Title: RNDCubesIFS Post by: Horbgorbler on September 30, 2013, 03:06:59 AM I have been playing around with the new formula: RNDCubesIFS. It is a wonderful algorithm!!! Thank you, thank you. This feature allows for the creation of amazing greebles or nurnias. Now that I have seen this, however, it makes me very greedy for even more flexibility. Some of the other IFS formulas in Mandelbulb 3D allow the user to select from a whole range of geometric shapes. I am thinking, for example, of the tiling formulas where the artist can change the “order” of the tiles and thus significantly alter the final shapes of the tiles that constitute a surface or volume – from triangles to pentagons to dodecahedrons.
1: Could an option be added to RNDCubesIFS to create not only cubes, but a whole spectrum of shapes: spheres, stars, specified n-gons, curved entities such as tubes, ovals, etc.? Would this require a huge mathematical effort? 2: Could the surfaces of the shapes be embellished with height-maps? Imagine what the final images could look like if the sets of the randomized cubes could be textured with patterns based on color images!!! I apologize in advance for my greed, but in my work in professional chemistry, I have often found that my math colleagues found it rather “trivial” (genius from my point of view) to compute things which I could easily imagine but viewed as astounding and quasi-magical. If you can randomize cubes, what is the impediment to randomizing other categories of shapes? If you can paint arbitrary bit mapped textures on a sphere, can you do this on randomized cubes and other types of geometrical volumes? Title: Re: RNDCubesIFS Post by: DarkBeam on September 30, 2013, 10:07:05 AM I have been playing around with the new formula: RNDCubesIFS. It is a wonderful algorithm!!! Thank you, thank you. This feature allows for the creation of amazing greebles or nurnias. Now that I have seen this, however, it makes me very greedy for even more flexibility. Some of the other IFS formulas in Mandelbulb 3D allow the user to select from a whole range of geometric shapes. I am thinking, for example, of the tiling formulas where the artist can change the “order” of the tiles and thus significantly alter the final shapes of the tiles that constitute a surface or volume – from triangles to pentagons to dodecahedrons. 1: Could an option be added to RNDCubesIFS to create not only cubes, but a whole spectrum of shapes: spheres, stars, specified n-gons, curved entities such as tubes, ovals, etc.? Would this require a huge mathematical effort? 2: Could the surfaces of the shapes be embellished with height-maps? Imagine what the final images could look like if the sets of the randomized cubes could be textured with patterns based on color images!!! I apologize in advance for my greed, but in my work in professional chemistry, I have often found that my math colleagues found it rather “trivial” (genius from my point of view) to compute things which I could easily imagine but viewed as astounding and quasi-magical. If you can randomize cubes, what is the impediment to randomizing other categories of shapes? If you can paint arbitrary bit mapped textures on a sphere, can you do this on randomized cubes and other types of geometrical volumes? Greebles? Nurnias? Whaaa? ;D Btw. You can actually create 4 variations of shapes getting the current version of the formula - spheres are already possible! Planned additions; 1. Levitation factor, to make shapes fly in z directon. 2. More shape types that I will decide soonish. I was thinking of n-agonal prisms too. But there are complications; because a triangular prism would look bigger than a square (also this problem may be present for pentagons - etc?). So edventually a fixfactor would be needed. This is annoying to me, ... A paraboloid variation would be greatly simpler. Also I was thinking of helical shapes generalized: BUT!!! this brings the same problem(s) as ngons. Normally, to write that in C would be easy as drink water; but I dothis in assembly with a limited number of resources available. So the greedy person would be me not you! ;D ;D ;D 3. Absolutely no!!! I will not add heightmap support - because it's far too complex for me. Title: Re: RNDCubesIFS Post by: Horbgorbler on October 02, 2013, 01:19:23 PM Thanks Darkbeam for all you do to improve MD3D.
I am completely unclear about which of the 10 parameters to change to alter the shape from cubes to something else like a sphere. Could you please explain? Also, just for the record, Greebles or Nurnies are actual terms used by some graphic artists. Wikipedia: "A greeble or nurnie is a fine detailing added to the surface of a larger object that makes it appear more complex, and therefore more visually interesting....Greebles are often present on models or drawings of fictional spacecraft or architectural constructs in science fiction and is used in the movie industry (special effects). All the best to you. Title: Re: RNDCubesIFS Post by: DarkBeam on October 02, 2013, 06:19:59 PM Really you don't imagine? :) "ShapeType = 0"... If you change zero to 1, the shapes won't look anymore the same...
Btw read the formula info pressing "i" button; Simple DIFS formula that (using pseudorandom numbers) generates an uniform pattern of cubes on XY plane. Vtn values near to zero may be a little noisy on edges due to some simplifications I made. You can choose different shape variations if you change Shape type; 0 = Rectangular prisms 1 = Cones (or cylinders) 2 = Spheres (or cylinders) 3 = Pyramids (in this case you can use Vtn > 1 for less sharp solids) 4 = Solid arcs (new added today!) ;D Title: Re: RNDCubesIFS Post by: Horbgorbler on October 04, 2013, 10:58:19 PM I feel very dense. My problem is that I see no field called ShapeType. I see 10 adjustable fields:
XVtn (0-1) Scale Z add Yadd Xadd YVtn (0-1) Apply scale + add ZmaxSz RandomizerX RandomizerY Am I using an older version? Or looking in the wrong place? Title: Re: RNDCubesIFS Problem Solved!!! Post by: Horbgorbler on October 04, 2013, 11:20:28 PM So sorry, I must have been using an older version. I downloaded the latest version and installed it in my formulas folder. Now I have 11 fields and ShapeType is at the bottom of the list. It's fantastic and works exactly like I imagined in my first post. Boy am I going to have fun exploring all the new shapes. You are a genius. Thanks for being patient with me. Now I am worried that I may not be using the latest versions of many other formulae... |