Logo by simon.snake - 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: Support us via Flattr FLATTR Link
 
*
Welcome, Guest. Please login or register. March 28, 2024, 06:37:06 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: Fractal Flames search algorithm?  (Read 738 times)
0 Members and 1 Guest are viewing this topic.
Softology
Conqueror
*******
Posts: 120


« on: March 07, 2011, 11:01:46 AM »

I have been getting back into rendering fractal flames.

The one thing I need advice with is how to determine if a random set of parameters is going to make a "nice looking" fractal or not.

At this stage I select random values for the probabilities and coefficient values and then do the following tests;

1. As the points are iterated I consider the test parameters bad if the values between the last and current point are too small (less than an epsilon value).  This helps weed out the images that have an instant single point attractor.

2. If the points grow in size too large from the origin then I assume the the attractor is always expanding and heading off to infinity in all directions so this is also a bad result.

3. If any of the points become NAN or INF or -INF then the parameters didn't work, so this also flags the skip this set of parameters flag.

4. During the test itrerations (I currently use 10000 iterations to allow them to settle down) I also count the unique pixels hit.  You can get a series of points that are not attracted to a single point but maybe 10 or so points.  This results in a flame with only 10 active pixels so the result is boring.

But even with these checks I do still get less than ideal results when testing random parameters.  Any more mathematical way to tell if the parameters are going to result in a better looking fractal image?

I tried the same log/log code as in the 3D strange attractors as shown here http://technocosm.org/chaos/attr-part2.html but that didn't seem to work with flames.  I am looking for a similar magic formula that applies to flame fractals.

Any ideas or tips?

I am getting some good results lately, but really want to save the user having to click a random button more times than necessary to get an aesthetically pleasing image.

Thanks,
Jason.







<a href="http://www.youtube.com/v/rfmfDGxYFeI&rel=1&fs=1&hd=1" target="_blank">http://www.youtube.com/v/rfmfDGxYFeI&rel=1&fs=1&hd=1</a>

<a href="http://www.youtube.com/v/cZILnDK87FQ&rel=1&fs=1&hd=1" target="_blank">http://www.youtube.com/v/cZILnDK87FQ&rel=1&fs=1&hd=1</a>

<a href="http://www.youtube.com/v/epTJb0dDzWg&rel=1&fs=1&hd=1" target="_blank">http://www.youtube.com/v/epTJb0dDzWg&rel=1&fs=1&hd=1</a>
Logged
Pages: [1]   Go Down
  Print  
 
Jump to:  

Related Topics
Subject Started by Replies Views Last post
JWildfire and sub flames (nested flames) Announcements & News thargor6 5 6261 Last post January 30, 2012, 11:50:45 PM
by thargor6
The beautiful world of fractal flames Film thargor6 0 1411 Last post May 16, 2012, 08:08:31 PM
by thargor6
Realtime fractal flames on GPU Fractal Flames « 1 2 » knighty 16 17820 Last post March 09, 2017, 05:05:37 AM
by mfeemster
fractal flames ? Mandelbulber Gallery mclarekin 0 1248 Last post May 14, 2013, 10:26:02 AM
by mclarekin
Are fractal flames really fractals? Fractal Flames pswanson 2 5773 Last post May 21, 2016, 05:26:37 PM
by DarkBeam

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 0.495 seconds with 24 queries. (Pretty URLs adds 0.038s, 2q)