7 replies to this topic

[nopcoder]

Hi,

I’m currently working on a game of my (chocking isn't it). Anyhow it’s a rather big project but it’s my first real attempt (all previous “unreal” attempts are collecting dust on the shelf). So your wondering where I’m going with this? Patience.

So I’m working on this game and its killing me, ok seriously thought: I get stressed and anxiety when I think about the game. The lack of experience finishing a big game is a concern but to solve that I need to finish my game? (irony).

What are your experiences? Have you seen the development process as a chore, and how did you turn it around (assuming its possible)?

But mainly: How do you work on your project (how do you keep it fun/interesting)?

[dave_]

There are plenty of books on the subject of software development. You don't have to get a massive tome, I've got extreme programming pocket guide which is a good starting point.



Test early test often. Twofold, try to do automated testing and from the game-play point of view you want to have a prototype as soon as possible.
Test driven development can help keep you focused on design and encourages modular design.

Most importantly: remember you're making a game and not a tech demo!
You've really got to keep focused on your game design.
Avoid feature-creep if you don't need it don't code it!

[Jare]

Iterate on the game's features. Start with the basics, get them right before moving to details and content. Make sure that you regularly stop to polish what you already have, so every month or two you have a sort of playable version that is good by itself.

Wrapping up a game is an exciting process, but rarely "fun".

[sputnikjon]

And IMO very important for game development: From time to time, get some voluntary testers to play your game and watch them playing (and maybe critisizing) it, although not finished yet. That's much fun, motivates me personally and helps to improve the final product.

[dave_]

sputnik_jon said:

And IMO very important for game development: From time to time, get some voluntary testers to play your game and watch them playing (and maybe critisizing) it, although not finished yet. That's much fun, motivates me personally and helps to improve the final product.

Thats very true, but good testers can be hard to find.
The people you need, need to be able to put up with the bugs and play it lots, they are part of the process.

[sirfalas]

What I would also suggest is to get into a habit of using code-versioning systems.

I started using SVN (with the help of TortoiseSVN. google it) to manage all my code with a previous project. I am sure I did not use it in the most effective way possible but I sure did find it very useful.

It saved my ass more than a few times when I did some crazy code modifications which basically resulted in the death of my project. All I had to do was hit revert and I reverted back to my old code where everything was working fine.

Before I used SVN, I would back up all my code basically by zipping them up and then renaming the file into "Project-Name_V0.1.zip". You can see how tedious that would have been.

Another thing I would also recommend is to start small. Split your tasks into sub-tasks and milestones. I am in the middle of writing my own engine right now and since I made the promise of not blindly copying code from a book or tutorial somewhere, my sub-tasks may seem small. It was something like

- Creating main engine file
- Complete Adapter/Device enumeration class
- Complete the required windows code (such as creating windows)
- Milestone 1: Application actually runs fine, checks all the possible display modes, allows you to switch between windowed/fullscreen mode.

I took a while to reach milestone 1. Most of my time was spent in the adapter/device enumeration class and besides the code was 'boring'. I am sure as hell not re-writing them for any of my next project.

Anyway, you get the point. One thing I realised from my failed attempts on a project is this. When I became overly enthusiastic with a certain idea and if I jumped straight into development, I would try to create it but because I did not put in attempt to get the basic foundation done, I would get discouraged and then never get around to completing it and it would be abandoned quickly.

So I tell you what I told myself, start small, split major tasks into sub-tasks and have patience. You will never be able to create the next game engine everyone is going to talk about in just one week.

[dave_]

sirfalas said:

I started using SVN (with the help of TortoiseSVN. google it) to manage all my code with a previous project. I am sure I did not use it in the most effective way possible but I sure did find it very useful.

Oh and if you need Subversion compatible hosting try http://code.google.com/hosting/

[sirfalas]

Oh. That is interesting. I did not know about the google projects hosting. Previously what I did was use TortoiseSVN to just create a repository locally on my harddrive.

But currently I have my own site (www.sirfalas.com [There are no ads so I am not making any money so I hope I am not breaking any anti-spamming rules, nor any proper content but I intend to change that very soon]) which I am hosting with dreamhost (www.dreamhost.com) and they provide SVN access too so I am using that instead.

But google code seems promising nonetheless.