>   Nonsense. You will never be able to achieve the same functionality that

Hey, i don't say that we should ditch whole scripting language. I just
mean that motion-paths should be given in script by giving POV control
points for bezier-curves. 

We have lots of examples for this in POV-Ray scripting language. As far
as i know, there is very few people that hand-code smooth-triangles or
bezier-surfaces in script. 

So motion blur paths should (of course), be inplemented on script, but
should only be acurately modelled by 2:nd hand utilites.



> : 3. While rendering , every time that ray intersects this motion-blur
> : bounding box, object should be jittered across it's motion path in
> : time-domain and re-tested with same ray. This would give us oversampled
> : Monte-Carlo approximation of it's blurred trail. It will also work
> : satisfactory with shadows, reflections and such.
> 
>   It will look very grainy. Like current media or focal blur with a small
> confidence (like the default one).
>   You can get a very good idea of how slow it will be by putting a plane
> at the focal_point, differencing a hole in it and putting some object behind
> it and then rendering with focal blur with a high enough confidence
> (like 0.999).

Well, this is not a good comparision. Focal blur shoots plenty of
redundant rays. Rays shot against motion-jittered objects would only
affect space surounded by it's bounding box. The smaller the object is ,
the faster it will go.

As far as i know, and judging to all computer graphic pappers i have
read up to date, this is easyest way to implement motion blur. At least
in complexity/visual appearence terms.

Fancy post-processing "look-alike" won't cut it in most of cases where
real-looking scenes are required.

Only better solution that springs to my mind is to construct algorithm
that is calculating integral over time-space for given object and in
such way determine visibility in certain point at certain time. This
would be really be a grandious project, not to mention different
algorithms for different types of objects. And even if we could do this,
it would probably involve heaps of numeric calculations anyway, so it's
probably better to shoot for Monte-Carlo approach anyway.


Cheers, Goran





 
> --
> main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
> ):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message