So, my design mistakes list made Polycast as well. They did a nice job going through all of my points. Once again, my story point sparked the most disagreements. To clarify, I am not against story in games. I am against the idea that having a story necessarily makes a game better. Many example exist where adding a story reduces the designer’s flexibility, such as in my Rise of Legends example. Everything you put into a game comes at a cost, and story is no exception.

Jeff Strain, co-founder of ArenaNet, gave a very interesting speech on the challenges of creating a successful MMO. Here’s an important point:

Before you start building the ultimate MMO, you should accept that “MMO” is a technology, not a game design. It still feels like many MMOs are trying to build on the fundamental designs established by UO and EQ in the late ’90s. In the heyday of Doom and Quake we all eventually realized that “3D” was a technology, distinct from the “FPS,” which was a game design. It’s time we accepted that for MMOs as well. We are finding ways to overcome many of the limitations of the technology that dictated the early MMO design, such as Internet latency and limited global scalability. These improvements can enable a new class of online games that break out of the traditional MMO mold and explore new territory. It can be a daunting proposition to willfully walk away from what seems to be a “sure thing” in game design, but lack of differentiation is probably the number one reason that MMOs fail, so we all need to leave the comfort zone and start innovating, or risk creating yet another “me too” MMO.

Also, similar to Civ4‘s development, they started an external alpha test years before release:

It’s crucial to get feedback from outside the development team at a very early stage. We started alpha testing over three years before Guild Wars was released. To say that the game was crude at that point is a bit of an understatement – I think we’re still tracking down screenshots from that period and trying to get them burned. It was a very controversial decision at the time, and generated a lot of heated debate within the development team, because it flew in the face of the traditional wisdom that you should never show anyone outside the company what you are working on until it is perfect. I wish I could tell you that every tester we brought into the alpha test was honest, abided by the NDA, and gave the development team carefully-considered and high-quality feedback after each of the tri-weekly play sessions, but that would not be the truth. There were several times after we launched the program that we revisited the notion and discussed whether the good outweighed the bad. But we kept at it, and by the time Guild Wars shipped in April, 2005 it was clear that the game had benefited from the alpha test program, and today we consider it an essential component of the development process.

So, I just gave an interview on tutorials, during which I had hoped to give a concrete example of a game which handled its tutorial poorly. Unfortunately, my memory failed me as most game tutorials eventually seem to blur together. Naturally, just today I saw a perfect example of how not to write one. The game is called Bloxorz, and it is a quite good puzzle game that feels a bit like a turn-based version of Marble Madness, if that makes any sense.

At any rate, when the game begins, the player is moved through 9 screens that give instructions on how to play. The problem is that this information is simply too much for the player to digest before he or she has even a tangible sense of how the game works. Simply put, gameplay cannot be described with just words. Did you understand my Marble Madness analogy in the paragraph above? Probably not. However, as soon as you actually play the first level, the basic gameplay becomes quite clear.

Thus, advanced features, like switches and teleports, are meaningless to the player until he or she actually understands the core game. The tutorial could be twice as effective if each of the instructions screens was simply placed before the level in which the new feature first appears. The designer is essentially forcing the player to read the entire manual cover to cover and then hoping that everything gets remembered. Information should be handed out to the player only when needed.

Give the game a try, it’s fun! Just not the best tutorial experience…

Much to my surprise, my articles on game design mistakes made it onto last week’s Games for Windows Podcast. I’m a regular listener, so it was cool to hear them talking about this blog. They discussed the first two points and the last, which was the one about stories. Just to be clear, I am not anti-story. I simply believe that designers should acknowledge that including a fixed story in a game comes at a cost to other potential features. Often, this trade-off makes sense – for example, RPG and adventure games would be hard to imagine without stories. However, sometimes games which could have open-ended goals (such as strategy games) limit their replayability by shoehorning in an unnecessary story.

Oh, and they mentioned that my blog is hard to read because the font is too small. Good point. I really need to actually figure out how to use Movable Type soon…

…this wedding cake is your goal:

Wow.

