We Keep Coding

Cylinder Snapping into place during Quaternion.RotateTowards

What I am trying to do is have a pillar in the center of a scene, and what i think is happening is the quaternion.RotateTowards is receiving a different starting/initial quaternion than the action which is causing it to snap/teleport into a different location which then starts to move to the next. I thought it may be because i'm misunderstanding how quaternions are handled in unity, so i've tried messing with normalizing it, but i can't seem to get any change on the teleport.
The goal is to attach this scrip to a simple 3D cylinder and have it wobble basically, where there will be a player on top of it trying to stay on it. However I can't seem to figure out why it is teleporting and was hoping for a second set of eyes.
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
public class PlatformWobble : MonoBehaviour
{
public float timeDelay = 0;
public float rRange = 0.5f;
public float maxRotation = 20.0f;
public float rotRate = 0.05f;
private bool wobble = false;
private Vector3 randomRotation;
private Quaternion rotation;
private Quaternion destination;
private float x, y, z;
private bool inPlace;
private void Start()
{
randomRotation = new Vector3();
inPlace = true;
}
// Update is called once per frame
void FixedUpdate()
{
if (inPlace)
{
destination = RotateRandomly();
inPlace = false;
}
transform.localRotation = Quaternion.RotateTowards(transform.localRotation, destination, rotRate * Time.deltaTime);
if (transform.localRotation == destination)
{
inPlace = true;
}
}
//This will grab the rotation of the object and move it randomly, but won't exceed maxRotation;
Quaternion RotateRandomly()
{
randomRotation = new Vector3(Random.Range(-rRange, rRange), Random.Range(-rRange, rRange), Random.Range(-rRange, rRange));
rotation = transform.localRotation;
x = rotation.x + randomRotation.x;
if(x >= maxRotation && x <= 360 - maxRotation) { x = rotation.x; }
y = rotation.y + randomRotation.y;
if (y >= maxRotation && y <= 360 - maxRotation) { y = rotation.y; }
z = rotation.z + randomRotation.z;
if (z >= maxRotation && z <= 360-maxRotation) { z = rotation.z; }
return new Quaternion(x, y, z, transform.localRotation.w);
}
}

Your quaternion calculations are incorrect: quaternions do not hold angles, instead, they hold the representation of a rotation about an axis. I emphasized "representation" because it is a little complicated...
The x, y, z components of a quaternion hold the rotation axis unit vector scaled by the sine of the half angle of rotation. The w component holds the cosine of the half angle. That is...
// NOTE: rotationAngle is in radians rather than degrees
Quaternion MakeQuaternion (Vector3 rotationAxis, float rotationAngle)
{
float c = Mathf.Cos (rotationAngle / 2);
float s = Mathf.Sin (rotationAngle / 2);
Vector3 v = rotationAxis.normalized * s;
return new Quaternion (v.x, v.y, v.z, c);
}
Now, the tricky part for your problem is coming up with the rotation axis and angle for the desired effect.
One solution (if you wish to stick to Euler angles) is to compute quaternions for each Euler rotation and then combine them:
Quaternion EulerToQuat (float XAngle, float YAngle, float ZAngle)
{
Quaternion X = MakeQuaternion (Vector3.right, XAngle);
Quaternion Y = MakeQuaternion (Vector3.up, YAngle);
Quaternion Z = MakeQuaternion (Vector3.forward, ZAngle);
// combine the rotations such that the object is first rotated about the Z axis,
// then about the Y axis, then the X (ie, reverse order of multiplication).
// Reminder: quaternion multiplicate is not commutative: order matters, so if this
// is not the order you want, just siwtch things around
rotate X * Y * Z;
}

FixedUpdate doesn't get called every render frame, only on physics frames. You need a way to tell Unity to have each frame showing the rotation change, rather than only updating the rendering when the physics frame has just run.
That is what Rigidbody.MoveRotation is for. Cache a reference to the Rigidbody, calculate the new global rotation it should have, then call MoveRotation:
private Rigidbody rb;
private void Start()
{
randomRotation = new Vector3();
inPlace = true;
rb = GetComponent<Rigidbody>();
}
// ...
// Update is called once per frame
void FixedUpdate()
{
if (inPlace)
{
destination = RotateRandomly();
inPlace = false;
}
Quaternion newLocalRot = Quaternion.RotateTowards(transform.localRotation,
destination, rotRate * Time.deltaTime);
Quaternion newGlobalRot = transform.parent.rotation * newLocalRot;
rb.MoveRotation(newGlobalRot);
if (transform.localRotation == destination)
{
inPlace = true;
}
}

FreeLookCam script with a different "world up"?

