Welcome to Fractal Forums

I am new to 3D fractals but I like them very much. However, in many fractals there are surface distortions, which look like the image attached as an example. This one was rendered with Mandelbulber 1.21-1 using the Mandelbox vary scale 4D formula, but it occurs with Mandelbulb 3D as well (sometimes even without zoom). Do you know any reasonable solution for this problem?

adjust the steping multiplier, it is a classy problem of overstepping, i am unsure how it is called in mandelbulber

several params deal with the problem:

in mandelbulb3d they are in the calculation tab
DE STOP= minimum distance where to stop marching
RAYSTEP_MULTIPLIER =current step is multiplied by that value to avoid overstepping

good luck

Yes, this is doing the trick. Thank you very much.  :)

In Mandelbulber, I have set the DE step factor from 1 to 0.01. Render time increased 3-4 times, but the image looks much better now.

nice, yes it is a severe quality setting, sometimes the rendering time increases big time, so the rule of engagement is to explore with low quality settings, and final overnight rendering with fitting values for the quality, not too much but not too low as well ;)

Yes, this is indeed a good idea. ;)

Some comments on DE Over Stepping / black holes with Mandelbulber 1.21.

Quote from User Manual

 https://sites.google.com/site/mandelbulber/user-manual
 “When DE step factor is too high, it is visible on image. The quality is then poor, some details are missed, and a lot of black spots are visible. But when this factor is too low, rendering speed is very low and quality is not better than when this factor is optimal”

(NOTE : I only have a limited understanding how the DE works and the following comments are from my suspect memory.)

The optimal setting of DE step factor can be somewhat ascertained by viewing the green histogram (refer User Manual) and Missed DE % while rendering, (not available in OpenCL mode).

As a general rule the Missed DE% generally reduces as you lower the DE step factor.

When I first tried to understand the DE step factor while I was trying hybrids, I would graph the results from DE step factor 1.0 down to 0.1. Generally  as the DE step factor was reduced the Missed DE% would start off reducing then roughly level out, after which any further decrease in DE step factor only achieved a minimal amount of reduced Missed DE% with little to none visual improvement, (not worth the extra rendering time.)

Now some lucky fractals start with Missed DE 0.000%.

With some others a useable Missed DE% can be achieved with significant visual improvement by just dropping DE step factor from say 1.0 to 0.7.

At other times the graph would roughly level out at numbers like, say 4.254% (generally a bad image), or say 0.169% (sometimes acceptable sometimes not).

BUT there seems to be no fixed rule as with some hybrid fractals the graph hit a minimum then increased again, e.g. DE step factor 1.0 = a Missed DE of 0.102% , then .. 0.5 = 0.012%  .., and  .. 0.3 = 0.086%. ..

With a few fractals DE step factors greater than 1.0 were tried, resulting in a quicker render time with a satisfactory image. Whereas with other fractals, even though the Missed DE% stayed low, the image quickly visually deteriorated as the factor increased above 1. Nothing beats a Visual check.

The Detail Level (Quality setting) affects the Missed DE% so sometimes you have to reduce the detail to obtain a low Missed DE%. (And for some images reduced detail looks better.)

Also the image improves when you render at the increased final resolution (say 3200 x 2400). However sometimes it improves greatly such that the Missed DE% doesn’t seem so critical, whereas other times the Missed DE% is critical to achieve a satisfactory image (depends on the fractal.)

I generally run at 0.7 DE step factor Linear DE Mode to start with, and maybe adjust later if necessary. It should be noted that at times there seems to be no apparent difference with having Linear DE Mode ticked, however with some hybrids the difference is very dramatic.