The Road to Zynga

So, I’ve got a new job, which was revealed publicly in an interview at GamesIndustry International. I’ve actually been at Zynga for nine months now, so most of my games industry friends (and regular friends, for that matter) had known about it for quite awhile, making this move one of those “worst-kept secret” things. Surprisingly, the news stayed out the press, with the exception of an off-hand reference at Gamasutra during my first week of work! The turn of events which got me to Zynga is somewhat head-spinning, especially since I am now back in Maryland, working with Brian Reynolds, Tim Train, and a bunch of other former Big Huge Games people. I can’t helped be a little bemused because I was originally hired in 2000 to replace Brian & co. to finish (or, more accurately, to restart) the development of Civilization 3 after the mass exodus to BHG. Indeed, at the time, emotions were still a little raw at Firaxis, so my initial, second-hand impression of them was perhaps not ideal! Over the years, I got to know Tim and Brian in person, and I could always tell that their design culture would be a great fit for me if the opportunity ever arose. I didn’t always expect that opportunity to come at Zynga, but so far, so good. I’ll have a lot more to say on that topic once I can actually talk about my current project!

Here are some relevant bits from the interview:

Q: You’re the latest in a string of former EA employees to join Zynga. What do you think this says about EA and what does it say about Zynga?

A: The shift we’re seeing with industry talent moving towards social and mobile games illustrates the challenges of the old model. That model demands selling a $60 box product, which requires exceptionally high production values to sway the consumer. Of course, AAA production means development costs measured in the tens (or perhaps hundreds) of millions of dollars, which then requires a game to sell at least several million copies at the $60 price to be worth developing. This business model can work, but the margins are not great, and they keep getting worse as development costs increase and retail sales soften. One of the bizarre paradoxes of this system is that some mid-tier games cost too little for traditional games companies to fund because the company needs to put all of its organizational weight behind a few key titles.

Mobile and social gaming suddenly became very attractive for game designers because they circumvent this problem entirely. As these games require a fraction of the cost of AAA games, radical new ideas can be tried out fairly easily and with little institutional pressure. Further, feedback is much easier to gather from the Web or connected smartphones, and teams can continually course-correct with regular updates. The old system of expensive, multi-year development, cut off from the oxygen of player feedback, suddenly became much less appealing.

Q: Given your experiences in the past on Civ IV and Spore, how well does what you’ve learned on traditional games apply to the social space and Zynga’s approach?

A: Fun still happens inside the player’s head, and that doesn’t change from platform to platform. What does change is that certain design possibilities are unlocked with each new technology, such as zero account overhead, assumed persistence, guaranteed connectivity, pre-made friends lists, built-in viral channels, universal access, and ever-improving game engines. These features let us explore areas of design that were previously unavailable. Indeed I feel like my work right now is a direct continuation of what I have done in the past, and I am simply using my twelve years of experience as a strategy game developer to take advantage of these new possibilities.

Q: Are you coding, or designing, or some of both?

A: I am doing the code and the design for the project, just as I did with Civ IV. I like working this way because I can spend less time explaining how I want a feature to function and more time watching how it works out in practice. I am using a somewhat unusual technology solution (GWT/PlayN) that enables me to write a HTML5 game in one language, including client and server, which means I have no duplicate, error-prone code and a very consistent style. Of course, doing code and design could make me a major bottleneck as the project grows larger, but one of the advantages of strategy games is that – because they are rules-driven instead of content-driven – a small amount of code can create a large amount of gameplay. Thus, extra engineering resources will be dedicated to features that work in parallel to the core game – scalability, UI, matchmaking, metrics, modding, and so on.

Q: Can you give us any hints at all about what your game project at Zynga entails?

A: I’m afraid that it’s too early to reveal much about what I’m currently working on, especially since a lot could change between now and our release date. However, I can say that I am working on a game that would not be a shocking departure from my development history, but one that is multiplayer-focused, free-to-play, persistent, and playable in the browser. Most importantly, the game is one that actually ends, with real winners and real losers. That’s a big departure for Zynga, but it’s a necessary element for many types of strategy games. Decisions matter in vastly different ways once games have an actual victory condition. Put another way, a bit of a level race does exist in persistent MMOs like World of Warcraft, but ultimately the race is just with yourself. No one else really cares how much time you put into your dwarf cleric. On the other hand, it’s a powerful feeling when you beat someone to a wonder in Civ IV.

Q: Do you see social games becoming more like familiar PC or console games in terms of mechanics?