I am trying to adjust the FreeLookCam script from the Standard Assets to work with my game, which has a different "world up" than standard unity projects. In my case, world up is -z, and I cannot seem to be able to adjust the script to fit this scenario.
Here is what it look like when I try it: https://www.youtube.com/watch?v=Ppb-tXvLZpI
and here is the code:
using System;
using UnityEngine;
//using UnityStandardAssets.CrossPlatformInput;
namespace UnityStandardAssets.Cameras
{
public class FreeLookCam : PivotBasedCameraRig
{
// This script is designed to be placed on the root object of a camera rig,
// comprising 3 gameobjects, each parented to the next:
// Camera Rig
// Pivot
// Camera
[SerializeField] private float m_MoveSpeed = 1f; // How fast the rig will move to keep up with the target's position.
[Range(0f, 10f)] [SerializeField] private float m_TurnSpeed = 1.5f; // How fast the rig will rotate from user input.
[SerializeField] private float m_TurnSmoothing = 0.0f; // How much smoothing to apply to the turn input, to reduce mouse-turn jerkiness
[SerializeField] private float m_TiltMax = 75f; // The maximum value of the x axis rotation of the pivot.
[SerializeField] private float m_TiltMin = 45f; // The minimum value of the x axis rotation of the pivot.
[SerializeField] private bool m_LockCursor = false; // Whether the cursor should be hidden and locked.
[SerializeField] private bool m_VerticalAutoReturn = false; // set wether or not the vertical axis should auto return
private float m_LookAngle; // The rig's y axis rotation.
private float m_TiltAngle; // The pivot's x axis rotation.
private const float k_LookDistance = 100f; // How far in front of the pivot the character's look target is.
private Vector3 m_PivotEulers;
private Quaternion m_PivotTargetRot;
private Quaternion m_TransformTargetRot;
protected override void Awake()
{
base.Awake();
// Lock or unlock the cursor.
Cursor.lockState = m_LockCursor ? CursorLockMode.Locked : CursorLockMode.None;
Cursor.visible = !m_LockCursor;
m_PivotEulers = m_Pivot.rotation.eulerAngles;
m_PivotTargetRot = m_Pivot.transform.localRotation;
m_TransformTargetRot = transform.localRotation;
}
protected void Update()
{
HandleRotationMovement();
if (m_LockCursor && Input.GetMouseButtonUp(0))
{
Cursor.lockState = m_LockCursor ? CursorLockMode.Locked : CursorLockMode.None;
Cursor.visible = !m_LockCursor;
}
}
private void OnDisable()
{
Cursor.lockState = CursorLockMode.None;
Cursor.visible = true;
}
protected override void FollowTarget(float deltaTime)
{
if (m_Target == null) return;
// Move the rig towards target position.
transform.position = Vector3.Lerp(transform.position, m_Target.position, deltaTime*m_MoveSpeed);
}
private void HandleRotationMovement()
{
if(Time.timeScale < float.Epsilon)
return;
// Read the user input
// var x = CrossPlatformInputManager.GetAxis("Mouse X");
// var y = CrossPlatformInputManager.GetAxis("Mouse Y");
var x = Input.GetAxis("Mouse X");
var y = Input.GetAxis("Mouse Y");
// Adjust the look angle by an amount proportional to the turn speed and horizontal input.
m_LookAngle += x*m_TurnSpeed;
// Rotate the rig (the root object) around Y axis only:
m_TransformTargetRot = Quaternion.Euler(0f, m_LookAngle, 0f);
if (m_VerticalAutoReturn)
{
// For tilt input, we need to behave differently depending on whether we're using mouse or touch input:
// on mobile, vertical input is directly mapped to tilt value, so it springs back automatically when the look input is released
// we have to test whether above or below zero because we want to auto-return to zero even if min and max are not symmetrical.
m_TiltAngle = y > 0 ? Mathf.Lerp(0, -m_TiltMin, y) : Mathf.Lerp(0, m_TiltMax, -y);
}
else
{
// on platforms with a mouse, we adjust the current angle based on Y mouse input and turn speed
m_TiltAngle -= y*m_TurnSpeed;
// and make sure the new value is within the tilt range
m_TiltAngle = Mathf.Clamp(m_TiltAngle, -m_TiltMin, m_TiltMax);
}
// Tilt input around X is applied to the pivot (the child of this object)
// m_PivotTargetRot = Quaternion.Euler(m_TiltAngle, m_PivotEulers.y , m_PivotEulers.z);
m_PivotTargetRot = Quaternion.Euler(m_TiltAngle, m_PivotEulers.y , m_PivotEulers.z);
if (m_TurnSmoothing > 0)
{
m_Pivot.localRotation = Quaternion.Slerp(m_Pivot.localRotation, m_PivotTargetRot, m_TurnSmoothing * Time.deltaTime);
transform.localRotation = Quaternion.Slerp(transform.localRotation, m_TransformTargetRot, m_TurnSmoothing * Time.deltaTime);
}
else
{
m_Pivot.localRotation = m_PivotTargetRot;
transform.localRotation = m_TransformTargetRot;
}
}
}
}

