Skip to main content
GameDev.net gamedev.net
🔒 Locked

The Industry Method

Started by Pirate_Lord Jan 31, 2008 at 8:24 PM 204 replies 37k views
Original Post
Pirate_Lord
Pirate_Lord
I have been asked to move my thread to this forum. I'd like to start a discussion about game designers, design documents, what game design is, and the methods used by the computer game industry to design games. Accompanying this thread is my web site, with an example of a game design document based on the game design konwledge of a bygone era. It is very different than what the computer game industry calls a game design document. The link to the web site is in my signature at the bottom of this post. I have other articles to post later in the thread, after this first one has been thouroughly hashed over, and it is my intention to keep this thread going for a very long time. So, please, join in the discussion and let's see what we all learn. I'm sure that I'll learn many, many things... and I hope to teach a few things of my own along the way. -- Pirate_Lord The Industry Method There are many different ways to design a game. The computer game industry settled into their way of designing games many years ago. The method they settled on was known to real game and simulation designers long before the computer game industry ever even existed. The method they settled on is known as “design by committee”. The ironic (or is it pathetic…) thing about this is that the phrase “design by committee” held a very specific meaning among professional game and simulation designers long before the computer game industry even existed. The phrase “design by committee” famously translates as “the worst possible way to make a game”. This is the method that the computer game industry has chosen to institutionalize. And they wonder why game design seems so “risky” for them? “Design by committee” essentially means that you don’t have a game designer. You might have someone with that title, but they are not acting as a game designer. They are more of an assistant designer, or level designer. The game itself is essentially designed through a democratic process. There is a reason why Hollywood does not make movies this way, and the computer game industry really should pay attention too that. They like to compare themselves too Hollywood, well, can anyone tell me the last time that Steven Spielberg was over-ruled by his Key Grip or outvoted by his cameramen? Of course not, Hollywood isn’t that incompetent. They know better than to attempt to direct a movie by committee. This is a lesson that the computer game industry truly needs to learn. When it comes to certain creative endeavors such as writing books, making movies, or designing games a single vision will utterly destroy “groupthink” every single time. Hollywood clearly knows this. Sid Meier is, of course, a perfect example of this in the computer game industry. The fact that his games were not “designed by committee” is actually the primary thing that sets them apart from most other games that the industry has produced. It is the source of that “thing you just can’t quite put your finger on” that make his games stand out. Sid’s games were, largely, his vision. There are a very few others like him, you know their names because they all stand out… because their games are the result, for the most part, of a single vision. That’s why their games stand out. That really is just about the only difference between them and the rest of you, their games are the result of a single vision and yours are not. But you still don’t see it. There are many ways to make a game. “Design by committee” is famously the worst way. The method that game and simulation designers had, long before the computer game industry even existed, come to a general consensus on as being the best method is known as the “benevolent dictatorship”. It was a very famous quote, and one that I know at least some of the older people in the computer game industry have heard before because I’ve heard them use it. “Game design is a benevolent dictatorship”. In most cases, if this phrase does not apply to your game designer when it comes to the actual design of the game (not to be confused with the production of the game) then you are using a substandard method. This method allows an experienced professional to create a game with a single vision. It allows that professional to do new and innovative things that would otherwise have been shot down in the “design by committee” process. It allows a true artist to do their thing, without that art being muddled and watered-down by people who are working outside of their field and don’t really know what they are talking about. It allows the resulting game to be unique, and not just an imitation of existing games. They even have the proof of this in their own industry, where they idolize people like Sid Meier and Will Wright. The only real difference between Sid and Will’s games and theirs is that Sid and Will’s games were largely the result of a single vision and their games were not. But they still don’t see it. The “design by committee” method always results in an imitation of an existing game. It works just fine if you are making a clone of a previous game, but it can’t achieve anything else. “Groupthink” will always gravitate towards the familiar. This is the primary reason, by far, that the computer game industry has been stuck in a 20-year long rut of crawling forward at a snail’s pace in imitating their own past games. Imitating existing games is all that their method can achieve. The method they use is actually incapable of doing anything truly new. There are very few computer games that are the result of a single vision, and those games serve as almost the entirety of the source material that they have to work with. Hence the snail’s pace of advancement. Where the computer game industry has wound up after over 20 years of almost exclusively designing by committee is not pretty. “Game Designer” has become a bastardized term in their industry. With very few exceptions, they don’t have game designers. They have people who have been given the title of “Game Designer” but don’t actually know the first thing about the subject. These people are actually more assistant designers, or level designers, the “game designer” is essentially everyone involved with the project in any way, shape, or form. They pretty much all have an equal say. In the end this group of people, that in most cases doesn’t include a single person who actually knows anything about game and simulation design, inevitably wind up “blindly blundering forward through trial and error, praying that things work out well in the end”. Then these same people explain to real game designers how “risky” making games is. I would agree completely, actually, it is very risky to make games if you are “blindly blundering forward through trial and error, praying that things work out well in the end”. It is far less risky, of course, if you just let a professional do it. But they don’t see that. You’d think that they would have a frame of reference to “get it”, since they would surely balk at the artists voting down the programmers on how something will be coded… but apparently they just don’t get it. The ugliness continues… in fact it gets handed to the gamers. Actually finishing a game or simulation is, by far, the hardest part. Starting about 10 years or so ago the computer game industry finally found a way to address this problem. They don’t finish their games anymore. They can’t. It is, by far, the hardest part and very few developers have a professional game and simulation designer who would be capable of achieving such a thing for them. Their very method precludes them from being capable of finishing a game; by that stage of the project there are too many visions in place for them to ever resolve them all. So they gave up trying about a decade ago, and they let the gamers finish designing their games now. They get it as far along as their amateur abilities are capable of and then they release the game. They plan, right from the beginning, on a series of patches that will complete the design of the game. They need the gamers to tell them how to finish designing the game, because they don’t have anyone who knows how too and the method they use makes it practically impossible for them to do it by themselves. This only further exacerbates their “design by committee” problem because it greatly expands the committee… to the entire audience. There is a difference between incorporating player suggestions into a published game and having the gamers finish designing the game for you. Computer game developers have the gamers finish designing their games for them because they are incapable of completing games on their own due to their method, and the fact that they don’t have a real game designer who is capable of doing the job. So here we are, after 20 years of “design by committee”. The people who so foolishly have devoted their lives to the study of game and simulation design are working in offices, driving cabs, and delivering pizzas. Meanwhile, groups of programmers and artists “blindly blunder forward through trial and error, praying that things work out well in the end” telling everyone else how “risky” making games is. Then when they get the game as far as they can, they release it to the gamers so that the gamers can finish designing it for them. All this to wind up, in almost every case, with games that would be considered “B-Games” in comparison too the ones that those office workers, cab drivers, and pizza delivery drivers who should be leading such efforts would be making if only the computer game industry were not so totally and completely clueless when it comes to game and simulation design. -- Marc Michalik (A.K.A. Pirate_Lord) Lost Art Studios – www.piratedawn.com
Hodgman
Hodgman
Ok, what evidence do you have to suggest that all (or most) games are "designed by committee"? Or that everyone has equal input into the design?


