Building games together?

Posted by (twitter: @_Tangrin_)
January 13th, 2012 9:37 am

So I have been ranting about this all day now on IRC. Building games together! This LD almost 900 games were submitted. If you think about it, the time and effort invested is mind-boggling. Roughly estimated when each participant invested maybe 30 hours in the weekend, that is about 30.000 hours of time invested in game creation in a weekend. But when the weekend ends many people walk away and don’t really look back until the next LD. I think this is a bit of a shame, because LD is the perfect opportunity to meet like-minded gamedevelopers.

So I am trying to figure out if it is possible for more people to collaborate on game creation, to meet more game developers and channel all this energy to create even more complete games. Because creating games is usually a team effort, and the more game developers you know, the higher the chance that individuals meet up that have fun together building games, or make the next big indie hit. I have even written a little word document that introduces the concept: Ludum Dare: Co-op:

Ludum Dare: Co-op

It is a rough concept at this point. The interesting thing is that many people on IRC like the idea very much, but are not sure on the execution. So discuss away! What would be the best way for you to collaborate with other developers, and together build something bigger than what we usually see in a LD?



13 Responses to “Building games together?”

  1. Jorjon says:

    Interesting idea, I like it! Although it needs more work.

    For example, you can’t expect to form a team with 1 programmer, 1 designer, 1 level designer, 1 sound designer. There are not enough sound designers / level designers to fill all the team, so you will have lots of incomplete teams. I would rather stick with a programmer and a designer, then the team will organize the roles of each one.

    Other thing is that this can’t be done in a weekend. When you have a team (that is not with you, but in another part of the world) problems arises. Communication, different ideas, syncing up files, etc. There will be much wasted time debating about a particular thing than actually making the game.
    I would make this like an October Challenge, that is a full month. That will be enough to people to know each other, know what strong points every one have, and know what time is good to talk via skype or whatever.

    So to summarize, I would make this a 2-people team, and over the course of a month. Otherwise, very nice idea, I’m really looking forward to it.

  2. kutuup1989 says:

    I would totally go for this with a little refinement! Just imagine, a team of 900 working on one 48 hour masterpeice! There would need to be some serious organization going on though to avoid people working on the same things and cohesion would be a a nightmare! Maybe it would be better to add a new contest each LD where there is no cap on team size? Maybe allow an extra day too, so 4 days total?

  3. Shigor says:

    Common misconception, often used by managers in software companies, is that 1000 developers will do in one hour what one dev needs 1000 hours to do. That’s as far from the reality as possible. One software design adage says that development speed increases linearly with quadratic increase in number of developers. (in my experience it’s slowing down after a while).

    While yeah, this is good idea, I doubt it would be much practical in LD 48 format (most of my objections were already voiced). Yeah, every small cell doing just one small part of the game makes it more practical, but still…

    You would have to prepare a LOT (the basic engine, basic shared art etc) before you could start grabbing people for levels. And I’m not sure if the end result wouldn’t look as a random mix of random games.

    • sfernald says:

      Yeah, honestly i think the game created this way would be an abomination and I doubt anything playable at all would be realized out of the effort. I would be massively intrigued with that abomination though. Maybe even more in the processes as documentated by chat logs and documents generated.

      I have done professional game development (for PlayStation, etc) on fairly small teams (12-20) and so much time is spent just trying to coordinate efforts and keep people working in the same direction. Also, there is so much dirty crappy work involved. That is why developers enjoy making the games by themselves. It gets them away from the mind-numbing bureaucracy.

      Having said all that, it would be fun to try just to see what if anything comes of it.

  4. Erifdex says:

    Now that, that would be incredible. Someone PLEASE set this up!

  5. Shigor says:

    Right, you have bottlenecks, you have one processing unit waiting for other to to give it access to resources… and these are processing units which are always (well… almost :) logical, reasonable and do what you told them to do…
    But with humans? 😀 Especially with the sort who likes to code / make games? Cooperation of few people can be quite complicated affair… dozen isn’t managable without someone managing it fulltime (and while good managers are worth their weight in carbon nanotubes, bad managers are worse then ton of lead… and good managers in game dev are like unicorns… they probably don’t exist at all :D)

  6. Vic says:

    Well, I think all concerns raised are valid.

    Let me give my thought on it, and why I think it can be managed.

    First, Management structures:
    I am very much against hierarchical structures. Sure, they look nice on paper, but it takes too much time to get decisions up and down the chain. More importantly, the message changes as it goes up and down the chain, there is usually not enough information available to make good decisions. So it needs to be very flat to work.