There are two lines you need to change, both in HandleRotationMovement, and they both have to do with swapping the angles for y and z. The first one is on line 83:
// Rotate the rig (the root object) around Y axis only:
m_TransformTargetRot = Quaternion.Euler(0f, 0f, m_LookAngle);
and the second one is on line 101:
// Tilt input around X is applied to the pivot (the child of this object)
m_PivotTargetRot = Quaternion.Euler(m_TiltAngle + 90, m_PivotEulers.z , m_PivotEulers.y);
On that second line I had to add 90 degrees to get the tilt limits in the expected range in addition to swapping the y and z angles.

C# How to make a smooth jump in unity3d without moving the X, towards the nearest object

I would like to make a smooth jump towards the nearest cube. I already have a script to detect the closest cube. I want that the X-axis is locked, so only the Y-axis and the Z-axis change when jumping. I would like to use a Jump animation when jumping. I already tried to use Vector3MoveTowards, but that didn't really work well, maybe I didn't use it properly.
Detect nearest cube where the player should jump to (C#)
void Update()
{
FindClosestCube ();
GameObject closestCube = FindClosestCube ();
Debug.Log (closestCube);
}
GameObject FindClosestCube() {
GameObject[] gos;
gos = GameObject.FindGameObjectsWithTag("cube");
GameObject closest = null;
float distance = Mathf.Infinity;
float position = transform.position.z;
foreach (GameObject go in gos) {
float diff = go.transform.position.z - position;
float curDistance = diff;
if (curDistance < distance) {
closest = go;
distance = curDistance;
}
}
return closest;
}
The tricky part is that at some cubes you have to jump up (y+1), with some cubes you jump towards the same Y (y+0) and with some cubes you jump down (y-1).
How do I do this?
Image of how it looks like:
EDIT: I have this code right now:
----------------C#-----------------
Rigidbody rb;
public int clicks = 0;
Vector3 target;
public Animation jumpAnimation;
bool jump = false;
float cubeDiffY;
bool movePlayer;
public float smoothTime = 0.3f;
public float yVelocity = 0.0f;
void Start()
{
rb = GetComponent<Rigidbody> ();
}
void Update ()
{
FindClosestCube ();
GameObject closestCube = FindClosestCube ();
Debug.Log ("Closestcube = " + closestCube);
target = closestCube.transform.position + new Vector3 (0f, 0.7f, 0f);
cubeDiffY = target.y - transform.position.y;
movePlayer = true;
Debug.Log("Cube Difference Y-axis = " + Mathf.Round(cubeDiffY));
if (Input.GetMouseButtonDown (0))
{
clicks += 1;
jump = true;
jumpAnimation = gameObject.GetComponent<Animation>();
//jumpAnimation.Play ();
}
if (jump == true)
{
Jump ();
}
}
void Jump()
{
float newPosition = Mathf.SmoothDamp (transform.position.y, target.y, ref yVelocity, smoothTime);
transform.position = new Vector3 (0, newPosition, transform.position.z);
}
I calculated the difference in Y-axis between the cube where the player is standing on and the closestCube. But the Jump() doesn't work. How do I fix that?

