Subtracting a float from another float does not equal a float?

Question

I am coding with the unity scripting API but I don't think that changes anything.

I have a variable playerMovementScript.rightWallX that is a float and is equal to -20.84.

I then have this line of code:

Debug.Log(-21.24f == playerMovementScript.rightWallX - 0.4f);

It always prints false. I feel like I'm missing something obvious cause I'd like to think I'm not THAT new to coding. Any help would be appreciated.

Floating point is not as accurate as you think. You need to check if it is approximately the value you want — BugFinder, Aug 27 '22 at 21:16
Does this answer your question? [Is floating point math broken?](https://stackoverflow.com/questions/588004/is-floating-point-math-broken) — Martheen, Aug 27 '22 at 21:17
https://docs.unity3d.com/ScriptReference/Mathf.Approximately.html — Retired Ninja, Aug 27 '22 at 21:20

score 1 · Answer 1 · answered Aug 27 '22 at 21:41

1

Floating points not accurate. Use them only for fast calculations. If u want to compare 2 floats use if (Mathf.Approximately(floatA,floatB))

answered Aug 27 '22 at 21:41

SntSdg

36
3

While your suggestion is spot on, the reasoning isn't quite right: floating point is very accurate. The problem is not accuracy, but the inability to represent most base-ten fraction numbers in base-two (binary) without infinite precision. This is because 0.1 in decimal is 0.001100110011... (repeats forever) in binary. However, 32-bit floats have only 24 significant bits (23 plus an implicit leading 1 bit), so 0.1d is represented as 1.10011001100110011001101*2^-3, with that final 1101 due to rounding. Thus multiplying by 10(decimal) results in a number slightly larger than 1.0. – taniwha Aug 28 '22 at 00:01
(continuation of above) However, one in 16-million (24 bits) is pretty accurate and even fairly precise. The problem is there are tricks for dealing with the rounding, including, as you suggest, not doing *direct* comparisons between floats, and thus goes against our intuition. – taniwha Aug 28 '22 at 00:04

score 0 · Answer 2 · answered Aug 28 '22 at 01:41

0

The reason it prints false is because you are using the == operator, which checks if 2 things are equal, therefore since you are doing -21.24f == playerMovementScript.rightWallX - 0.4f, the result is unequal and then it prints false. I am pretty sure you meant =, but we all make mistakes.

answered Aug 28 '22 at 01:41

Abyssal

1
3

The second part of your answer makes no sense, you cannot assign something to a value. – mikyll98 Sep 01 '22 at 11:46

Subtracting a float from another float does not equal a float?

2 Answers2