Question [SOLVED] : Explicit match on x and y velocities of a rigid body and jump function

hello group,

I need help to understand the difference between two approaches to execute a jump function in a 2D game-
and why the first approach fails to execute

Approach 1- Check input button pressed, get x and y velocities of the rigid body.
Observation 1- Jump fails to execute while the rigid body is in motion on the x axis
Code1 below -

Code (CSharp):

1.        if (Input.GetButtonDown("Jump") && rb.velocity.y == 0f && rb.velocity.x !=0f)
2.         {
3.             Debug.Log("jump while running");
4.             Debug.Log("while running:" + rb.velocity.x);
5.             if (IsGrounded() == true)
6.             {
7.                 //Debug.Log("value of isgrounded after jump is " + IsGrounded());
8.                 //rb.velocity = new Vector2(0, 12f);
9.                
10.                 doublejump = true;
11.                 jumpsoundeffect_audiosource.Play();
12.                 rb.velocity = new Vector2(directionX*rb.velocity.x, jumpforce);
13.             }

Approach 2- Check input button pressed, do not perform any additional checks
Observation 2 - Jump executes with rigid body is in motion on the x axis
Code2 below -

Code (CSharp):

1. if (Input.GetButtonDown("Jump"))
2.         {
3.             Debug.Log("jump while running");
4.             Debug.Log("while running:" + rb.velocity.x);
5.             if (IsGrounded() == true)
6.             {
7.                 //Debug.Log("value of isgrounded after jump is " + IsGrounded());
8.                 //rb.velocity = new Vector2(0, 12f);
9.                
10.                 doublejump = true;
11.                 jumpsoundeffect_audiosource.Play();
12.                 rb.velocity = new Vector2(directionX*rb.velocity.x, jumpforce);
13.             }

Explicit conditional match on x and y velocities of a rigid body leads to non-execution of jump func

 

This is obviously because rb.velocity.y == 0f is not true. It could be 0.000001 or -0.0002
Try replacing it with

Math.Abs(rb.velocity.y) < 1E-4f

just to see if that's the case.

 

Never do direct comparisons between floats and a specific number. These are very unlikely to return true.
Always check for approximations, as suggested by @orionsyndrome. That's also why Approximately exists to automatically check if the values are within Epsilon of each other, which is simply a predefined small value.

Generally speaking, why does velocity.x matter when jumping? Why check velocity.y, if the ground check fulfills the same purpose? Seems unnecessary.

 

so unlike a guaranteed true or false, this tells me that there is a transient state of motion when jump is pressed.

 

you are correct, there is no need of redundant checks in production code. but today, you both made me realize that unity system works for approximations.

 

You use it in a condition check.

Ergo:

Code (CSharp):

1. bool noYMovement = Math.Abs(rb.velocity.y) < 1E-4f;

The point is you just shouldn't check floating point numbers for equality.

It's a general programming thing and not at all specific to Unity.

 

Just so you realize that's an extremely small amount... 0.0001 to be exact.

This means it is possible to be going UP faster than this and in a single frame be going DOWN faster than this if gravity is "strong enough," in which case it will not trip.

Back-of-envelope checks tell me that earth's gravity at -9.8 m/s^2 at even 120fps (which is high!) is WAY strong enough to miss this tiny micro window, like nearly three orders of magnitude too large. You will be extremely unlikely to ever detect y velocity less than that tiny amount unless you just get lucky.



Instead, every frame, compare if its sign changed in the direction you wanted compared to the previous frame.

Floating (float) point imprecision:

Never test floating point (float) quantities for equality / inequality. Here's why:

https://starmanta.gitbooks.io/unitytipsredux/content/floating-point.html

https://forum.unity.com/threads/debug-log-2000-0f-000-1f-is-200.1153397/#post-7399994

https://forum.unity.com/threads/why-doesnt-this-code-work.1120498/#post-7208431

"Think of [floating point] as JPEG of numbers." - orionsyndrome on the Unity3D Forums

 

I just copied orion's coe so I wasn't paying too much attention to the actual value. It was the principle I was trying to bring across.