Where I work, a designer comes either comes up with their own concepts, or they take a theme that an artist has presented and create a concept around that theme. In both cases the designer specs out the entire game by themselves, and then sees it all the way though production to make sure it adheres to their original plans.

The designers (and separately, the artists) periodically meet for brainstorming sessions, where they don't all create concepts together, but use the combined "group-think" to get past designer's block, or to get feed-back on their ideas.

The only invasion of vision that they suffer is the approval of the "game design executives", who act as quality-control and have the power to reject a spec.


Also, you say the title "game designer" has been bastardized - can you please offer a definition of what a "real game designer" is.
Here, the requirements for designer positions is very strong math skills, focusing on statistics / economics / game-theory, and basic-programming (for testing your own simulations) and psychology are pluses.
Trapper Zoid
Trapper Zoid
Quote:
Original post by Hodgman
Ok, what evidence do you have to suggest that all (or most) games are "designed by committee"? Or that everyone has equal input into the design?

I'd like to know this too, as your entire argument hinges on this assertion. From my experience, this is stretching the truth somewhat.

From my brief time in the computer game industry, developers would input ideas at meetings and make suggestions to the designer based on their areas of expertise, but the designer would have the final word on what goes in the game.
Trapper Zoid
Trapper Zoid
Heh, although the first thought I had was "You're using Papyrus as your window title font?"