Okay I set up a quick version of your game and got what you wanted to work, it is not exactly a quick solution, because what your doing doesn't have built in functionality for other than using animations.
Here is the character script that has all the code you need and commented thoroughly so it should explain itself.
using UnityEngine;
public class Character : MonoBehaviour
{
//the collider for the player
private new BoxCollider collider;
//the jump box collider on a empty game object that is a child to the player object
public BoxCollider JumpBox;
//the offset of the cube so it doesn't stop inside of it
public Vector3 cubeOffset;
//how high the jump will be
public float JumpHeight;
//how fast the jump will be
public float JumpSpeed;
//holds the change in position the jump will produce
private Vector3 jumpDelta;
//holds the destination cube the jump is attempting to hit
private Cube destinationCube;
//true if a jumping animation is currently playing
private bool jumping = false;
//used to swap the jump direction from up to down
private bool jumpDirection = true;
//used to hold the position of the jump so it knows when to stop
private float jumpPosition = 0;
// Use this for initialization
void Start()
{
collider = GetComponent<BoxCollider>();
}
// Update is called once per frame
void Update()
{
if(jumping)
{
//move straight towards the cube
transform.position = transform.position + (JumpSpeed * jumpDelta);
//move up and down to simulate a jump
//check the current move direction
if (jumpDirection)
{
//add to the jump position twice product of the JumpHeight the JumpSpeed so that it will
//rise and fall the same amount of time it takes to move to the destination
jumpPosition += JumpHeight * JumpSpeed * 2;
//if it has passed the jump height reverse the jump direction
if (jumpPosition >= JumpHeight)
jumpDirection = !jumpDirection;
transform.position += transform.up * JumpHeight * JumpSpeed * 2;
}
//the jump direction is going down
else
{
jumpPosition -= JumpHeight * JumpSpeed * 2;
transform.position -= transform.up * JumpHeight * JumpSpeed * 2;
}
//check if the character collider intersects witht he cubes collider
//if it has then stop jumping and set the final position as the destination position
if (collider.bounds.Intersects(destinationCube.BoxCollider.bounds))
{
jumping = false;
transform.position = destinationCube.transform.position + cubeOffset;
}
}
//detect a jump
if (Input.GetKeyDown(KeyCode.Space))
{
//detect all hits on the jump box
Collider[] hits = Physics.OverlapBox(JumpBox.center, JumpBox.size * 0.5f);
//get the closest collider with the right tag
Collider result = GetClosestColliderWithTag(hits, "Cube");
//if we have a result then begin the jumping animation
if(result != null)
{
//gets the destination cubes cube component(the custom class you have on your cubes)
destinationCube = result.gameObject.GetComponent<Cube>();
//calculate the jump delta
jumpDelta = (result.transform.position + cubeOffset) - transform.position;
//remove the left and right components so the jumping doesnt move to the left or right of the player
Vector3 component = Vector3.Project(jumpDelta, -transform.right);
jumpDelta -= component;
component = Vector3.Project(jumpDelta, transform.right);
jumpDelta -= component;
//setup the jump animation control fields to the initial values
jumpPosition = 0;
jumpDirection = true;
jumping = true;
}
}
}
private Collider GetClosestColliderWithTag(Collider[] colliders, string tag)
{
//just gets the closest collider
float distance = float.MaxValue;
int result = -1;
for (int i = 0; i < colliders.Length; i++)
{
if (colliders[i].tag == tag)
{
float distanceTemp = Vector3.Distance(transform.position, colliders[i].transform.position);
if (distanceTemp < distance)
{
distance = distanceTemp;
result = i;
}
}
}
if (result != -1)
return colliders[result];
else return null;
}
}
And here is my cube script which has some things you will need to add
using UnityEngine;
public class Cube : MonoBehaviour {
//these arent important just fields I used to set up a quick version of your game
public GameObject StartPoint;
public GameObject EndPoint;
public float Speed;
private Vector3 directionVector;
private bool direction;
//YOU WILL NEED THIS!!
[HideInInspector]
public BoxCollider BoxCollider;
// Use this for initialization
void Start() {
//not important
directionVector = EndPoint.transform.position - StartPoint.transform.position;
directionVector.Normalize();
//DONT FORGET TO SET YOUR BOX COLLIDER
BoxCollider = GetComponent<BoxCollider>();
}
// Update is called once per frame
void Update()
{
float distance = 0;
if (direction)
{
distance = Vector3.Distance(EndPoint.transform.position, transform.position);
transform.position += directionVector * Speed;
if (distance < Vector3.Distance(EndPoint.transform.position, transform.position))
direction = !direction;
}
else
{
distance = Vector3.Distance(StartPoint.transform.position, transform.position);
transform.position -= directionVector * Speed;
if (distance < Vector3.Distance(StartPoint.transform.position, transform.position))
direction = !direction;
}
}
}
Previous Answer
I would say you need to calculate the perceived position of the object in the future.
Vector3 futurePos = cubePos + (cubeMoveDirection * cubeMoveSpeed);
Once you have the future position, even if it is not exact, you should aim your animation towards that position. To do this I would have the animation change a speed vector instead of an actual transforms position that way we can rotate this speed vector in any direction you want while keeping the orientation of the block. Otherwise you have to rotate the entire block to point towards the direction you want. If this is what you want then put your block under a empty gameobject, rotate the empty gameobject to point to where you want and do the speed calculations only.
Next your animation should have a net move vector which should be pre-calculated and scaled down or up to meet the distance to the future position. It will look something like this(note this is not tested)
//class fields
Vector3 AnimatedSpeed;
Vector3 AnimationDelta;
//basic calculation
//get the direction vector from the players current position to the future
block position
Vector3 dirVector = futurePos - transform.position;
//find the rotation from the current orientation to the direction vector
Quaternion rotation = Quaternion.FromToRotation(transform.forward, dirVector);
//calculate the distance from you to the cube and scale it with the magnitude of the AnimationDelta
float result = Vector3.Distance(transform.position, futurePos);
result = result / animationDelta.magnitude;
//finally rotate the forward vector by the rotation and multiply it by the
//animation speed and the result to get the step by step movement as
//the animation plays. NOTE: The animation should be based on forward direction
transform.position += (AnimationSpeed * rotation) * result * Time.deltaTime;
Hopefully this does it, like I said I haven't tested it at all so you may have to do some tweaking based on your particular case as this is essentially psuedo-code.
Good luck! I'm off to bed I'll check back when I wake up.

