Rebuilding a Game

Hey guys, just wanted to know what your opinion was on rebuilding a game. eg. Is it worth it to start all over (Programming-wise) on a game you've already worked on for a few months? Or do you feel it's better to go back and manually edit your programming? Have you ever done so before?

I'm leaning towards rebuilding a game I was working on recently, since I've now learned all kinds of new things over the months, and I get the feeling that going back and manually changing the event sheets would actually take up more time.

If the current events meet all of your requirements and works, do not rewrite them. If you have been working on the game for months there are probably dozens of difficult to detect bugs that have been fixed and you'll have to find all of those bugs again. Nicely arranged events are good and something to strive for, but not at the expense of putting finishing your game back by months. The best game is the game that gets finished.

If your events do not meet your requirements and especially if they don't work a rewrite may be in order.

I don't think it's a good idea generally to completely start over, but refactoring is an important part of software development. I think you need to look at it and try to estimate the amount of hassle/work it would be or save refactoring it or not. For example, with my game, I started it before learning what families were. Refactoring the enemies code to use them saved me a ton of time and frustration. In retrospect, I should have taken the time to refactor everything that needed them.

From what I've read on the internet along time, rebuilding from scratch is not something exceptionnal or rare.

A lot of developers tend to say by the end of their game that they rewrote it at least twice along the development process (from the prototype to the finish product), and it proves especialy true in my experience when, as you've mentionned, you learn new tricks and "good ways" along time.

You find yourself thinking about the game and the code of the game in the new light of the new knowledge/experience you have.

Arima's focus on refactoring is a way to cope with already a lot of written code when going back from scratch is hard on the moral/motivation level and that some of this code works as an engine or is good/stable/strong enough to be reused.

On the other hand, in C2, I've noticed that actualy the less code you have often means the better it is coded.

It might be the same in every code languages, but I only experienced it in C2.

makegames.tumblr.com/post/1136623767/finishing-a-game

See point 11. If you're really sure that fixing the problems in the code you have now will take longer, then feel free to start over, though.

CyberDagger "11. STOP MAKING EXCUSES FOR STARTING OVER" that cycle is exactly what I was about to say! Great article. I work as a software developer and at the end of every project I think "this is a mess!", but that's just because I'm a better programmer by the end. If I decided to rewrite every time I got better I'd never ship or at least never ship on time.

Some more food for thought:

Things You Should Never Do, Part I

Seth Godin: Quieting the Lizard Brain

How I Saved the Gaming Industry Overnight By Being Awesome

* This page froze my browser for 20 seconds while opening, but it does work.

It's probably a good idea to keep your events organized so that you can re-use them or revisit them.

That lizard brain speech was really well-done, though I think my reasons for leaning towards a rebuild are different. To be more specific, this game was a class project that I had help with from a friend. Now that classes are over, I'd like to change some concepts of the game and have all my own original work in it, in addition to the better programming I'd like to implement.

But it seems the majority feel like a complete rebuild is a bad idea. I suppose I could just edit event sheets. Don't want my lizard brain thinking for me. <img src="smileys/smiley36.gif" border="0" align="middle" />

On that note, here's the game if anyone feels like playing:

dl.dropbox.com/u/42313097/Boss%20Man_Final_Corrected.exe

Though this is the final version for classes, it still has its fair share of glitches and bugs, but it's perfectly playable. I figure maybe I can get some kind of critique before I start working on the new version of the game.