-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path1
32 lines (26 loc) · 1.1 KB
/
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
## [inactive] [1] BUG Narrow phase collision solving order
### Recursive descent narrow phase collision solver
I need to build a map of collisions indexed by each entity.
With this data I can apply this algorithm.
```
A is moving into B.
Does B have any unique collisions it must Solve? Yes
Are we here because of this unsolved collision? No
B is moving into C.
Does C have any unique collisions is must Solve? Yes
Are we here because of this unsolved collision? No
C is moving into D.
Does D have any unique collisions it must Solve? Yes
Are we here because of this unsolved collision? No
D is moving into A.
Does A have any unique collisions it must Solve? Yes
Are we here because of this unsolved collision? Yes
Then we must solve `D is moving into A` with the information we have.
- We could make blanket assumption that D cannot move.
This means that D will lose a movement race from E.
At first glance, I think this would be fine.
```
... Wait
If D can't move than none of the entities can move.
At second glance this is the correct behavior...
This makes recursive descent exactly what I want.