A Battle Lost Through Attrition

Soldier doll lying on ground with chalk outline.

Photo credit: Eisenbahner, flickr.com

This article originally appeared at Tea Leaves in October of 2010.

Hardcore operational-level computer war games have a serious problem: most of them are unplayable.

Not all war games, obviously. And "unplayable" means different things to different people. A game that crashes whenever you start it up, for example, is literally unplayable. I mean something softer, here: I mean that many games which should be brilliant, which should capture the hearts and minds of every war gamer on the planet are instead consigned to a dusty corner of the market by unacceptably primitive user interfaces, obtuse mechanics, and poor communication of results.

"War game", in the context of this article, means a combat simulation, turn-based or real-time, that models a hypothetical or historical conflict but that strives to achieve "realism" in addition to entertainment.

An "operational level" computer war game is a game that models conflict typically at a level between battalion, division, or corps in size, and where lines of supply are typically an issue. A game where individual soldiers or squads are maneuvering and setting up lanes of fire, such as Close Combat, is tactical in scale. In an operational game, you're moving entire divisions around the front.

What sparked my desire to discuss this is a trio of games that I've been evaluating lately. All three games are published by Matrix: Gary Grigsby's War in the East, Norm Kroger's The Operational Art of War III, and Advanced Tactics World War II. This article is not meant to be an in-depth review, but rather an exploration of some of the ideas that occurred to me while I played these games.

Advanced Tactics brought some of these thoughts to the surface. The game is brilliant, but here is the thing: the UI is so tragic that I needed a week off from work to discover that the game is brilliant.

This is a problem. Perhaps it is just a problem with me, but I doubt it. I have had more exposure to war games than your average computer game player. I think it's a problem with the game. A New Yorker interview with Shigeru Miyamoto, Nintendo's lead game designer, puts the issue squarely in the spotlight:

Earlier, Miyamoto, a bluegrass fanatic, had suggested that learning to play a game is like learning to play a musical instrument. “Take the guitar,” he said. “Some people, when they stumble over how to accurately place their fingers in an F chord, they actually give it up. But once you learn how to play an F chord you become more deeply absorbed in playing the guitar.” The F chord, as he sees it, is a kind of bridge between indifference and pleasure. “If the bridge is too easy to pass by, it’s called ‘entertainment.’ If it’s rather difficult, it can be called ‘hobby'.

If we continue the analogy, the war games I'm talking about are asking you to play an F chord but, for the best sound quality, would like you to put a new set of strings on the guitar every time you play it.

The knee-jerk Internet forum response is to claim that the UIs are this complex because the underlying subject matter is complex. This is false. The subject matter is complex, of course, but the failures of the operational-level games are not due to this particularly complexity. They're not even really game design failures. They are failures of software development. Let's get specific.

A Brief History Of Not Dating Girls

The operational games I'm describing derive their look and feel from the classic "hex and counter" war games of the 1970s, published by companies such as Avalon-Hill and SPI. Those games varied widely in complexity, but once you got past the fiddly aspect of having to manipulate lots of tiny counters, 95% of them boiled down to three simple rules:

  • Each counter has two numbers on it. One is the combat value, one is how far it can move.
  • Units (usually) have a "zone of control" next to them. If you move your counter next to an enemy counter, it will have to stop.
  • After you move all your counters, you resolve the attacks. Add up the number of points of all counters attacking and defending to come up with a ratio (2-to-1, or 5-to-1, or 1-to-2). That ratio will be a column on a "combat results table" printed on the map. Roll a die to pick which row on the table to use. Look up the row and column, and that tells you what happened in the battle.

These games were not without flaws — in particular, the "I move all my pieces, then you move all your pieces" turn order, while simple, meant that the non-moving player would have time to go out for lunch (and possibly dinner) while waiting for his turn to play. But most of these games, with some notable exceptions, tried hard to abstract out the details of combat for the purposes of playability.

At first blush, a computer implementation of this sort of game should only get better. After all, now the computer can tell me exactly where I can and can't move, and I no longer have to look up the results on a combat results table. In fact, you can play literal translations of these games at HexWar, and they are better.

Alternatively, instead of using the computer to just automate board game mechanics, developers can increase the complexity of those mechanics. In theory, this is a reasonable choice. When done properly, it makes the game deeper without making it more confusing. But often, the results are simply depressing.

