In article <slr### [at] parkerrfwicom>, 
ron### [at] povrayorg wrote:

> On Thu, 27 Jan 2000 20:30:04 -0500, Chris Huff wrote:
> >One thing I have thought of: Modify Trace() (or make a separate 
> >function) to take one sample, and from the distance of the intersection 
> >point to the ray start figure out a jitter amount to jitter additional 
> >samples.
> 
> Two of the sample images on the site I referred to use distance-based
> antialiasing.  The problem is that the image in the magnifying glass
> gets more blurred than it should because of the length of the rays.
> Differential tracing would fix that.

Yeah, but it has one big flaw: I don't understand it, so I can't 
implement it. This "distance-based antialiasing"(I think what I am 
talking about would be better described as depth or ray-length based 
antialiasing) is something I have a chance of understanding and 
implementing. :-)

Can you give a simplified explanation of how it works? I have downloaded 
the paper, but can't make a whole lot of sense out of it.

-- 
Chris Huff
e-mail: chr### [at] yahoocom
Web page: http://chrishuff.dhs.org/

Post a reply to this message