The Discreteness of Video Games (and sarcasm)

Video games always and fundamentally operate in discrete, absolutist terms. At their cores, they are quantifiable abstractions of real-world systems, or else systems that are more or less intuitively comprehensible to someone. In a video game, something either is one thing, or else it is another thing. There are no in-betweens.

In Mario Bros., Mario is Mario. He can walk, run, and jump, and occasionally shoot fireballs. Sometimes he is big (if he ate a mushroom). Sometimes he is small. But he cannot skip, or walk swiftly, or crawl, or be of medium height. A programmer (with the help of an artist) make a Mario that can walk quickly and crawl. But those are two additional behaviors. Mario would crawl at the same rate. This too you could program, but then the game might not properly express the feelings of desperation, or cunning, or whatever, that Mario feels as he is crawling–whether he is crawling because he is injured, or because he is sneaking up on Bowser or… This hypothetical Mario game could have as many layers of complexity as there is hard drive space in the universe, and still it would be dealing in discrete units.

That said, games have gotten very good at dealing with complex simulations. Physics engines can mimic physics reasonably well: flying objects bounce, dead soldiers go ragdoll. But in-game physics is always approximation of real-world physics. The trick to a good physics system is to be very clever with your smoke and mirrors as you establish illusionary verisimilitude.

I was speaking to a friend recently about one day making a game about 19th-century African colonialism. It would be a game of grabbing land before other European powers do, and surrounding unoccupied territory, like in Go. But it would also be a game of very hazy borders, of establishing outposts far inland and then really having no idea what sort of control you’ve obtained over the area, whether some other European is controlling it, or what any of the political and economic implications of that could possibly mean. It would be a game of strategy, but also a game of confusion, a kind of confusion that you would never be able to ‘fix’ and probably often wouldn’t be able to ‘just deal with’. This is so far a very difficult system to imagine in video-game terms, since it would be entirely about very hazy, very long-term dynamics, whose effects you wouldn’t know then, and hundreds of real-world years later still might not know–in fact, so hazy that there wouldn’t be one true objective thing to ‘know’.

So… like history, then.

History, the way it was taught to me and the way I continue to understand it, fits very poorly with models and frameworks (and video game systems)–not so much as a square peg in a round hole, but rather it’s like the metaphysical impossibility of trying to shove the human condition into that round hole. It can, however, be fun and interesting to learn (at least for me), and in that respect it does align with the second of video games’ two basic properties (the first being discrete systems). To make this game about Africa, I probably will have to pick one, or several, of the relevant dynamics, pretend it’s a discrete process with neat little building blocks, make it really fun to play while being informative, while still allowing the player to be aware that I’m taking some gross liberties with the way the world works.

In my current game, Neocolonialism, there are eleven discrete regions, each with discrete extractive or processing facilities. The game is not, then, intended to be a simulator. I think the simulation is a very silly genre, for all of the reasons stated above; it aims to re-create reality in a medium that is fundamentally incapable of just that (though they can often be hilarious). To that end, I have designed Neocolonialism to make some pretty significant approximations of what in reality is an exceedingly complicated system. Each region, for example, has only ten sovereign bonds, which players can buy or sell. It is a simple system that in no way reflects how sovereign bonds function at a technical, legal level, but it does reflect how they are wielded by finance capitalists as a kind of tool or weapon against their peers (and also the rest of the world).

My hope is to make a game so wide-spanning and abstract that issues of discreteness are less relevant. I could have, for example, tried to model individual cities and mining centers across the entire world. Instead, I have 2-5 such facilities per region. All of Africa may have two Factories and one Mine, for example. When zoomed out to that level, the nuances of global economics wouldn’t affect you, the capitalist’s, decision-making much anyway. So goes my line of thinking.