To be fair to Pirate_Lord's argument, I do see some of the "too many cooks" problem in some games with large development teams. But I don't think this is because everyone has design input; rather it's result of lack of vision by the team as a whole (or possibly dictated from higher above; there's not much a developer can do against a publisher's wishes unless they've got a lot of industry clout). A good game design has a strong vision behind it, and for a large team you need a very charismatic designer sell that vision to all the team members. In this case I'd be surprised if every team member didn't also suggest design ideas, it's just they're all aiming towards the same goal.

I strongly doubt a dictatorial designer would achieve the same ends. Computer game development teams are based on highly creative people, and if you trample over that and treat them like cogs in a machine then they'll either revolt or oblige your expectations with soulless work.
boolean
boolean
Know what game was the result of "group think" with a bunch of different developers all pitching ideas to make one game? Chrono Trigger.

Know what game was the result of one person controlling every facet of game design to achieve his goal? Daikatana.

Know what movie had a ton of people re-writing the script, changing characters, actors asking not to be killed off, and changing just about everything the director wanted? Star Wars IV.

Know what movie was the result of one person controlling every facet of movie production because he wanted his vision and his vision only? Star Wars I.

Your thread makes no Goddam sense.
superpig
superpig
All the games I've worked on have had dedicated designers who control the design of the game, one of whom has always had total executive control over what does and does not make the cut (the lead designer). We were all able to pitch ideas to the design team, but there was never any guarantee that any of those ideas would be accepted if the design team didn't think they'd fit with the game (or if the producer didn't think it was achievable within the schedule and budget).

So, simply put, your facts are wrong. I'm not even going to bother reading the rest of your post after such a complete disconnect from reality.
Richard "Superpig" Fine - saving pigs from untimely fates - Microsoft DirectX MVP 2006/2007/2008/2009
"Shaders are not meant to do everything. Of course you can try to use it for everything, but it's like playing football using cabbage." - MickeyMouse
mittens
mittens
One of the major problems with your theory about the power of the unified design mindset (of a single individual) is that you're basing it off of false ideas or misconceptions of other industries. You mention Steven Spielberg as a sort of movie set dictator -- which is fundamentally different from an auteur. An auteur is a director (or producer) whose films wholly reflect their ideas/vision. This does not, however, mean that the director is the only person with "say" on the set of the movie. Steven Spielberg, for example, does not often handle the writing and cinematography of his movies; he typically sticks to his role as director and producer. This is even ignoring the fact that the movie industry has been around far, far longer than the games industry -- it has had time to mature, evolve in its creative method, and so on.

So, with that point rendered null, on to the next one: design by committee is never, inherently, a bad thing. Design by committee where everyone has a completely equal say and, as a result, no one has the power to say "Enough, this is how it should be," however, generally is. There needs to be one individual with the lead creative vision (unless it's a team of two-to-four people) who can override an excessively democratic but that in itself does not make the committee-based design bad by any means.

I wonder if it's coincidental that the one game you're attributed to have had some part of was made around the same time that you said "games are never finished." It was also a game which was highly touted and recognized for it graphics and not its gameplay. Surely this isn't the paragon of design innovation and ingenuity. Of course, maybe Pirate Dawn is your attempt as redemption; it surely must be an original and unique title. Oh, hrm. It appears to be an MMO. Isn't that the kind of genre that all of the people in the Help Wanted forum are trying to make games for? Isn't that the same genre which hasn't seen true innovation since the days when they were still MUDs?

In short, I blame you for everything since JFK's assassination.
Promit
Promit
Also, you are really boring. It's no wonder no one listens to you. If you were any more sleep inducing, they'd have to regulate your speaking in order to keep society functioning.
SlimDX | Ventspace Blog | Twitter | Diverse teams make better games. I am currently hiring capable C++ engine developers in Baltimore, MD.
zer0wolf
zer0wolf
A lot of the best games today are "designed by committee", though not quite in the sense that you're talking about. As others have pointed out, there is generally a lead designer on a project who gets the executive say. Any smart designer isn't going to just hole himself off though. Others in your team are always going to be able to see things that you didn't see, and getting that input is invaluable for making a better game.

