Saturday, June 11, 2005

Nintendogs: The case of the non-game that barked like a game

A game design review of an innovative game

My beautiful lass recently borrowed a copy of Nintendogs from a friend (Thank you Porter!). This title has been burning up the charts in Japan and managed to get a perfect score in Famitsu magazine, a feat matched by only 4 other games in history. Why is it so successful and what can we learn from it as game designers?

This is a game design review, not a game review. A game review typically is written for the consumer and is intended to help them decide if they want to purchase the title. A game design review is written for other game developers and is intended to highlight successes and failures of the various layers of the game design. The hope is that we learn from the successful design practices and apply them in an intelligent fashion to our future titles.

Not a game?
The first thing that arises whenever someone talks about Nintendogs is the claim that it is not a game. "Dude, it is just a stupid Tamagotchi clone with dogs!" Such a vehement response is indicative of a larger trend in both the designer and gaming community. Unfortunately, it is a trend that has its roots in a fundamental misunderstanding of game design.

To generalize, there are two camps of designers in the world

  • Craftsman designers: Craftsman designers look at existing titles on the market and build up these to create improved titles. Craftsman designers are always hardcore game players with intimate knowledge of their preferred genre. Craftsmen know details, but fail to see deeper patterns.
  • Theoretical designers: Theoretical designers looks for common elements across all game genres and builds their designs from these fundamental parts. Where a craftsman designer might say "My FPS (First Person Shooter) needs a double shot gun to replace the single shotgun", a theoretical designer might say "Player enjoyment is dropping at this point, so we need to add a new risk/reward schedule."

The craftsman definition of a game.
Each of these two groups looks at a title like Nintendogs in very different ways. A craftsman designer classifies a title as a game if it fits into a pre-existing category. Is it a RTS game, a FPS, an adventure game, etc?

  • Is the game title part of a pre-existing genre?
  • If so how does that title compare to my personal enjoyment of other games in that genre?

This method works well when you operate within well defined genres (as is the case with most hard core gamers.) It breaks down when a title doesn't fit into a pre-existing genre. They'll still apply the same analysis, but generally end up shoe horning the title into a genre that is a poor fit.

There is nothing on that market that compares to Nintendogs. If you dig into the game mechanics at an abstract level, it has surprisingly more in common with a RPG than most virtual pet games. Yet hardcore gamers make a snap judgment and instantly assume it must be a Tamagotchi-style game. This is an unfortunate mistake that limits our understanding of the game design.

The theoretical definition of a game
The other way of looking at it is to look for the key elements that make up any game.

  • Are there psychological risk / reward systems?
  • Are there overlapping reward cycles on different timescales?
  • Can the game design be classified into standard game design elements such as tokens, verbs and rules?
  • Can the various layers of the game design be separated out so that the title can be examined in terms of core mechanics, metamechanics, contextualized tokens, plot, etc?

Nintendogs has clear game mechanism in each of these areas. There are clearly specific elements throughout the game that match existing game system that have been used throughout the history of game design. The theoretical designer realizes that a powerup is a powerup whether you call it a 'Quad Damage' or a 'Doggy Brush'.

To get this point across, let's look at Nintendogs by identifying the various game play elements that make it such an addictive experience and then compare those to the same mechanics used in other genres.

Overall Structure
Nintendogs follows the path of many successful titles. You start out with a novice character and through a variety of challenges and adventures, you grow the character in strength and power. Along the way, you gather treasure, gain new abilities and discover far away places.

In Nintendogs, it just happens that your character is a puppy and the world you are exploring is the city around your house. The battle sequences where you gain experience are dogshows. The powerups you get a hair care products and doggy snacks. Strip away the setting and plot of Nintendogs and you are still left with an innovative title, but it is one that bears more than a passing resemblance in structure to many other games.

This big structure is composed of smaller elements

  • Core mechanics: Layers of risk / reward activities
  • Meta Mechanics: Additional activities that tie together groups of core game mechanics.

Core Mechanics: Learning tricks
The primary activity you do with your dog is tell it commands and pet it. The first level of risk / reward cycle goes something like this:

  • Action: Say a phrase in a clear concise manner.
  • Reward: The dog responds with an attractive animation.
  • Example: Say "Lucky!" and the dog comes up to you. Say "Sit" and the dog sits.
  • Comparisons to other titles: In a fighting game, you hit a button with specific timing and it kills an enemy. The enemy crumples to the ground in an attractive animation. For Nintendogs, voice is certainly a fun new control mechanism, but in many ways it is no different than hitting the attack button in the fighting game. From a game mechanics viewpoint, pressing a button that means 'sit' is identical to saying 'sit' and having a voice recognition sub-system return the value 'sit' to the game.

The second level of risk / reward cycle builds on the first.

  • Action: Once you have gotten the character to perform a 'success' animation, you can now rub your stylus on the dog.
  • Reward: The dog plays an additional success animation, and gains 'happiness'
  • Example: After the dog sits, you can pet the dog and he arches his back contentedly. Pet long enough a sparkle appears that states you have improved your dog's happiness.
  • Comparisons to other games: In our fighting game, the dead enemy drops a heart. You need to run up to the coin and collect it, thus improving your 'health'. The meter that records your dog's happiness and the meter that rewards your character's health serve nearly identical purposes despite their very different names.

