Checking Godot's source to confirm suspicion about AStarGrid2D
Marcos' Mysterious Maze » Devlog
In this video I fix another routing bug, related to Godot's AStarGrid2D algorithm. Sometimes one makes assumptions about the behavior of an engine's feature that are simply incorrect. In that case, it helps that the engine is open source!
Marcos' Mysterious Maze
a puzzle roguelike-like
Status | In development |
Author | Marcos Donnantuoni |
Genre | Puzzle, Platformer |
Tags | Procedural Generation, Roguelike, Roguelite, Sokoban |
More posts
- Stitching levels again: corridor and moat shenanigans1 day ago
- Plates and bridges working! Except for a fun bug3 days ago
- Plates and bridges, part 5: fall is coming4 days ago
- Plates and bridges, part 4: design conundrums7 days ago
- Pressure plates and bridges, part 38 days ago
- Pressure plates and bridges, part 29 days ago
- Pressure plates and bridges, part 110 days ago
- Simplifying the bounce code using tween_method14 days ago
- Juicing the game with bouncy animations15 days ago
- Smoothing transitions from corridors to levels16 days ago
Comments
Log in with itch.io to leave a comment.
This is interesting and I think you should make an issue if you have the time. I have a completely different use-case in my point and click game, where sometimes if you dig the player into getting themself stuck, they still try to walk to the point even though they are standing on a non-navigable tile. This causes them to jitter back and forth endlessly until you unstick them. I was stumped on why this happened but I think this answers it!
Thanks for sharing! Interesting. So perhaps more people have this issue... I'll try to make the report today.
I found out other person raised the same issue:
https://github.com/godotengine/godot/pull/86983