Develop Reference

Referencing Instantiated objects after their creation in Unity

Hi!
After the discussion with Ruzihm in the comments. I've now created a simple version of my game to better ask the question I'm having.
The question now is, since I'm not able to manually create a connection to the testObject field in the inspector. How do I now tell Unity to use my instantiated objects while the game is running?
And is this a good solution for a RTS game that may have 100s of Units active at a time? The end goal here is to apply this force to a radius around the cursor. Which I was thinking of using Physics.OverlapSphere
Here's the minimal scenario of what I have:
New Unity scene
Attached the InputManager to the main camera.
Created a capsule and a plane.
Added ApplyForce to the Capsule
Created a prefab from the capsule and deleted it from the scene.
In the InputManager I added the ability to press space to Instantiate a capsule with the ApplyForce script attached..
Drag the capsule prefab to the InputManager "objectToGenerate"
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
namespace GL.RTS.Mites
{
public class InputManager : MonoBehaviour
{
public GameObject testObject;
public ApplyForce onSpawnTest;
public GameObject objectToGenerate;
void Start()
{
onSpawnTest = testObject.GetComponent<ApplyForce>();
}
void Update()
{
if(Input.GetKeyDown(KeyCode.Space))
{
Instantiate(objectToGenerate);
}
if (Input.GetMouseButton(0))
{
onSpawnTest.PushForward();
}
}
}
}
The ApplyForce script that I attach to the Capsule:
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
namespace GL.RTS.Mites
{
public class ApplyForce : MonoBehaviour
{
public float moveSpeed;
Rigidbody rb;
void Start()
{
rb = GetComponent<Rigidbody>();
Debug.Log("A Mite has spawned!");
}
public void PushForward()
{
rb.AddRelativeForce(Vector3.up * moveSpeed * Time.deltaTime);
Debug.Log("A force of: " + moveSpeed + " is being added.");
}
}
}

Well, you are creating your new instances of your object, but your input manager immediately forgets about them (note that you do nothing with the return value). The InputManager only knows about the ApplyForce that was created in its Start (and then interacts with it depending on mouse input) and your ApplyForce script knows nothing about any InputManager. So, it should come as no surprise that only the first instance reacts to the mouse input.
So, something has to be done to your InputManager and/or your ApplyForce. Your InputManager could remember the instances it creates (which isn't enough, because what if for example, a map trigger creates new player controllable units) or it could go looking for units each time.
Your ApplyForce could register with the InputManager when they are created, but then you would need to loop through the units and find out which ones are under the mouse, anyway.
Since you only want to select ones based on what is near or under your cursor and only when input occurs and not like every frame, I would go with the simplest approach, just letting your InputManager find the units when it needs them. Something like below, explanation in comments:
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
namespace GL.RTS.Mites
{
public class InputManager : MonoBehaviour
{
public GameObject testObject;
public ApplyForce onSpawnTest;
public GameObject objectToGenerate;
private Camera mainCam;
// which layers to consider for cursor detection
[SerializeField] LayerMask cursorLayerMask;
// how big for cursor detection
[SerializeField] float cursorRadius;
void Awake()
{
// cache main camera
mainCam = Camera.main;
}
void Update()
{
if(Input.GetKeyDown(KeyCode.Space))
{
Instantiate(objectToGenerate);
}
if (Input.GetMouseButton(0))
{
Collider[] colls = FindCollidersUnderCursor();
// check each collider for an applyforce and use it if present
foreach( Collider coll in colls)
{
ApplyForce af = coll.GetComponent<ApplyForce>();
if (af != null)
{
af.PushForward();
}
}
}
}
Collider[] FindCollidersUnderCursor()
{
// find ray represented by cursor position on screen
// and find where it intersects with ground
// This technique is great for if your camera can change
// angle or distance from the playing field.
// It uses mathematical rays and plane, no physics
// calculations needed for this step. Very performant.
Ray cursorRay = mainCam.ScreenPointToRay(Input.mousePosition);
Plane groundPlane = new Plane(Vector3.up, Vector3.zero);
if (groundPlane.Raycast(cursorRay, out float cursorDist))
{
Vector3 worldPos = cursorRay.GetPoint(cursorDist);
// Check for triggers inside sphere that match layer mask
return Physics.OverlapSphere(worldPos, cursorRadius,
cursorLayerMask.value, QueryTriggerInteraction.Collide);
}
// if doesn't intersect with ground, return nothing
return new Collider[0];
}
}
}
Of course, this will require that every unit you're interested in manipulating has a trigger collider.

