Difference between revisions of "User:Kunigami/GSoc2011/Reports"

From BRL-CAD
(Reports)
(Reports: Week 5)
Line 1: Line 1:
 
= Reports =
 
= Reports =
 +
 +
== Week 5 ==
 +
 +
# Implemented an osl shader, which seems not to give correct results
 +
# Implemented a stand-alone application to render BRL-CAD scenes with OSL shaders [[http://kuniga.files.wordpress.com/2021/06/osl-rt-2011-06-25.png screenshot]]
  
 
== Week 4 ==
 
== Week 4 ==

Revision as of 13:13, 27 June 2011

Reports

Week 5

  1. Implemented an osl shader, which seems not to give correct results
  2. Implemented a stand-alone application to render BRL-CAD scenes with OSL shaders [screenshot]

Week 4

  1. Discovered that the crashing was due to multi-threaded issues. I'm currently using -P 1 on my tests.
  2. Re-defined the OSLRenderer interface, so that it considers recursion rays needed for reflection and transmission
  3. Start studying how the rt application works.

Week 3

  1. Adapted the OSL raytracer so that it can be called by the osl shader.
  2. The rt application is currently crashing. Inspecting with valgrind I suspect that it is due to memory leaks.

Week 2

  1. Not very productive week. With the help of the mentors I managed to solve the compile error.

Week 1

  1. Implemented a (poor) polka dot for BRL-CAD, to get familiarized with the shaders syntax. [polka dot goblet]
  2. Trying to compile the OSL raytracer outside OSL build, but getting compile errors, that I have no idea on how to solve. I've emailed the dev list asking for help.

Week 0 (Community Bonding)

  1. Compiled OSL sources both in Ubuntu 10.10 and Mac OS X Snow Leopard. (had trouble with 11.04, mainly due to LLVM)
  2. Found a raytracer developed by Erich Ocean and Brecht [1] that uses OSL-written shaders. It currently renders refraction wrongly. [screenshot]
  3. Discovered the cause of the above error: I was using the wrong shader for glass. Now, the image is correctly rendered. [screenshot]