"The Skill of Finishing a Game"

[Jackolantern]

This is a repost, but it is such a good read that really every indie game dev should read it. And if you are one of the millions who start game projects like crazy but have not finished one, or have finished less than 10% of the ones you started, this should be considered required reading before you write a single line of code more.

It is a blog post by the creator of Spelunky, called "Finish a Game". It outlines in perfect language why devs eventually abandon so many projects just to start up another one they abandon for a new project, and on and on. It also gives tons of tips on digging-in and finishing the game. If you haven't read it in the course of the 2+ years it has been circulating the game development community, read it now!

"Finish a Game"

[hallsofvallhalla]

Nice!

[Callan S.]

I think it doesn't really cut into the commonly held idea of there is a 'right way'. And the commonly held idea you have to figure that right way first. The commmmmonly held idea that lots of people know about this right way and if you spend lots of time listening to them, you'll figure the right way. Worse, that while all these people think they know the right way, even if you ignore them, then - well, unless you're just writing for yourself, then you're writing for these other people.

I don't think it gets into how to get out of the 'right way' false economy, while still engaging an audience.

[a_bertrand]

Actually, I do have also quiet a few "dead games" of mine. Sometimes those are just rough prototypes, some times I went much further with load of work invested in it. However, I do feel the blog post resumes common errors, of over-thinking and never reaching the end. Yet, if you don't try, you will never succeed. So it's better todo 100 prototypes of possible games and maybe finish one than never start anything

[Jackolantern]

I don't think he was really talking about prototypes designed to test a game idea. Those obviously may be dead-ends that are better left behind, thus you don't go all out with the art, music, complex organization, etc. He was more aiming for the scores of game devs in every community who start a project with the full intentions of getting it to completion who, after a couple of months, convince themselves that there are problems that can't be overcome or the technology just won't work so they can go back to the design phase. Oftentimes I believe it is just a way to hit the eject button when the hard development grind starts (I know because I have done it a couple of times myself). This blog post is considered important to the wider game development community because it was one of the first well-known articles that dealt with this very-common-but-hardly-talked about phenomenon and tried to help those who keep finding themselves in that position

[a_bertrand]

Well, I made as well a couple of projects like Nebula Rider which I simply killed after maybe 1 or 2 years of development without finishing it. Why? Because the technology I choose died in the mean time (Silverlight), because players wasn't all that much keen, and somehow my idea wasn't working all that well at the end. Is it a bad thing? No, I do that for the fun. If you see that you don't reach any point, and you don't have fun, simply leave it. Yet don't give up after a couple of issues and after 2-3 month, a game may take years to reach a good point, specially hard to implement games like RPG, and any MMO game.

[Callan S.]

Makes me think of another approach - one where you have a community like here at indie resource - of starting a thread and having really basic games - not much more than hello world - perhaps one choice in it. That'd be a stage one completed game. Then, if you want, you can level up to stage 2, and write a game with two choices and link that one. Say they have to be finished within a couple of days at low level, maybe higher levels a week (have some sort of chart)

It'd be pretty vague what constitutes a choice (ie, if you can move a character on a 2D field, is that four choices or not?). But probably what would nail it is if you didn't finish it, it's probably overshot your level!

I'm kinda thinking out loud for some sort of thread I might start.

[Jackolantern]

That would be cool! One of the biggest stumbling blocks for new devs is starting too large.

[a_bertrand]

Indeed many just look too far, without having the skills even to implement a register / login page. Yet, the issue is, game development like it would have been at the first age of video game like pong or even packman are over, and we all EXPECT something more complex. Great graphics (3d), music, sound effect, MMO, etc.... all those add up. Sure that's what you EXPECT from a today game. But if you start, I doubt you have the skills in all those fields (and honestly they are all as hard as coding if you want good results),and you may not even know what is essential and where to start.

Sometimes, a really simple game, which could be all text, without much code, can be fun. An example? Look at this "game" I tried a couple of years ago:
http://temp.nowhere-else.org/webquest/

There is NO code at all! Why did I made something like that? To prove that it is actually possible to have a game without code, and what counts is an idea. I wanted also to test the idea and see how I could create it.

[Callan S.]

Interesting concept, bertrand! But your games first question is too cryptic for me!