Unity 2D - zoom in =ok, zoom out = borked

I hope someone can help. Im trying to create a little script that zooms in to my player and back out - toggling.
The zoom in works fine, but when I try to zoom back out it doesn't work, it gets stuck. I've created a bool to ensure it only runs the code when it needs to and I'm wondering if that is what's causing the error.
using UnityEngine;
using System.Collections;
public class CameraZoom : MonoBehaviour
{
public float zoom = 10f;
public float normal = 3.471398f;
public float smooth = 5f;
private bool isZoomed = false;
public Camera cam;
public GameObject player;
// lock the camera settings
public float LockedX = 0f;
public float LockedY = 0f;
public float LockedZ = 0f;
private bool hasBeenZoomed = false;
Vector3 targetPos;
private Transform playerTransform;
// Use this for initialization
void Start()
{
targetPos = transform.position;
playerTransform = GameObject.FindGameObjectWithTag("Player").transform;
}
// Update is called once per frame
void Update()
{
if (Input.GetKeyDown("z")) { isZoomed = !isZoomed; }
if (isZoomed == true)
{
ZoomInToPlayer();
hasBeenZoomed = true;
}
else
{
if (hasBeenZoomed)
{
ZoomOutFromPlayer();
hasBeenZoomed = false;
}
}
}
void ZoomInToPlayer()
{
// By default the target x and y coordinates of the camera are it's current x and y coordinates.
float targetX = transform.position.x;
float targetY = transform.position.y;
// ... the target x coordinate should be a Lerp between the camera's current x position and the player's current x position.
targetX = Mathf.Lerp(transform.position.x, playerTransform.position.x, smooth * Time.deltaTime);
//Debug.Log("player x is " + playerTransform.position.x + " and TargetX is " + targetX);
// ... the target y coordinate should be a Lerp between the camera's current y position and the player's current y position.
targetY = Mathf.Lerp(transform.position.y, playerTransform.position.y, smooth * Time.deltaTime);
//Debug.Log("player y is " + playerTransform.position.y+ " and TargetY is " + targetY);
// Set the camera's position to the target position with the same z component.
cam.transform.position = new Vector3(targetX, targetY, transform.position.z);
// Change the size of the camera viewport
cam.orthographicSize = Mathf.Lerp(cam.orthographicSize, zoom, Time.deltaTime * smooth);
}
void ZoomOutFromPlayer()
{
// By default the target x and y coordinates of the camera are it's current x and y coordinates.
float targetX;
float targetY;
// Change the size of the camera viewport
cam.orthographicSize = Mathf.Lerp(cam.orthographicSize, normal, Time.deltaTime * smooth);
// ... the target x coordinate should be a Lerp between the camera's current x position and the original x position.
targetX = Mathf.Lerp(transform.position.x, LockedX, smooth * Time.deltaTime);
// ... the target y coordinate should be a Lerp between the camera's current y position and the original y position.
targetY = Mathf.Lerp(transform.position.y, LockedY, smooth * Time.deltaTime);
// Set the camera's position to the target position with the same z component.
cam.transform.position = new Vector3(targetX, targetY, transform.position.z);
}
}

