Torque X 3D Update - June 14

I finally got around to writing this after promising it a couple weeks ago. My intent with this is to catch everyone up on what we have so far so that we can start giving incremental updates (that aren't nearly this long). This is going to be a basic overview of what is done or in progress in Torque X 3D. Everything I'm mentioning here is in addition to what you already know exists in the engine. The rendering core, component system, gui, object model, etc of Torque X 2D is all the same in 3D with various minor changes and improvements.

In 2D, the component system is used to implement most of an object's functionality, but there are still some things handled via the standard inheritance model. For instance, T2DStaticSprite, a child of T2DSceneObject and grandchild of TorqueObject, handles rendering itself. In 3D, everything is a component. Everything. There isn't even a T3DSceneObject. Components are added directly to TorqueObjects. And honestly, we're not just trying to push components. It actually just works out really well this way. Here is a list of the base level core components that have been implemented so far and what they do:

T3DSceneComponent - In other engines, this would be the scene object. It manages the object's transform, scene container data, bounds, and visibility.

T3DRenderComponent - Rendering. This is subclassed to implement specific types of rendering. For example, there is a T3DTSRenderComponent that renders DTS shapes (Torque's model format).

T3DRigidComponent - Rigid body physics and collision. I'm not going to go into details on this, I'll just say that the collision system is fast, accurate, and easy to work with.


A bunch of spheres rolling around on the terrain and colliding with each other. The boxes around each sphere are just rendered so rotation can be visualized.

T3DMountComponent - Mounts scene components to other scene components.

T3DCameraComponent - A camera. The camera is set on a scene view and will focus on the object that it is attached to, with the exact transform and control defined by subclasses.

T3DPhysicsComponent - This is higher level physics than what is implemented by the rigid component. It is more for implementing game object specific interactions, like a player moving across a terrain.

T3DAnimationComponent - Handles several different types of animations. Includes a finite state machine for managing state based animations.

There are a few more not as important components, that I won't mention for brevity's sake.


Projectiles.

With these components, we are working towards making a simple FPS game. In an attempt to make the component descriptions actually mean something, I'm going to give a basic overview of how we implemented the FPS player.

There are two physics states added to the physics component. OnGround and InAir. Each one handles the player's movement when the player is in its state. There is an animation component that has a finite state machine for animating the player's movement. There is also an animation that watches the value of the look direction and maps it to the player's head and arms so they follow the camera. A dts render component handles the actual rendering of the player.

A FPS camera type was implemented by deriving from the base camera component. It has the ability to toggle between first and third person and follow the look direction of the player. A mount component was added for mounting flags and weapons, and a rigid component was added for collision. That's pretty much it.


The player standing in a poly soup interior.