GPULightmapper exclude back-face triangles while calculating bounces #52618
Conversation
|
Does this close #51638? |
It did not remove it completely although together with increasing the setting for |
|
I found an issue in the implementation of the exclusion. I will commit an improved version of this fix. |
|
I imagine that this PR fix this, right @williamd67? The house is empty inside with the boards with only one side
|
| @@ -118,13 +118,17 @@ bool ray_hits_triangle(vec3 from, vec3 dir, float max_dist, vec3 p0, vec3 p1, ve | |||
| bool trace_ray(vec3 p_from, vec3 p_to | |||
| #if defined(MODE_BOUNCE_LIGHT) || defined(MODE_LIGHT_PROBES) | |||
| , | |||
| out uint r_triangle, out vec3 r_barycentric | |||
| out uint r_triangle, out vec3 r_barycentric, out bool backface | |||
There was a problem hiding this comment.
By convention, parameters which are modified by the function to return values use the r_ prefix (like the two other out parameters here).
There was a problem hiding this comment.
Thanks for the feedback, I updated the PR and removed the extra parameter.
Edges that are at the edge of a plane, may get behind the scene and will hit back-face triangles which where included in the lighting calculations. This caused leaking of light at the edge of planes. In case a ray hits back-face triangle, it is skipped in the bounce calculations.
williamd67
force-pushed
the
GPULightmapper-bounce-improvement
branch
from
1e3ec4c
to
7c19684
Compare
|
I fixed the issue 'distance checking for back-faces was skipped' I mentioned before. Now it should fix #51638. |
|
Thanks! |
Edges that are at the edge of a plane, may get behind the scene and will hit
back-face triangles which where included in the lighting calculations. This
caused leaking of light at the edge of planes.
In case a ray hits back-face triangle, it is skipped in the bounce calculations.
Now it should fix #51638.