Those Who Can't, Don't Teach

Every war game should have an in-game tutorial. Advanced Tactics, War in the East, and The Operational Art of War don't.

This, for me, is such a basic element of good game design that I'm constantly surprised when I see it violated. A game without a tutorial is like a movie without popcorn; like sex without foreplay; like a hamburger without French fries. A game without a tutorial is incomplete.

Sure, let me skip the tutorial if I'm sure I don't need it. But in a sufficiently complex game the tutorial can do so much heavy lifting that to skip it is madness. It can introduce you to the user interface. It can acquaint you with the basic controls of the game. It can allow you, as a software developer, to hide more advanced controls and still be confident that the user will be told where to find them.

All three of the games I'm looking at in this article do "Tutorial by manual". This means that they provide a "tutorial scenario" which you use by reading the manual and following along. This is inadequate on multiple levels. First, it means that you are told what to do instead of being shown what to do, which is less effective. Second, it opens the possibility that the player may go off-script. Going off-script is a vital part of learning how to play a game well, but is a big mistake during the early parts of a tutorial. Many games split the difference by tightly scripting the beginning few turns of a tutorial, and then at some point telling the player "OK, from here on out you are on your own." Lastly, I have never seen a manual-based tutorial that adequately helped focus the player's attention on what was important, as opposed to what is simply present.

Advanced Tactics also ships with several "training scenarios" which set up simplified battles. For example, the first battle is on a featureless plain, with identical numbers of troops, but with one side using armor and the other using infantry. The next scenario is the same, only this time in forested terrain and giving the infantry player bazookas. The third scenario introduces artillery, and so on. This set of teaching scenarios did more to teach me about the game than the manual-driven tutorial. In my fantasy world, these scenarios would have been presented as an in-game tutorial that also introduced me to the various UI elements I needed to use to (for example) perform artillery attacks. The tutorials would have explained the effect that being out of supply has on a unit's readiness. The tutorial would have explained how to generate political points and how to use them to create new headquarters, and how to transfer staff between them. As it is, I groped through the dark, eventually figuring all these things out.

Despite what the abused customers of ultra-grognard developers might tell you, figuring these things out by yourself does not make you a better person.

Information Wants To Stay The Hell Out Of My Way

Imagine this scenario. You are the Supreme Allied Commander for Allied forces in Europe during World War II. You walk into your office and your aide-de-camp says "Good morning, General Eisenhower. Your general staff awaits you in the conference rooom to discuss Operation Overlord."

"Excellent. I'll be right there."

"One moment, sir. Before that, you should be aware that Fox company of the 506th has run out of condoms in their survival kits."

"Uh, well, get them replacement kits."

"Very good sir. Also, a truck was destroyed in the Ardennes. Should I requisition a replacement?"

"Yes."

"I'll get right on it. A number of toothbrushes have gone missing in a training camp in North Carolina. How many staff sergeants would you like to assign to investigate the crime?"

At this point, in real life, Eisenhower would court-martial his aide-de-camp for being a Nazi spy.

There is important information, and there is unimportant information. Unimportant information should be completely invisible unless specifically asked for. If it becomes important, show it to me then. Some designers of operational games seem to be completely unable to decide how to prioritize the information they show you: large tables full of numbers, all in the same font size, are the order of the day.

This extends to user interface, too. Consider this, the "command bar" in Advanced Tactics (click to enlarge):

Advanced Tactics Gold

Considering just the items in the middle of the bar, we've got Move, Transfer resources, Strategic Transfer, separate buttons for various air and sea options, buttons to change the headquarters of the selected unit, buttons to display the supply overlay, and a number of others. All of these buttons are the same size. Buttons can sometimes be disabled, but are still present even when irrelevant (why even show me the naval subformation button in a scenario without ships or water?)

Clicking on an enemy unit gives a similar-but-different view of various options that might or might not be relevant to the current situation. Since the icons are fairly small targets, the only way to find out what they do (before you have memorized them) is to hover the mouse briefly over each of them. And then go read the manual.

Here's another example: this is the control panel that you use to control units in Norm Kroger's Operational Art of War III:
kroger

So in the world of TOAW, "Next unit" is as important as "resolve battle" which is as important as "show/hide hex grid" which is as important as the twenty-five other miscellaneous options, some of which you will never use. It is to weep.

