travelTime = 0.0f;
while (Vector3.Distance(Vector2.Lerp(startPos, endPos, travelTime), transform.position) >= maxDistance) {
locationToGo = Vector2.Lerp(startPos, Input.mousePosition, travelTime);
travelTime += 0.1f;
}
// SHOW
// Smoothly move the camera towards that target position
transform.position = Vector3.SmoothDamp(transform.position, locationToGo, ref velocity, smoothTime);
So, this code is meant to use Lerp to return a Vector2 a certain amount of 'steps' towards the mouse pointer. And then give that value to the SmoothDamp to move the square to the pos. But the Lerp is not calculating correctly. Does anyone know what is wrong, or any better working alternatives?
Since I don't really know what exactly you are trying to achieve i'm just gonna guess you want to move a GameObject to the MousePosition every Frame.
You could do it like this (it is out of my head, not compileproof):
GameObject ourGameobject;
Vector3 ourTarget;
void Update()
{
ourGameobject.transform.position = Vector3.lerp(ourGameobject.transform.position, ourTarget, Time.DeltaTime);
}
For faster or lower speed just multiply Time.DeltaTime with sth.
Also Note that Input.MousePosition returns a coordinate on the screen.
Related
I've been told that Rigidbody.MoveRotation is the best way in Unity 3D to rotate the player between fixed positions while still detecting hits. However, while I can move smoothly from fixed position to position with:
if (Vector3.Distance(player.position, targetPos) > 0.0455f) //FIXES JITTER
{
var direction = targetPos - rb.transform.position;
rb.MovePosition(transform.position + direction.normalized * playerSpeed * Time.fixedDeltaTime);
}
I can't find out how to rotate smoothly between fixed positions. I can rotate to the angle I want instantly using Rigidbody.MoveRotation(Vector3 target);, but I can't seem to find a way to do the above as a rotation.
Note: Vector3.Distance is the only thing stopping jitter. Has anyone got any ideas?
First of all MoveRotation doesn't take a Vector3 but rather a Quaternion.
Then in general your jitter might come from overshooting - you might be moving further than the distance between your player and target actually is.
You can avoid that bit by using Vector3.MoveTowards which prevents any overshooting of the target position like e.g.
Rigidbody rb;
float playerSpeed;
Vector3 targetPos;
// in general ONLY g through the Rigidbody as soon as dealing wit Physics
// do NOT go through transform at all
var currentPosition = rb.position;
// This moves with linear speed towards the target WITHOUT overshooting
// Note: It is recommended to always use "Time.deltaTime". It is correct also during "FixedUpdate"
var newPosition = Vector3.MoveTowards(currentPosition, targetPos, playerSpeed * Time.deltaTime);
rb.MovePosition(newPosition);
// [optionally]
// Note: Vector3 == Vector3 uses approximation with a precision of 1e-5
if(rb.position == targetPos)
{
Debug.Log("Arrived at target!");
}
Then you can simply apply this same concept also to rotation by going through the equivalent Quaternion.RotateTowards basically just the same approach
Rigidbody rb;
float anglePerSecond;
Quaternion targetRotation;
var currentRotation = rb.rotation;
var newRotation = Quaternion.RotateTowards(currentRotation, targetRotation, anglePerSecond * Time.deltaTime);
rb.MoveRotation(newRotation);
// [optionally]
// tests whether dot product is close to 1
if(rb.rotation == targetRotation)
{
Debug.Log("Arrived at rotation!");
}
You can go one step further and use a tweeting library to tween between rotations.
DOTween
With that you can call it like this:
rigidbody.DoRotate(target, 1f) to rotate to target in 1 second.
Or even add callbacks.
rigidbody.DoRotate(target, 1f).OnComplete(//any method or lambda you want)
If at some point you want to cancel the tween yuou can save it on a variable and then call tween.Kill();
So, you want to animate the rotation value over time until it reaches a certain value.
Inside the Update method, you can use the Lerp method to keep rotating the object to a point, but you will never really reach this point if you use Lerp. It will keep rotating forever (always closer to the point).
You can use the following:
private bool rotating = true;
public void Update()
{
if (rotating)
{
Vector3 to = new Vector3(20, 20, 20);
if (Vector3.Distance(transform.eulerAngles, to) > 0.01f)
{
transform.eulerAngles = Vector3.Lerp(transform.rotation.eulerAngles, to, Time.deltaTime);
}
else
{
transform.eulerAngles = to;
rotating = false;
}
}
}
So, if the distance between the current object angle and the desired angle is greater than 0.01f, it jumps right to the desired position and stop executing the Lerp method.
The camera is supposed to follow a 2D character, here is code
void LateUpdate ()
{
var to = target.position;
to.z = transform.position.z;
var newPos = Vector3.Lerp(transform.position, to, speed * Time.deltaTime);
transform.position = newPos;
newPos.z = to.z;
Debug.DrawRay(newPos, Vector3.up, Color.green, 5);
}
I also draw positions of the character and the camera.
