HOME | DD
lyc — 3d mandelbrot
Published: 2008-04-30 06:58:04 +0000 UTC; Views: 29826; Favourites: 56; Downloads: 1750
Redirect to original
Description
designed to work as a mug
credit for the fractal iteration goes to twinbee over at fractalforums.com!
Related content
Comments: 63
this was cpu rendered, a good few years ago it was one of the first renderings of the original mandelbulb.
it'll make its way into indigo renderer sometime, for damn sure
👍: 0 ⏩: 0
amazing object, I think I prefer the non-metallic version though..
👍: 0 ⏩: 0
because it's not art (no composition) and has many many technical flaws too; it really is a scrap.
👍: 0 ⏩: 1
The underlying code is the composition. It really is about how beautyful programmer's art can get
👍: 0 ⏩: 0
Is this thing a free element in 3d object meaning? Or some generate code which can input in some Fractal application and then output (to export) 3D mesh/object. I feel very interesting to try out setup light n render it, but seem it's not easy to get 3D model for standard 3d application (I visit that forum and don't see any chance to download or request that thing in 3d object...). Maybe it's not free... (?)
If it's free and available for almost people, pls I ask you about it. Can you please give me a 3d object or the available way to get it from code/script related to Fractal application? Thanks, not sure I'll get a answer while you put this in scrapt.
👍: 0 ⏩: 1
oh i always appreciate interest, and don't leave comments ignored
the object is given by an equation from a friend (name is "twinbee" on fractalforums.com), which i have rendered with my own program (everything in my gallery is made with my c++ programs, most of the time without photoshop). there is no real 3d mesh, no polygons, it's just one equation like for example x^2 + y^2 + z^2 = r^2 (which defines a 3d sphere) but more complex. unfortunately, this is extremely slow - an image will take many hours even on the best computers using 100% of all cpu cores. this also means it is basically impossible to make a 3d mesh of it, since it never exists in my program (it can be done, but this is a lot of work and i unfortunately have little time for fractals these days).
currently i am rendering (since most of today and last night!) a new image, with a similar equation but modified so it gives a different surface. at this time tomorrow, it should be here in my scraps too - i put it in my scraps since i feel it's not really "art", just a test / experiment without any thinking about design.
thank you for taking the time to look at my scraps, and i'm sorry i can't be of more help right now. if you really really really want it, maybe i can try to plan some time for making a small program that will (with a LOT of cpu time!) export a high resolution .obj format mesh.
👍: 0 ⏩: 1
Well, I want it a lot but I think it's not enough to take your limited time for building a custom program to export hires obj. Thanks so much for your kind, I really appreciate it. If sometime you get back interest and free time enough for building that program, it'll be great
Honestly, is there any way to made out a mesh from that equation via some existed Fractal generator? Or they can't allow people input that kind of equal for generating n exporting? Fractal programs're numerous but contain export feature not that much, and complex equation and so on... I guess no way like that(?)
Below code is full content of necessary thing or just a part of full?
// A triplex number is a '3D' number
triplex multiply(triplex a, triplex b) {
triplex n;
double pi=3.14159265;
double r = pow(a.x*b.x + a.y*b.y + a.z*b.z , 0.5);
double yang = atan2( pow(a.x*b.x + a.y*b.y, 0.5) , a.z ) ;
double zang = atan2(a.y , b.x);
n.x = (r*r) * sin( yang*2 + 0.5*pi ) * cos(zang*2 +pi);
n.y = (r*r) * sin( yang*2 + 0.5*pi ) * sin(zang*2 +pi);
n.z = (r*r) * cos( yang*2 + 0.5*pi );
return n;
}
👍: 0 ⏩: 1
well, the first thing is that a mesh will never really be able to represent a fractal, since a fractal has technically "infinite" detail (in practice it is limited on a finite computer, but still extremely "sharp"). so a mesh of any resolution will always have insufficient detail when looked at too closely.
the way i would do it, is to make a very big grid (it would have to be at least 1024 * 1024 * 1024) and then, for every cell in the grid use that "triplex multiplication" to decide if the cell is inside or outside the surface. so if you have a triplex number t, then you do something like "new_triplex = multiply(old_triplex, old_triplex); old_triplex = new_triplex", so it's just like the standard mandelbrot feedback (and the initial triplex is given from the cell co-ordinates).
sorry if that's a bit too detailed, but the next steps once you have the inside/outside should be more familiar: this can be used to define an "isosurface" or "implicit surface", so there should be many tools available that can turn it into a mesh, using the "marching cubes" algorithm. (all these terms are good for googling to find information and tools)
so there's two reasonably big parts that need to be done... if you can find a tool to convert a grid of data into a mesh (should be available somewhere), i can probably make you a program that will save such a file... it would be very slow, running maybe several days, but i don't need any memory while calculating it, only the storage on disk. well, you'll maybe have problems with the program that converts it to a mesh, but let's see first how that goes
👍: 0 ⏩: 1
OK, I get it. And no I don't get technical details in your answer but I get your idea and kinda processes to done that work. I knew basically about fractal 'cause it's very interesting n beautiful, even since I was a 6 grader I like to read about it (over 12 years ago). It's somewhat fractional dimension, not 2D, 3D or any integer, and it's self-similar in infinite level of zooming, etc...
