StrafesNET/strafe-project
7
Fork 4
Code Issues 8 Pull Requests 1 Packages Projects Releases 2 Wiki Activity

integration tests: Simultaneous collisions do not occur in a consistent order #17

New Issue
Closed
opened 2025-11-18 12:25:18 +00:00 by aidan9382 · 1 comment
aidan9382 commented 2025-11-18 12:25:18 +00:00
Member
Copy Link

Meshes created in the way as seen in fn test_scene() from test_scenes.rs do not collide in a consistent order when hit simultaneously. This is especially noticeable in a 3-way collision where one part is a floor, as this means the test may or may not fail due to a bnum div-by-zero crash related to ground contact

It's hard to know if this is also the case outside of the test_scenes setup, or if it's just the way this specific piece of code does it that leads to the inconsistent behaviour, as I've not yet tried to pin down what exactly is the inconsistent part of this process internally

Repro code can be found in the commit df1b224 at https://git.itzana.me/aidan9382/strafe-project/src/branch/simul-collision-3way

Meshes created in the way as seen in `fn test_scene()` from `test_scenes.rs` do not collide in a consistent order when hit simultaneously. This is especially noticeable in a 3-way collision where one part is a floor, as this means the test may or may not fail due to a bnum div-by-zero crash related to ground contact It's hard to know if this is also the case *outside* of the test_scenes setup, or if it's just the way this specific piece of code does it that leads to the inconsistent behaviour, as I've not yet tried to pin down what exactly is the inconsistent part of this process internally Repro code can be found in the commit df1b224 at https://git.itzana.me/aidan9382/strafe-project/src/branch/simul-collision-3way
aidan9382 commented 2025-11-18 19:02:09 +00:00
Author
Member
Copy Link

There is a proposed fix at https://git.itzana.me/aidan9382/strafe-project/src/branch/consistent-collision-order (I can't PR it due to gitea 500ing on me)

This implementation will prioritise all contacts over intersects, and if they're the same type, collisions are prioritised by model id

There is a proposed fix at https://git.itzana.me/aidan9382/strafe-project/src/branch/consistent-collision-order (I can't PR it due to gitea 500ing on me) This implementation will prioritise all contacts over intersects, and if they're the same type, collisions are prioritised by model id
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
AxisAngles CI itzaname Quaternions
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: StrafesNET/strafe-project#17
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?