 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Rune" <run### [at] mobilixnet dk> wrote:
> > Not so long ago, I saw a plugin for a commercial
> > raytracer, that featured AA without moire patterns.
>
> AA without moire patterns can't exist AFAIK? It can only be reduced in some
> cases, but not be prevented in all.
<theoretically>
probably for an infinite number of samples/pixel (+some jitter) could
remove moire ...
</theoretically>
At least, if you do forward-raytracing, there's a chance, to go
without moire (look at WinOSI) ...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Mon, 1 Apr 2002 22:16:34 +0200, "Rune"
<run### [at] mobilixnetdk> wrote:
>AA without moire patterns can't exist AFAIK? It can only be reduced in some
>cases, but not be prevented in all.
Moire patterns can be detected by Fourier analysis and compensated for
using post-processing filters. Maybe such filters can be incorporated
into anti-aliasing as well?
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG e-mail : pet### [at] tagpovrayorg
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I searched again for the plugin I talked about.. I found it.. It's the
famous "final render".. Direct link:
http://www.finalrender.com/finalrender/the_product11.htm
Though of course they say nothing about how it's accomplished.
Regards,
Hugo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3ca8ce46@news.povray.org>, "Jan Walzer" <jan### [at] lzernet>
wrote:
> <theoretically>
> probably for an infinite number of samples/pixel (+some jitter) could
> remove moire ...
> </theoretically>
I don't think so. There will still be visible artifacts due to the fact
that pixels are square areas of color, even if each pixel is exactly the
color of the area under it. In my raytracer, I made an antialiasing
method that just made multiple passes over the image. After rendering
overnight, enough passes that the difference between the pixel color and
the "real" color was too small to represent on the monitor, there were
still visible moire artifacts. Sampling each pixel with a spherical
distribution instead of a flat square distribution helped some.
--
Christopher James Huff <chr### [at] maccom>
POV-Ray TAG e-mail: chr### [at] tagpovrayorg
TAG web site: http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Christopher James Huff" <chr### [at] maccom> schrieb im Newsbeitrag
news:chr### [at] netplexaussieorg...
> In article <3ca8ce46@news.povray.org>, "Jan Walzer" <jan### [at] lzernet>
> wrote:
>
> > <theoretically>
> > probably for an infinite number of samples/pixel (+some jitter) could
> > remove moire ...
> > </theoretically>
>
> I don't think so. [....] Sampling each pixel with a spherical
> distribution instead of a flat square distribution helped some.
Hence I added the "+some jitter"-part ... I assumed the jitter to
be spherical ...
OTOH: Even if, it would produce an blurred image, probably a rectangular
jitter, spanning over 3x3 pixels should suffice, to avoid moire ...
The image will loose sharpness, but moire should be gone (for infinite
number of samples)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