    I suggest having one facilitator on each team, the facilitators can communicate but there are no levels above that. A facilitator is responsible for one or more teams.

    Now, second point:

    Common misconception, often used by managers in software companies, is that 1000 developers will do in one hour what one dev needs 1000 hours to do. That’s as far from the reality as possible.

    This is true up to some point. Complexity only increases with dependency. I think the best way to tackle this is build a game that has areas that are as isolated as possible. Each team gets instructions to get up and running, and then they are free to do within their isolated area as long as they follow the basic instructions they get at the start of the challenge.

    You would have to prepare a LOT (the basic engine, basic shared art etc) before you could start grabbing people for levels. And I’m not sure if the end result wouldn’t look as a random mix of random games.

    Yes, I agree. There needs to be preparation. The current idea is to build a basic framework for a metroidvania game. Each team builds an area of the game and the link ups of each area are predetermined. There would be predefined tile sets for developers to use, but people can expand on the basic set. Maybe there can be a styleguide. I think the randomness will need to be a strength of the game, instead of a weakness. It will lead to a very unique and diverse experience.


  7. triplefox says:

    I think there are definitely two kinds of events here. One is “make a game in a tiny team instead of solo” – which is not too different from what is developing with main LD’s Jam. The other is “architect tools to crowdsource a game” – which doesn’t rely on having a event, necessarily, but it could be a good “kickoff” to someone’s project.

  8. ImaginAudio says:

    I’m a composer and sound designer, and I’m up for helping anyone with their project!

  9. 7heSama says:

    The plan as you all are discussing it: Hundreds of people divide into dozens of teams coordinated from the top down to make a linear game, with each team being responsible for a level (or development area, i.e. art, code, etc) that we all stick together. Or Jorjon’s idea, which is basically a co-op October Challenge
    (basically stealing my idea for next year’s october challenge! That developers who have already sold games/ads in games try something new, be it a new platform, language, or working with a partner! Cuuuuurse you jorjon!).

    I think a better take on this would be as follows;

    A 2-week (more? less?) game jam. Dozens of independent teams (2-5 people) each create a game featuring the same protagonist (voted on before hand) and somewhat similar settings (also voted on before hand). Then they are all bundled together. All games must be written in flash (probably. something else could work).

    Essentially, what I want to do is take the 1000-dev game idea, but:
    – reduce the reliance on managers
    – increase the coherency of the game
    – increasing freedom of individual teams.

    The ramp-up features several different categories (and probably rounds, though voting burn out could be a problem) to vote on.
    At least for the first game, it would be 2d with pixel art style assets only (this could be voted on in the future, maybe we could get our hands on a voxel framework?!).

    – Game style: top-down, horizontal, etc (not sure what else to put).
    – Common element: user-submitted ideas (img or text, with size/character limits). Ideas must be what the element is, not how it’s used; so an 8-legged armored robot could be an enemy, a terrain object, the battleground itself, collectible items, etc. But each game would have to feature the winning common element somehow.
    – Themes: These are simply to jumpstart ideas and simplify team building. After a few rounds of elimination, people can plant their stake in the remaining ~20 themes or so, and then are split into teams of 2-5 and get cracking. (Should team creation after individuals choose themes be automatic or handled by individual theme people?)

    SO to review; take the small teams from the original idea, extend the time available, open up people’s options in terms of design (not just sidescroller/castlevania clone/etc) but make them use certain content to maintain coherency. Then stick them all together in one package (probably should have short blurbs on each game written so people aren’t blindly choosing which game out of dozens to play).

    • sfernald says:

      This idea seems a lot more realistic to me.

      I mean, it could be pretty simple really and the whole thing would sort of work as one project.

      For example, everyone uses flash.

      The fixed game screen size is 800×600.

      Everyone uses same default keys to control game.

      Everyone has to use same character in the game (all teams provided with character graphics and animations).

      Same theme across all games.

      Something like this might really work because it would remove all management requirements. Would probably be fun to play too. Like reading a book of short stories for various authors all set in the same world. Some games would be zany, some deadly serious, some beautiful and artistic. It might just come together into something wonderful.

  10. I would definitely like to do a Jam, but the biggest problems seem to be 1.) a centralized place to find people and organize, and 2.) timezones. I talked to a couple of artists, but they were all 5 or more timezones ahead/behind me which makes things really difficult.

Leave a Reply

You must be logged in to post a comment.

[cache: storing page]