A: After a couple years the term “social gaming” came to mean different things to different people, and many traditional game developers simply use the term to describe “games on Facebook that we don’t like.” On the other hand, some of the biggest hits (Words with Friends, Bejeweled Blitz, Draw Something) are not really thought of primarily as “social games” but simply as great games that take advantage of the possibilities of social gaming – guaranteed connectivity, account persistence, meaningful friends lists, etc. Therefore, the best-case scenario for social gaming is for the term itself to disappear as all games become more social, even ones that are traditionally thought of as primarily single-player. For example, consider how friend-based leaderboards drive the experience in console games like Trials Evolution or Burnout Paradise.

Ultimately, gameplay is determined by the input and output of the devices we use, which enables and encourages certain types of play, and there are currently three important formats – the consoles (defined as gamepad, high-def TV, couch-based), the PC (keyboard, mouse, monitor, desk-based), and mobile (small screen, touch-based, omnipresent). Consoles will continue to dominate avatar-based games as controllers are built to “pilot” characters or vehicles. PCs are the natural home for more complex games because of their precise input and “lean-in” environment. Mobile is perfect for asynchronous gaming, and the touch-based user interface creates entirely new types of games just like the Wii did with motion-based controls. The best games will take advantage of the unique power of their formats, which is why we shouldn’t necessarily expect Facebook games to suddenly become Halo.

Q: What’s your take on the impact of social and mobile games on the traditional console business model?

A: What’s interesting about the games business right now is that we have four different business models operating at once. For console games, retail still dominates with increasing revenues from microtransactions. On the PC side retail is dead, but traditional, single-purchase games still thrive on Steam. Furthermore, free-to-play games from CityVille to League of Legends can succeed purely via microtransactions. On mobile devices, the app store dominates with the big money coming from in-app purchases. Over time, these four models will almost certainly merge into one consistent system. We have already seen plenty of convergence; Steam made a big push to support microtransactions and free-to-play games while in-app purchases were an important later addition to Apple’s App Store. The big question is the next generation of consoles. The benefits of fully embracing an app store model are obvious, but there are many entrenched interests and historical relationships slowing down that transition.

App stores, in general, are very interesting simply because they make it possible to actually charge players for small-scale games. For example, there is no reason why I shouldn’t be able to buy Ascension for $4.99 to play in my browser, but the tradition simply does not exist to “buy” web pages. On the other hand, the Apple App Store combines frictionless purchasing, flexible pricing, automated submissions, and an enormous audience to create a market for games which didn’t exist previously. At some point, the importance of the browser (with open standards and cross-device support) needs to combine with the ease of the app store (with a standardized purchasing model), to hopefully create an even better system.

Q: Do you think the kind of hardcore PC player that you targeted with Civ IV will want to play your next game at Zynga, or do you have to change your design philosophy to cater to a new, more casual audience?

A: I expect Civ fans will absolutely be interested in my current project at Zynga; in fact, I anticipate them to be among the first wave of players to try out the game. As with Civ IV, I will be trying to make the game as simple as possible to keep it accessible for any kind of player. Remember that Civ itself is no typical strategy game and has an audience which is an order of magnitude larger than any other turn-based strategy game. The reason is that the series is infused with “Sid’s design parsimony” (to borrow a phrase from Chris Crawford), refusing to pile on more and more rules and mechanics in the name of realism. I hope to carry this tradition forward to make a game for every gamer.

Dragon Age Legends: Guilds Explained

My current game, Dragon Age Legends, was released last month. I wrote the following post as designer notes on the upcoming guild system.

One of the design goals of Dragon Age Legends is to have meaningful social mechanics. Many social game are social only in the sense that they use the player’s social graph to help spread the game virally through her network – by encouraging the player to invite her friends as neighbors to grow her farm, for example, or trading items with one another through the gifting system to finish a building. While these mechanics help grow a social game’s visibility, they don’t necessarily make the actual experience of playing the game more fun.

Our core social feature in Legends is borrowing friends’ characters to fight alongside one’s own character in combat. As one’s friends level up their own characters, they make the game more fun by providing new skills and stronger characters to use. Unlike most RPGs, players of Legends will be able to try out the entire skill tree depending on how their friends upgrade their characters. Borrowing a friend’s character also provides that friend with a small gold bonus, which creates some interesting dynamics – encouraging players to have the most appealing characters (to earn more gold) while also giving veterans a charitable reason to bring along low-level friends they want to help.

However, a genuinely meaningful social mechanic can create its own share of problems. Facebook friends are not necessarily one’s actual friends. Players often announce their names and character details in various forums, hoping to find “fake friends” to fill out their list. Doing so creates three advantages. First, the more friends the player has, the more opportunities for his character to be borrowed and thus earn friend gold for the player. Second, high-level friends make combat far easier because of their high stats and upgraded skills.

