nesdev.org

MetalSlime wrote:...I think this approach leads to a lot of "rewrites" and "reorganization" of code I find myself doing every few months or so.

Ack, same here!
I do like quick results, though. It keeps me motivated over a long period of time. Like, I'd be on a coding spree, spending all of my free time on the project for a month and get a lot done, then progress stalls for two months because the next hurdle is too big or I'm just burned out. Todo lists work fine for a while to get back into the action, but after a week or two, when I'm in the middle of my "coding mania" any attempts to update them slow down the pace of work so much.
I keep getting back to it, though, so that's definitely a step up from all insta-failed projects before it.

marvelus10 wrote:Something like this before you start coding.

http://dl.dropbox.com/u/2590713/SMB1.jpg

Yeah, if you're in a team it's obviously a good idea to have elaborate design documents.
If you're working alone, at least outline the general idea of your game, even for small projects. That way you will never lose the big picture when your mind needs to focus on the details of the actual implementation.

I'm a huge sucker for that kind of footage by the way, where is that picture from? Did I miss another Iwata Asks interview?
EDIT: Oops, just saw the new topic in General Stuff.

I have an interesting thought to add that I've found helpful for maintaining momentum over time: Always do slightly less than you *want* to do. It's like you have an energy bar that is constantly refilling on its own very slowly... (I can think of a few games with something like that...). Don't use it up too fast!

Gradualore wrote:I have an interesting thought to add that I've found helpful for maintaining momentum over time: Always do slightly less than you *want* to do. It's like you have an energy bar that is constantly refilling on its own very slowly... (I can think of a few games with something like that...). Don't use it up too fast!

I like this. I think I actually subconsciously do this.

I am in the middle of entering all of my bugs and feature ideas into a web-based bug tracking system (trac) that integrates with my revision control system (subversion). I should have done this sooner.

I find that having a granular todo list helps keep the momentum going. I've got a private beta-tester who has given me some excellent feedback.

I find that my development effort breaks down into four phases:

1) Initial design and prototyping. Lots of fun. Can't stop coding. I can't sleep at night because my mind is racing with what I want to code but haven't gotten to yet.

2) Slow slog to getting a usable prototype that implements enough of the game that its worth playing. This is where I tend to lose the most stream, as this phase is mostly grunt work.

3) Beta-testing. I now have a playable product, but it needs lots of bug fixes, tweaks and other enhancements. This phase is both mundane and a bit exciting. I can see the end, I have a playable product. Momentum picks back up. I just entered this phase with my NROM game-compo entry.

4) "long tail". You know the saying, "The last 10% takes 90% of the time". Lots of grunt work, some boring. It is easy to get burnt here and "take a break", but lose interest while away. I have a great feeling that this won't happen this time around, as the compo is coming up (ob-bitch: Is it?)

Unless someone else decides to host this compo, I don't see it happening. Jeroen doesn't appear to be into it at all.

Is Jeroen (nesdev bbs) the same as "Jero32" on nesdev irc?

I asked b/c Jero32 is on IRC often, so we could go ask him there...

As far as I can tell, one and the same.