The third level of risk / reward cycle adds an additional layer.

  • Action: Periodically, the dog will perform a new action. You can enter into a minigame in which you must say a particular phase associated with that action over and over again.
  • Reward: The dog plays an additional success animation, and gains a new ability that will let you take on additional metagame challenges.
  • Example: During petting the dog rolls over on it's back you are presented with the option to train it to do the 'rollover' command. You say 'rollover' repeatedly until the dog understand what you are saying. Music plays and your dog learns a new trick. As a higher level reward, this is immensely satisfying.
  • Comparisons to other games: In our fighting game, the character fights their way past a boss enemy (think Megaman). At the end battle, an animation plays and you gain the 'Ice attack." Woot!

Metagame: Competitions
Once you've gained a few skills, you can participate in dogshows. Think of this as a game of Simon. You must perform certain actions within certain time limits. If you succeed, you win money that can be spent on additional goodies and powerups.

There are a variety of other competitions ranging from agility contests to Frisbee tossing events. These events help put your skills to use and provide a clearly defined set of challenges for goal oriented players. The 'win challenge stage, get money' is a system found in most games. Just because it involves a puppy doesn't mean you aren't dealing with old school proven game mechanics.

Metagame: Walking your dog
Walking your dog is very similar to the overworld in a typical RPG. This how you discover new places such as training courses and special encounters.

One innovation that Nintendogs adds here is that the distance you can travel is dependent on the strength of your character. This creates a challenging mini-game. Given a limited amount of energy, what path exists through the city that lets your dog hit the maximum number of bonus points and special areas? This is a minor variation on the Traveling Salesman problem that most computer science students run into in their undergraduate courses. There is no easy solution to this class of problem, which makes it a constant challenge.

The special points on the map are randomly generated so the player is required to come up with new routes each time. This helps prevent burn out. Also, there is an explicit 30 minute timer in place that reduces the rewards if you play this section too often.

Other game systems
There are numerous other common systems involved in Nintendogs.

  • Random reward schedules: When your dog brings you a present, it can be something good or something useless. By having intermittent reinforcement, the designer maintains the enjoyable addictive nature of the reward for a longer period of time.
  • Sandbox Levels: The interior area that you start out in acts as a sandbox area for introducing new actions (like the Frisbee). When the goal oriented challenges appear, the player is already trained in the basic use of their character's abilities.
  • Time penalties: When you don't play for very long, your character loses some of it's powerups. The dog gets dirty, hungry and thirsty.

Everywhere you look are standard game design techniques. The initial structure of the game, how skill are introduced and then used in challenges, the entire broader meta-game that drives the player forward...all these can be found in great games going back decades. Nintendogs is a good game because it has a solid game design, not just because older Japanese women like dogs.

However, the reason Nintendogs is a great game is because older Japanese women like dogs. Nintendogs is also a glorious example of game anthropology at work.

The game anthropology of Nintendogs
A solid game design is one piece of a successful game. However, a successful break out design needs to also identify an unmet need in the society at large.

Game Anthropology: Game anthropology is about watching how ordinary consumers go about their lives; what sort of things do they do, what do they want to do, how do they use the things they have? Amidst all this, what opportunities exist to play games?

It turns out that Japanese folks love dogs, but due to their cramped apartments they are rarely able to own them. Every child dreams about owning a dog, yet very few ever will. There are actually dog renting services that let you walk a dog just for a little while so you can capture the thrill of pet ownership.

The designers of Nintendogs recognized that if you can build a game that lets people experience the joys of owning a dog you can tap into a market rarely touched by traditional games. This is a very different thought process than "Let's make a game like X except better". Instead, it requires game designers to go out in the field and understand how games an be applied to broader trends in their culture. The designer of Nintendogs asked the simple question "How do games apply to the world outside of me?"

Such big picture thinking that is utterly alien to people who think of games as the sole domain of elite geeky guys. The good news is that people who successfully apply the techniques of game anthropology are rewarded beyond their wildest dreams.

Conclusion
Nintendogs is a game that get two key elements right:

  • It addresses a niche need within the broader culture that is highly underserved.
  • It understands game design theory well enough to build an original new game experience out of proven game design techniques.

Kudos to the theoretical developers behind Nintendogs and their mastery of the game design process. Anyone who calls themselves a game designer should pick up the title when it comes out in English. It is a perfect case study of how to build a highly innovative game that achieves impressive market success.

Ignore the fools who claim that it isn't a game and dismiss it's success as a random chance. The myth that game designers must be derivative to be successful is fundamentally false. Creating successful innovative games like Nintendogs is wholely a matter of hard work and a deep understanding of the game design process. The phenomenon of Nintendogs is completely reproducible if we heed its design lessons.

take care
Danc.

PS: There is one poor design choice in Nintendogs that I wanted to call out. Time penalties for not playing are a horrible way to encourage people to start playing. The person may feels obligated to play but games should not be a guilty activity. I much prefer systems like those found in World of Warcraft that give the player bonuses if they return after not playing for a while. The player feels "I haven't played in a while, but if I play now I'll get something good!"

Time penalties can backfire on the designer. Often, the player assumes that their work has decayed so far (the dog has forgotten words, etc) that it is simply not worth playing again. The player who has fallen off the wagon feels "I am so far behind now, why try?"