In short, I don't think your treatise on the industry makes any sense whatsoever.
slayemin
slayemin
Quote:
Original post by Promit
Also, you are really boring. It's no wonder no one listens to you. If you were any more sleep inducing, they'd have to regulate your speaking in order to keep society functioning.


I <3 u Promit!!
Pirate_Lord
Pirate_Lord
I am really not saying that a game designer should be a "dictator". On that specific point, another way of saying what I am trying to say might be this. That everyone needs to actually understand the game before beginning to decide how it needs to be changed. You can't have that understanding based on "20-40 pages of vague notes". Actually, I have another article that addresses exactly this issue so I'll post that one after this post as a continuation of a response to this point. This issue is all about what I call "The Glorious Vision".

The disconnection between what I mean, and the interpretation in most of the responses, I think comes from my having a vastly different defintion of "design document" than the computer game industry does. Someone said that you do have game designers that work out the game ahead of time in detail. In reality, I'm sure that is true. One problem with "game design" is that almost everyone is self-taught, so almost everyone has their own method. I'm not saying mine is the only correct one, only that it is a very effective method... and beyond that an exceptionally "safe" one.

I've been lectured for 20 years about how "risky" this all is. Well, if you truely take in the Pirate Dawn design document, which serves as an example of what I am talking about, that should help make my point here. As will an understanding of what I call "The Glorious Vision".
Pirate_Lord
Pirate_Lord
The Glorious Vision

All games start out as a “Glorious Vision”. Game design is the art and science of making this “Glorious Vision” a reality. The final game rarely equals the “Glorious Vision” upon which it is based. Any game that comes anywhere close to matching the original “Glorious Vision” is probably a timeless classic. On the other hand, “The Glorious Vision” is very dangerous and it has been the undoing of many a game project.

Any time anyone has a game idea they either decide it is a “worthy” idea, or they discard it from their mind. When the decision to remember a game idea is made, a “Glorious Vision” begins to evolve in the person’s mind of exactly what the game will be. But games are too complex to resolve in your head. So, eventually, if you really are serious about a game idea, you have to start writing things down. It is almost a certainty that a part of “The Glorious Vision” simply cannot work as envisioned. How far do you need to go with the design document to know that the game that will actually exist, as opposed to “The Glorious Vision”, will be a good game? I would imagine the answer would be based on how much was being risked on the making of that game, wouldn’t it? If you want to be sure that “The Glorious Vision” is for real, you’ll need to design the game to near-completion on paper first. How high a value, exactly, does it take for this to be considered necessary?

There is no way of making any meaningful estimate of how many failed computer games were the end result of a “Glorious Vision” gone wrong, but I would imagine that it is a very high number. Most of those games probably would have been at least profitable had an amateur with a “Glorious Vision” not been allowed to dream their way through “designing” a game. I’d bet big money that “The Glorious Vision” has been the undoing of many, many computer game projects.

A detailed game design document guarantees that a project will not fall victim to “The Glorious Vision”. A design document that takes the game far enough, in all of its elements, to know all of those elements will actually function within the game guarantees that you are making an actual game… and not just a “Glorious Vision” that you will eventually learn cannot exist in reality. Without a detailed design document, that proves that the game can actually exist, all you have is a “Glorious Vision”. And there is another problem.

If your design document is “20-40 pages of vague notes”, then all you really have is a small part of one person’s “Glorious Vision” on paper. All this achieves is to inspire a new, slightly different, “Glorious Vision” in everyone who reads it. It doesn’t define the game, it merely describes it. The defining is left too the reader. So a “game design document” that is really nothing more than a vague overview probably does more harm than good. It establishes multiple visions of what the game will be among those who are making it.

“The Glorious Vision” is the bane of the computer game industry’s existence. It really is hard to understand why they shun the means of solving that problem.

-- Marc Michalik (A.K.A. Pirate_Lord)
Lost Art Studios – www.piratedawn.com

JBourrie
JBourrie
Quote:
I think comes from my having a vastly different defintion of "design document" than the computer game industry does.

No, I think you just have a vastly bigger ego and superiority complex than the computer game industry does.

Do you realize that both of your arguments so far are based on completely faulty assumptions?

