ntropy2007

I proudly finished my first 4k. It's a flame-fractal [1] sampling program. While it does not produce the most pretty fractals of all times due to a lot of kludges in the code (e.g. I kicked out the variation weight normalization), I managed to pop in all of the 16 variations of [1]. The package contains four binaries:

ntropy2007: full random. prettyness comes with low prob.
ntropy2007.horseshoes: focuses on the horseshoe variation.
ntropy2007.linear: focuses on linear variations (a renderer that supports linear xforms only could be crunched into less than 2k, btw).
ntropy2007.sjd: focuses on the spiral, julia and diamond variation. In my humble opinion it produces the most beatyfull images.

I reduced the resolution to 640x480, so the renders are not to noisy.

images

The first image is a true capture, as you can grab when running on your box. The three big ones a are hardcoded and big renders, you can make such, too, if you tweak the code a bit.
hmm, no image....

controls

Only two keys: [Enter/Return] to jump to next fractal. [Esc] to leave.

sys requirements

Should run on every linux with SDL (ya, I used SDL. I don't really see an idealistic problem with it since I only used it for pixel-dumping. but feel free to tell me via mail why it's bad). You probably must have the SDL and libc libraries installed.
My box is:

Debian GNU/Linux Etch, kernel 2.6.18.3 (32 bit)
AMD Athlon XP 1800+ (32bit, 1-core)
gcc 4.1
SDL + libc 6 libraries

compile

(from within src directory, I made no makefile)

# gcc main.c -o fourk0001 -Os -mfpmath=387 \
-mfancy-math-387 -fmerge-all-constants -fsingle-precision-constant \
-fno-math-errno -Wall -ldl -ffast-math -nostartfiles -nostdlib  \
-fno-unroll-loops -fshort-double

After that massage the file:

# ls -l fourk0001
# strip -R .note -R .comment -R .eh_frame -R .eh_frame_hdr -s fourk0001
# ls -l fourk0001

(the working title was fourk0001)

four k

# ls -l ntropy2007
-rwxr-xr-x 1 smach smach 4096 2007-07-15 14:25 ntropy2007

Thanks to Thierry Berger-Perrin (ompf.org) here's an embroided dump of the binrary: click here. And, of course, thanks to Thomas Ludwig (lyc.deviantart.com), for teaching me ifs (even if some time has passed by, I want to thank him because this is my first released ifs-prod).

license

references

[1] The Fractal Flame Algorithm, http://flam3.com/flame_draves.pdf, Scott Draves