- This topic has 2 replies, 2 voices, and was last updated 10 years, 3 months ago by Karko.
-
AuthorPosts
-
-
August 29, 2014 at 5:26 pm #4200KarkoMember
Luca,
I can’t figure through experimentation what some of these functions do. I get perlin, simplex, and worley, sum, and remap. But what do absolute, fractal, interpolate, lerp and the others do?
-
August 31, 2014 at 9:48 am #4217deltalucaKeymaster
So each of the noise functions outputs a value -1 to 1:
The absolute noise combinator takes the input noise, and maps its output x into 1 – 2abs(x), you effectively end up with sharp peaks at 1, wherever the original function passes through 0.
The absolute2 noise combinator, is equivalent to chaining the absolute combinator twice, the effective result being you get sharp peaks at 1 and also sharp peaks at -1.
The fractal noise combinator, takes a single noise input, and you sample it at various frequencies, getting a weighted sum. You would generally use this one something like: octave 1 having scale=1, multiplier=0.65. octave 2 having scale=2, multiplier=0.25, octave 3 having scale=4, multiplier=0.1. So that you take the same noise function, at higher and higher frequencies, with small and smaller multipliers. The effective result being to turn a smooth function into one that is more textured. Eg: it takes http://bit.ly/1qrVKX4 and turns it into http://bit.ly/1wVAdMd
The interpolate noise combinator, takes a single noise input, and will sample it only at discrete grid points (set on the node), perfoming a linear interpolation for inbetween values. This is mostly an ‘optimization’ node, so that if part of the generation is particularly expensive because of having crazy complex noise functions, you can stick an interpolation node at the head to reduce the quality of the noise for better performance. But you can also use it to get some nice looking features as well by using particular large grid size and using something like fractal on the output.
The lerp noise combinator, takes 2 noise functions, and does a linear interpolation between them based on the output of the 3rd noise function. Eg you could have a noise function that gives nice smooth features, and one which gives really crazy high frequency spikes, and with the lerp noise, end up with a noise function that in some areas is nice and smooth, and in some areas is high frequency crazy based on the third noise function.
-
August 31, 2014 at 2:00 pm #4220KarkoMember
Thank you very much!
-
-
AuthorPosts
- You must be logged in to reply to this topic.