YOU GUYS!
I've done it!!!

I listened to your collective advice and finished an albeit rough working TETRIS clone!!!!

You guys rule!

Thank you so much.

Quote:Original post by Chad Smith
You can always expand it. It's a lot easier to expand something then it is to decrease the size of a design half way through the game

While this may work for you, and may work for many [hell, lets just suppose it works for everyone but me], I've found this to be the magic bullet that can kill any project. As soon as you start expanding on a design with a code base that has already been written, you're going to be pushing on the boarders of that code base. In many instances this is fine, but as soon as you add a feature that forces you to rewrite portions of your code, you'll rapidly find yourself scrapping parts of your project. This [in my experience] leads to re-writing LOTS of your code, and after having worked on a project for months and months with absolutely nothing to show for it [nothing, because you just had to undo everything you did have], the magic is gone, and the project dies. In the same vein, as soon as you try to add a feature that you cannot add because you've already coded over where that feature *would* have gone, you become disappointed with what you're doing, and the project dies. That, and it's really tough to let one new feature in, and not let in a hundred.

It's important to make, and stick to, the plan. If you're not really happy with the plan upfront, don't even start working on it until you've hammered the plan into something you *are* happy with. Sleep on it for a few days, kick around the final version of the plan for a week. If you're still madly in love with the idea, and everything works great on paper, then start on it.

I don't mean to say you're [Chad Smith] wrong [actually, we'll just assume that your method has worked and will always work for you and many others], but in my experience, nothing kills a project faster than establishing a precedent that the plan of attack is at all flexible. Once you start coding, you should be able to see the entire map out in front of you of what is going to go where, and have confidence that a month from now, that map will be exactly the same. If you start having to redo things you *thought* you had finished weeks ago, you'll loose interest fast.

*edit* oh, and Warlax, congrads.

[Edited by - Drigovas on September 17, 2007 1:34:46 AM]

I got this issue... and it's because my ideas are just way too big for my own good. I've been choking away for three years trying to realize something... it goes back farther then that spanning multiple languages and decades... delayed due to work, school and family all alike. Games is a hobby of mine.

I'm almost done... in programmereeze I'd say 90%... but I suspect it's more like 50%. Just when I think that I'm near completion of something, I find that I need to tweek something about it... which causes a cascade of many lines of fresh new code. It took a long long time, but I've got code that it should be easy to add to... that was the biggest nugget.

I suspect that my code is mostly complete... but the wall I've got is content. So much to draw... so little time. Which is okay for me... because I do have something put together, it's just very skeletal for now. I need to toss on some layers of skins or something to that effect.

My only solution/remedy/patch to my issue is joining up to a programmer site, and reading about all of the other folks with a similar issue to complete... that way I don't feel so alone in my undertakings.


Don't look at me... I'm really just here to vent :/

Quote:Original post by Warlax
YOU GUYS!
I've done it!!!

I listened to your collective advice and finished an albeit rough working TETRIS clone!!!!

You guys rule!

Thank you so much.

Damn, that was fast. Congrats! A tetris clone was a good place to start. It's simple, but not completely trivial, so it covers a lot of the components of a basic game framework. Good luck on your future projects!

Break it down into managable parts.

I originally started my space shooter 7-10 years ago, but I wanted to do it all and it became too great a project and I got frustrated.

I'm starting again, new ideas, new plans, broken down into managable phases:

Phase 0: make a rigid body physical data ganerator.
Phase 1: develop free 1v1, multiplayer co-op version in deep space.
Phase 2: Develop SP campaign for retail, w/ phase 1 included.
Phase 3: Develop MMO version. Should accept P1 and P2 models of game.
etc, etc, etc.

If necessary, break the phases down into even smaller parts;

P1.a: develop lobby program
P1.a.1: develop UDP/TCP pkt communication subsystem and protocol.
P1.b: develop engineer's panel and program.
P1.c: develop science officers panel and program.
etc, etc, etc.