I have come across examples in three.js that use:
renderer = new THREE.WebGLRenderer( { canvas: document.querySelector( 'canvas' ) } );
This relates to a <canvas></canvas> element.
On the contrary, there is another method:
renderer = new THREE.WebGLRenderer();
var container = getelementbyid("container"):
container.appendChild( renderer.domElement );
This refers to a <div></div> element.
Both approaches use WebGLRenderer and can style the canvas. Are they exactly the same or are there advantages and disadvantages such as device support, performance, and limitations - why would one choose one over the other?