More than that, though, is that I find the abstractions I have decided to include to be humorous. The game is very sarcastic, at least for me–it is more of a satire than a straightforward ‘serious game’. I think the way bonds translate directly into parliamentary votes is hilarious. The effectiveness of my game as a form of teaching of political expression will depend entirely whether players ‘get the message’, despite the fact that video games can only approximate any ‘message’. Through sarcasm, players will hopefully realize that the processes as presented in-game are not Cold Hard Truths but rather are Amusing And Yet Informative Simplifications.


Assets Table

I wrote a few weeks ago about the difficulty in providing a meaningful visualisation of Neocolonialism. My game is notable (though this is probably true for others as well) that it could be played with a series of complex Excel spreadsheets. Having done it this way as the initial prototype, I can confirm that the process is more tedious and painful than it sounds. (My friend and I agreed, though, that it was the most ‘corporate’ way possible to play the game.)

That said, spreadsheets are necessary for the games, or at least one spreadsheet. There is just too many numbers floating around, I think, that having a table to hold them all is a strategic asset. I personally hate looking at that many numbers in a game, but this IS a game about finance capitalism, so it’s more a matter of choosing the least of evils.

From time immemorial (i.e. since I started working on this in late September-early October), the game has had an Assets window. It’s generally been very ugly. This is how it looked until yesterday:

Oh man, this looks so bad...

Yuck.  Having a table of all this information probably is very useful–this could be a way to see at a glance who is making more money and why–but THAT table is totally unreadable. Worse yet, I never tried updating it because I had written the code in October, without documenting it much. And now it’s February, and I can’t make sense of what the hell I was thinking.

So, this weekend, I rewrote the assets window. Now, it looks like this:

Much, much more readable. The highlighting of cells and coloring of cells makes much more sense. Plus, although it’s not obvious from this post, this table is much easier to modify than the previous version, and formats text smarter. Phew. What a relief.

I’m still not totally satisfied with it though.

The chosen aesthetic is meant to look like Excel. Like I said above, when the game was in its Excel-spreadsheet-embryonic state, it actually had a distinct corporate feel to it, and I’m trying to replicate that here.

There are two foreseeable problems with this plan. First, the new assets window clashes with the general game aesthetic:

And the second, related problem is that I don’t know how many people will think “Oh, it looks like Excel!” and how many people will think, “Oh, it’s a rough, crappy-looking piece of the game whose aesthetic was never ironed out!”

So, as an experiment, I inverted the colors:

Now, while this has the game’s coloring scheme, I personally think it is harder to read, so I might just stay with the black text-on-white background look. I also think that this one is clearly an in-game tool. No one would ever make a table with that scheme unless it was for a game. The black-background assets window might in fact make your gameplay experience less immersive, and might on top of that be less useful than the white-background assets window.

The next step might be to fiddle with minor visual details in the white-background color scheme, in order to make it look more like Excel. The gray cells, for example, may need a slight gradient effect. Maybe there needs to be menu buttons that are basically non-functional but look pretty. The problem will be that I’m starting to discover an uncanny valley effect exists in graphic design. The current design is extremely straightforward, and if it actually looked like Excel, that would be awesome too, but if it only looks somewhat like Excel, it’ll just look very wrong. And I’m no graphic designer, though I can get by, evidently.

So the actual next step would be for other people to comment on the current look, and then take it from there.

Other news: The game is almost in a playable state, with a few lingering but important bugs (like that it doesn’t check to see if two players have the same name… but it really, really should). I’ve also been editing the manual extensively, and I will probably post it here soon as a pdf.


Cuban Africa

I recently played Victoria II, a Paradox game that’s a lot like other Paradox games, only even less fun. It’s a grand-strategy game that attempts to simulate world history from 1836 to 1935. In practice, there isn’t much to do, and even less that is comprehensible.

What Victoria II does have is one pretty map. There are many other maps, but they are all ugly and useless, such as the one displaying administrative efficiency (I don’t know what that means), one displaying railroad construction (they get built so fast I tend to not need to know where they’re getting built), a terrain map (I don’t think this one ever affect my decision-making, even though it’s the default), and more. But there is one good map, and it’s the world map:

Wow. Look at that map. Look at that text wrapping around that land. Wow.

