In the #gamedevelopers channel, the topic of game development contests came up. I mentioned that I would like to join one. The last one I was in was GiD #13 in June of last year, so I should do something similar again. One person mentioned that she didn’t like such contests as she didn’t have time for them. I argued that I could get some good practice in, but she responded with:
<turbo> no practice. just do and redo
<turbo> we like to call them polishing passes 😀
…
<turbo> it does get better every time round
I really liked this idea. It’s basically how I was treating Oracle’s Eye. When I started it, it was just supposed to be a quick project to give me more experience with game development. Eventually, I realized that I could make a commercial project with it.
Now I was aiming for a bigger target. I assessed what I already had, and I realized that I needed to up my game (no pun intended…ok, it was) in more ways than one. So I updated and changed and added and deleted until I had a better version.
Was I gaining experience? Yes. But I didn’t consider it practice. I was simply doing. And when I revisited some aspects of my code, I ended up making it better most of the time. Each pass made it better. It is kind of like writing a term paper for class. If you get it done early enough, you have time to go through it and check for errors, awkward sentences, and possible confusing ideas.
Polishing passes. Do, then do again. Each pass should result in more polish.
EDIT: Now I no longer misquote turbo. B-)
2 replies on “Polishing Passes”
This is the way that I write most of my code. A lot of the time I don’t have a clear understanding of what it is I have to do, so I will get something working, and than I look at it and try and decide how I could make it better. This is similar to your previous post about planning. Planning is important, but if you are not sure how something should fit together, you can’t really plan it.
I’ve always viewed writing code like writing a paper. It’s never usually done in just one pass, and doing a couple drafts is bound to give you something better than you had before.
Thanks, Casey! Most people can’t write a paper in one sitting and get a high grade for it. Similarly, I can see that game development might take multiple passes before a game might be considered high quality. It’s easier for smaller projects, of course. I don’t see Half-Life 2 being scrapped and redone after years of development.