Planetside Software Forums

General => Open Discussion => Topic started by: mhaze on October 01, 2016, 12:27:56 PM

Title: Simplex noise
Post by: mhaze on October 01, 2016, 12:27:56 PM
Interesting page with some great noise examples. https://github.com/simongeilfus/SimplexNoise (https://github.com/simongeilfus/SimplexNoise)
Title: Re: Simplex noise
Post by: AP on October 01, 2016, 06:29:07 PM
Always more noises to fuel the imagination.
Title: Re: Simplex noise
Post by: Dune on October 02, 2016, 02:15:58 AM
Looks like most of these can be made with some mixes in TG.
Title: Re: Simplex noise
Post by: René on October 02, 2016, 06:09:57 AM
I agree, even without blue nodes. The merge shader can do wonders.
Title: Re: Simplex noise
Post by: Tangled-Universe on October 02, 2016, 10:34:17 AM
Quote from: Dune on October 02, 2016, 02:15:58 AM
Looks like most of these can be made with some mixes in TG.

I would disagree:

https://en.wikipedia.org/wiki/Simplex_noise
Title: Re: Simplex noise
Post by: AP on October 02, 2016, 07:17:16 PM
I would be curious how well Terragen would possibly benefit having Simplex noise.
Title: Re: Simplex noise
Post by: René on October 03, 2016, 01:41:48 AM
So all in all it's better and faster. Still i think it's possible in TG, albeit without the 'better and faster' part. ;)
Title: Re: Simplex noise
Post by: Dune on October 03, 2016, 02:42:19 AM
Yes, with that information in background, I agree with your disagreement, but also still agree with René and myself. But why don't we have that noise in TG? Off limits? Too hard to implement? No complaint of course, just curious.
Title: Re: Simplex noise
Post by: AP on October 03, 2016, 04:41:31 AM
Also, here...

https://en.wikipedia.org/wiki/OpenSimplex_noise

Why use simplex noise?

Classic noise has problems with non-uniformity throughout its domain of definition, particularly for 2D planar slices of 3D and 4D noise, it has visible axis-aligned artefacts, it is expensive to compute for 4D and up, and its derivative in 3D and 4D is a very complicated high order polynomial.

Simplex noise is several times faster to compute, particularly for 4D and up, it does not have nearly as many visual problems with non-uniformity and axis-aligned artefacts, and it has a simple polynomial derivative everywhere, even for higher dimensions.

Simplex noise looks better, but different, and is thus visually incompatible with classic Perlin noise. The difference in feature size and range of values can easily be compensated for by a few simple scaling multiplications, but the different visual character might change the visual result of shaders that depend heavily on one or two components of noise. (Fractal sums of several noise components should still look about the same, though.)

http://uniblock.tumblr.com/post/97868843242/noise

https://gist.github.com/KdotJPG/b1270127455a94ac5d19

https://cmaher.github.io/posts/working-with-simplex-noise/

http://www.6by9.net/simplex-noise-for-c-and-python/

https://briansharpe.wordpress.com/2012/01/13/simplex-noise/

http://www.java-gaming.org/topics/simplex-noise-experiments-towards-procedural-generation/27163/view.html



Title: Re: Simplex noise
Post by: René on October 03, 2016, 09:18:16 AM
A few side by side examples.


https://clojurefun.wordpress.com/2012/08/05/perlin-noise-vs-simplex-noise/

http://imgur.com/a/pcDgj