Finally, a surplus of friends allows the player to bypass the rest time restriction. Borrowed characters enter a rest state after combat finishes for a certain period of time (often for a few hours, depending on the level and damage sustained). This rest period exists so that players will not be able to reuse a single friend’s character over and over again (and, thus, feel no incentive to invite other friends into the game). However, if a player has a huge list of fake friends playing the game, perhaps numbering in the hundreds, this mechanic becomes irrelevant as there will always be a ready supply of fully rested characters.

These three advantages can greatly distort the balance of the game. In particular, if a player has a huge list of high-level friends, which is not difficult if one looks in the right places, the combat becomes trivially easy. High-level characters can wade through most monsters easily, which prevent us from finding an ideal difficulty level. We could boost the strength of all monsters across the board to compensate for endless high-level friends, but that change would ruin the game for the average player who only uses her real friends. Besides, we want the player to experience the power of a high-level friend mowing down waves upon waves of darkspawn from time to time, just not in every battle.

To fix this issue, we are creating a new system – Guilds. A Guild is a select group of 16 friends who are playing Legends. The player can only borrow characters for combat from this group of 16. The composition of a Guild can be changed at any time (as long as the character being removed is not currently resting), so a player is not restricted to whichever friends are first added to the Guild. Also, a new castle room – the Great Hall – will allow players to expand their Guilds.

Guild membership is one-way; I might have my friend Ethan in my Guild, but Ethan does not need to have me in his Guild. However, if we are both in each other’s Guilds, we receive a bonus of a shorter rest time when using each other’s characters. This feature creates some interesting social pressure, forcing players to choose between using their friends with the best characters and using their actual best friends.

From a design perspective, the greatest benefit of Guilds is tuning as we can now balance the game for a single target – a player who has 16 friends, with a mixture of low- and high-level characters. Because rest time is proportional to a character’s level, players might not want to fill their Guild with only high-level character who would often be unavailable.

Ultimately, the player should be making interesting decisions during all parts of the game, including when deciding which friends to use for combat. Perhaps a certain battle looks fairly easy, so a player might want to use a couple low-level friends or maybe to even try it solo. Perhaps a looming boss battle makes the player hesitant to waste his highest-level friend on a normal encounter. With infinite high-level friends, these dynamics disappear, to the detriment of the gameplay.

Moreover, we want players to be interacting as much as possible with their real friends, as these are the most important social bonds tying the player to the game. Guilds encourage this behavior via the reciprocal membership bonus to rest times as well as the simple ability to build a subset of friends most relevant to the player. Guilds are a small but important step towards creating meaningful and balanced social mechanics within Dragon Age Legends.

Dragon Age Legends: Economy Explained

My current game, Dragon Age Legends, was released this week. I wrote the following post as designer notes on the economy system.

Many successful Facebook games are persistent management games. In FarmVille, the player plants and grows a prosperous farm. In Millionaire City, the player designs and constructs a bustling city. These games give players a certain number of resources to spend on development and then reward their smart moves with more money to invest, creating a positive feedback loop. The money players earn enables them to buy more things that will earn them even more money for more things and so on.

While this mechanic forms the solid underpinning of most management games, it creates two problems for long-term, persistent play. First, as the gameplay is stretched over months (instead of a few hours in the case of non-persistent games like SimCity), the amount of interaction required for each visit is quite small – place a building here, drop a road down there, etc. Therefore, most social games add UI busywork to fill up the player’s time – plowing, seeding, and harvesting every plot for farming games or clicking on each energy, food, and wood icon that appears in FrontierVille. In actual time spent, these often mindless activities comprise the bulk of each play session.

Second, these games often lack a sense of purpose outside of simply growing the player’s ability to continue growing. If the only reason to earn more money is to invest in items that will earn more money, the game eventually loses players not interested in pursuing purely aesthetic goals. At some point, maintaining a city begins to feel more like clearing the weeds or doing the laundry than playing a game.

The turn-based combat of Legends solves both problems at once. The tactical battles occupy the majority of the player’s time within the game, and they are clearly not mindless click-fests, challenging the player and rewarding smart moves. Furthermore, the battles give the persistent castle an actual purpose; expanding the castle is not an end in and of itself. Instead, the output of the castle is consumables the player can use in combat: health potions, mana salves, shard bombs, and so on.

Indeed, the way consumable are handled in Legends is meant to improve on their use in traditional RPGs as well. Traditionally with these type of games, hoarding is quite common as the player is uncertain what is around the corner – what if she uses too many potions and will not be able to tackle the later game as the difficulty increases? Turn-based RPGs suffer even more as the player is usually choosing between a repeatable skill and a consumable item, which means that the power of the latter has to greatly outweigh that of the former to be worth the permanent loss of the item.

