Help with the burning ship 3d variant

[pierre]

Hello all,

I've asked a question before about rendering the mandelbulb, and got some nice help, so i thought i'd come back to the forum, now that i have (a lot of) trouble rendering the fractal described here:

http://www.fractalforums.com/theory/here-are-the-first-true-3d-mandelbrot-images/

(I mean what people refer to as "the second formula" in this thread).

However this time my question is quite different: i'm not struggling to get proper shadowing or something, i just get a picture that is *completely* wrong. I'm using a ray-tracer which I wrote myself, and now it's capable of producing an almost decent mandelbulb, and very nice quaternion julias, so it's kind of working. However with the n=2 version of the fractal just mentioned, i get the following ridiculous result (camera at (5, 5, 5) pointing at (0,0,0)):



(It's in a very small resolution because I've tried a very slow way of computing, which uses no tricks for speeding up, in the hope that the picture would become right... and no luck)

Because i'm so ashamed of the above, here's a julia set that vaguely looks like something:



Terrible.

I've tried many things and i'm out of ideas. So besides asking for any suggestion as to what may go wrong here, what would really help is a picture of the "mandelbrot" (as opposed to julia) for this fractal, together with the coordinates of the camera and where it is pointing at.

A picture of a julia set together with the julia coordinates (and camera position etc) would be awesome, too.

Any test i could run to find out what is going on?

thanks!
pierre

[pierre]

I'VE GOT IT!!!!

I must have spent 10 hours on this at least today. It turns out that

THE C FUNCTION "ABS" IS NOT THE ABSOLUTE VALUE.

Instead, it returns the nearest integer to the absolute value.

AAAAAAAAAAAARGHHHHHHH!!!!!!!!!!

Surely this was the source of everything. I'll try tomorrow...

[pierre]

this is more like it:


what a complicated shape!

while i'm at it, does anyone know good places to put the camera, or good julia coordinates?

[miner49er]

Aw man, don't be ashamed! I like that pic!

I really must get back into coding up some fractals...trouble is, I'm having too much fun using Mbulb3D :-)

[pierre]

thanks!

it's working fine now. Here's another shot of the same object, much zoomed in:

[visual.bermarte]

I like it  is it available somewhere?

[pierre]

I like it  is it available somewhere?

If you're asking whether a high-res version of the picture is available, then no, sorry...

[pierre]

... but thanks for your comments

[DarkBeam]

Look here too; http://www.fractalforums.com/gallery/3d-burning-ship/

[M Benesi]

Look here too; http://www.fractalforums.com/gallery/3d-burning-ship/

  Also, I thought you or Jesse implemented it for Mandelbulber.  Additionally, I wrote a formula for ChaosPro which is available at the ChaosPro database.

[DarkBeam]

Use _absz or _absy or both
else there is a general formula... hours of coding and nobody used.