I did an interview recently at Boing Boing Gadgets on tutorials. Here’s an excerpt:

So what’s the best real world example of tutorial you’ve ever come across?

I’ve seen lots of good tutorials, but I’m finding it hard to think of great ones. Making a great tutorial may be the hardest part of the developments process; it’s certainly the part I find the hardest. I would like to mention one interesting thing that Prince of Persia: Sands of Time did which served as a tutorial even though it didn’t feel like one. Between levels, you would see a black-and-white dream sequence which showed some of the moves you needed to make to pass the upcoming area. The visuals were not specific enough that it spoiled the puzzles, but they did introduce you to the advanced moves you would need so that you were better prepared for a new challenge. I had never done a wall run before, but when I saw one during the dream sequence, I immediately became aware that there was a new skill I should master in order to pass the next level. The game still took the time to teach me the literal button presses needed to do a wall run, but the dream sequence did a great job of making me want to learn this new move because I saw the context for it. Finding a way to show off cool features to encourage learning is a great idea—Google seems to be doing this as well with their product video demos for Street View and whatnot.

Continuing on from my previous post, here are four more common mistakes made by game developers.

5. Hidden code/data
Protecting your code and data is a very natural instinct – after all, you may have spent years working on the project, developing unique features, pushing the boundaries of the genre. Giving away the innards of your game is a hard step for many developers – especially executives – to take. Nonetheless, we released the game/AI source code for Civ 4 over a year ago, and – so far – the results have been fantastic. Three fan-made mods were included in the Beyond the Sword expansion – Derek Paxton’s Fall from Heaven: Age of Ice, Gabriele Trovato’s Rhye’s and Fall of Civilization, and Dale Kent’s WWII: The Road to War – and so far, these mods have been heralded as one of the product’s strongest features. To be clear, these mods would have been nowhere near as deep or compelling (or even possible) if we had not released our source code. I should specify that for many PC developers, I’m preaching to the choir, so I’d like to be very specific about which genre I am calling out – strategy games. For whatever reason (perhaps the lack of a pioneering developer like id?), strategy developers have been much more closed off to modding than their shooter and RPG brethren. Sure, there are exceptions, like Blizzard’s fantastic scenario editor for WarCraft 3, but by and large, strategy modders do not have many places to turn for platforms on which to work, which was one reason we felt compelled to focus on modding for Civ 4. Giving stuff away can feel good. It also feels smart.

6. Anti-piracy paranoia
The damage that piracy does to our industry is impossible to calculate but also impossible to ignore. Few company heads can be as brave as Brad Wardell and just leave out copy protection altogether. Thus, having some sort of mechanism to stop casual piracy is a given but what is not a given is the hoops companies will make their customers jump through just to be able to start the game. The most important question to ask when considering these protections is “will this added security actually increase our sales?” A good place to be lenient, for example, is with local multi-player games – in other words, can players without the disk join a multi-player game hosted by a legitimate copy. Starcraft let you “spawn” extra copies of the game that could only join LAN multi-player games. (Interestingly, this is the same model that Ticket to Ride employs on the Net. It is always free to join a game but only paying customers can host.) Allowing unlimited LAN play was our unofficial policy for Civ 4 as well. The game does a disk check when you start the EXE but not when you actually launch the game; thus, a group of 4 friends could just pass one disk around for local multiplayer. We do not believe players are willing to buy extra discs just for the ability to play multiplayer at a LAN party, which are rare events. However, we would love for new players to be introduced to Civ in these environments, encouraged by their friends who are already fans. At some point, they are going to want to try single-player – in which case, it is time for a trip down to the local Best Buy.

