Is the pic above (which also looks really nice by the way) the equivalent of my 2D ortho projection? Doesn't quite look like it, but then I wouldn't know what to expect from a different angle!
Oh so it is?? But that previous grey one hardly looks like the next gold one. Not even slightly symmetric for instance. Are you sure?
i rendered a quick top-down preview (low precision, different materials):
Excellent! This is definitely a 3D version of the ortho one I did
Can I ask you a favour? Can you render it at higher resolution, and also veer the light source to one side, so that the left part of the fractal is brighter than the right, so that we can see a more '3D' like picture. Maybe also use the grey material too, as it better reflects the subtle blue sky light source.
btw, that's not simple ray tracing it's simulating all possible light/surface interactions (based on metropolis light transport),
That's all well and good, but I won't be happy until you render with full brute force photon motion simulation
to accurately emulate what really happens. Lol, just kidding...
it's simulating all possible light/surface interactions (based on metropolis light transport), spectral skylight (based on a practical analytic model for daylight) in a completely correct way, which requires 100-1000 or more samples per pixel for clean images...
How does that compare to photon mapping or radiosity out of interest?
.. for me fractals are just fodder for my first love, rendering systems
On a tangent, it's be interesting to experience 3D in general (or this world with special goggles) with inverse perspective, so that nearby objects are small, and more distant objects are big (the larger more distant objects would still be 'behind' the nearer stuff). I bet it would look really cool...
i don't think we'll ever see a 100m times speed improvement, there are physical limits (if you don't believe this, ask yourself: why stop at 100m? why not 1000000000000 quadrillion times faster?) i'd say we have another 1000x or so to go, at most.
Yes that prediction for 2040 is looking slightly optimistic now (that article was written 2 years ago). It's quite a shame the increases have been so small lately
At least there's parallel processing which will good for raytracing, and radiosity type effects (I hope).
oh and about physical limits, do you really think that the speed of light will be broken in the year 2600?
Yeah it probably won't, unless we manage to cheat space-time somehow. Haha, one can dream
it's also definitely the mandelbrot set, here's a thin slice (from y = -0.15 to +0.15) of the object:
Nice!! I'd die to zoom right into that top-left little cave part next to the spindly part to see how much of the detail has survived the move to 3D. If you have time of course. I don't want to eat up all your CPU time!
and doing analogous transformations in the other spherical co-ordinate; because of this there are actually entire families of 3d mandelbrots, depending on your choices of angle coefficient and offset.
I wish!... As good as it looks, this beast is far from the real thing, because most of the infinitely complex details haven't survived the Z axis. Unfortunately, they still look as though they've been 'smeared' over or lost completely. However, there's a chance some of the infinite detail has nearly survived in smaller portions of the set. For example, that cave part near the top looks as though there could be some interesting stuff. Again, I doubt it though.
moreover, i've been thinking of visualising the whole 4d object. yes, 4d: actually, at each point in space there is a scalar potential, which can be interpreted as density (like a cloud). since i plan to simulate atmospheric physics (in particular rayleigh scattering - which is what's responsible for blue skies and red sunsets etc.) anyway, this will serve as a really nice volumetric dataset. i'll definitely pre-render it to a grid though, because it's SO damn slow! that ought to look pretty cool, and would actually be way faster than the method i'm using here.
Ace. Can't wait to see some more pics of this thing!