Thanks for the comments. I like these formulas because they generate fractals with a lot of variety and they aren't fussy about generating an image. Pretty much anything goes with the two (or more) terms. You can use different terms or the same one repeated and get good results. I have tried lots of combinations of terms and I like the ones with a trigonometric term mixed with a z term the best. Using a trigonometric term generates fractals that have much more intricate features. I forgot part of the equation I posted and the actual equation has an offset constant added to the pixel constant (see PAR file below). The offsets move the origin of each term so they overlap but don't sit right on top of each other. To find this fractal, I was exploring zooming into an area with the "wrong" symmetry. This fractal has z^3 as its dominant term so zooming into 3 way symmetry will always end in a z^3 minibrot. Over the past few years I have been exploring deep zooms into areas with other than the dominant symmetry. For this fractal, I zoomed into an area with two way symmetry to see what was there. In the past, I have found lots of interesting stuff when doing that and this fractal continued the pattern. You would be right if you guessed that there cannot be a minibrot at the center of the two way symmetry. In most of the cases I have looked at the fractal died out but not until you zoom far in. Once the area died out I picked an area of 3 way symmetry off to the side and zoomed into that. All of the good detail along the way was "copied" into the zoom of the 3 way symmetry. The fractal I posted was one of the nice areas along the way. I originally didn't post the PAR file because the image is a deep zoom and needs arbitrary precision math for a formula type that is not available in FractInt. Zooming out until the depth is compatible with double precision allows FractInt to be able to generate the fractal. Here is an image of the two way feature I zoomed into and a FractInt compatible PAR file with the exact equation. You can zoom out more if you want to see the parent fractal. Here is a link to an image: http://dl.dropboxusercontent.com/u/33642054/image/post_20140219_zout.jpg The Fractint compatible PAR file for the image is: 20140219_zout { ; Exported from Fracton. reset=2004 type=formula formulafile=fracton.frm formulaname=F_20140220_0805 passes=1 float=y center-mag=-0.1900505137602805/-1e-51/44246396.029\ 66739/1/0/0 params=-1/1/0/3/-1/0/1/0/0/0 maxiter=2000 inside=0 periodicity=6 colors=0000000000000000000000000000000000000000000\ 000000000000000000900J20V40d40n6cze0q70g60W30M30C2\ 00007D0FR0Of8at4_s0Nc0FQ06C302F0BT0Nf0YtCls6ie0ZS0\ ME0C00007D0FR0Of8at4_s0Nc0FQ06C302F0BT0Nf0YtCls6ie\ 0ZS0ME0C00007D0FR0Of8at4_s0Nc0FQ06C302F0BT0Nf0YtCl\ s6ie0ZS0ME0C00000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 00000000000000000000000000000000000000000000000000\ 0000000000000000000000000 } frm:F_20140220_0805 { ; Similar to the parallel resistance formula a=real(p1),b=real(p2),d=imag(p1),f=imag(p2), z=0,c1=pixel-p3,c2=pixel-p4: z=1/(1/(a*z*(cos(z)-1)+c1)+1/(d*(z^f)+c2)), |z|<100 } -- Mike Frazier www.fracton.org