Checking rotations

I'm having a few troubles relating to euler rotations.
Put simply if I want to check whether the current rotation is at '0', but it is actually at '-4.325711e-06', which is -0.000004325711 and if Unity is being rightly mathematically pedantic, it does not equal '0'.

I am getting the initial rotation amount, which is set a '0', and it returns the above value everytime.

Is the correct way of figuring out if the rotation is '0' on this axis is use a fuzziness and add 0.01 ie

Code (CSharp):

1. rotationFuzziness = 0.01;
2.  
3. if (rotation < (initialRotation + rotationFuzziness))
4. ...

Code (CSharp):

1.     bool CheckForkMaxHeight ()
2.     {
3.         return (currentEulerAngle > (maxRotation + heightFuzziness));
4.  
5.     }

BTW, the rotation works, I'm trying to figure out a shorthand way of comparison, and if you have better ways of doing it - I am converting Quaternions to euler angles.
Thanks.

 

Just discovered this cool Mathf function - I wonder how 'similar' is similar though?

https://docs.unity3d.com/ScriptReference/Mathf.Approximately.html

 

I like your thinking. They use an arbitrary epsilon that I don't think you can change.

I prefer this construct:

Code (csharp):

1. const float myEpsilon = 0.01f;
2.  
3. if (Mathf.Abs( quantity1 - quantity2) < myEpsilon)
4. {
5.   Debug.Log( "Close enough for gummint work.");
6. }

 

Thanks, I was reading about Epsilon. There's this as well which runs really well:

Code (CSharp):

1.     bool CompareWithThreshold (float a, float b, float threshold)
2.     {
3.         return ((a - b) < 0 ? ((a - b) * -1) : (a - b)) <= threshold;
4.     }

 

I goggle at code like that... do not want!

(It actually does exactly what mine does, except less-than vs less-than-or-equal)

I would imagine that also has two branches in it instead of just one.

But hey, your codebase sir!

 

If it’s a matter of it being finicky you could always check a range.

Code (CSharp):

1. if (transform.rotation.x > -0.0001f && transform.rotation.x < 0.0001f) DoTheThing();

 

Except you do not want to do anything with the .x field of a Quaternion.

Perhaps you could do stuff with the .eulerAngles.x field however.

https://docs.unity3d.com/ScriptReference/Quaternion.html

Those are NOT the same thing.

 

You really don't want to put branching in a generalized float equality check though...