Legends solves this problem with a few simple changes. First, consumable use does not end a character’s turn; instead, it is an optional step. The character can either shoot an arrow or drink a potion and shoot an arrow. Thus, by passing up the use of a consumable, the player is forfeiting an opportunity and needs only to weigh the effect of the item versus its permanent loss.

Second, the game’s core stats, health and mana, do not regenerate as they do in most RPGs. Health does replenish but only outside of combat and only in real time – one point of damage per hour. As for mana, characters receive only one or two free points at the start of combat, enough to use a couple skills per battle. Thus, if players want to restore their health and mana, they need to rely heavily on their consumable items – health and mana potions, injury kits, and mana salves. As long as battles are balanced correctly, consumables become the fuel that players use to power though DAL’s combat.

However, Legends also differs from traditional RPGs because the player is actually in control of the supply of consumables, via the castle. The core of a player’s castle are the workshops – the apothecary (for crafting potions), the infirmary (for salves), and the alchemy lab (for bombs) – which can create consumables over certain timed increments. For example, the player can place a worker in the apothecary to create 2 health potions in 30 minutes of real time. Giving players this control frees them from the anxiety of depleting their limited supply of items. With Legends, players can always invest time and gold into their castle to start rebuilding their supply.

Indeed, the two halves of the game – the castle and the combat – create a self-sustaining economic loop. Gold earned from fighting battles and completing quests can be invested into expanding the castle and upgrading its rooms. Accordingly, maintaining the castle and tasking workers to create items provides the potions, salves, and bombs the player needs to defeat the increasingly difficult monsters encountered over the course of the game.

Thus, the two parts of the game fit together and buttress each other. A well-supplied character will lose less battles, earning more gold that can be invested in the castle. A well-maintained castle will create more consumables for the the player to use in combat, increasing the odds of success. The combat and the castle provide context for each other, motivating the player to keep fighting and to keep building.

Dragon Age Legends: Combat Explained

My current game, Dragon Age Legends, was released this week. I wrote the following post as designer notes on the combat system.

One of the most lauded features of Dragon Age: Origins was its tactical combat system, which encouraged players to plan ahead and make interesting choices during battle. Indeed, although the system ran in real-time, the player could set a number of triggers that paused combat, to give time for deciding which skills to use and against which enemies. This feature allowed some to play Origins as almost a turn-based game that emphasized smart tactics over fast reflexes.

The Dragon Age Journeys Flash game built in parallel with Origins brought elements of the franchise to an actual turn-based game, in which battles played out on a hex-based grid. For the new Dragon Age Legends Facebook game, we built on top of what worked within the Journeys system (which itself was based on Daniel Stradwick’s Monsters’ Den Flash RPG).

For example, we borrowed the interleaved turn queues from Journeys, which means that the heroes and monsters take turns one at a time instead of fighting together as alternating groups, as is common in many group-based RPGs. Giving the player knowledge of which exact characters will move next creates some interesting tactical decisions. A monster who will attack sooner might be a better target than one who is a greater threat overall. Skills which disable or freeze enemies can be used intelligently to keep the most dangerous monsters at bay.

However, we simplified other mechanics from Journeys. Instead of using a hex-based grid, we adopted a simpler layout familiar to fans of Japanese RPG series like Final Fantasy in which the heroes and monsters line up on opposite sides of the screen is static slots. We built a 2 X 3 grid for each side, with front and back columns so that characters in the back column are protected from melee attacks by characters in the front column. This arrangement allows players to plan ahead by attacking a specific monster in the front line that, when dead, would expose a weak but dangerous blood mage in the back.

Another big change involved how we handled health and mana, the two most common stats from RPGs. Instead of using a bar-based system, in which a character might have 45 of 80 health points, we adopted a more chunky icon-based system. A character’s health is represented simply by 4 hearts, which are measured in halves, identical to the health system from the Zelda games. Mana is represented by single icons and as most skills cost just one mana, the number of icons a character has is shorthand for how many skills she can use.

This simplification had a number of advantages. First, the system had much greater clarity than opaque health and mana bars. These bars generally have the same size on the interface for the sake of consistency, which unfortunately obscures their true values. Although one character might have 20 hit points while another has 200, their health bars look the same on the game screen. In fact, most modern RPGs have superimposed text on top of health bars to give the actual values to avoid any confusion. With the iconic hearts, Legends avoids this problem as the graphics do not obscure any game data – what you see is what you get.