Your methods ZoomInToPlayer and ZoomOutFromPlayer are written in a way that suggests that they should be called once per frame for the duration of the zoom in/out animation. However, ZoomOutFromPlayer will only be called once, because Update, when ZoomOutFromPlayer is called, the hasBeenZoomed is immediately set to false.
What you're trying to do here, essentially, is a simple Finite State Machine. I suggest researching this design pattern a little more — it will help you noticing the sources of such problems and structuring your code in a better way.
In this particular case, a good way to prevent this problem when designing your code would be to write something akin to "API documentation" for yourself, when writing your methods. For ZoomOutFromPlayer, it would read something like this:
Call every frame when you want to perform zoom-out animation, until the animation is complete.
After you written (and read) such a description, you should immediately notice a red flag — "until the animation is complete"? So, the code that calls this method should somehow take track of whether the animation is complete or not, during a separate mechanism? Wouldn't it make it really easy to use this method incorrectly? Well, that's exactly what happened here.
Instead, what you could've done, is to create two different methods, ZoomInUpdate and ZoomOutUpdate, with descriptions that would read something like this:
Call every frame when the camera should be zoomed out/zoomed in.
This way, using this methods is a lot easier, and you can safely throw out additional logic with hasBeenZoomed out. Just call these methods every frame, and ensure (inside these methods) that they change the camera settings with a certain speed, if these settings need to be changed, or otherwise do nothing.

Try this
using UnityEngine;
using System.Collections;
public class CameraZoom : MonoBehaviour
{
public float zoom = 10f;
public float normal = 3.471398f;
public float smooth = 5f;
private bool isZoomed = false;
private bool isZoomFinished = true; // the animation zoom is over ?
public Camera cam;
public GameObject player;
public float LockedX = 0f;
public float LockedY = 0f;
public float LockedZ = 0f;
private bool hasBeenZoomed = false;
Vector3 targetPos;
private Transform playerTransform;
void Start()
{
targetPos = transform.position;
playerTransform = GameObject.FindGameObjectWithTag("Player").transform;
}
void Update()
{
if (Input.GetKeyDown("z") && isZoomFinished) {
isZoomed = !isZoomed;
isZoomFinished = false;
}
if (isZoomed && !isZoomFinished)
{
ZoomInToPlayer();
}
else if (!isZoomed && !isZoomFinished)
{
ZoomOutFromPlayer()
}
}
float delta = 0;
void ZoomInToPlayer()
{
delta += smooth * Time.deltaTime;
//Cam size
cam.orthographicSize = Mathf.Lerp(cam.orthographicSize, zoom, delta);
//Cam pos
float targetX = transform.position.x;
float targetY = transform.position.y;
targetX = Mathf.Lerp(transform.position.x, playerTransform.position.x, delta);
targetY = Mathf.Lerp(transform.position.y, playerTransform.position.y, delta);
cam.transform.position = new Vector3(targetX, targetY, transform.position.z);
// is animation over ?
if(delta >= 1) {
isZoomFinished = true;
delta = 0;
}
}
void ZoomOutFromPlayer()
{
delta += smooth * Time.deltaTime;
//Cam size
cam.orthographicSize = Mathf.Lerp(cam.orthographicSize, normal, delta);
//Cam pos
float targetX;
float targetY;
targetX = Mathf.Lerp(transform.position.x, LockedX, delta);
targetY = Mathf.Lerp(transform.position.y, LockedY, delta);
cam.transform.position = new Vector3(targetX, targetY, transform.position.z);
// is animation over ?
if(delta >= 1) {
isZoomFinished = true;
delta = 0;
}
}
}

