I love it. Gambit-C Scheme is my favorite also (although MzScheme has lots of great libraries). I have an Android phone - any chance of an Android version? I've seen some links of compiling to Java byte code, and there is a native SDK.
I don't have access to any of the Android platform yet, so not any time soon. There should be no problem targeting it though. This was highly an experiment, and I already had enough on my plate.
In the next year or two I hope to move away from the iPhone in favor of Android.
Neat. Could you talk more about Scheme. In retrospect how did it work out? If you could do it over again would you use Scheme or just jump into Objective-C? What were the pros and cons of using Scheme.
Sure! I plan to write a good post-mortem on my blog soon, so I will go into all of that in detail then.
In general, I'm really glad I used Scheme. Gambit Scheme worked great, and let me develop the game much much faster using incremental develop commonly found in the Lisp family of languages. View this video of mine to see what I mean:
There were some downsides. The main one being that people didn't understand what I was doing, and I couldn't have hired another programmer if I wanted to. I also developed it from scratch because I didn't want to bother with writing bindings (but that was also to learn more about graphics). Also, I feared that the GC would make it too slow, and in the end, it was hard to control what the GC was doing. It worked out pretty well, but although it runs decently fast, the GC still incurs a pretty hard performance hit.
All said, I would do it again in a heartbeat, but I would probably make bindings to an existing framework and use it more as a lightweight scripting language. That way, the GC isn't run as often, and a lot of developers would be familiar at least with the API.