Unity navmesh agents

I'm working on a RTS game. I did a normal move script and I increase the agent's stopping distance so they don't look at each other and pump that much. But they still hit each other.
I can't find the way to make agents avoid each other and not push each other. Or someway to ignore the physics while still trying to avoid each other. Here is the code for the click to move
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
using UnityEngine.AI;
public class moveTest : MonoBehaviour {
NavMeshAgent navAgent;
public bool Moving;
// Use this for initialization
void Start () {
navAgent = GetComponent<NavMeshAgent>();
}
// Update is called once per frame
void Update () {
move();
}
void move()
{
RaycastHit hit;
Ray ray = Camera.main.ScreenPointToRay(Input.mousePosition);
if (Input.GetMouseButtonDown(1))
{
Moving = true;
if (Moving)
{
if (Physics.Raycast(ray, out hit, 1000))
{
navAgent.SetDestination(hit.point);
navAgent.Resume();
}
}
}
}
}

From the following link (which sadly don´t have images anymore)
Modify the avoidancePriority. You can set the enemie´s value of
30 or 40 for example.
Add a NavMeshObstacle component to the "enemy" prefab
Use this script for the Enemy Movement
Script
using UnityEngine;
using System.Collections;
public class enemyMovement : MonoBehaviour {
public Transform player;
NavMeshAgent agent;
NavMeshObstacle obstacle;
void Start () {
agent = GetComponent< NavMeshAgent >();
obstacle = GetComponent< NavMeshObstacle >();
}
void Update () {
agent.destination = player.position;
// Test if the distance between the agent and the player
// is less than the attack range (or the stoppingDistance parameter)
if ((player.position - transform.position).sqrMagnitude < Mathf.Pow(agent.stoppingDistance, 2)) {
// If the agent is in attack range, become an obstacle and
// disable the NavMeshAgent component
obstacle.enabled = true;
agent.enabled = false;
} else {
// If we are not in range, become an agent again
obstacle.enabled = false;
agent.enabled = true;
}
}
}
Basically the issue that this approach tries to solve is when the player is surrounded by enemies, those which are in range to attack (almost touching the player) stop moving to attack, but the enemies in the second or third row are still trying to reach the player to kill him. Since they continue moving, they push the other enemies and the result is not really cool.
So with this script, when an enemy is in range to attack the character, it becomes an obstacle, so other enemies try to avoid them and instead of keep pushing, go around looking for another path to reach the player.
Hope it can help you some how

making a 2D GameObject react to the position of another 2D GameObject

So, my boss moves in a specific direction when it detects the player. The problem I'm having is how to get the boss to move depending on where the player is within a certain proximity. So if the boss is on the player's left, he'll move to the left. If he's on the player's right, he'll move to the right. But I can't figure out how to make him react based on distance. Right now I'm just doing a Debug.Log to save a few seconds.
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
public class phantom : MonoBehaviour {
private Rigidbody2D rb;
private Animator anim;
public Transform Target;
void Start ()
{
rb = GetComponent<Rigidbody2D> ();
anim = GetComponent<Animator> ();
}
void Update ()
{
if (transform.position.x > Target.position.x ) {
Debug.Log ("left");
}
if (transform.position.x < Target.position.x ) {
Debug.Log ("right");
}
}
}

You could use the Vector3.Distance method to determine the distance between your two object based on thier respective transform. That way, you can modify your boss's behavior according to his proximity to the player. The smaller the magnitude value is, the closer your two transforms are.
Ex:
int distanceYouWant;
if(Vector3.Distance(transform.position, Target.position).magnitude < distanceToDoStuff)
{
Debug.Log("Boss do stuff!");
}
Here is the link to the Unity scripting API doc: https://docs.unity3d.com/ScriptReference/Vector3.Distance.html
Hope this helps!

I figured it out. I just made the function not in Update but with an OnTriggerEnterStay2d (Collider2D other). Then I placed a trigger collider on the same GameObject and only when it detects the Target (the player) does the debug come up.

Why cant I move my camera in (Unity)C#?

camera transform
I am trying to move my camera based on the players' movements on Y axis in Unity.
However, it does not work...
What did I do wrong? I have attached image of my script (C#) here.
and, Yes, I did attach this script with Main Camera.
using System.Collections;
using System.Collections.Generic;
using UnityEngine;
public class CameraController : MonoBehaviour {
GameObject player;
// Use this for initialization
void Start () {
this.player = GameObject.Find("cat");
}
// Update is called once per frame
void Update () {
Vector3 playerPos = this.player.transform.position;
transform.position = new Vector3(
transform.position.x, playerPos.y, transform.position.z);
}
}

Make the player GameObject public and just drag and drop your player in the inspector in unity see if that works? Are you getting any exceptions? Also add Debug.Log (player.transform.position.ToString ()) to see if it is showing the right values. Are you sure you player object name is cat and not Cat, it is case sensitive. Check on those things and let me know if you figured it out!

