Project Premise
The idea for my THREE.js game is the have the player navigate though a maze of objects and reach an exit. I need collision detection not just for cubes and spheres, but also models imported from Blender. I thought about using RayCasting but I decided to go with Physi.js.
The Problem
,The idea is to give the camera a "body" so that the player can't pass though wall or other objects like they can when they are just a camera object. However, My "player" object seems to loses all of its Physi.js attributes as soon as I give it THREE.PointerLockControls(player). What I'm doing is creating a Physijs.BoxMesh, the player, and adding the camera to that. After that I pass the "player" object to my setupControls() function.
var geometry = new THREE.BoxGeometry( 500, 500, 500 );
var material = new THREE.MeshPhongMaterial( {color: 0x0000FF} );
player = new Physijs.BoxMesh(geometry, material, 1, {restitution: .9, friction: .1});
scene.add(camera);
scene.add(player);
player.add(camera);
//give player control of THIS mesh.
setupControls(player);
The Controls work as expected, I can control the player mesh with the camera stuck to it creating an FPS view, but I when I do this I can still move though walls and other objects. I've even moved the camera back on the Z position to confirm whats going on and I can see that the player mesh is just passing though objects instead of being hindered or knocking them over.
Solutions?
If I remove setupControls(player), then the Physi.js physics begin to work on the player mesh! It will fall from gravity, bounces around and everything. It looks really cool, but now I can't control it! It seems like I can only have one or the other lol. So does anyone know what I could do to solve this problem? Is what I'm preposing even possible? I'm new to THREE.js so any input is much appreciated!
