How to check "if (float != 0)" if floats "are not precise"?

So,
if floats are not precise and could be 0.000001 instead of 0, how do you use them in an if-statement, when you want to chek for !=0 or ==0?

Can you maybe truncate the float? Would that even work?

 

It kinda depends on how you land on zero. Most unity methods will take care of that and shouldn't get floating-point errors. Your solution depends on the amount of precision you're aiming for.
You can round and cast to int (really not precise), using Math.Round(), or using a percentage margin, or using arithmetics (a + precision <= b && b + precision <= a).
It really depends on your use case and the types of calculations that leads you to floating point errors.

 

Unity has a method for comparing floats, Mathf.Approximately(float a, float b)

 

Well, I didn't know that. Funny thing is, they say they compare within epsilon precision, and Epsilon docs explains the arithmetics solution I've suggested above. Quite cool.

 

What exactly are you trying to do? The best approach here depends heavily on what you are trying to accomplish.