Wednesday, February 23, 2011

Polarized Light Fractal Microscopy

More explorations of tweaking negmandel to look like colorful stained glass or polarized light microscopy.

Unfortunately even at 2048x2048 there's not enough detail for my taste, so I may return to these to crank it up a bit.











This is what the ore looks like for this type of fractal:

4 comments:

  1. These are incredibly beautiful, but I'm confused about what they are. Do you use polarized light microscopy to make these images, or are they fractal math designed to imitate the appearance of pol light microscopy? (Or some combination of fractalized modifications to pol light images??). They're absolutely fascinating... I'd love to hear more about their construction!!

    ReplyDelete
  2. Sure thing Clover, these images are purely synthetic, unmodified output from UltraFractal. The formula is in fact the common 'generalized mandelbrot' fractal, but using a nonstandard power setting of (-1.5555,0), and these locations are on the 'inside'.

    The result from using a negative-fractional-power setting is something that I call 'negmandel' and it creates these intensely varied and fractured forms all full of cuts.

    The polarized light look is achieved through careful layering (in UltraFractal) of multiple copies of the fractal at the same location, with differing iteration limits and coloring methods.

    I try to only use non-destructive layer modes, such as multiply, add, overlay and difference. The nice chromatic effect here is mostly due to difference coloring being used on one of the layers.

    There is one particular coloring method that creates a gradient that ramps towards the cut boundaries. This allows the cut edges to be darkened or lightened independently, which helps to add depth. You can see the effect of this layer in all of these images in the dark edges to the fragments.

    I hope that illuminates the process a bit! If you're keen to play with them yourself you could get UltraFractal and I would be happy to send you some UPRs so you can see how they're constructed.

    ReplyDelete
  3. Thanks for the detail. I can't say I followed very well, at this point, as I'm rather a novice to fractals, mathematically, but at least I have something to chew through now! So far, my only delve into fractals was to spend some time writing a simple fractal script in matlab... a rather inefficient script that took forever and a day to compute even a tiny 400x400 pixel monocoloured image! I might have to check out UltraFractal and see what that is... I started in matlab so that I could really force myself to go back to the most basic principles of fractal construction... which is how I tend to learn best.... but I ache at the beauty of what can be created with these more powerful scripting engines!

    ReplyDelete
  4. Amazing. These are my new wallpaper.

    ReplyDelete