About the step that you make grid to limit it into a finite loop of computing, depend if it reach the grid bound (sorry porbably I use wrong words/meaning here, I haven't study anything related to programing and almost forget math that I get in university 
And you say it doesn't need any memory while calculating, how about CPU usage, does it need some? Storage on disk, that's fine (unless it take too large space for storage
And for something easy n comfortable, can you make some parameter that control grip size. Example when I start to let my PC calculating that 10243 grip data, and too long for waiting, then I want stop to try lower size, is it possible for me simply change some clear number or press some button? It's quite important 'cause my personal PC's weak for now.
👍: 0 ⏩: 1
Hi, If you really want to export fractal objects to Bryce, 3D Studio etc. then you should take a look at programs by Terry Gintz such as QuaSZ - just Googling for Terry Gintz should find them I think.
👍: 0 ⏩: 1
Thanks for info, sir. I got that QuaSZ and try but can't import this 3d mandelbrot to it. I lack of many knowledge to manually build it in QuaSZ, can you spend a bit time to give me suggestion how to build this 3d mandelbrot in QuaSZ (some fractal programs also have export function but I can't manage to build suitable data for them, since format of code different for each other. I don't have experience to deal with custom formular or some way to write external data that useful in those programs). Thanks again n hope get your feedback soon.
👍: 0 ⏩: 1
Hi, unfortunately I can't help much as I don't actually use QuaSZ or any of Terry's other software. I believe there may be a Yahoo group or a Google group or a Usenet newsgroup or a mailing list relating to Terry's software so that may help - failing that try posting a question in Usenet sci.fractals
👍: 0 ⏩: 0
lol i try not to think about it, my poor poor light transport code...
👍: 0 ⏩: 1
hmmm you've somehow tried the mug through the internets?
👍: 0 ⏩: 1
nope, your fractal mug is infinitely big, so all I had to do was reach out.. now, I'm infinitely drunk..
👍: 0 ⏩: 1
i mean, ..., mother .... *finds not the words*
👍: 0 ⏩: 1
*BUMP TOPIC * , ... , that is so awesome
can you not make more of it, pls ?
👍: 0 ⏩: 1
dude, have you not realised yet that i don't get any notification if you're talking to yourself?
👍: 0 ⏩: 1
oh shit :S
was too much in love with image
👍: 0 ⏩: 1
Fantastic!
I would like to take a crack at a 3D render like that one day.
👍: 0 ⏩: 1
It would be a major project, perhaps this summer.
👍: 0 ⏩: 1
it doesn't have to be ray tracing, practice with simple functions first; do 2d inside/out testing to rasterise triangles, make some cool shapes, use barycentric co-ordinates to interpolate colours, add motion blur, hdr blooming, ...
👍: 0 ⏩: 0
I love it!! Very cool liquid metallic effect on the fractal.
👍: 0 ⏩: 1
it's copper btw (measured reflectance data)
👍: 0 ⏩: 1
👍: 0 ⏩: 0
I'd love to see a mandelbrot sculpture like this in a gallery somewhere. Only I'd be too tempted to poke it.
Nice reflections from the coloured walls.
👍: 0 ⏩: 1
it would hurt tremendously.
the reflections on the wall are due to the laws of Physics rather than any plays of my own, thus i can't take credit for it!
👍: 0 ⏩: 1
You put the walls there 
And applied the laws of physics
I think some credit is yours?
👍: 0 ⏩: 1
if i have peed further, it is because i stood on the shoulders of giants
👍: 0 ⏩: 0
How is it generalized to 3D? It's amazing
Do you think, you could get caustics from something as complex as a 3D-fractal? I know, you somewhere already wrote that such things are extremely hard to do with infinite detail... But I'm certain, such a 3D-Mandelbrot would deliver amazing caustics, when it's made out of water
👍: 0 ⏩: 1
eeeuhm i didn't create the iteration, a guy on fractalforums did, and while he hasn't explained his methods i think it's just preserving the action of a complex function z -> z^2 + c in terms of its polar representation, and doing something else (?!) with the remaining (3rd) degree of freedom available (also in the polar representation).
perhaps the code could be analysed and optimised with this kind of framework; it's extremely slow, which makes me want to pursue other approaches...
👍: 0 ⏩: 1
I see.... Well, adding a 3rd dimension isn't all that fast ^^
The interesting part with this 3D simulation is that it looks so smooth and round. All the 3D-simulations with solid bodies, I saw yet, where four-sided-radial-symmetric...
👍: 0 ⏩: 0
looks fantastic dude. have you submitted it yet? (to the comp.) really good stuff
greetz
mat
👍: 0 ⏩: 1
yeah i submitted a different version for the compo, more directed at their self-coded entry. i just submitted this different version because i'd actually like to get a mug with this printed on it!
👍: 0 ⏩: 1
haha is that the prize 
greetz
mat
👍: 0 ⏩: 1
sweet. stoked u got the job man. good for you. what exactly are you gonna be doing? as long as its not a drag
👍: 0 ⏩: 1
| Next =>