7. Black box mechanics
Sometime during the late-90’s, around when Black & White was being developed, the concept of an interface-less game came into vogue. The idea was that interfaces were holding games back from larger, more mainstream audiences. Ever since then, I have noticed a discernible trend to hide game mechanics from the player. Age of Kings shipped in 1999 with an incredible reference card listing every cost, value, and modifier in the game. With most modern RTS’s, however, you’re lucky if the manual actually contains numbers. I want to emphasize that the answer here is not to bathe the players in complicated mathematics in the name of transparency. Instead, designers should think of their interfaces as having two levels: a teaching level and a reference level. The teaching level focuses on first-time players who need to know the basics, like how to build a tank and go kill the bad guys. The reference level should answer any question the player can think of about how a game mechanic works. It is perfectly fine, by the way, to put this info inside of a separate in-game resource, like the Civilopedia. Rise of Legends implemented an interesting version of this two-interfaces idea. Most of the popup help in the game had an “advanced” mode that you could unlock by holding down a key, giving you significantly more details about the game’s underlying mechanics.

8. Putting story in the wrong places
I was tempted to come up with 7 things not to do and just leave off the story one as I’m sure it’s my most controversial point. A bunch of people will disagree with me over the place of story in games, so let me just say up front that I know that I am wrong. I still want to make my point, though. I don’t like story in games. I don’t like the boring cut-scenes. I don’t like the stereotyped characters. I don’t like the plots that I have no control over (and, sorry, the Bioware you-are-either-God-or-Satan twists count too). I especially don’t like it when games stop me from fast-forwarding through the crappy dialogue (I’m looking at you, Japan). But what I really hate is when a story gets stuck somewhere it really doesn’t belong. Like in a strategy game. After all, strategy games are the original games. Humans first discovered gameplay with backgammon and chess and go; it’s a noble tradition. The “story” in a strategy game is the game itself. Layering a story onto an RTS campaign is like putting a copy of Hamlet in my pie. I mean, sure, Hamlet is a great play, but my pie would also sure taste better without it! Put another way, how much better of a game would Rise of Legends have been (and it was already a great game) if they had given up on creating a story-based campaign and instead iterated on the cool Conquer-the-World mode from Rise of Nations? Ironically, the campaign mode was my favorite way to play RoL. I loved that you could only acquire technologies and advanced units on the strategic map between missions, which helped to simplify the core RTS game. However, I enjoyed the campaign in spite of the story, not because of it. The key point here is that, for the sake of chasing a story, Big Huge Games missed a big opportunity to match a great core RTS game with a simple, overarching strategy layer that could be infinitely replayable. They are not alone; almost every other RTS developer seems to be falling into the same trap, and I don’t know why.

Of course, if I ever made an RPG, I would probably name the bad guy Foozle, so what do I know?

Well, for better or worse, these are the eight things I hate seeing in games, especially strategy games. What about you?

Fable
The Darkness Demo
Heavenly Sword Demo
Psychonauts
Selling Out

Wonder how long he can keep it up…

Inspired by Troy S. Goodfellow’s list of the Eight Greatest Features he values in strategy games, I started thinking about the opposite question: what are the greatest mistakes that I hate to see done over and over again in game design? In no particular order, here are my first four:

1. Hard-core game conventions
One of the most common pitfalls for a game designer is to fear that the game is not hard enough. This fear often leads to hard-core game conventions, like restrictive save systems and unlockable content, that only put roadblocks in the way of the mainstream gamer who is just looking to have a good time. If you feel your game needs the tension of a restrictive save system, go ahead and implement it… but only as a feature of a higher difficulty level. Difficulty levels are the key to making a game accessible to both the casual and the hard-core gamer; we could never seem to add enough difficulty levels to Civ to keep our wide variety of fans happy. Trauma Center (DS) is a good example of a great game that was ruined by having no difficulty levels whatsoever. The surgery game is a brilliant use of the DS touch-screen, but the linear challenges get so hard by the fourth or fifth level that most people get hopelessly stuck after only a couple hours. Considering that the levels were timed, it wouldn’t have taken them more than a week to implement a difficulty system that simply extended the time limits at the easier settings. A Trauma Center with difficulty levels would have enjoyed similar success to Elite Beat Agents – another great touch-screen game but one not afraid to let the player start at an easy difficulty level.