Again, let's be clear: this interface is not hard to use because the problem space is hard. This interface is hard to use because it is poorly designed.

My contention is that it is poorly designed because of a feedback loop wherein the designers never test the games with new players, so new players never play the games, because they are unplayable, so the games are only tested by people whose entire attitude is "As long as it is easier than pushing 600 cardboard chits around a paper map, it's a massive improvement!, so the designers never realize the games are unplayable.

I am complaining about this not because I don't like these games, but because I do. I love them. And sometimes, when someone you love has a serious problem that is ruining them, you have to stage an intervention.

Finding Out What the Hell Just Happened is For the Weak

In the same way that input methods for these games are not constructed with usability in mind, resolution of critical events like combat is often more confusing than enlightening. This is actually an area where Advanced Tactics performs better than most: it provides you with a graphical display of combat that, round by round, moves soldiers and equipment from a "Attacking" or "Defending" box to a "Casualty" or "Retreating" box. At least, that's what you get if you attack. If you are attacked on your turn, you just have to notice that your men are missing (unless you use the ‘history replay’ option, which still doesn't give you a clear explanation.)

The worst of the three games, as far as this aspect is concerned, is Norm Kroger's Operational Art of War III where every single enemy's turn can be fairly described as "And then, over the course of about 30 seconds, 10 attacks are resolved and you are shown an incomprehensible screen of text." I'm sure there is some way to get a higher level summary of what happened on a given turn, but I'll be hornswoggled if I can figure it out. Here's a picture of a combat results screen from TOAW III:

The Operational Art of War III

Consider what's going on in this dialog box. Off the top of my head:

  • The font is stupidly small (and ugly) making it harder to read.
  • Compounding the mistake, the "newsprint background" makes it even harder to read
  • The text is center aligned, so you can't easily use visual pattern matching to find the parts of the report you care about.
  • In the laundry list of sentences telling you which units attacked, defended, or retreated, the word "attacks" "defends" and "retreats" are in no way highlighted
  • There is a freaking laundry list of sentences telling you which units attacked, defended, or retreated. That could be eliminated and replaced with one picture of unit counters
  • The only way to tell which unit in that laundry list belongs to which team is to read a word. A picture of the unit or even the judicious use of coloured text could do this more effectively
  • And of course the laundry list interleaves attackers and defenders willy-nilly, so you can't even use a heuristic like "Attackers listed first, defenders afterwards" to decode it.
  • Arguably the most important information in the combat results dialog — how much equipment and personnel were lost! — are tucked away at the bottom.

I could go on, but I've made my point.

The most recent of the three games, Gary Grigsby’s War in the East, is in some ways the most approachable, although the subject matter, Germany’s invasion of Russia along the Eastern Front, is inherently daunting. In places the UI is almost acceptable, with buttons that look like buttons, and movement and simple attacks controlled largely by pointing and right-clicking. In other places, such as the Order of Battle screen which controls strategic force creation and reassignment, it’s typically obtuse. Even the tutorial on the issue (manual based, of course, not in-game) left me more confused than when I started. But the game tries hard to let you ignore those screens completely if you just want to follow the historical path.

It Doesn't Have To Be This Way

So what's my point? Simply this: strategic depth does not require user interface complexity. Back in the bad old days of the 1980s, many sophisticated computer users believed that GUIs were "dumbed down" interfaces, and that command-line interfaces were a prerequisite for effective use of a computer. These people were wrong, and those who believe that an inscrutable UI is necessary for a sophisticated war game are wrong. Some developers are learning this lesson, and others are still in the dark.

The main effect of an inscrutable UI is to confine interest in the software to a small, shrinking minority of users. That minority is ridiculously loyal and desperate for content, but it’s a self-limiting market and, in my opinion, an evolutionary dead end.

Only developers who figure out how to make strategically deep games that are easy to use will avoid that end. Those developers, not the ones preaching to the choir, will define the very meaning of the words “war game” in the coming century.

Wouldn't you rather be one of those developers, instead?

Advanced Tactics: World War II, ($26.99), Norm Kroger's Operational Art of War III ($26.99), and Gary Grigsby's War in the East ($79.99) are all published by Matrix Games and are available for purchase and download on their web site.

Disclosure statement: Matrix Games graciously provided Tea Leaves with review copies of these games.