As far as I can tell, the main reason to play Victoria II is to generate screenshots like that one, i.e. to create giant ahistorical labels on a map. People online apparently frequently seize extra territory just so that the lettering formats better.

That bit of whimsy aside, Victoria II does do a good job of giving a cursory depiction of colonialism.

Classic colonialism was the process of extracting resources from somewhere and sending them back to a central somewhere else (usually in Europe or North America). This almost always involved the overt seizure and nationalisation of foreign territory. The “Dutch East Indies” and “New England” and, in this case, “Cuban Africa”, were all clients of a parasitic relationship wherein there was one overall host–the Netherlands, England, or Cuba–that,  far more than any other nation-state or governing body, controlled the extraction of profit. Colonialism was a question of which government of a  nation-state or shadow government of a nation-state would reap the benefits, and land and empire were the fundamental metrics used.

Victoria II does a reasonable job of depicting this. As the president/autocrat of Cuba, I and I alone control Cuban Africa. Maybe someone will conquer it, or maybe a wave of French worker will migrate to there, but it is still my land. I get to make policy decisions in accordance to my economic needs. That screenshot is not my save file, but if it was, I would probably be focusing my attention on expansion, in an attempt to seize the mineral assets of Congo, south of the current administrative borders (the actual process of colonization in this game is, sadly, abstracted to a progress bar analogous to one you’d see when installing a new program).

So that all works, basically. Victoria II is the 19th century history of nation-states, and 19th century colonialism is nationalistic colonialism, so it’s easy to map. That’s because nations are easy to map. Nations have discrete colors on a map, with discrete boundaries, and large labels over them. What was colonialism? This. Like everything in life, it’s more complicated than that, but that map is a good place to start.

Neocolonialism, by contrast, is much harder to map.

Neocolonialism is a complex cocktail of background financial and political decisions conjured by a global potpourri of rich people. It’s far more subtle than its ancestral waves of colonialism were. If you look only at the geographic boundaries of nation-states, it’s invisible. For example, African countries nowadays don’t have any foreign prefixes attached to their names. If you didn’t know better, you’d assume their governments are similarly autonomous.

But it is real. This is neocolonialism:

Look at how wordy that damn image is. Without the text, it’s a picture two people, probably in some magazine ad because they look way too happy. The whole irony of it is that the effects of neocolonialism are blatantly on display, and we don’t notice. But with the text, it’s a complicated mess.

And this is the closest analogue we have to that brutally elegant map of colonial Africa.

I’m struggling with this phenomenon in the development of my game’s UI. Currently, this is the basic look:

Southeast Asia is red because a crisis has developed there. Europe has a white outline because it’s Europe’s turn to make a policy decision. (We are, as the bottom text explains, waiting for the Prime Minister to propose a policy, which we’ll then vote on.) Africa looks shiny because I’m mousing over it. The bottom left panel holds all of Africa’s information, including a pie chart of the distribution of player-owned assets. Currently, no one has bought African bonds. The various icons, and the white line between Europe and Southeast Asia, all affect bond value and industrial output.

It’s foreseeable that the UI will change, and often, but for now this is my best effort at depicting neocolonialism. This is not a game where your name or your country’s name will span a swath of terrain, because you don’t directly control any land, and almost always have to share control with other players. Power is equated with how many of a country’s 10 bonds you currently possess, and whether you have more than other people. Each bond then corresponds to a vote in Parliament. All of this makes the pie chart on the bottom-left the most informative depiction of how well you’re doing in any particular country.

It looks really different than Victoria II, and it plays very differently too, even though they’re about related processes. I’m trying to make a game wherein making decisions for nationalistic reasons is not only a bad idea (which it is, if you want to make money), but also a nonsensical idea. I’m trying to make a map that in no way implies that you actually control these territories, but does strongly imply the tools at your disposal to control a country’s economic policies for your personal benefit. The UI design for my game is not just a design challenge for me, but an attempt to visually represent neocolonialism in the most immediately meaningful way possible.