keekerdc.com

essays and pithy thoughts

Statecraft: the simultaneous turn paradigm of 'Diplomacy'

in Statecraft

There’s a board game that goes by the name Diplomacy that I studied a fair bit a while back; unfortunately I’ve never been able to assemble enough like-minded people for a game, and playing online seemed a rather dull and slow way of going about it. In any case, the reason that it gripped me as interesting was that it took the global conquest aspect of Risk and did away with the dice rolls, replacing it instead with a sumo wrestling sort of combat mechanic, one which relied solely on your ability to bamboozle your opponents in order to be successful.

In simple terms, it goes something like this: All players are fighting for control of a fixed number of supply depots. Each supply depot you gain allows you to put another unit on the board. Units can either be ordered to stand pat, move into an adjacent territory (and possibly try to ‘dislodge’ any other units that might already be there), or support the action of another unit trying to do something in a neighboring territory. The twist is – instead of each player having an individual, isolated turn all to themselves, where they move their stuff and nobody else is acting – everyone plays each turn at the exact same time. A turn consists of a fixed amount of time for discussion, planning, and calculated misdirection, after which everyone submits orders for their units, which are all reconciled against each other according to a relatively straightforward process.

Unit A has two supporters for an order to move into a territory, while Unit B only has one supporter for the same territory? Unit A gets it. It’s through manipulating other players into sharing their plans, spreading false plans of your own, asking for favors to support moves that you have no intention of making, and stringing people along as much as possible can you get the other players to unwittingly assist you into winning the game.

Applying it

In applying this paradigm to a Civ-like strategy game, the combat system itself can be left to Diplomacy. What I’d like to drag along is the simultaneous turn mechanic.

Multiplayer in Civ has always been an awkward proposition, and it’s something I hadn’t messed with all that much. Turning on the ‘simultaneous turns’ option makes a deliberately paced strategy game into something that feels more like an RTS with someone constantly jamming on the pause button. It creates the possibility for two types of scenarios that have no place in a turn-based-strategy game: instances where a player that can get their command into the server a split second earlier gets an advantage, and situations where a player is waiting on an opponent to make their moves before making their own. Conversely, playing multiplayer without the ‘simultaneous turns’ option meant a game could drag on for days, and without reliable means to let it do just that while letting people play at their own pace, the only real option for this are LAN-type situations where all participants are going to be in the same house for a weekend.

Adopting just the turn structure of Diplomacy, in which players issue orders for what they control for a single turn, after which they’re reconciled against each other, and the game board is updated with the new state for the start of the following turn, offers attractive advantages over a cascading progression of individual player turns.

The root benefit this brings is the elimination of the advantage derived from getting one of the earlier slots in a turn cycle. In a game where a single tile can be the difference between landing that crucial settlement site and being screwed the rest of the game, it shouldn’t come down to random chance that one player was able to arrive just in time to squeeze out an opponent because they’re ahead in the turn cycle. This dribbles down into the combat dynamic, and can actually make the one-unit-per-tile rule make a modicum of sense.

It also allows the ‘simultaneous turn’ multiplayer dynamic make sense both practically for players as well as within the gameplay. This is actually the only mode I plan to run all games in, allowing for any mix of human players and AI players in any game without having to bend the turn framework in order to accomodate them.

Finally, it fits perfectly with the cloud-based paradigm I’m adopting for this game, where all games are stored on the server, and all communications between the game client and the game server are done through a REST API. By deliberately constraining myself to a game environment where nothing can be done in real-time, I think I’ll end up building a much more solid game.