You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Because we store a 3D velocity vector in ray data that is separate from the direction stored on the ray the velocity vector will not always be consistent with the direction if a ray were to change trajectory during a trace. This shouldn't really happen except in the case where a particle might reflect off a boundary. In this case the 3D velocity vector we store on the ray will be inconsistent with the direction the ray is actually traveling and we will need to update this vector on a trajectory change.
Steps to Reproduce
Create a particle with some velocity pointed at a boundary, set the maximum distance to be large enough that the particle will hit the boundary and condition with a different direction. The ray direction and the velocity stored in ray data at this point will be inconsistent.
Impact
This may happen on occasion but since we have not yet established the boundary conditions that particles will have when they interact with a boundary and current mapping is still underdevelopment the impact of this is unclear but it should be sorted out to ensure true consistency.
Suggested Design
Since the only time a particles trajectory should ever change is when it would hit a boundary or during collisions and velocity updates, which do not occur during tracing, it probably makes the most sense to create an extended ReflectRayBC which also ensures consistency between the velocity vector stored in ray data and the direction the ray is traveling in.
The text was updated successfully, but these errors were encountered:
Bug Description
Because we store a 3D velocity vector in ray data that is separate from the direction stored on the ray the velocity vector will not always be consistent with the direction if a ray were to change trajectory during a trace. This shouldn't really happen except in the case where a particle might reflect off a boundary. In this case the 3D velocity vector we store on the ray will be inconsistent with the direction the ray is actually traveling and we will need to update this vector on a trajectory change.
Steps to Reproduce
Create a particle with some velocity pointed at a boundary, set the maximum distance to be large enough that the particle will hit the boundary and condition with a different direction. The ray direction and the velocity stored in ray data at this point will be inconsistent.
Impact
This may happen on occasion but since we have not yet established the boundary conditions that particles will have when they interact with a boundary and current mapping is still underdevelopment the impact of this is unclear but it should be sorted out to ensure true consistency.
Suggested Design
Since the only time a particles trajectory should ever change is when it would hit a boundary or during collisions and velocity updates, which do not occur during tracing, it probably makes the most sense to create an extended
ReflectRayBC
which also ensures consistency between the velocity vector stored in ray data and the direction the ray is traveling in.The text was updated successfully, but these errors were encountered: