Tea Leaves

I’ve been playing a lot of Advance Wars lately. It is a perfect little gem of a game, and I’d like to use it to make some points about good game design.

Good game design increases richness, but eliminates complexity. Good game design emphasizes content over form. And, all things being equal, good game design favors mainstream technology over the cutting edge.

I actually travelled back in a time a bit, and played the very first of the series, Famicom Wars for the original Nintendo Entertainment System. It’s an instructive comparison. Since that game was introduced in the late 1980s the game mechanics have not changed at all. Nearly everything about the game, except the graphics, is the same. And yet the Advanced Wars games are noticeably more fun than Famicom Wars. Why?

The answer is simple: the latter games are better because they became simpler.

Richness is Good, Complexity is Bad

If you’ve ever been involved in designing and implementing software, you’ve participated in the following conversation. Two engineers will disagree about some aspect of how the product should work. Implementing either behavior is easy. Both engineers will present their arguments, and are convinced of the rightness of their position. Neither can convince the other. Eventually — perhaps to avoid further conflict, perhaps just to get the issue behind them — someone will suggest “I’ve got an idea. How about we just put a knob on the product, and let the user decide which behavior they get.”

This is almost always the wrong decision. But often, if all of the engineers are young and inexperienced, and there’s no one around to issue a smackdown, knobs like this make it into the final product. The product is made uglier and more complex, the end user is presented with choices that she doesn’t care about, and you sell fewer units.

A game is rich when it presents you with a lot of interesting choices to make. A game is complex when it presents you with choices that you don’t care about, or when the mechanics of making those choices are intrusive. When designing a game, it’s apparently easy to confuse complexity with richness. Richness is a valuable trait in a game. Complexity is the enemy of fun. The ideal game provides a maximum amount of richness with a minimum amount of complexity. Ernest Adams has written about this in some detail.

My favorite recent example of complexity in action would be a game I recently reviewed, Commandos 2. Let’s review my description of the UI from that game:

For example, A readies an attack, unless you want to use your fists, in which case you press Q. In the example I gave earlier, the Sapper would use the I key to cut the wires, D to detect mines, click on the mines to retrieve them, and then P to place mines. If the Sapper wanted to place a satchel charge, he uses the B key, but to throw a grenade, he’d hit G (but if his friend the Driver wanted to throw a molotov cocktail, he’d hit S).

It’s clear that when develping this title, a large number of tragiocomically poor UI decisions were made. This is a common syndrome in PC games; my working theory is that the developers assign keys to functions early in the production cycle. Then, due to time pressure or thoughtlessness, they never return to ask the question “How should these choices be presented to the user to make the game playable?”

Compare this to the UI in Advance Wars. In that game, to attack with a tank you hit the “A” button. To assault a city with infantry, however, you hit the “A” button. To launch an air assault against enemy bombers, you hit the “A” button, but contrariwise to use your own bomber to bombard enemy troops, you hit the “A” button. The odd man out here is shelling with artillery. I always keep a cheat sheet handy for that one; you have to remember to hit the “A” button.

One reason that Advance Wars has a better UI is that the constraints of the console require it: there’s simply a limit to the amount of user interface inanity one can have when you’ve got only 4 buttons, a joypad, and a few triggers. This isn’t to say that all console games are better than all PC games, or that PC games can’t have a good UI (in fact, the core mechanics of Panzer General are very similar to those of Advance Wars). It’s just that since PC-based games can have needlessly baroque UIs, the realities of schedules and software development tell us that they sometimes will.

The take-away lesson for software developers, I think, is this: in terms of user interaction design, unless you have a damn good reason to do otherwise, design your game as if it will be targeted at a Gameboy Advance. Protect richness, but destroy, annihilate, and eliminate complexity as if you are Genghis Khan. You provide richness in a game by increasing the number of interesting choices the player has to make. Complexity, on the other hand, is created by making it harder to make those choices, or by hiding the interesting choices in a sea of boring ones. Richness is a virtue. Complexity is just retarded.

Content Trumps Form

Returning to Advance Wars for a moment, the difference between it and its predecessors is clear: the maps are smaller. The battles are smaller. Concepts are introduced a few at a time. Yes, there are a few “systemic” improvements, such as the introduction of fog of war, but those improvements aren’t used on every map.

In other words, the main difference between the earlier and later games is not that the latter games have better mechanics, but that the latter games have better content.

The importance of this can’t be overstated. Let’s look at one of my favorite genres, the RPG, specifically the Baldur’s Gate series. When Bioware (and later Black Isle) poured time and energy into improving the content, they ended up with Baldur’s Gate II and Planescape: Torment, two of the most emotionally engaging and memorable RPGs of their era. When Bioware poured the same amount of time and energy into the “system,” they ended up with Neverwinter Nights, which for all of its bells and whistles is fundamentally, a scriptable multiplayer 3D mechanism for delivering hollow and empty experiences to players. There are things I like about Neverwinter. But an hour of Torment is better than 100 hours of Neverwinter Nights.

Conversely, the NWN engine only became usable with Knights of the Old Republic, a game that focused on content to the exclusion of everything else, and which took much of the user-visible flexibility of the NWN engine and tossed it out the window, strictly limiting what the user could do (for example, restricting the extent to which the user could reposition the camera).

We can see the same dichotomy in the Diablo family of games. Blizzard delivered Diablo II, which was more or less the exact same game as the first Diablo but with more plot and more items, and created one of the best selling games of all time. Microsoft delivered Dungeon Siege, emphasizing all of the improved features of their engine, and nobody — except, ironically, modders — cared.

There are groups of users — hardcore gamers, and modders — who care about the details of your game engine. But the average player just wants to be entertained. It’s great that I can buy a DVD with a director’s commentary on it talking about what camera, film stock, and lighting was used to make a certain shot. But we’d think any studio that put a discussion of that on a poster was unwise. If you find that you can’t explain to the public why your game is worth playing without resorting to talking about the underlying technology, then you have lost.

The Cutting Edge is for Bleeders

Diablo II was the top-selling game of its era. When you look at the other games that were released in 2000 that didn’t make enough money for their creators to roll around naked on huge piles of gold, an interesting observation can be made: from a purely technological standpoint, Diablo II was primitive. No true 3D rendering, for example.

The fact that Diablo was “primitive” and the fact that it sold 8 hojillion copies are related: it meant that anyone, even people with a somewhat out-of-date computer, could buy it and play it. Furthermore, Diablo II had longevity: as people realized that it played well, it continued to sell like hotcakes over the life of the product. This is atypical; your typical “best seller” game, like a hot movie, sells well when it is released and then declines over time. If you, as a game designer, are absolutely convinced that you have to design a turn-based strategy game that requires an eight-CPU system with a thousand-dollar videocard to play well, I can’t stop you. But the lesson of Diablo II shouldn’t be missed.

Sometimes, it’s true, being on the cutting edge can make the difference between selling a game and not selling it; World of Warcraft is a great example of a game where the content and technology combine to create something greater than the sum of its parts. But the cutting edge, as they say, is also the bleeding edge, and being on it has two important side-effects: it reduces the size of the market you are selling to, and it increases the risk that you will ship garbage, or not ship at all. If your game doesn’t require cutting edge features, then you’re assuming risk in your schedule for an uncertain payoff.

As a hardcore PC gamer, I have a soft spot in my heart for the bleeding edge. Many of us remember the first time we saw Doom, and we instinctively want to root for companies that push the edge of the envelope. But for every game that pushes the envelope and succeeds, there are 10 painful, timewasting flops. Don’t be one of those: unless you have an absolutely compelling product requirement to not run well on 90% of your customers’ machines, use mature, well-tested technology for your game.

The Payoff

These are what I see as the main problems facing modern PC games: first, too much complexity, particularly in user interfaces. Next, there is too much design focus on form, and not enough on content. Lastly, games often use cutting edge technology for no sensible reason. In discussing them, I’ve sometimes phrased the payoff for developers and publishers in terms of sales. But that’s just a crude approximation of the real issue: I’m a PC gamer. I want to play your games. I can’t play your games if they suck. There is too much competition from too many other forms of entertainment for me to waste my life downloading the latest “Detonator” drivers to try to eke out another 1.5% framerate improvement in your product. I want to spend my brain power figuring out how to circumvallate Vercingetorix at Alesia, not trying to understand why in the world the developer thought I needed to use one keyboard command to move my Hastatii and a completely different one to move my Principes.

One final note, in closing. I’ve been using Advance Wars as my example throughout because it is a Game Boy Advance game. This handheld console, which has sold about 100 million units, is essentially a repackaged Super Nintendo system. In other words, the best turn-based strategy game on the market was created with technology that dates from 1990. Keep that in mind when sitting down to create the product requirements for your next game.

Not every game is a strategy game, but every game has to maintain a balance between richness and complexity. If your game isn’t as good as it should be, maybe you shouldn’t be asking yourself what you can add.

Maybe you should be asking yourself what you can cut.

Thanks to Peter Su and Nat Lanza for their suggestions on this article. The title was inspired by Donald Norman’s book The Design of Everyday Things.