Red lines are character's positions, and green lines are camera's positions
What do I do wrong?
UPDATE:
I've figured out something interesting. On the picture below green lines are positions of the Camera that is moved by Vector3.Lerp inside a LateUpdate method. Yellow lines are positions of the character that I set to the character's Rigidbody2D inside FixedUpdate method, and the red lines are positions of the character's transform as they seen from inside the camera's LateUpdate.
What I want to say is that the actual position of the character is driven by its Rigidbody2D component. By changing Rigidbody2D's "Interpolate" option we can get different results.
The problem is that even if I add to the Rigidbody2D's position the same value every FixedUpdate tick, the result isn't so consistent. Sometimes the distance between new and old position is bigger than it should be.
Add to that the we set position of the camera in the LateUpdate method, which has different update rate than FixedUpdate, so even if we set the new position to the character's transform, and not to the Rigidbody2D, the movement of the camere still won't be smooth, because speed of the character will be different every frame.
For now I have only one solution.
Set the new position of the character to its transform, and not to the Rigidbody2D
Change position of the camere in its FixedUpdate, and not in the LateUpdate.
Then the positions will look like this
But since position of the camera is set in the FixedUpdate, it won't be so smooth as it might be, and also i'm not sure whether collision detection of the character will work good, since we set its position directly to its transform.
the problem could be coming from how you are using interpolation to determine how far to move the camera.
I don't know if Vector3.Lerp's behavior would be to extrapolate if the third parameter (its fraction) is higher than 1.0, but i suspect this could be the problem (specifically if there is a bit more time between frames, and speed * Time.DeltaTime becomes higher than 1.0)
A better way (eliminating the lerp) could be to do the interpolation of distance over speed and time yourself ;
void LateUpdate ()
{
var to = target.position;
to.z = transform.position.z;
//you can just multiply a Vector3 with a float
//so we can do the interpolation maths ourselves like this :
var distanceToMove = (to - transform.position) * speed * Time.deltaTime;
var newPos = transform.position + distanceToMove;
transform.position = newPos;
newPos.z = to.z;
Debug.DrawRay(newPos, Vector3.up, Color.green, 5);
}
(if that got confusing, here is cleaned up version to make it more concise)
void LateUpdate ()
{
var to = target.position;
to.z = transform.position.z;
transform.position += (to - transform.position) * speed * Time.deltaTime; ;
Debug.DrawRay(transform.position, Vector3.up, Color.green, 5);
}
What I want to do is to make a kind of 2.5D runner game in Unity, which the character's all three rotation axises are frozen and the position on Z axis is also frozen. I don't know how to make the character moving forward nicely on the seesaw. (I create the seesaw by using HingeJoint.)
I create a struct to detect the CapsuleCollider status by using Physics.Raycast() function and that works fine.
private struct ColliderStatus
{
public bool headed; //colliding up
public bool footed; //colliding down
public bool onPlane; //colliding down && the obstacle colliding does not have slope angle
public bool lefted; //colliding left
public bool righted; //colliding right
public bool inAir; //not colliding anything
}
I've tried these ways:
Add force on Rigidbody to move forward
//To move character rigidbody move forward automatically in runner game
//when the speed is lower than the minimum speed and it's on plane or in air.
if (rigidbody.velocity.x < minForwardSpeed && (colliderStatus.onPlane || colliderStatus.inAir))
{
rigidbody.AddForce(20f * Vector3.right);
}
//Add gravity to player
Vector3 gravityForce = new Vector3(0f, -gravityOnPlayer, 0f);
rigidbody.AddForce(gravityForce);
It doesn't work well because the character continue going up when it's on the seesaw though the seesaw starts to tilt. And there will be a velocity loss when the character fall to ground from a higher plane or after jumping and what it looks like is that the character will stunned for a little moment on the landing point and then begin to accelerate.
Use transform.Translate() to move forward && change the way of adding gravity
//Use transform.Translate() to move forward
//I recognize that by this way, there will be no velocity loss
//when the character falling down to the ground at the landing point
//If I don't use this condition, my character will stuck on the
//right vertical wall
if (!colliderStatus.righted)
{
transform.Translate(new Vector2(minForwardSpeed, 0f) * Time.deltaTime);
}
I don't know why I can't write like this since it will cause the velocity doesn't react correctly:
//Use transform.Translate() to move forward
if (!colliderStatus.righted && rigidbody.velocity.x < minForwardSpeed)
{
transform.Translate(new Vector2(minForwardSpeed, 0f) * Time.deltaTime);
}
To change the way of adding gravity, I use a function SlopeAngleVector() to calculate the slope vector the character is running on.
