Choosing a game engine with valid evidence

We have an edge piece to lay down
When putting together a puzzle, I always start with the edge pieces. (No, not the corner pieces. And I have no philosophical reason why not.) I feel like we have an edge laid down now.

It’s good to start out with a game idea. Bill had the idea.

However, during our brainstorming sessions, we dreamt up all aspects the golden version of the game. I pictured a detailed, opulent 3D multiplayer world with amazing cooperative features and a deep story woven into the gameplay. The learning aspect would be so subtle, so full of analogy, and so graduated, that people wouldn’t even notice they were learning theory as well as practice. I pictured eventually moving the game to Xbox and using the new NATAL input device where people could make the motions of turning the wheel of the press and would see their character doing it on screen. I had definitely left reality.

When I started trying to implement the game in SecondLife, then Unreal, I got bummed out cause I didn’t see how those engines could fulfill our golden vision. Then, doubt crept in when I heard things like Bill lamenting that he couldn’t get anyone in his demographic to log onto SecondLife. Or that they might not have the hardware required for Unreal.

Postmortem: I think if I knew last year what I know now, I would’ve asked for data from our player demographic as soon as we finished with the game concept and before we got emotionally invested in gameplay details. Now we can choose a 3D game engine and begin detailed design and development with total confidence. What a difference that makes!! I’ve been working more on the game in the last 48 hours than in the last couple months. I’ll be showing Bill a demo in Silverlight on Sunday. When we discuss what we like and want to change, I’ll know ahead of time what’s possible.

Game engine choice
First, since I already had it, I looked up the system requirements of Unreal 2004. Funny. If I used Unreal 3, only 40% of Bill’s survey respondents would’ve been able to use the game. With the older version of Unreal, almost all of them could’ve used it. But now that I’ve seen Unreal 3, it would break my heart to use Unreal 2004.

I won’t go near the Source engine because of the massive frustrations I experienced, with both Valve’s Steam delivery mechanism, and using Source in a school environment. Of course, I was reminded that I hadn’t tried Steam since 2006, so I tried it again last week. It STILL brings my computers to a standstill.

Last night I found a game engine that looked positively awesome! Game Studio. It was inexpensive and came with lots and lots of assets. Also, it boasted rapid and easy development with the use of behaviors that you can tack onto game objects.

But then I found a discussion comparing Game Studio with Torque 3D here. Wowsers. I wasn’t thrown by the negatives people raised about Game Studio because they really weren’t that bad. I assumed some loss of detail and flexibility in a rapid application development (RAD) tool. However, the comments people made about features Torque 3D were illuminating.

With Torque3D, I’d get everything I wanted, plus some things I didn’t expect. You can publish your game to the web!! Holy crap! Since when?! That would help for Bill’s customers who can only use a library or internet café computer, with which they wouldn’t be allowed to install anything. Also, if we build an installable game, you can build for the Max and Linux as well as the PC. Awesome!! And, if we use TorqueX 3D, we can build for Xbox 360. In school, I preferred Unreal to Torque, but I’m gonna have to get over that. Yes, Torque has less assets and it takes more time to get things going, but on the up-side, Torque was not built to favor shooter games. Yay! That satisfies Bill’s concern about “Engines conceived *outside* of socially redemptive cultures (cultures based in competition, aggression, exploitive imperialism, etc.) are less likely to serve.”

OK. On with developing the demo. I have 5 days.