However, the most important reason to adopt the icon system is that it gives players a very tangible understanding of the consequences of their actions; the interface tells the exact effect of every possible action. For example, mousing over the “Power Strike” skill button next to a Hurlock shows that the action will take one-and-a-half hearts away from the monster while draining one mana from the player’s character. Mousing over a health potion shows that two hearts will be restored to the character. Mousing over a shock bomb shows that one heart will be taken away from all enemies. This transparency enables the player to plan ahead and make smart moves.

One reason so many RPGs go with a bar-based system is that hits points need to scale up over the course of a game, ranging from 10 hit points for starting monsters to over 1000 for bosses. With icons, Legends can never support hit point values that scale so high because the interface can only show so many hearts. Instead, the underlying character stats – attack, defense, agility, and luck – provide the scaling we need as players progress through the game.

Damage is calculated by creating a simple ratio between the attacker’s attack and the defender’s defense values. As this ratio increases to 2:1, 3:1, and higher, the damage value in hearts goes up. However, if the ratio remains constant, so does the damage. Thus, characters battling with 10 attack and 8 defense do the same amount of damage as characters battling with 50 attack and 40 defense, which means that the relative value of hearts stays the same. Similarly, agility and luck determine the odds of glancing blows and critical hits, by comparing the attacker’s luck with the defender’s agility.

As for mana, most skills cost just one point, with a few costing two and a handful costing more. The skills are all upgradeable up to ten levels (similar to the Japanese RPG Etrian Odyssey), which means that their power increases over the course of the game. However, their costs do not, so that players can expect to use the same number of skills per battle through all phases of the game. The costs can remain fixed because the better skills are simply an extension of the character’s growing power as he levels up.

Thus, the game’s simple and discrete health and mana values are maintainable across the game’s various levels. The players quickly gain an intuitive sense for how the game works (“shard bombs do one-and-a-half hearts of damage”) without having to memorize formulas or manage large, crooked numbers. Being able to think in small increments – a point here, a couple of points there – allows the player to fit a whole battle into her head at once, making combat a fun, tangible experience instead of a chaotic, stressful one.

My Favorite Week: 2011 Edition

GDC 2011 start on Monday, and I am even more excited than normal this year because I helped program the conference. The GDC Advisory Board invited me to join them last summer, and it’s been very interesting watching how the sausage gets made. I’m looking forward to seeing which ideas and sessions work out and which ones don’t. (If you don’t like that the “fuzzy” sessions – the rant, the microtalks, and the challenge – are scheduled during lunch, you can blame me!) It’s been an honor to be part of the process – I am as much a GDC junkie as ever.

I’m taking part in one talk this year, a panel on strategy gaming, for which I hand-picked the members. Hope to see you there!

Strategy Games: The Next Move

SPEAKER/S: Tom Chick (Quarter to Three)Ian Fischer (Robot Entertainment)Soren Johnson (EA2D)Dustin Browder (Blizzard Entertainment) and Jon Shafer (Stardock)

DAY / TIME / LOCATION: Friday 11:00-12:00 Room 134, North Hall
TRACK / FORMAT: Game Design / Panel

DESCRIPTION: Strategy games have one of the longest traditions within the industry, including two of last year’s biggest games, STARCRAFT II AND CIVILIZATION V. In what direction is the genre heading? What are some of most important, and possibly overlooked, gameplay innovations of the last few years? How has the growth on online, persistent play affected the way strategy games are developed? Has the rapidly expanding mainstream audience changed how strategy games are targeted, or is the genre at risk of turning into a ghetto? As the market moves towards free-to-play, micro-transaction-based gaming, how will strategy gaming adapt while maintaining fairness of play? Is there still room for traditional, boxed strategy games?

TAKEAWAY: Several experienced strategy developers will share their own perspectives on the future of the genre, offering insights on both game design and the challenges facing the genre in the coming years.

INTENDED AUDIENCE: Although primary of interest to strategy game developers, the session will also be relevant to anyone interested in how a game genre evolves and reinvents itself in the face of a changing market.

Talking about Dragon Age Legends and Social Games

I’ve been on a couple podcasts recently in which I talk about my newest game – the Facebook-based Dragon Age Legends. The first was the strategy-focused Three Moves Ahead, which also addressed my talks/columns on theme and meaning in games. The second was the The Digital Life, in which I was paired with veteran design Brenda Brathwaite to talk about the emerging format of social games. I can’t promise that I don’t repeat myself between the two, but they are both worth a listen for developers interested in this field – and for fans interested in my next game!

