Text Dump: Maneuverability, StarCraft, and human override

Post-heavy week this week- Unit Operations excerpts tomorrow and Wednesday, and then if I can clean them up, maybe some more coherent posts at the end of the week.

Below, some loosely related threads that are incubating.

Posts like these are an opportunity to share some recent thoughts/readings and to use bits as fodder for later exploration as I decide what it is that I’m trying to say to begin with.

 

I. Pants: Mobility and Obscurity

II. Seeing like a StarCraft Player (Or, Macromanagement): On reading other player’s intentions and managing resources

III. Human Override (Micromanagement): Turning off default autonomous behaviors, and occasionally destroying meaning, to game the system. (More StarCraft).

IV. The Elder Game: Growing meta-games and adding meaning to the system.

V. More on Anti-System Parties: Fascism, Spengler, and the Archdruid (again). I need to relax on the Archdruid sharing.

 

I. Pants

Long ago (in Fogbanking time- literally one of my first few posts), I wrote a post where I talked about pants as a barbarian technology, adopted by settled civilizations as they were forced to adopt horse-riding out of military necessity.

More “innovation in pants tech”: I’m not certain that it’s true, but it has been said that a Samurai wore hakama pants to obscure their leg movement without sacrificing leg mobility. Having many options, and having your desire (in this case, to move any particular way) hidden from view, can give you an advantage in adversarial interactions. I’ve always been fascinated by this kind of thought- it doesn’t have to do with speed or strength so much as with enabling high maneuverability and reducing the enemy’s response time.

ooda_diagram1

OODA loop: In a two-player adversarial game the advantage goes to the player who can operate within the loop of the enemy- that is, the player who can take stock of the situation and act to change the situation before the enemy reaches the decision phase, effectively forcing him to either react prematurely or reassess the situation and lose control to affect events intelligently.
Image credit: http://strategy4sustainability.wordpress.com/2006/06/21/ooda-cycles/

 

Was the pants segue a bit hackneyed? Yes, maybe. That’s fair.

 

II. Seeing like a StarCraft Player (Macromanagement)

Personally, when I think of the importance of visibility to the enemy, I think of StarCraft II. What a beautiful game. Wish I were any good.

I’m about to geek out pretty hard.

StarCraft II is a real-time strategy game featuring three races with different play-styles. The objective is to destroy the enemy by gathering resources and raising an army. The smartest way to achieve victory is to hit the opponent’s supply line (i.e. her mineral gathering operation) and reduce her capacity for growth.

In the early game, resources are almost entirely reinvested into building new workers to increase the rate of resource extraction. Soon, players must balance this need to reinvest for future resource extraction with the reality that raising an army redirects time and resources.

All good players scout [that is, use a unit to explore the map]. At the lowest skill level, scouting grants the player information about the location of the enemy, the nature of possible paths to the enemy, and the enemy’s defenses. But the smart player is also making a beeline to a particular feature of any player’s base in the opening three minutes of plat: the vespene geyser.

All players, when they start, are located near two resources in predictable arrangements: mineral (the basic resource, the only resource you need to build workers, and the resource that is spent the most) and “vespene geysers” (henceforth, “gas”). Unlike minerals, which can be mined by any worker unit at any time, gas must be “developed” by constructing a special building on them before workers can collect any. Gas development presents an opportunity cost of potential mineral collection and workforce growth. More expensive, specialized military units require some gas to construct. If my scout sees that no gas is being developed in the early game, then in my mind certain branches of the enemy’s technology tree are closed off. If my scout sees that gas is being developed, then I know that the player necessarily has a mineral disadvantage relative to anyone who isn’t developing gas.

A good player knows his options (his own technology tree), and he has a feel for what the enemy is pursuing (via their resource behavior + knowledge of their technology tree). This player should have a feel for which units would trump which in a match-up, and ought to adjust his own build [units being developed] accordingly. Switching builds can be costly in time and resources, if there are building construction requirements.

Some branches in the technology tree are extremely high-risk high-reward and can only be effective if their development is hidden from the enemy. The building required to generate invisible units are very expensive, and the invisible units themselves are very weak and expensive, but they are literally un-counterable unless the enemy invests in moderately expensive detection infrastructure. Same issue with flying units, which are generally very expensive but can ignore terrain (and various non-shooting military units). Flying units are precious and can become fodder for other kinds of expensive anti-air infrastructure (or the paper-scissors-rock dynamic of other kinds of flyers).

