!

I’ve hosted it on my SkyDrive for now:

http://cid-109c9e6c1fce824a.skydrive.live.com/self.aspx/.Public/TextureGeneration.7z

I’ll look over the weekend and sort out all the broken-links. If you need any thing else, just give me a shout.

-John

Any chance you can check the filehost for the source code link? It reports that the file is not available, and I’d really like to get my hands on it.

Best,
Byron

We have talked before on Ziggyware, and you see on my link here that I’m the guy behind the Liero3d projekt

Anyway, I don’t know if i mentioned it the last time, but the reason i started working in XNA, and started with 3D, was so that i could write a planet-to-surface engine one day, and when i finally decided to start on the project, I found your tutorials on the subject extremely useful

So i wanted to give you a big -thank you- for all your great work on these Tutorials, and i of course hope to see more of them on your page in the future

Also, i think you made a good choice when moving your project to a space-themed game, as it opens up for more options with a space-to-surface engine

take care

I have to say I searched the net and didn’t found much, but I don’t know much about the subject like you, so I might of found it and didn’t know it was that.

I don’t have the book you mentioned, but I’ll look into it.

Also, if you have the time, I would love if you could comment or explain the grad function you currently have, all of it. The other parts of it are not understandable to me either.

And by the way, great article! It helps understand perlin noise (most of it), and it’s the best one I could find, so thanks!

I guess you are right in that you could put this value directly in the permutation table, when I’ve a little more time I’ll check it. There are also other functions which can be used to calculate the actual gradient, so I’ll look around the net and ad a few more links. (If you have the book ‘Texturing and Modelling: A procedural approach, it talks about ways to do this).

in your grad function you take the hash argument and use bitwise AND with 15 (1111.b). That mean that any number above 15 won’t be used fully, and doing that bitwise AND 15 can also be done on the table since you don’t seem to use the actual hash value for anything else, but rather the result hash value.

I have to also say I don’t really understand what you did in that function, or rather why. I didn’t found you explaining that function.

Minor optimization on your permutation table generation:

uchar perm[256];
// initialize with unique values 0..255
for(int i = 0; i < 256; i++) {
perm[i] = i;
}

// seed the random number generator
srand(mSeed);

// randomize the permutation table
for(int i = 0; i < 256; i++) {
// for each value swap with a random slot in the array
uchar swapIndex = rand() % 256;

int oldVal = perm[i];
perm[i] = perm[swapIndex];
perm[swapIndex] = oldVal;
}

That should make it O(n) (or is it O(1) argh!)