private Vector3 SlopeAngleVector()
{
Vector3 nextStepPositon = new Vector3(transform.position.x + 0.01f, transform.position.y, 0f);
Ray nextPosRay = new Ray(nextStepPositon, Vector3.down);
Ray nowPosRay = new Ray(transform.position, Vector3.down);
RaycastHit nextPosHit;
RaycastHit nowPosHit;
Vector3 slopeAngle = Vector3.zero;
Physics.Raycast(nowPosRay, out nowPosHit, 5f, obstaclesLayerMask);
if (Physics.Raycast(nextPosRay, out nextPosHit, 5f, obstaclesLayerMask))
{
slopeAngle = new Vector3(nextPosHit.point.x - nowPosHit.point.x, nextPosHit.point.y - nowPosHit.point.y, 0f).normalized;
}
return slopeAngle;
}
Then I add the gravity by calculate the gravity projection on the slope vector:
private void AddGravity()
{
Vector3 gravityForce = new Vector3(0f, -gravityOnPlayer, 0f);
//my character could be collided by the long vertical wall(colliderStatus.righted)
//so I set the condition as "!colliderStatus.footed"
//otherwise, I would use "colliderStatus.inAir"
if (!colliderStatus.footed)
{
gravityForce = new Vector3(0f, -gravityOnPlayer, 0f);
}
else
{
gravityForce = Vector3.Project(Vector3.down * gravityOnPlayer, SlopeAngleVector());
}
rigidbody.AddForce(gravityForce);
}
Now my character can slide down from the seesaw but it will keep going backwards. And it cannot make it through when on the low slope angle seesaw.
How to make a good behavior script for the runner on seesaw?
I'd suggest looking at some of the Unity standard asset character controllers, I believe they take slopes into account for their character movement. It may give you some ideas.
I'd also recommend modifying the way your code calculates the angle of the slope. The raycast hit will give you back a surface normal, you should then be able to use the Vector3.Cross to figure out the angle of the slope.
It'll be something like: Vector3.Cross(normal, (vector that points away from screen)).
You may need to tweak it to get it working correctly but this can give you the slope angle in one raycast. It may also eliminate potential issues of your move to position being just below the see saw.
As a general tip, try not to mix transform and rigidbody stuff together, if you want to move the rigidbody, move the rigidbody directly, not indirectly through the transform.
I have a ball which rotates around the point 0,0,0 in the Z-axis. When the space button is pressed, the ball has to go inside the large circle. Now my code looks like this. When you press space, the ball does not behave as they should. I want to know how to make a balloon down exactly down
that's how the ball should behave ->
behavior image
my code:
void Update () {
if (Input.GetKeyDown (KeyCode.Space)) {
transform.position = new Vector3 (transform.position.x - 1, transform.position.y - 1, 0);
} else {
transform.RotateAround(new Vector3(0,0,0), new Vector3(0,0,1), 2);
}
}
Your code to 'jump' the orbit doesn't do what you want because Transform.RotateAround modifies both the rotation and the position of the object's transform.
So jumping to (position - 1,1,0) in the world is going to return wildly different results every time.
What you want to do instead is calculate the (Vector) direction from the object to the centre of orbit (the difference), then scale that down to how far you want it to move, then apply it to the position.
private Vector3 _orbitPos = Vector3.zero;
private float _orbitAngle = 2f;
private float _distanceToJump = 2f;
void Update()
{
if (Input.GetKeyDown(KeyCode.Space))
{
var difference = (_orbitPos - transform.position).normalized * _distanceToJump;
transform.Translate(difference);
}
transform.RotateAround(_orbitPos, Vector3.forward, _orbitAngle);
}
This will move the object to be orbiting 2 units closer when space is pressed immediately.
If you wanted to have a smooth transition instead of a jump, look into using Mathf.Lerp, Vector3.Lerp and the routines involved.
I'm making a basic 2D space shooter game in unity. I have the movements working fine and the camera follows the player. I've been scratching my head over how to give the camera a slight delay from the player moving to the camera moving to catch up to it without it teleporting. I was told to use a Vector3.Lerp and tried a few things from stackoverflow answers but none seemed to work with the way my code is set up. Any suggestions?
(myTarget is linked to the player)
public class cameraFollower : MonoBehaviour {
public Transform myTarget;
void Update () {
if(myTarget != null){
Vector3 targPos = myTarget.position;
targPos.z = transform.position.z;
transform.position = targPos;
}
}
}
If you linear interpolate (Lerp) you risk that Time.deltaTime * speed > 1 in which case the camera will start to extrapolate. That is, instead of following it will get in front if your target.
An alternative is to use pow in your linear interpolation.
float speed = 2.5f;
float smooth = 1.0f - Mathf.Pow(0.5f, Time.deltaTime * speed);
transform.position = Vector3.Lerp(transform.position, targetPos, smooth);
Mathf.Pow(0.5, time) means that after 1/speed second, half of the distance to the target point will be travelled.
The idea with Lerping camera movement is to gradually and smoothly have the camera make its way to the target position.
The further away the camera is, the bigger the distance it will travel per frame, but the closer the camera is, the distance per frame becomes smaller, making the camera ease into its target position.
As an example, try replacing your transform.position = targPos; line with:
float speed = 2.5f; // Set speed to whatever you'd like
transform.position = Vector3.Lerp(transform.position, targPos, Time.deltaTime * speed);