Community Project Mark 562

Business and Project Management Discussions Community & Volunteer Projects
#101

When I was at the last GDC, there were some good points in Satoru Iwata’s keynote about how they develop game ideas at Nintendo:

  1. First Miyamoto gets an idea for a game by observing things that people find fun, and analyzes them to find out what is fun about them. Quite many of his ideas come from his hobbies (gardening produced Pikmin, getting a pet produced Nintendogs and even Wii Fit came from his hobbies), so Iwata said that he has made Miyamoto sign an NDA for not talking about his hobbies outside work ;D.

  2. Then when there is an idea, a small team of maybe just one developer will work for months or years on building a prototype game which concentrates on what makes the game fun. At the same time, multiple prototypes are in development on their own pace (Iwata will never ask how a prototype is progressing, for fear of giving pressure to finish it more quickly). Only after the prototype is fun, the production of a publishable game begins together with a big development team. Then they know that the core concept has been proved to be fun, so they always have something to which to return if the actual game is not anymore fun.

  3. During development, hallway usability testing is used much. Miyamoto kidnaps a random Nintendo employee (non-developer) and forces him to play one of the games, while Miyamoto silently sits by and watches what the player does - what is fun and what is frustrating.

I think that by following these ideas, especially the point about first creating a fun prototype, would help a community project like this to succeed.

#102

Yeah, those are great pointers, trouble is that’s not very easy at all to do unless you’re in a big company that can absorb the cost of messing around with something until it’s fun, with no deadline and no constraints. Similarly for a community project you need people willing to devote an unspecified amount of time to get to this point of being fun.

That can be difficult!

#103

The way I see it, if you wanted to do the same thing as a hobby/community project then you’d have to do that whole process on your own (or possibly with one other person). Only when you get to the last point (“Mass production stage”) have you got the concept and the fun nailed can you ramp up to being a full community project with loads of people.

The 4k competition actually serves as a good game prototyping method, since there’s much less focus on graphics and polish and the winners are usually the ones with great gameplay. Similarly, Rescue Squad 2 was originally a one-week competition game that was unexpectedly fun so I’ve been turning it into a full game (worryingly it’s taken an extra two years and it’s still not finished…)

#104

We could do so, that everybody will work on their own little game prototype at their own pace, and then if one of them produces something fun and the community agrees that it is fun, then a community project will be started.

#105

Or pick an existing 4k game as the prototype? (with the author’s permission, of course)

#106

That’s a good idea. I might have fun messing around with some sort of sandbox game to make something fun as a prototype, then everyone else could add a lot of content to it.

#107

Funny that my initial suggestion of working towards a techdemo, as the only option to get something started, is now where we’re going :slight_smile:

#108

Yeah yeah.
:stuck_out_tongue:

#109

Couldn’t help myself :persecutioncomplex:

Anyway, when I finally have some free time (maybe this year…?) I’ll try to contribute something here and there.

#110

The prototype phase is definitively something we need to do before we start the mass production, but there is still a problem.

How do you make the transition between the prototype and the mass production phase?

#111

You let others play your prototype, and if they find it fun, they may be willing to join the project. The code from the prototype is not necessarily reused, probably there will be a clean start (unless it’s more like a maintainable tracer bullet than a throwaway prototype). The project will have the prototype’s core concept as their focus in gameplay.

#112

Ok but here is a situation.

We are a couple of months later. Demonpants has worked on his sandbox prototype and now it’s very cool. He show it to us and everyone find it fun and decided to make a bigger game based on it.

What now? The bigger game won’t just build itself and it’s not just about having better graphics.

#113

Create a SVN rep.

Projects in their early stages are always a big mess, that happens to not crash on the dev’s machine.

Things will work out, even if it requires to rewrite the whole base.

#114

I always like to have at least a minimal design phase… Don’t think it could work if you don’t have a plan of what to do.

The problem with a game is that you can’t plan everything at the beginning because the goal is to make something fun and what is fun is very subjective. That’s why making a prototype first is so important, because it’s the only way to find out what is fun.

But after that, you need to have a plan. You can’t simply gather everyone around the prototype and tell them ‘‘GO, do some code’’. The only way I can see something like that working is if 1 or 2 people do a big chunk of code alone, then other people willing to read through that big chunk of code might understand the direction that the project is trying to take from that code. (That code would probably be different code from the prototype). So it’s like someone working alone again.

#115

After that it’s like managing any other software project where the team is distributed, which makes communication much harder than if they would meet every day. Maybe http://producingoss.com/ or some other place has ideas on how to make that happen.

#116

Thx for the link. I begin to read it and it seems full of good tip :slight_smile:

#117

anything new related to this?

#118

Wow didn’t even know this existed… :o
Kev posted request, people threw in a 4 page worth of suggestions, and then it just stopped as if it never existed, until this day.

As a suggestion to other readers, if you want to make an impressive 3D game, you have jMonkeyEngine 3: HDR, Real-time shadows, advanced bump mapping, per-pixel lighting and more. I want to know a C++ programmer who wouldn’t be impressed when Java does that.

EDIT: Lol nevermind, I just noticed the other threads in this forum XD