Microbenchmark test for MonoBehavior vs RobotArms Component + Processor

gistfile1.txt

2 prefabs consisting of the default Unity cube were created, each with 1 of the two scripts attached. A test harness then instantiated 1000 of each cube at the origin.

Profiler results:

Overview                          Total         Self        Calls     GC Alloc      Time ms     Self ms
----------------------------------------------------------------------------------------------------------
Camera.Render                     44.0%         0.2%        1         0B            6.47        0.03
BehaviourUpdate                   42.6%         6.0%        1         160B          6.27        0.88
  - MonoBehaviourJiggle.Update()  18.6%         17.9%       1000      0B            2.74        2.64
  - RobotArmsCoordinator.Update() 18.0%         17.7%       1         160B          2.64        2.61
Physics.Processing                5.4%          0.3%        1         0B            0.79        0.04

MonoBehaviourJiggle

using UnityEngine;

public class MonoBehaviourJiggle : MonoBehaviour {
	public Vector3 Amount;

	void Update () {
		transform.Translate(new Vector3(
			Amount.x * Random.Range(-1, 1),
			Amount.y * Random.Range(-1, 1),
			Amount.z * Random.Range(-1, 1)
		));
	}
}

RobotArmsJiggle

using UnityEngine;
using RobotArms;

public class RobotArmsJiggle : RobotArmsComponent {
	public Vector3 Amount;
}

RobotArmsJiggleProcessor

using UnityEngine;
using RobotArms;

[ProcessorOptions(typeof(RobotArmsJiggle))]
public class RobotArmsJiggleProcessor : RobotArmsProcessor {
		
	public override void Process(GameObject entity) {
		var component = entity.GetComponent<RobotArmsJiggle>();
		component.transform.Translate(new Vector3(
			component.Amount.x * Random.Range(-1, 1),
			component.Amount.y * Random.Range(-1, 1),
			component.Amount.z * Random.Range(-1, 1)
		));
	}
}