From: Marcin Romaszewicz (marcin++at++sgi.com)
Date: 10/16/2002 13:31:57
The problems you are seeing are back-projection issues that plague most
graphics systems. The only machines that I know of that do this correctly
are sgi's IR series. When projecting textures, the hardware needs to do a
test to check if geometry being rendered is behind the center of
projection, but in many cases, it does not. I'm sorry to say that there is
no way to work around this in a simple way. The only way would be to make
sure that geometry behind the light is not rendered at all, thereby hiding
the back projection problem. This is currently not supported in
performer's projective lighting implementation; you would have to write
your own draw callback.
-- Marcin
On Wed, 16 Oct 2002, Frank R. Hartman wrote:
> pfGreetings,
>
> I have been experimenting with projective pfLightSources and I noticed a
> few odd behaviors:
>
> First there seems to be some light "leaking" so that some lighting is
> taking place at around 90 degrees off the lights spot direction. I have
> a large black border around my projected texture because I thought that
> the texture clamping might be to blame somehow but it did not help.
>
> Second I notice a second porjected texture exactly 180 degrees from the
> spot direction. I am not quite sure but it seems that this light
> affects polys which are facing the opposite direction. I would guess it
> has something to do with the lights pfFrustum(?).
>
> I have been able to observe these artifacts under Linux and Irix by
> simple modifications to the pfDCSes in the projective.C sample program.
> Has anybody else seen this or have any info about what I am doing wrong?
>
> Thanks for any help,
>
> Frank Hartman
> NASA/JPL
> Mars Exploration Rover Project
>
> -----------------------------------------------------------------------
> List Archives, Info, FAQ: http://www.sgi.com/software/performer/
> Open Development Project: http://oss.sgi.com/projects/performer/
> Submissions: info-performer++at++sgi.com
> Admin. requests: info-performer-request++at++sgi.com
> -----------------------------------------------------------------------
>
This archive was generated by hypermail 2b29 : Wed Oct 16 2002 - 13:32:01 PDT