Building an orthographic camera for Unity3D, how can I keep four specific targets in sight?

I have a 2D Orthographic camera for a game that is built similarly to Tennis. I've been working out the code for this, and I think I'm only part of the way so far. At the moment... it... sort-of keeps the two players in frame, with the frame leading more towards the human player - Player One. However, it also has a tendency to drop below the ground level - especially when it zooms out to keep the targets in view. There are four specific targets: The Ground, Ball, Player One, and Player Two.
The ground should always be at the 'bottom', otherwise when the camera goes below ground it shows sky - not really a desirable effect. This is the code I've developed so far... the two players are effectively paddles, but it only shows the center between them:
using UnityEngine;
using System.Collections;
public class CameraController : MonoBehaviour {
// Moving Camera, This Camera tries to follow Player One, Two, and Ball //
public static CameraController Instance;
public Transform PlayerOne;
public Transform PlayerTwo;
public Transform Ball;
public bool Ready = false;
private float MinX = -46.0f;
private float MaxX = 46.0f;
private float MinY = 12.0f;
private float MaxY = 35.0f;
private float LimitFromPlayerX = 20.0f;
private float LimitFromPlayery = 20.0f;
private float DistanceToGround = 0.0f;
private float DistanceFromCenter = 2.0f;
private float Height = 4.0f;
private float Damping = 5f;
private Vector3 EstimatedCameraPosition = Vector3.zero;
private Vector3 WantedPosition = Vector3.zero;
void Awake()
{
Instance = this;
}
public void SetCameraPosition()
{
Vector3 PlayerPosistion = PlayerOne.position;
PlayerPosistion.z = -10.10f;
transform.position = PlayerPosistion;
}
// Update is called once per frame
void FixedUpdate () {
if (Ready) {
Vector3 PlayerPosistion = PlayerOne.position;
PlayerPosistion.z = -10.10f;
Vector3 CameraPosition = transform.position;
CameraPosition.z = -10.10f;
// Get the Distance from the Player to the Ball //
Vector3 WantedPosition = PlayerTwo.transform.position - -PlayerOne.transform.position;
float WPMag = Mathf.Abs(WantedPosition.magnitude);
WPMag = Mathf.Clamp(WPMag, 20, 40);
Debug.Log(WPMag);
Camera.main.orthographicSize = WPMag;
transform.position = Vector3.Lerp(CameraPosition, WantedPosition, Damping * Time.deltaTime);
}
}
void LateUpdate()
{
Vector3 Posit = transform.position;
Posit.x = Mathf.Clamp (Posit.x, MinX, MaxX);
Posit.y = Mathf.Clamp (Posit.y, MinY, MaxY);
transform.position = Posit;
}
}
WPMag was my attempt to zoom out and match the players at least, but so far it's only partially working.

To keep all three objects in view, you'll want to focus the camera on the midpoint or "average" position of the three objects. Getting the midpoint of three objects works just like getting the midpoint of two objects: (PlayerOne.position + PlayerTwo.position + Ball.position) / 3.0f
You'll then want to set orthographic size to the furthest distance from the midpoint of the three tracked objects. This will ensure that all three objects stay in the frame.
Finally, to keep the camera from showing underground skies, you should enforce a lower limit for the camera's Y position. The orthographic size is half the vertical height of the viewport, so you'd use CameraPosition.y = Mathf.Max(CameraPosition.y, groundHeight + camera.main.orthographicSize)