Manta Interactive Ray Tracer Development Mailing List

Text archives Help


Re: [Manta] r2077 - trunk/Interface


Chronological Thread 
  • From: Thiago Ize <thiago@cs.utah.edu>
  • To: ollie@lanl.gov
  • Cc: manta@sci.utah.edu
  • Subject: Re: [Manta] r2077 - trunk/Interface
  • Date: Fri, 15 Feb 2008 14:44:28 -0700
yeah, I see the surface acne along the shadow boundary. I'm guessing what is happening here is a self shadowing artifact. If that's the case, then I would have expected my raising of T_EPSILON from 1e-5 to 1-4 would have actually improved this problem for you.

When you hit 'c', what is the camera position? Also, what are the dimensions of your scene? (Some of the acceleration structures will print the bounding box for you during the build).

By the way, if you want nice looking images, try super sampling by adding one of the following command line arguments:
  -imagetraverser deadline"()" or
  -pixelsampler "jittersample(-numberOfSamples 4)" (feel free to vary the number of samples to taste)
this will make things slower, but will give nicer looking results. It will also hide some of these surface acne bugs -- but fixing the root cause of the problem is always better :-)

Thiago

Li-Ta Lo wrote:
1203110353.3959.43.camel@exponential.lanl.gov" type="cite"> 
On Fri, 2008-02-15 at 13:52 -0700, Thiago Ize wrote:
  
Author: thiago
Date: Fri Feb 15 13:52:53 2008
New Revision: 2077

Modified:
   trunk/Interface/Parameters.h
Log:
Compromise T_EPSILON between the two previous revisions. If this
casuses artifacts, let me (thiago) know and I'll try and see what can
be done to fix it.

    

Take a look at the twin light zone in the snap shot. Is it counted as
artifacts or just unavoidable?

Ollie

Archive powered by MHonArc 2.6.16.

Top of page