An amateur might be fooled into thinking that it might be useful to hold back and stock up on resources while determining a plan, and then to deploy resources when a plan solidifies. In fact, one of my amateur problems is that once my attention is spread across a wide variety of tasks, it becomes hard to spend resources quickly enough. Unspent resources are wasted time. Having too many resources is better than having none, but just barely. Turning resources into anything takes time and attention, and buildings have queues and limits. The only resources worth having are resources that are moving: from the ground to your inventory, from your inventory into actualized units, buildings, and upgrades.

The accumulation problem goes even farther. Unused units are wasted units. Unused buildings are wasted buildings. These are bottlenecks that separate good players from bad ones. A good player (not even a great one, I seriously mean “a player who is not a joke”) has all of his buildings constantly producing units. Most reliable builds depend on the interaction between two or three complementary unit types, and you’ll need many. As the resource collection rate grows, the player scales up by either putting down unique infrastructure for higher-tier units, or putting down more of the low-tier buildings to increase the flow of low-tier units (or building a new base at another set of resources). Workers should always be working, and fighting units accumulating in your base do you less and less good- there are population limits, there are other players identifying and countering your idle fighters- there are other players are looking to expand into new bases as their local natural resources begin to dry up.

Finally, unspent attention is wasted resources.

 

III. Human Override- (Micromanagement)

Without your explicit attention, units are at the mercy of their natural pathing and default behaviors, which are intelligent enough to function but are perfectable with human attention.

Many aspects of good “micro” involve using twitch-action and rules awareness to push the limits of a unit’s capabilities beyond what their default behaviors suggest. For example, the marine unit’s actual attack is shorter than its attack animation, so you can command the unit to move after delivering the attack, cutting the animation off, and then command the unit to attack again immediately, and in doing so you can increase the marine’s rate of attack. A committed player can take attack ranges, attack speeds, unit speeds, splash damage, and terrain into account to drastically outperform default-behavior armies of similar strength.

Units attack with the same strength no matter how damaged they are, but the game AI does not try to game this aspect of the rules, either. The micromanaging player can focus fire on individual enemy units to remove them and their attack strength from the battle sooner, and the micromanaging player might draw back damaged units and move fresh units forward to absorb enemy fire while the damaged units, no longer targeted, can return to attack anew.

Of course, this kind of behavior takes a great deal of attention. It is difficult to game these limits while maintaining your Macro (the resource-allocation and development I mentioned before).

In other games, where the fiction is meant to be stronger than the rules, this kind of behavior typifies the “Munchkin”, one who focuses on the numbers at the expense of the illusion of the game (ex. spamming cheap moves, denying the marines their stylish “firing weapon” animation). The relationship between the on-screen objects and the fictional beings they mean to represent is threatened by this class of play. In games where Agon is stronger than Mimicry, such as StarCraft II, the fiction may be only a convenient teaching device for a complex rules system, and this is simply high-level play.

 

IV. The Elder Game

I’m plagiarizing myself from Facebook about four months ago:

There is a concept in massively multiplayer online games called the “Elder Game”. The Elder Game initiates when players have advanced through all of the linear game content and have mastered the game mechanics sufficiently. Many early Elder Games were created accidentally as a result of arms races between advanced players in PvP environments, or as a result of supposedly impossible quests that can preoccupy creative teams. Elder Games are sort of cyclical- you aren’t permanently advancing any further by game metrics necessarily, but you might by social metrics (you have useless but ultra-rare items, for instance- or your rank is up at the perilous #1 spot). Elder players run clans and make up meta-games. They invent cheap strategies to dominate games, and invent counter-strategies. They build wikis. Good elder players are talked about after they’re gone.

Some MMO games feed themselves pretty easily by fostering communities that generate these sort of elder-game opportunities by trading tricks. Other such games die, because they’re too restrictive and there’s nothing to do once you’ve hit the ceiling.

Game designers are particularly interested in the Elder Game in subscription-based MMOs for obvious reasons.

 

V. Anti-System Parties

In a recent mulling post, I was expressing interest in the idea of the “anti-system party” and its apparent predictability in national politics. The Archdruid Report has been running a fascinating series on Fascism, its policies and its relationship to the few politics that we can now recognize in our Cold War bipolar lens. It smells strongly of Spengler’s analysis of the inevitable return of Caesarism and the public want for blood to break the stranglehold of money that characterizes a late-era Civilization in her death throes.