Argument #1: Design By Committee

Very, very few games are designed by committee, and haven't been... well... ever. Most big games have a creative director (who is the vision holder and makes the final decisions), a lead designer (whose job is to manage the everyday processes to make sure they follow the vision), and a design team (whose jobs are to implement the vision. This isn't "design by committee", this is "there's too much work for one man to handle". In the end, whoever is the "vision holder" decides what's good or bad for the design.

Quote:
can anyone tell me the last time that Steven Spielberg was over-ruled by his Key Grip or outvoted by his cameramen?"

You say "over-ruled", as if they were going to vote him off the island if he didn't do what they say. If somebody has a good idea that would make the movie better, Spielberg would be an arrogant prick to not at least listen to it. His job is to make the best movie he can make, and good ideas can come from anywhere. That doesn't mean his movies have to become a mish-mash of random ideas where everybody has their say, but you're just ignorant if you think he works alone in a bubble. He takes any ideas that comes up, weeds out the bad ones, and weaves in the good ones where he sees fit. Similarly, the vision holder for a game might get a really good idea from the UI coder, and it would be a disservice to not at least consider tieing it into the design. Just because you're the vision holder doesn't mean your ideas are the only way.


Argument #2: The "Glorious Vision"

So you've described Peter Molyneaux and... hobbyist MMO teams? Most of the game industry takes too few risks when it comes to creating a grand vision, yet you're trying to claim that they're still biting off more than they can chew? Seriously, name a game made by a real commercial developer that has been canceled specifically due to this supposed "glorious vision". I'm sure you can find one... but name two?

Quote:
If your design document is “20-40 pages of vague notes”, then all you really have is a small part of one person’s “Glorious Vision” on paper.

Then it's a good thing that any decent game design is already more than 20-40 pages of vague notes.

Quote:
Well, if you truely take in the Pirate Dawn design document, which serves as an example of what I am talking about, that should help make my point here.

I don't have time to go through the whole thing, but I think you really need somebody to comment on your doc before your head swells so big it snaps your neck.

- From the table of contents it looks like this doc is pretty detailed. That's a good thing.

- Your page formatting, on the other hand, is not a good thing. Page breaks in the middle of paragraphs, a general aversion to whitespace, and a complete lack of sample images make it very difficult to read and understand your vision. Since the point of the document is for the rest of the team to know exactly what you want them to make, your document has to be clear to all types of people. Some people understand big blocks of words easily, others will get frustrated because they can't envision it. Don't be afraid to leave some whitespace at the bottom of a page in order to avoid splitting up sections... that whitespace may later be filled with expanded sections or images. If you don't have an artist, I'm sure you can find some shots from film, tv, comics, art, etc that can help sell the vision.

- Your game overview has alot of notes (written in parentheses) inserted all through the (confusing to read) paragraphs. Many of these notes would be better off mentioned in their own separate sections.

- You have merged target demographic, genre placing, and competitive product analysis into Overview of Play, further confusing the point of that section.

- Now that you are into the actual gameplay details, you've certainly put alot of thought into this. However, I'm so confused by the bland and cluttered page formatting that I can't follow it. I'm sure I'm not the only one, and remember I am the person you're writing this doc for, not you.

- "See the “Shipyard” screen template for a full layout of this screen… and then talk too me about how it works. This will save us 3 or 4 pages in this document"
Is this not written by the same person who chastised other devs for writing "vague notes".

- Ok, I'm bored now.

Congrats, this design doesn't appear to suck completely. However you've gotten so lost in your details that you've forgotten that other people have to read this monstrosity too. For example, programmer #6 needs to write an AI for character X. How does he quickly find exactly what he needs to do this job without having to decode what is and is not useful to him. The AI sections have multiple page lists that should have been formatted to fit on a single page, it's not clear where one section ends an another begins... This is a good example of the many places within the doc that you need to consider the reader when writing.

So instead of coming here to GameDev and telling us all we're bad designers (which is basically what you did with this thread), why don't you spend that time making yourself a better designer and prove that the "Pirate_Lord" design document is a shining example of what all other designers should follow. Because as-is, it (appears to be) a nicely detailed design that is presented in a way that will only confuse and/or bore the majority of your team.

