Title: Interesting result from well-known idea on ff Post by: David Makin on January 05, 2012, 03:34:03 AM Trying a new algorithm - initially I thought it would be a good fast alternative to standard Perlin noise/fBm and with some adaptations it does get close, but the basic idea is the best 2D pattern generator I've ever come across - this image is using a slight modification but the basic algorithm is a well known idea here at ff.
Anyone care to hazard a guess at the algorithm ? (http://nocache-nocookies.digitalgott.com/gallery/9/141_05_01_12_3_27_29.jpeg) http://www.fractalforums.com/index.php?action=gallery;sa=view;id=9772 (http://www.fractalforums.com/index.php?action=gallery;sa=view;id=9772) Will let y'all know anyway as soon as I release the formula to the UF database ;) Title: Re: Interesting result from well-known idea on ff Post by: David Makin on January 05, 2012, 10:41:21 AM Here's one with the adjustment switched off....
(http://nocache-nocookies.digitalgott.com/gallery/9/141_05_01_12_10_39_27.jpeg) http://www.fractalforums.com/index.php?action=gallery;sa=view;id=9774 (http://www.fractalforums.com/index.php?action=gallery;sa=view;id=9774) Title: Re: Interesting result from well-known idea on ff Post by: Sockratease on January 05, 2012, 11:01:33 AM My wholly uneducated guess would be...
Triplex Perlin Noise? Rendered in 2D of course, so probably not... Title: Re: Interesting result from well-known idea on ff Post by: DarkBeam on January 05, 2012, 11:16:28 AM Aperiodic tilings :dink:
Title: Re: Interesting result from well-known idea on ff Post by: David Makin on January 05, 2012, 11:56:03 AM Well it definitely produces aperiodic tilings - in fact a little modification could make them periodic of course ;) But I don't know if it's a standard way of producing such - it's actually just plain iterated 2D folding *but* applying randoms so that the folds are not at fixed positions, however they start at large offsets from the origin and reduce over iteration until they finally stay within the range +/-0.5 :) Oh - of course "Galaxy Test" has self-rotation added and "Pattern Test" includes fixed rotations on each iteration. |