Torus DM

http://www.deviantart.com/deviation/2809465/

Its a big floating space torus of the DM variety, rather than SSC, and its a *real* quartic torus and not some fake-o intersection of quadrics.

Enjoy!

Great Work!

Btw, just before you see the ring 90°from the side, the shading is a bit odd.

btw. i find it partially amusing that you guys want everything to be *real*. Why is it just a big deal if faking the damn thing works just as well? (even better at some cases)

Anyway, nice work J 👍

Originally posted by Magic.X
Great Work!

Btw, just before you see the ring 90°from the side, the shading is a bit odd.

I know... its an artifact of the edge fader... beside it looks better with it than without.. otherwise the hole looks really weird from shallow angles.

And tug... the fake torus method I used in pack 9 creates a 'lemon' type intersection which looks nasty and not very circular. There is no better faking method that I know.

Jheriko, right. Maybe i was just thinking about too simple things like raytracing a plane instead of using the d=d/(y+1) method or something like that 🙂

Blah, i don't even know what im talking about 😉

Very nice indeed. You pulled of a nice job on that one 🙂

yeah cool,hehe nice work!👍

Now time to be a math pioneer and create a septic solver.😛

Tug: the reason I personally like doing stuff the 'real' way is because it gives you control over everything. If you know exactly how something works, you can easily change things around.

Originally posted by anubis2003
Now time to be a math pioneer and create a septic solver.😛

Actually the method utilised here can be applied to any surface which can be rearranged to be in the form f(x,y,z)=0. I have raytraced x^6+y^6+z^6=1 - a sextic, a double torus (x²(ax²-b)+y²)²+z²=c - an octic, a trigonometric surface sin(x)=cos(y) - infinite degree. The limits of DM raytracing are now your knowledge of geometry, your understanding of numerical methods and the ability to keep it all running at a half-decent framerate.

Most septic surfaces that are known and have been studied are horrible nasty shapes with hundreds (literally) of singularities or are nothing more than planes with lumps or kinks in them.. so I don't expect we'll see many of them. Even order polynomials are the way to go for the most interesting geometry, n-torii, chmutovs, lots of strange cylinders. The problem is that searching the net will only turn up the same old shapes again and again... the study of higher degree geometry is very incomplete, not much work has been done on surfaces of degree>4 so we'll have to investigate the shapes ourselves if we want to create new 3D dms. There are hundreds of types of cubic though, and an even larger number of known types of quartics, so there are thousands and thousands of surfaces out there waiting to be found, raytraced and transformed into incredible AVS presets.

Yeah, but like you said - many of those aren't doable in the present state of AVS. In order to make it look right the gridsize would have to be over 100x100 which is really damn slow.

Singularities are just plain nasty in AVS.. otherwise someone would have bothered to raytrace a cone by now. But even if they weren't you would still need a very large number of iterations and a very complex camera code to do something decent with these sorts of shapes.

yes...but do you clean toilets?

Atero is proud of his new job 😉

Jheriko, AVS handles singularities nicely, via a "Zero Output" or Illegal Operation lol.

Nice Torus Jheriko.

Yeah very impressive. I like some of your early stuff when you first came here and I wliked you stuff but you have impressed loads too.

ALthough a zip would be handy as I can't stand deviants sometimes slow, make you confirm you email site, etc site.

Especially for Rovastar: a zip file.

🙂