How to attach the camera to a player object instantiated by HLAPI Network Manager?

In short, I have a very simple multiplayer game. It's the Roll A Ball game (Unity3D tutorial). So right now I have the players etc spawning perfectly and everyone is able to control their own balls perfectly fine.
But here's the problem: I've got a default Main Camera. Since it's only the local player itself that needs to see it, I figured there's no point in trying to spawn a seperate camera for each player on the server.
However, to make the camera follow the player, I need to attach it the player gameobject. Obviously I can't attach it to the player prefab as it's a clone the camera needs to follow. But since the player is being spawned by the Network Manager component, I have no idea on how to refer to this clone.
What I've tried myself:
public class CameraController : NetworkManager
{
public GameObject playerPrefab;
public Transform target;
private Vector3 offset;
public override void OnServerAddPlayer(NetworkConnection conn, short playerControllerId)
{
GameObject player = (GameObject)Instantiate(playerPrefab, new Vector3(0, 0.5f, 0), Quaternion.identity);
target = player.transform;
NetworkServer.AddPlayerForConnection(conn, player, playerControllerId);
}
void Start()
{
offset = transform.position - target.position;
}
void LateUpdate()
{
transform.position = transform.position + offset;
}
}
But this resulted in:
Which I find extremely odd since as you can clearly see, there's no NetworkIdentity component on the NetworkManager object. I've been trying A LOT of things for the past 4 hours now and I just can't do it. So now I'm hoping you guys can help me out.
Edit: This is how the Network Manager normally spawns a player. As you can see, there's no code for it:

I had the same issue and figured out the followig solution. Seems like you already got a solution, but maybe it is interesting to share some possible ways for other people in the same situation.
This is a way to do it without a camera attached to the prefabs.
I'm using a NetworkManager to instantiate Player-prefabs. (Same as you)
I solved the problem of finding references to the clone objects by letting the clones tell the camera, who they are (or which transform belongs to them).
The Player has the following script:
using UnityEngine;
using UnityEngine.Networking;
using System.Collections;
public class PlayerController : NetworkBehaviour {
public override void OnStartLocalPlayer()
{
GetComponent<MeshRenderer>().material.color = Color.blue;
Camera.main.GetComponent<CameraFollow>().target=transform; //Fix camera on "me"
}
void Update ()
{
if (!isLocalPlayer)
{
return;
}
var x = Input.GetAxis("Horizontal") * Time.deltaTime * 150.0f;
var z = Input.GetAxis ("Vertical") * Time.deltaTime * 3.0f;
transform.Rotate (0, x, 0);
transform.Translate (0,0,z);
}
}
On my default main Camera (there is no camera attached to the player prefab, just the default camera) I put the following script on. It takes a target which I initialised with the prefab using the inspector.
using UnityEngine;
using System.Collections;
public class CameraFollow : MonoBehaviour {
public Transform target; //what to follow
public float smoothing = 5f; //camera speed
Vector3 offset;
void Start()
{
offset = transform.position - target.position;
}
void FixedUpdate()
{
Vector3 targetCamPos = target.position + offset;
transform.position = Vector3.Lerp (transform.position, targetCamPos,smoothing * Time.deltaTime);
}
}
After starting the game, each clone tells the camera who he is, so the target changes to the individual clients clone with this line from the Player's Script:
Camera.main.GetComponent<CameraFollow>().target=transform; //Fix camera on "me"
This way you don't need to create one camera per instance of player-prefabs (I'm not sure if this makes big differences in performance) and you don't have to deactivate all cameras which don't belong to your client.
If you host the game in the editor you can see that there is just 1 camera instead of one camera per connected client (like when you attach it to the prefab).
I think this is a good use of this method, you can use it to put things in it, which should be applied to the Local Player only.
public override void OnStartLocalPlayer()
{
}
I tried by starting the game in the editor and in a build and it seems to work well.

I would add a camera to the prefab and then write a player script like this:
using UnityEngine.Networking;
public class Player : NetworkBehaviour
{
public Camera camera;
void Awake()
{
if(!isLocalPlayer)
{
camera.enabled = false;
}
}
}

I've not really worked with networking but what if you do this after you spawn the local player
Camera.main.transfor.SetParent(the transform of the local player here);
As I understand the problem each separate instance of the game has a main camera.

Thanks to Rafiwui's point into the right direction, I've finally managed to get it working. All I had to do was adept his code a bit. The end result was:
public Camera camera;
void Awake()
{
camera.enabled = false;
}
public override void OnStartLocalPlayer()
{
camera.enabled = true;
}
Thanks A LOT to you all for helping me out! This has been quite a day for me.