Mace Griffin, Bounty Hunter

Warthog Games | 1999-2003

Between 1999 and 2003 I worked as the sole lead animator on Warthog’s internal space shooter/FPS hybrid ‘Mace Griffin, Bounty Hunter’ which was developed in the wake of, and stood a lot of comparison to Bungie’s ‘Halo’.

Early trailer cut

There were many many good, and a few bad things about the development of ‘Mace Griffin, Bounty Hunter’, some of which I’ll briefly cover..

The Good

The team. The bulk of the team, generally the people who are still involved in games dev now, were a really talented bunch of friendly people and it was a privilege to have worked with them. Most of these people have gone on to become prominent figures within the industry.

The Company, Warthog Games. By far the best games developer I have worked at during my career to date. A really friendly family atmosphere that encouraged creativity.. unlike some of the more cut-throat firms I’ve worked at in more recent times.

The game itself.. The general concept, style and genre; A large scale sci-fi FPS space opera. All of which just about ticked every box for everyone involved.

The fact that I pretty much single handedly designed, planned, created and implemented all of the games animation – which gave me total creative freedom and a warm fuzzy feeling that every bit of animation was down to me. (see also ‘bad’ 🙂 )

The Bad

Most of the team hadn’t any previous experience on a FPS game (this included yours truly); this ended up showing in the final product and sadly resulted in average review scores which in turn killed any hope of a follow up.. The sad fact here is that by the end of the project we could all see the mistakes and had we been gifted that second chance would most certainly have nailed it.

The fact that I pretty much single handedly designed, planned, created and implemented all of the games animation – For such a large game with what at the time was considered a huge animation footprint it would probably have been wise to have had a small team of animators on board.

Yooouu Neeeed Hands…

The first person hands/weapons were one of the first areas I tackled on Mace Griffin (then known only as ‘Bounty Hunter’ in-house).

I opted to use 3D Studio Max’s Biped for the task as it would allow me to very easily blend individual actions together. It would later also prove useful in handling the full-body and partial-body motion capture data that covered all of the bipedal characters within the game so made a good all round choice of motion-edit/keyframing tool at the time.

Below is an early First Person weapons test that I stitched together to proof the motion and transitions. All animation was hand key-framed by myself using 3D Studio Max’s Biped.

The animation is quite long and basically served to allow a visual appraisal of a very large number of possible branching of weapon animations/transitions.

The first person hands/weapons are the only animation sets within the game that aren’t driven via AI-logic or designer script, they’re driven by player input through a simple state graph and thus pretty much just plugged straight in as I had animated them without any ugly unpredicted blending or lack of.. for this reason they stand out as some of my best work when playing the actual game.

..and here’s what IGN’s Hilary Goldstein wrote in his review of the game back in 2003 shortly after its release..

“Mace is, in general, a very attractive game. The real-time lighting on the weapons and the awesome weapon animations are among the best I’ve ever seen. Mace makes reloading a gun seem like an art form.”.

Ahhh, bit of self indulgence there 🙂 It’s nice to see that something about Mace was appreciated as the game provoked some quite scathing reviews at the time. The brown envelope is in the post by the way Hilary 🙂

Characters | Mocap

All bipedal character motion data was meticulously planned out prior to going to motion capture at Granada Television studios in central Manchester (UK). Mocap is expensive to both capture and to post process so I made doubly sure that I had it all planned out fully prior to hiring performers and studio time.

We spent a total of around 7 days on principle shooting and a further 4 days on pick-up shots later down the line.

The bulk of the motion data was good quality. The performances were by and large good. We’d been pretty careful to grab multiple takes during the shoot so could narrow down the best selects.

Characters | In-Game implementation

The characters within the game are all driven either by AI or via script, both of which in turn drive a state-machine which is set up to handle blending between states. Sadly the state machines ended up being quite primitive and a great deal of the motion data that I’d prepared to be part of these state and transitioning networks, simply wasn’t used due to various constraints and issues during the projects life, which I won’t go into but there are some interesting anecdotes here..

It didn’t help that the AI coder, who’d spent around 2 years developing the unique AI system for the game ended up leaving and causing the AI to be re-written from scratch as we approached the final phase of development.

We were also expecting a ragdoll physics system for the game to cater for all impact\death states, this was to be coded in-house and didn’t receive development time until very late into the project cycle.

Unfortunately the ragdoll system didn’t really work out 😀 so it was decided at this point that we needed a whole bunch of impact and death animations.

All of these, pretty major, setbacks to the character work for the game ultimately took it’s toll and can sadly be seen in the locomotion of the final game characters.

BadSmallCat!