“Fear and Loathing in Farmville”

GDC 2010 is now in the books, and it will be a hard one to forget because the whole conference seemed to be obsessed with one thing, which I summed up in this tweet. Or, as Sirlin puts it here: “Facebook, Facebook, Facebook, Facebook, Facebook, Facebook, Facebook, Facebook, Facebook, Facebook.” Off the top of my head, here are the highlights and lowlights of this fixation:

  • The long-running Casual Games and Virtual Worlds Summits have vanished entirely from the conference, presumably eaten up by the new Social Games Summit.
  • Ngmoco’s Neil Young describing the growth of free-to-play online games as “the most significant shift and opportunity for [game developers] since the birth of this business.” This shift fundamentally changes the way game are made because developers can now launch early and adjust based off play patterns and user metrics.
  • Zynga’s Mark Skaggs, formerly of EA, praised metrics as the answer to most game design problems. Much has been made about their discovery that pink was the best color for advertising Zynga’s other games, but the telling point was when Skaggs said that “if a player repeats something, it’s fun.”
  • My old Spore teammate Chris Hecker railed against external rewards as a true motivator as they can mask an otherwise dull game. Further, focusing primarily on metrics can actually make the game worse because they can overvalue external rewards, which are easier to measure. Chris also leveled this broadside at metrics-focused companies: “If you are intentionally making dull games with variable ratio extrinsic motivators to separate people from their money, you have my pity.”
  • Carnegie Mellon’s Jesse Schell walked back from the ledge of his now infamous DICE talk on pervasive rewards systems, saying that doomsday is not inevitable. He went on to explicitly draw the line in a new war between persuaders (developers who want players’ money) and the rest of us (who want to give the players joy). When addressing persuaders, Schell actually used the phrase “you know who you are.”
  • Zynga’s Bill Mooney offended the entire independent games community in his acceptance speech for Farmville at the Choice Award by defining the Facebook game as “just as indie” and then trying to recruit everyone in the audience, many of whom have open disregard for Zynga. Josh Sutphin had a message for him: “Learn some fucking tact.”
  • Brian Reynolds, who is now Zynga’s Chief Designer, showed up on no less than three panels to point out repeatedly that social games need to be social first and games second. Farmville‘s crop-withering mechanic, in particular, was referenced as a not-fun mechanic that compels people to play out of a sense of shame. (What if my real-life friends see how poorly I am maintaining my own farm?)
  • Daniel James of Three Rings puzzled over the phrase “social gaming” as he felt that his old games (such as Puzzle Pirates) were far more social than Farmville, which is a primarily single-player game in which players pass around “tokens.” At multiple times during the conference, James expressed his serious ethical qualms over the path social gaming was laying for the industry. So many of the methods for making money are thinly-veiled scams that simply exploit psychological flaws in the human brain.
  • At a panel on why “dinosaur” designers are flocking to social games, Reynolds, Slide’s Brenda Brathwaite, Noah Falstein, and Playdom’s Steve Meretzky all praised social gaming as a new frontier where radical and rapid innovation exists, in contrast to the more conservative world of AAA retail games.
  • Will Wright pointed out that the astonishing growth of Facebook (and Facebook gaming) more likely resembles an S-curve than a power law curve. Thus, although this new market is indeed enormous, the upward sloping curve will level off at some point, so we should be careful not to make exponential predictions.
  • Sid Meier only briefly touched on Civilization Network, his new Facebook project, in his conference keynote, but what else needs to be said? Sid Meier is making a Facebook game! (Quite literally, in fact, as Sid is doing his usual designer/programmer thing.) Further, the three primary designers of  the Civilization franchise (Sid, Brian, and myself) are all now making social/online games.

What is to be made of all this? Meretzky made a key point in the dinosaur panel that, with free-to-play games, there is no more separation between game design and game business. Every change to a game’s balance might immediately and significantly affect revenue. Will it go down because the virtual items for sale are now less desirable compared to the free ones? Or will it go up because the player is now inconvenienced enough to buy a boost? Or will it go down because the inconvenience has driven away enough of the core fanbase? (I made a similar point in my Nov 2008 column on designing free-to-play games.)

The question on most developers’ minds is the following: what is the role of the game designer in this new world where business and design mix in such fundamental ways? The answer to this question drives fear in the heart of the boy or girl beating inside most professional game developers. Brian Reynolds himself often pointed out that the role of Zynga’s Chief Designer is not actually as important a position as one might imagine. At the VCON Summit, Eric Goldberg of Crossover Technologies suggested that companies “use the tactics that make the most money possible… that your staff can live with.” At that summit’s keynote, David Perry talked about the morally dubious “treasure chests” of ZT Online, which are engineered to prey on gambling addicts and provoked a visceral response from Sirlin:

This egregious, unethical practice is the kind of thing he should have presented as extremely dangerous. If you are “playing to win” in business, yeah, you’d do that. But doing so is damaging to the lives of our own customers… I mean personally, I’m embarrassed to be part of an industry that so blatantly manipulates people like rats in a skinner box, and isn’t he embarrassed about that too?

