Restore legacy pmove snapping around the world origin #62
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting
| static void PM_SnapPosition() { | ||
| pm->s.velocity = pml.velocity; | ||
| pm->s.origin = pml.origin; | ||
| pml.origin = PM_SnapToLegacyGrid(pml.origin); | ||
| pml.velocity = PM_SnapToLegacyGrid(pml.velocity); | ||
|
|
||
| pm->s.velocity = pml.velocity; | ||
| pm->s.origin = pml.origin; | ||
|
|
||
| if (PM_GoodPosition()) | ||
| return; |
There was a problem hiding this comment.
The new snapping logic only applies when PM_GoodPosition() succeeds. When G_FixStuckObject_Generic adjusts the player's origin, the code returns immediately without snapping the corrected value back to the 1/8‑unit grid. That means any fix made for a stuck player can be lost when the state is later serialized and implicitly quantized again, putting the player back inside geometry near the world origin—the exact scenario this change tries to avoid after removing the z<0 tolerances. Consider snapping pm->s.origin (and velocity if needed) again after a successful stuck fix so the exported state remains on the legacy grid.
Useful? React with 👍 / 👎.
Summary
Testing
https://chatgpt.com/codex/tasks/task_e_68e44840602883289ce9c34476306be2