Saturday, October 13, 2007

Inefficiency of big teams

Tyler Bello sent the following comment on size of games directly to me:

"I noticed your blog post's about team size/development time and wanted to add something. The business world ticks very slowly, this applies to big game studios. If you compare indy projects to big projects of the same caliber (regardless of sales/studio etc.) you will notice that the indy titles develop 10x faster than the ones from the big studios and with much smaller teams (or, at the same pace, with much smaller teams).

The first example that comes to mind is Project Offset. In only 1.5 years a team of 3 (1 programmer 2 artists) created an incredible engine. I would attribute this to the efficiency of small teams, the bigger the motor the less efficient it is with gas usage.

You can view the video under the downloads tab.

Another very small team company , whose games happen to sell wildly, is Introversion (Uplink 1 programmer, Darwinia/Defcon 2 programmers). They started with Uplink, which recieved mild sales but much praise. They then moved on to Darwinia which was a smash hit and now Defcon which went straight to Steam and is also very popular.

Minuscule teams can still make games that are just as good/beautiful/whatever as the big dudes, it's just not as common.

I just don't think that teams should be as big as they are and games don't have to take as long as they do to make. The process is drawn out by bureaucracy."

I've seen a few magazine ads for Darwinia, but of about 40 students I asked, only one had played Darwinia, and that only a demo (verdict: not so good). Evidently it isn't quite in the same category as Oblivion, Halo 3, Rainbow 6, and other very well-known games.

When a big publisher plans to publish a major game, they intend to spend a large sum on marketing. Introversion, as a smaller publisher selling games in less-than-top outlets, may not need to spend that kind of money.

Whenever large sums are at risk, companies will want to manage that risk, and monitor it. Managing risk is very important: if your project depends on very few people, then the risk that one or more of them will quit, or become incapacitated, or simply not be up to the task, is very significant. There's every incentive to spread the risk amongst more people, hence a larger group. There's also the notion that more people will finish faster, though this is sometimes regarded as a fallacy where programming is involved, generating a classic book (which I have not read, I must say) called The Mythical Man-Month.
Further, there will be more people whose job is to monitor and coordinate, because there are more programmers.

I have a saying: "The level of chaos increases with the square of the number of people involved. The level of chaos increases with the CUBE of the number of people IN CHARGE." And more chaos means less efficiency.

I'm also guessing that Introversion self-financed their games, that is, made the prototype and then found a publisher. If so, then Introversion assumed much of the risk, and could choose to risk depending on one or two programmers.

We need to remember also that in the "beautiful" games, there are many more artists than programmers involved.

1 comment:

Ian Schreiber said...

Darwinia isn't very well-known (although fans of the wonderful PS2 title Okami will recognize the "paint" mechanic -- and you can note that Darwinia did it first).

My students have enjoyed the other two Introversion titles (Uplink and Defcon) much more; aesthetically they are much better at bringing to mind the late-80's hacker movies like War Games, and this aesthetic is much easier to connect with than the abstract landscapes of Darwinia.

It should also be noted that in all three games but especially Uplink, "art" is fairly nonexistent. This is, IMO, part of the brilliance of the game design: they apparently made a game on purpose that only required one or two programmers and very few art assets. And this serves as a great exemplar for what we're trying to tell students: design to your team! If you've got yourself and two friends, don't make World of Warcraft! Make something you can actually finish! (And if you have a notable lack of programming or art or audio talent... design a game that turns that lack into a strength, as with Uplink's minimalist UI.)

My understanding is that yes, Introversion is self-funded; they described themselves as "the last of the bedroom coders".

And yes, Mythical Man-Month is an awesome book and everyone even remotely involved in making any type of software should consider it required reading.

"Always do right--this will gratify some and astonish the rest."Mark Twain
"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away." Antoine de Saint-Exup'ery

"Not everything that can be counted counts, and not everything that counts can be counted." Albert Einstein

"Make everything as simple as possible, but not simpler." Albert Einstein

"The worst form of inequality is to try to make unequal things equal." -- Aristotle