Collision normal accuracy #89
RileyRosell
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Truth be told, I’d been trying to reverse engineer the Sonic slope mechanics on my own project. However, I’ve had some VERY frustrating issues with consistency in collisions, and I was wondering if you had any similar issues while building this engine, and if so, what fixed it?
specific issues I've had:
-collision normal for a given flat slope varies by +/-2 degrees when things work well, but sometimes by more
-"Left handed" slopes (incline when moving left) have significantly MORE variation, by as much as +/-10 degrees
It’s possible this is caused by implementation failure on my end (I use the Hitbox with a test_move rather than sensor raycasts, and the Sprite+Hitbox is rotated by 22.5 degree increments rather than having unique sprites for each rotation value). That said, I still want to check if I’m alone in needing to circumvent these quirks.
I tested this project for a bit, and found no real “slope twitch” like exists in my project. With how smooth it runs, there MUST be something very different in how these things are implemented…
Beta Was this translation helpful? Give feedback.
All reactions