This debate over business-vs-design spawned a thread at Quarter to Three in which game developers are expressing their feelings over Farmville and its ilk:

It’s not social games as a threat to game design, it’s money-driven treadmill games that’s a threat to game design. A coworker identified a similar problem with a money-driven free-to-play social game, in which they specifically destroyed the balance in key ways at times in order to persuade the players to pay money to fix their own game balance. It is a war. It’s suits versus the creative people. (link)

I can’t believe one of the most important figures in strategy gaming [Brian Reynolds], the guy who had a major hand in bringing us absolute classics like Civ 2Alpha CentauriRise of Nations, and Rise of Legends is now Chief Designer for those creeps at Zynga. (link)

I don’t like that at all. It turns my art into a business intent only on making as much money as possible. And while making money is the goal for the large industry, the fact is that we’re still as much about creating great experiences first and foremost, and the money is a happy second. With Farmville and such, the premise is to make a lot of money, and that is the drive that informs every single decision. (link)

Making the game worse can make it generate more revenue. The lesson is to focus on generating fast bucks over improving the artistic quality of your game. Enjoyment isn’t as important as long as they keep paying and playing. The dividing line flaring up is an old one; are games an artistic endeavour furthering culture or are they just slot machines to be designed for revenue maximization? (link)

Farmville makes overt use of known psychological techniques to influence and control behaviour and ties that directly into revenue generation. . . . When you have games industry professionals from large companies arguing that we shouldn’t worry about making a game less enjoyable as long as it generates more revenue – to me that is something to be concerned about. (link)

Farmville‘s formula is simple. Make it easy to scream forward to the point where you can’t properly spend your coins anymore without spending real money. . . . Do not misunderstand me, I am saying, without any ambiguity, that doing this is wrong. I see very little difference between this and tactics at stores such as raising the price of something, removing functionality, and slapping a “On Sale 40% Off!” sign on it. (link)

The question will be, when it comes to tuning Brian Reynold’s Facebook game, will the guiding principle be increasing Zygna’s revenue or making the game more fulfilling? (link)

The Zynga guy said, you need to identify what people are doing most often in a game, because that’ll be the most fun activity. If that were true, the funnest activity in Starcraft is building Zerglings and the funnest in late-game Civ IV is clicking END TURN. (link)

Obviously, developers are wary of how Facebook gaming will change the industry in the years ahead. (Compare the importance of business metrics now with 1997’s Ultima Online, which lead designer Raph Koster points out “wasn’t designed around any business model in particular.”) The irony is that Facebook games typically share four characteristics that really do promise great things for both gamers and designers:

  • True friends list:  Gaming can now happen exclusively within the context of one’s actual friends. Multiplayer games no longer suffer from the Catch-22 of requiring friends to be fun while new players always start the game without friends.
  • Free-to-play business model:  New players need not shell out $60 to join the crowd. Consumers don’t like buying multiplayer games unless they know that their friends are all going to buy the game as well. Free-to-play removes that friction.
  • Persistent, asynchronous play:  Finding time to play with one’s real friends is difficult, especially for working, adult gamers. Asynchronous mechanics, however, let gamers play at their own pace and with their own friends, not strangers who happen to be online at the same time.
  • Metrics-based iteration:  Retail games are developed in a vacuum, with designers working by gut instinct. Further, games get only one launch, a single chance to succeed. Most developers would love, instead, to iterate quickly on genuine, live feedback.

These four pillars are the reason why many game developers are flocking to Facebook. (Of course, many of these characteristics are not exclusive to Facebook, but combining them together with such a large audience makes Facebook the obvious choice right now.) However, Jesse Schell is right; a war is brewing over who will call the shots. However, the question is not simply one of suits-vs-creatives. The question is will designers take the time to learn the business, to learn how to pay the bills while also delivering a fantastic game experience? As BioWare’s Ray Muzyka put it during a panel on connected gaming, ultimately all decisions are made with a goal to make money, but the goal may be short-term revenue (“can we sell more blue hats tomorrow?”) or long-term growth (“does our community believe in what we are doing? are we creating life-long fans?”). The successes will not come from open conflict between design and business but from developers who internalize the tension and attack the problem holistically.

