[Merged by Bors] - Fixes Timer Precision Error Causing Panic #2362
Conversation
jacobgardner
changed the title
NathanSWard
added
C-Bug
|
which would imply the panic was at self.set_elapsed(self.elapsed() - self.duration() * self.times_finished);(Specifically the substraction). EDIT Ahhh ok, now I see. The value of elapsed is a result of |
|
bors try |
tryBuild failed: |
|
@jacobgardner it looks like you need to run |
|
@NathanSWard sorry about that. Should be formatted now. |
|
bors try |
tryBuild succeeded: |
NathanSWard
added
the
S-Ready-For-Final-Review
| t.tick(duration); | ||
| assert_eq!(t.times_finished(), 33); | ||
| t.tick(duration); | ||
| assert_eq!(t.times_finished(), 33); |
There was a problem hiding this comment.
shouldn't the extra tick be on this one?
There was a problem hiding this comment.
So my logic was roughly this:
0 + 33.3333 = 33.3333 (Rounds down to 33 times finished, leaving the remainder as .3333)
.3333 + 33.3333 = 33.6666 (Rounds down to 33 again, remainder fo .6666)
.6666 + 33.3333 = 33.9999 (Rounds down last time to 33 with remainder of .9999)
.9999 + 33.3333 = 34.3332 (Rounds down to 34 this time)
There was a problem hiding this comment.
It will add the extra tick after the fourth iteration, then it will correctly add it every three iterations... I would have liked it to be every three from the start, but that's probably not possible when using those kind of numbers 🤷
There was a problem hiding this comment.
Gotta love floating point numbers 🙃
|
bors r+ |
# Objective Fixes #2361 ## Solution Uses integer division instead of floating-point which prevents precision errors, I think.
|
Pull request successfully merged into main. Build succeeded: |
bors
bot
changed the title
# Objective Fixes bevyengine#2361 ## Solution Uses integer division instead of floating-point which prevents precision errors, I think.
Objective
Fixes #2361
Solution
Uses integer division instead of floating-point which prevents precision errors, I think.