David Fontaine wrote:

> light_source { .... area_light <axis_1> <axis_2> size1 size2 [adaptive
> [n]] [jitter [n]] }
> The axes define the rectangular area the light_source occupies. The
> "sizes" are actually resolutions, size1*size2 is the number of lights POV
> must calculate for, more is better but slower.
> jitter "jitters" the lights a bit for each pixel to eliminate banding, but
> can sometimes give grainy results. A number after it is a multiplier used
> on the default ammount of jittering.
> adaptive makes POV use further subsampling based on contrast between
> neighboring rays to further eliminate banding. Adaptive will not go beyond
> <size1,size2> in resolution. Because adaptive can give innacurate results
> by not filling in all areas, a number after it specifies the minimum
> number of times the array is subsampled before adaptive sampling is used.
>
> example: light_source { .... x*10 z*10 5 5 adaptive 1 }
> this creates a 10x10 light along the x-z plane at whatever location, uses
> a minimum of 9 rays per sample (3x3; the corners [adaptive 0] and then one
> subdivision [adaptive 1]) and may use up to 25.

Thank you for explaining!

> if you are still confused read the docs

Just a little bit confused, but I'll think that I will try out your and Bob
Hughes' suggestions with some (or many :) renderings.

Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message