So there we were – two months into the crazy adventure of running our own games company. We had announced Hogrocket to the world a few weeks earlier, and (fortunately) had been fighting off interview requests from the press and e-mails from potential partners. Our website was doing the business; sign-ups to our mailing list were gradually rising as was our Twitter follower count. We even got a couple of CVs through! So far so good.
As for development itself, we had a few interesting iOS prototypes in progress at this stage. We'd spent a while looking at gameplay mechanics that would work on a touch screen. We were particularly interested in games that had the player trace a route, or drag things around, or rotate stuff, or other motions that were particularly unique to this kind of platform. We soon gravitated toward one prototype in particular; a little demo we called Trains.
From prototype to game
The concept of the game was quite simple: a little train moved along its track at a set speed. You tapped the junction points to change the route of the train, and if you drove near any of the little sheep alongside the track they would hop into one of your train's carriages. You completed the level by taking these sheep back to your shed, rewarding players who chose the fastest or most efficient route around the level. What we had was a simple but addictive collect 'em up – ideally suited to mobile platforms.
After deciding to concentrate on Trains, we made a conscious attempt to schedule our time effectively. This was met with, well, varied success. Right off the bat we set up a detailed multi-stage pre-production plan, which went something like this:
1. Tech development, engine and gameplay
First we build all the necessary technical gubbins to make the thing work effectively on all devices. This also included supporting elements like building an art pipeline, a level editor, custom online analytics tracking, and stuff like that.
2. Adding game elements: time, medals, level progression
Once we had a decent base mechanic, we worked to enhance it and add a bit of depth. This included adding a time limit, medals for completing levels in a set time, and also a tree-based progression through the levels, so they could be unlocked in a unique order.
3. Trying out some different control mechanics
We then revisited the core mechanic once more, and tried a few different control methods. Tapping the train to apply a set boost was a favourite, but we also tried analog boost (hold to go faster) and following the train with your finger to activate the boost.
4. The great 3D distraction
Then we did something stupid. We did some research and ended up getting pretty addicted to Train Conductor (an awesome game, by the way). We especially loved its 3D trains and how they moved convincingly over the 2D background. We figured that something like that would work well for our train game too, so we started moving down this road. We built some 3D tech and interviewed some 3D artists with the intention of having them build us some prototype models. This was quite an involved process, considering there's a lot of code to write and a lot of stages involved in contracting an artist. We needed to find someone, interview them, explain the concept of the prototype and where we wanted to go with it, introduce them to our codebase and what we needed technically, and then have them actually do the work.
Then we figured, "Hey, if we're going to all this effort we might as well get some 3D trees, bushes and buildings made up too". That led to experimentation with a 3D landscape, and before we knew it the entire game had moved to 3D.
This bought a whole bunch of problems, namely that the game took longer to load, wouldn't support older devices as effectively, was harder to play (due to the perspective and varying sizes of hit zones), and was a much more time-consuming (and thus expensive) thing for us to create art and levels for. In the end we dumped 3D completely and went back to 2D. All in all we lost about a month and a half's worth of work to this great distraction…
Pages — 1 2