I have to admit my own reservations about this transformation; game design itself simply might be not as much fun as it used to be. I cannot easily sum up how enjoyable brainstorming a game is during the early, heady days of blue skies and distant deadlines. With a release-early-and-iterate mentality, these days are now over, for good. Games will no longer be a manifestation of an individual’s (or a team’s) pure imagination and, instead, will grow out of the murky grey area between developers and players. The designer-as-auteur ideal is perhaps incompatible with this model, but I believe the best game designers are the ones willing to “get dirty” – to engage fully with a community to discover which ideas actually work and which ones were simply wishful thinking. Loss of control is never fun, but as Sid is fond of saying, the player should be the one having the fun, after all, not the designer.

Theme is Not Meaning: The Slides

My keynote on theme-vs- mechanics went well. I’m getting positive feedback so far although I am curious where people may have felt my arguments were not as strong. I’ve posted the slides in the sidebar, but here is a direct link. Also, the talk was written up a couple places online, with the Destructoid piece almost being a transcript:

My GDC Keynote (and Panel)

This year at GDC, I am giving one of the keynotes at the Serious Games Summit. Here’s the description:

Theme is Not Meaning
Speaker: Soren Johnson (Designer & Programmer, EA Maxis)
Date/Time: Tuesday (March 9, 2010)   11:15am — 12:15pm
Location (room): Room 133, North Hall
Experience Level: All
Summit: Serious Games Summit
Format: Summit Keynote

Session Description
A game’s theme does not determine its meaning. Instead, meaning emerges from a game’s rules – the set of decisions and consequences unique to each one. What does a game ask of the player? What does it punish, and what does it reward? What strategies and styles does the game encourage? Further, when theme and rules are in dissonance, the designer cheats the player while also compromising his or her original vision. This can often lead to significant failures in the serious games space. In this opening Serious Games Summit Keynote, game designer Soren Johnson explores the critical question of how to produce meaningful experiences that are uniquely tied to gameplay vs. the content wrapped around that play?

Readers of my Game Developer column will recognize that this talk is an offshoot of my recent set of columns on how a game’s theme and meaning need to work together or risk alienating the audience. Hopefully, the two formats (print and lecture) will support each other instead of being redundant. I’m looking forward to the Q&A as I expect quite a few people will differ from my conclusions.

I am also participating in a panel at the AI Summit on supporting game designers:

Answering the Designers’ AI Wish List
Speaker: Brett Laming (Lead Programmer, Rockstar Leeds), Richard Evans (Lead Simulation Engineer, Maxis), Soren Johnson (Designer & Programmer, EA Maxis), Chris Jurney (Senior Programmer, Double Fine Productions), Adam Russell (Games Studio Manager and Lecturer, University of Derby)
Date/Time: Wednesday (March 10, 2010)   5:15pm — 6:00pm
Location (room): Room 310, South Hall
Experience Level: All
Summit: AI Summit
Format: 45-minute Panel

Session Description
We asked designers from all across the industry to answer a questionnaire of probing – and even outright crazy questions. The intent was to get their heads and assemble a sort of wish list. We then present their answers to a panel of top-notch AI designers and programmers and ask them… how would you go about granting this wish? In what promises to be the most forward-looking session of the AI Summit, this panel should give us all a look into not only what the designers would like in their games, but some ideas on how to address the difficult obstacles in AI.

Idea Takeaway
An impression of how much AI technology can accomplish, an insight into the technical design process of experienced developers. This session should be kind of fun and wacky!

Hope to see everyone at GDC, my favorite week of the year!

So this is what Twitter is for…

A few hours ago, I twittered about Denis Dyack’s Develop talk on how narrative is going to overtake gameplay in importance:

SorenJohnson: Hey Denis, if you put the narrative in front of the gameplay, you are no longer making a game. You’re making a movie.

Harvey Smith responded to my tweet, challenging me on defining everything as either a game or a movie. Then Clint Hocking jumped in. Followed by Rob Fermier. Then Brenda Brathwaite suggested we start using the #gamedesign tag. (I had just started using #dyackrant, but too late…) Then Ian Bogost. And David Jaffe. And Damion Schubert. Even John Romero made an apperance. It felt like a virtual post-GDC Fairmont chat. The discussion is now spiraling out to larger and larger circles – just search for “#gamedesign” – and I’m not sure how long it will carry forward but it was an interesting experience. Twitter is quite poor at helping outside readers understand the context of most tweets, but as a pure social activity for members of the discussion, it is currently unrivalled on the Net.

Update:  in_orbit did some voodoo with the Twitter API to produce a threaded version: Hey, this Web thingy is pretty cool!