Maybe perturbation could go with a slope

[Alef]

Maybe perturbation theory could be united with slope formula which create pseudo 3D effect. This is colour (pallette) dependent but looks as full 3D or maybe even better than real 3D raytracers. Slope formula calculates pixels just next to calculated pixel and from that calculates ups and downs.
There alredy are Kalles Fractaler not shure are there so much place for another one but this could give an edge;)
 
Did a small Ultra Fractal examples with included parameters, so then UF formulas can be seen:





mbrot2_slope {
fractal:
  title="mbrot2_slope" width=1024 height=400 layers=1
  credits="Alef;12/4/2014;Frederik;7/23/2010"
layer:
  caption="Background" opacity=100
mapping:
  center=-0.61777095064903/0.6790465094562 magn=19180.711
  angle=156.5545
formula:
  maxiter=1000 filename="Standard.ufm" entry="SlopeMandel" p_start=0/0
  p_power=2/0 p_bailout=1.0E20 p_offset=0.000000000000001
  p_zmode=potential p_xfer=linear p_zscale=1.0 p_zscale2=0.005
  p_everyiter=no
inside:
  transfer=none
outside:
  transfer=linear filename="Standard.ucl" entry="Decomposition"
gradient:
  comments="slightly changed standart. IMHO good sky." smooth=no
  rotation=1 index=0 color=6303744 index=64 color=12085789 index=168
  color=16777197 index=257 color=33023 index=343 color=512
opacity:
  smooth=no index=0 opacity=255
}


mbrot3_slope {
; copyright Kerry Mitchell 15sep98
; 
; sample image to illustrate
; embossing effect
fractal:
  title="mbrot3_slope" width=1024 height=512 layers=1
  credits="Alef;12/4/2014;Kerry Mitchell;9/15/1998"
layer:
  caption="Layer 1" opacity=100 method=linear
mapping:
  center=-0.81810721128409/0.19884197484006 magn=4118.0018
  angle=34.1619
formula:
  maxiter=1000 percheck=off filename="Standard.ufm"
  entry="SlopeMandel" p_start=0/0 p_power=2/0 p_bailout=1.0E20
  p_offset=0.000000000000001 p_zmode="distance estimator"
  p_xfer=linear p_zscale=1.0 p_zscale2=0.005 p_everyiter=no
inside:
  transfer=none
outside:
  transfer=linear filename="Standard.ucl" entry="Decomposition"
gradient:
  comments="slightly changed standart. IMHO good sky." smooth=no
  rotation=1 index=0 color=6303744 index=64 color=12085789 index=168
  color=16777197 index=257 color=33023 index=343 color=512
opacity:
  smooth=no index=0 opacity=255
}

[Botond Kósa]

Interesting idea. The slopes are calculated on the distance estimates, not the normalized iteration values, so I will have to implement DE first.

[Kalles Fraktaler]

You can do it pretty well also with ordinary iteration bands:

[TheRedshiftRider]

Alef thats a very interesting idea.

That is something I want to implement!
(I only have effort to do what I feel interesting myself. Maybe I am selfish, but I did share the code )

Maybe the palette you used is a bit too dark for this.

It would look really cool to have this applied to the area Im exploring now. It is very dens but with a large image size it would look ok. Movies would also look great.

[Chillheimer]

You can do it pretty well also with ordinary iteration bands:

in kf? how?! I want to do that!!
mind sharing the palette, to show what you did?