Reply With Quote
Surprisingly good looking mountains you have made here. The problem I always have is that the mountains will not be exactly where I want them. but these sure look better than the ones I had created with the basic clouds filter.
Oh, yeah. I've just never used any of the third-party filters very much. I don't know why, I really don't, cuz some of them are quite good.
If the radiance of a thousand suns was to burst at once into the sky, that would be like the splendor of the Mighty One...I am become Death, the Shatterer of worlds.
-J. Robert Oppenheimer (father of the atom bomb) alluding to The Bhagavad Gita (Chapter 11, Verse 32)
My Maps ~ My Brushes ~ My Tutorials ~ My Challenge Maps
Surprisingly good looking mountains you have made here. The problem I always have is that the mountains will not be exactly where I want them. but these sure look better than the ones I had created with the basic clouds filter.
Looks suspiciously like a ridged multifractal type noise function with maybe an exponential post effect. I keep meaning to write some filters for Photoshop if I can get a few weeks of free time (too many other things in the channel ahead of the Photoshop filters).
I did the attachments with Wilbur's Filter>>Fractal Noise filter followed by Filter>>Mathematical>>Exponent with the default values. Generating a few large ones of these and loading them will allow for the same effect without direct synthesis in Photoshop. Wilbur also lets you save the images as 16-bit PNG surfaces so that might get you a slightly better effect in some cases.
Do you know, Waldronate, I really wish I knew what all these functions meant. I've been playing with mapzone today and 'abs sum' and other weird names mean nothing to me. The only way I know what they do is to use them and play with the sliders, and then it's only really a vague feel rather than thinking, 'ah I need this effect' I need an 'abs sum' function. I really admire how you can see a pattern and immediately are able to spot its provenance to a mathematical function.
20+ years of looking at such things and implementing them in code will tend to do that to you... Of course, I'm not good for much else, but that's a personal problem.Originally Posted by ravells
In this case, the diagnostic features are the winding ridgelines. The most common noise function basis out there is the Perlin lattice noise, which has a nice cubic interpolant (slices look like cubic curves). The absolute value function gives ridges where the basis goes from positive to negative. From there it's just scale and add as shown in the animation below (just a change in octave). An exponential will bring the mid-range parts down lower while leaving the highest parts intact.
There are suprisingly few primitive operations involved in many of the standard filters. Add, multiply, convolve, displace, FFT noise synthesis ("tileable clouds"), voronoi, median, and a half dozen or so others. It's fun to decompose the common filters to see the operations involved. The jiggle filter, for example, is very roughly a low-octave noise used as input to a displacement filter. The "bathroom glass" type of filter is the same sort of thing but with a higher-frequency noise pattern. The mosaic filter is a random Voronoi image colored by the average color in that area of the image. The stained glass filter is a mosaic filter blended with the underlying image and with a glass displacement filter applied. And on and on.
Beautiful, pal! I love it.
Posting Permissions
