Logo by Fiery - Contribute your own Logo!

END OF AN ERA, FRACTALFORUMS.COM IS CONTINUED ON FRACTALFORUMS.ORG

it was a great time but no longer maintainable by c.Kleinhuis contact him for any data retrieval,
thanks and see you perhaps in 10 years again

this forum will stay online for reference
News: Follow us on Twitter
 
*
Welcome, Guest. Please login or register. February 26, 2020, 11:42:17 PM


Login with username, password and session length


The All New FractalForums is now in Public Beta Testing! Visit FractalForums.org and check it out!


Pages: [1]   Go Down
  Print  
Share this topic on DiggShare this topic on FacebookShare this topic on GoogleShare this topic on RedditShare this topic on StumbleUponShare this topic on Twitter
Author Topic: Another way to make a "3D Mandelbrot"  (Read 732 times)
Description: A suspension of the Mandelbrot set
0 Members and 1 Guest are viewing this topic.
Tater
Navigator
*****
Posts: 71



« on: March 28, 2016, 07:01:11 PM »

This is not a method to make "the grail" but rather a different way of making a 3D object from a standard Mandelbrot set.

Calculate many points in the boundary of the M set. These are the points c for z_n+1 = z_n^2 + c that neither go to zero or infinity under iteration. These points map to other points in the boundary of the M set when iterated, so for each c value stored, calculate c^2+c, and make a continuous interpolation between c and c^2+c. This object shows, in some sense, what happens to the boundary of the M set if there was a continuous smooth transformation of M under the suspension of the map.

If the continuous interpolation is linear, (1-t)*c + t*(c^2+c) you will get a Mandelbrot set at the t=0 end of a cylinder and the Mandelbrot set at the t=1 end of the cylinder, and a continuous transformation of the Mandelbrot set into itself in between. One could also map the cylinder to a torus, or animate the process. I suspect the animation would be fascinating.

As with other 3D chaotic sets, the suspended M set should be organized by its periodic orbits, so one could compute important skeletal points by just suspending the low order periodic points

http://www3.amherst.edu/~rlbenedetto/talks/mhc_ug14.pdf
Logged
bkercso
Fractal Lover
**
Posts: 220



« Reply #1 on: June 01, 2016, 03:47:40 PM »

Interesting idea!

(If you calculate these interpolated points after every iteration and do the further iterations with them, you get the coupled Mandelbrot set (two sets with linear coupling), which also can be considered as a 4D object:
http://www.fractalforums.com/mandelbrot-and-julia-set/coupled-mandelbrot-sets-%28cms%29/

There are some 3D pictures in the articles I linked in Reply #11 and #12, but not high quality ones.)
« Last Edit: June 01, 2016, 04:38:11 PM by bkercso » Logged
claude
Fractal Bachius
*
Posts: 563



WWW
« Reply #2 on: June 01, 2016, 04:43:24 PM »

Calculate many points in the boundary of the M set. These are the points c for z_n+1 = z_n^2 + c that neither go to zero or infinity under iteration.

Sure.

Quote
These points map to other points in the boundary of the M set when iterated

I don't think this is true at all, the dynamical plane z is different to the parameter plane c, and numerical testing seems to show no matching.

Examples:
c = -2, z = 0, -2, 2, 2, 2, 2, .... --> 2 (not in M or its boundary)
c = i, z = 0, i, i - 1, -i, i - 1, -i, ... --> cycle of length 2 (and i - 1 is not in M or its boundary)
c = 0.25, z = 0, 0.25, 0.3125, 0.34765625, 0.3708648681640625, ... --> 0.5 (not in M or its boundary)
c = -0.75, z = 0, -0.75,-0.1875,-0.71484375,-0.2389984130859375,-0.6928797585424036, ... --> -0.5 (in M but not on the boundary)
Logged
Tater
Navigator
*****
Posts: 71



« Reply #3 on: June 02, 2016, 12:59:51 AM »

"I don't think this is true at all, the dynamical plane z is different to the parameter plane c, and numerical testing seems to show no matching.

Examples:
c = -2, z = 0, -2, 2, 2, 2, 2, .... --> 2 (not in M or its boundary)
c = i, z = 0, i, i - 1, -i, i - 1, -i, ... --> cycle of length 2 (and i - 1 is not in M or its boundary)
c = 0.25, z = 0, 0.25, 0.3125, 0.34765625, 0.3708648681640625, ... --> 0.5 (not in M or its boundary)
c = -0.75, z = 0, -0.75,-0.1875,-0.71484375,-0.2389984130859375,-0.6928797585424036, ... --> -0.5 (in M but not on the boundary)"

Yes, you are right. I need to rethink this, but it is certainly true that the Mandelbrot set interpolated to its first iterate would be interesting to see. Thanks for straightening me out.
Logged
Pages: [1]   Go Down
  Print  
 
Jump to:  


Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Valid XHTML 1.0! Valid CSS! Dilber MC Theme by HarzeM
Page created in 0.121 seconds with 29 queries. (Pretty URLs adds 0.009s, 2q)