Game in a Bottle founder Peter was kind enough to answer some questions for me about his creative and development process. He talked about some more general topics earlier, but this time we dug into his process as a developer some more.

Kongregate: What IDE/development tools do you use?

Peter: Flash CS5, FDT (an Eclipse plugin) for writing the code, Photoshop CS4, Sound Forge 9, and Notepad++ for my to-do list.

Kongregate: The GemCraft games have tons of levels. What kind of process do you use to try to balance that much content?

Peter: Having that many levels was a great mistake; it didn't give the game more variability, only made the development (and especially the testing) time longer. I didn't find a smart method to balance all the levels, all I did was play them over and over a few times.

Kongregate: They say the hardest part of game development is stopping. How do you decide when a game is finished?

Peter: In fact, by the end of the project I was very tired and anxious, so it wasn't that hard to stop. On the other hand, the final game I had in my plans, always turned out to require much more time to reach than I thought. So the "last few weeks" lasted for some 3 months.

Kongregate: You successfully introduced microtransactions with Gemcraft Labyrinth, which is often difficult to do for a single player Flash game. Do you have any advice or lessons learned from that experience for other developers who hope to follow the same path?

Peter: It was a great experiment, I had no idea if it would succeed or not. Luckily, the player reception was much better than I thought, many players didn't have any problem sparing a few bucks on the game. My main thought for pricing was not to be greedy, but to give real value for a reasonable price. But maybe I'm wrong and the game could have earned much more if there were unlockable cheats and whatnot for money. Maybe next time, although I feel that it could ruin the reputation of the GemCraft series.

Kongregate: Do you have any advice or suggestions for young, aspiring game developers?

Peter: Start with a small, simple game, and get higher until you're certain you can make a game that can stay on the front page of the major portals for at least 2 weeks.

Making a game demands tons of devotion, giving up your free time, and working late hours, if you want the game to be ready to launch sooner than 3 years from now (or rather, never).

You can work alone, or in a group, each has its pros and cons. These days I'm working alone, that means I have to do everything, from sounds to art to coding. A group can advance much faster, but it needs lots of meetings (chat sessions), there's a chance of miscommunication, or a member leaving the group.

As the months go by, the project will start to rot, and you'll have to get even more determination to stay on target, or to keep to group together.

It's not enough it you enjoy playing your game, show it to others, let them sit down and start playing, and watch them as they try to get past your main menu. You'll be amazed how much you can still improve the ease of use of your game.

If one minute passes, and the player is still not playing the game itself, but adjusting some avatar colors or inventory, your game is bad. Start with an intro level, toss the player right into the action. Picking weaponry and fancy looks for the hero can wait.