I have heard fantastic comments about the Physics library and Mathematics libraries and so I am currently using it in my current game project for XNA.

My problem is that I am used to using engines with a GUI such as Unity and I am not used to programming complete 3D camera movement and I'm not 100% certain on the best approaches. An example of this is that I'm trying to not use the Matrix44F.CreateLookAt instead favouring a more simple approach by multiplying the position matrix to the rotation matrix. Is this a good approach?

Currently I am writing a base camera class. I want it to be able to move freely but have no input logic although it will contain all the methods to control movement through 3D space.

I am storing the current position as a Vector3F and the rotation as a QuaternionF. I am able to get rotations and simple movement but I am new to this and so I am unsure of how to move in a local direction. Currently it is very easy to move around in the global space but I don't know how to feed a local direction such as Vector3F.Forward * speed to a method to move the camera in a local forward direction instead of simply globally moving along the Z Axis.

I am sorry if this is an extremely basic question but just thought I would ask the math magicians over here for any help to aid me in my pursuit for a nice Camera class. Any recommendations or points I should consider would also be nice. Don't worry I don't expect anyone to give me all the code just hints and bits of help.

Thank you :D

Hi freakycode,

usually I would also point you to other tutorials (thanks winterkewl for providing the link!).

Except in this case DigitalRune Geometry offers a special type called Pose. You won't find it in XNA.
A Pose defines a position + orientation. We here use it almost everywhere to position objects in a world (rigid bodies, models, cameras, etc.). It was designed to handle the 90%-case. I have written a short two-part blog series that should give you an introduction and hopefully answers some of you question. (Part 2 explains how you can use Pose to control a camera.)  

Part 1: Pose = Position and Orientation
Part 2: How to use a Pose to define World and View matrix

I don't want to imply that this is the right way to do it - but it works great for us.
Let me know if it was helpful.