[Edited by - JBourrie on February 3, 2008 1:10:32 PM]
superpig
superpig
Quote:
Original post by Pirate_Lord
Everyone needs to actually understand the game before beginning to decide how it needs to be changed. You can't have that understanding based on "20-40 pages of vague notes".
Sure you can. You can tell me that the game involves driving in a straight line for 20 minutes along an empty road and then doing battle with housecats that have rocket launchers, and I can tell you that driving in a straight line for 20 minutes along an empty road doesn't sound very fun and that housecats are too small to hold rocket launchers.

One of the skills a good designer has to have is the ability to communicate information in a concise, structured, and accessible way. 20 pages is far more than you should need to describe the core elements of your game, and it's entirely possible to talk about things that need changing with only the core elements. Sure, you can't talk about the details, but generally, the details don't matter. You sort them out later. Knowing that Ship Type A has attack statistics XYZ and Ship Type B has attack statistics PQR is not nearly as important as knowing simply that ships can attack one another. It is the broad, easily-summarize-able concepts that give the game its prominent character and focus.

Quote:
The disconnection between what I mean, and the interpretation in most of the responses, I think comes from my having a vastly different defintion of "design document" than the computer game industry does.
Maybe you should try using the established terminology instead of subverting it for your own purposes and then being surprised when we disregard that? If the established terminology doesn't cover what you've got then pick a new term.

Quote:
I'm not saying mine is the only correct one, only that it is a very effective method... and beyond that an exceptionally "safe" one.
No, it's not remotely safe. Just because you've written down the details doesn't mean the game will be fun. It's only "safe" in the sense that once the document is written it doesn't make much difference to the project if you get hit by a bus or the company fires you. If somebody makes the game exactly as to your specification, and they put it through playtesting and people don't enjoy playing it... what then?
Richard "Superpig" Fine - saving pigs from untimely fates - Microsoft DirectX MVP 2006/2007/2008/2009
"Shaders are not meant to do everything. Of course you can try to use it for everything, but it's like playing football using cabbage." - MickeyMouse
JBourrie
JBourrie
Oh, one other comment on your doc, something that didn't dawn on me right away but is a huge negative for your doc. It was never clear to me what the player does. There was alot of detail about things the in-game characters can do, and modes they can play, but what does the player do: the guy holding the controller. How is he interacting with this big elaborate set piece? What do the buttons that he presses do? What kind of UI does the player need? How does he interact with that UI? What does the game feel like to play? What is it that makes this game fun?

That information may already be in the document... once again, I'm not going to take the time to decode it, but what the player does should be first and foremost in any good game design.
MaulingMonkey
MaulingMonkey
Quote:
Original post by Promit
Also, you are really boring. It's no wonder no one listens to you. If you were any more sleep inducing, they'd have to regulate your speaking in order to keep society functioning.


Kaze
Kaze
Why does the expression:
"a camel is a horse designed by committee"
exist.
A horse is a general purpose animal where as a camel is specialized to very specific requirements, a horse would be like c++ where as a camel more like pl/sql.


Trapper Zoid
Trapper Zoid
Just to echo JBourrie's comments: I skimmed through your design doc from a link in one of your previous closed threads, and I had pretty much the same opinion he had. The design doc looks like it's got a lot of info in there, but I had a lot of trouble finding the info I was looking for - specifically, what the player actually does.

I'm not entirely sure what your point is about the "Glorious Vision". Most professional dev teams have means for tempering down a vision into a workable document. I agree that one of the largest flaws of game dev teams is their tendency to aim for more that they can handle given their schedule, but it's not as if too ambitious designs or "feature creep" hasn't been discussed and fought with for decades; it's a well known problem in the industry.
OrangyTang
OrangyTang
Quote:
Original post by Pirate_Lord
Actually, I have another article that addresses exactly this issue so I'll post that one after this post as a continuation of a response to this point. This issue is all about what I call "The Glorious Vision".

1. A few paragraphs of rambling, nonsensical statements are not an "article", no matter how much you wish it was.

2. GDNet is not your personal blog. Go here instead, and if your writing is actually interesting enough then people will come and read it. Posting random rants and then ignoring all the replies and critisism isn't what a discussion board is for.

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.

Sign in to reply to this topic.