利用者:DingTo/GSoC 2013/Weekly Reports/Week3

提供: wiki
< 利用者:DingTo‎ | GSoC 2013‎ | Weekly Reports
2013年7月5日 (金) 21:56時点におけるwiki>DingToによる版 (Week 3)
(差分) ← 古い版 | 最新版 (差分) | 新しい版 → (差分)
移動先: 案内検索

Week 3

What I did this week

  • I finished the Vector Transform node, which is now working in both SVM/OSL, and it behaves identical. It took me some time to learn about the Transform matrices we have (especially the Object space transform case) but after some explanation by Brecht in IRC I understood the concept behind it.
  • I started with the Ray Depth feature, and I have looked into 2 different ways of implementing it:
    • Ray Depth output inside the Light Path node. This way, we can get the current bounce (0, 1, 2...) as info output for the shader nodes.
    • Per shader bounce limiter. With this method, we can limit the ray bounces to an object on a per shader level, instead of only globally for the entire scene.

Next week

I still have an exam next week, so will spend some time on preparations again. As I am done with all my nodes though, which was my goal for the first half of GSoC, I am far ahead in schedule.

I will keep working on the Ray depth feature next week and hopefully have a better implementation then.

Questions

Mainly about the Ray Depth feature, am I on the right track with these or would another way of implementing them be better? And why do I have these artefacts with the feature as part of the Light Path node?