2. Repetitive interface tasks
I am currently enjoying the old-school dungeon crawler Etrian Odyssey quite a bit on my DS, enough so that I can’t help day-dreaming about how much fun it would be to remake Bard’s Tale or Legacy of the Ancients for the DS. Unfortunately, the game’s interface does a terrible job of enabling the player to skip over needlessly repetitive tasks. Want to sell your loot? You have to click on every single Hare Tail in your inventory not once, but twice for confirmation! After a long excursion, this can often lead to around 100 presses of the A button when you get back to town. A simple “sell all of item X” would be an invaluable time-saver. Likewise, as a typical party-based RPG, there comes a time when your group no longer has to fear the lower-level creatures. However, for every random encounters, you still have to select ‘Attack’ and target a creature for all five of your characters even though there is literally ZERO danger to your party. (That’s ten presses of the A button for those of you keeping score at home.) A “party auto-attack” command for these battles would have saved me literally hours of play time. Always remember, your player’s time is valuable.
Fun Factor = Interesting Decisions / Actual Time Played.

(UPDATE: Yeah, so I blew this one. There is a “sell all” option in Etrian Odyssey, and it’s even shown on the interface. The point is still valid, but I targeted the wrong game.)

3. Limited play variety
No matter how good your game is, it is going to get stale after awhile. It’s a real shame when a great game doesn’t take the few extra steps necessary so that the player can mess around with the settings to create alternative play experiences. Company of Heroes is an incredible tactical RTS, a watershed moment for the genre – but there is no way to have an Axis vs. Axis battle or even a game with more than two sides. This design choice may fit the fiction of WWII, but it significantly reduced the game’s play variety. A good example of an RTS that got this right is the Age of Empires series. Not only could you mix-and-match any combination of civilizations and players and teams, but you could also design your own map scripts. I remember one interesting Age of Kings map designed by Mike Breitkreutz, a Firaxis programmer, that had almost no wood and tons of stone and gold, turning the game’s economy upside-down. You could even have multiple players controlling the same single civilization (one player could control the military, the other the economy, for example). Thus, I’ve played 2-vs-3 games of AoK where the sides with 2 civs was actually controlled by 4 players (and guess which side won?!?) These simple variations probably doubled the life-span of AoK amongst my group of friends.

4. Too much stuff
The temptation to pile extra units and buildings and whatnot onto to an already complete design is strong. Indeed, I have seen many people describe games as simply a collection of stuff (“18 Weapons! 68 Monsters! 29 Levels!”) Needless to say, this is a wrong-headed approach. A game design is a collection of interesting decisions, as Sid would say, and the “stuff” in the game is there not to fill space but to let you execute decisions. Games can provide too few options for the player but – more commonly – games provide too many. How many is just right? That’s simple enough to answer, it’s 12! (it’s definitely not 42…) OK, obviously there is no magic number, but 12 is a good figure to keep in mind. It’s an excellent rule-of-thumb for how many different options a player can keep in his or her mind before everything turns to mush. It’s the number Blizzard uses to make sure their RTS’s don’t get too complex. StarCraft averaged 12 units per side. So did WarCraft 3 (not counting heroes). And you can bet your bottom dollar that StarCraft 2 is going to be in that neighborhood as well. In fact, Blizzard has already announced that, for the sequel, they will be removing some of the old units to make room for the new ones.

Next time: pirates, modders, and black boxes…

This afternoon, I gave a talk to some Korean game developers and academics on developing the AI for Civilization. One of them asked about a strategy game that I developed back in another lifetime which used genetic AI. I promised that I would add a link on my blog to this project, which was titled Advanced Protection. It’s a little MFC app that demonstrates how genetic algorithms can be used to encourage more adaptive AI behavior (although I should mention that the game mechanics were designed largely to favor an environment in which a GA could be used…)

As an aside, there was actually a Genetic AI project for Civ4 that was ongoing for a number of months. Too bad it is currently defunct…