Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 23, 2014
arrowPress Releases
October 23, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Abstraction & Civilization
by Jon Shafer on 12/07/12 12:27:00 pm   Expert Blogs   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


You can read more of Jon's thoughts on design and project management at his website. You can also find him on Twitter.

Abstraction is utilized by every game ever made. Not only is perfectly mimicking reality impossible – it shouldn’t even be the goal. Interactive entertainment sets itself apart by offering players interesting decisions, testing their skills or immersing them in a unique world. Games fail to achieve these goals when they prioritize realism above all else.

What does abstraction do for a game? How can it be used for both good and ill? What does a designer need to consider? We’ll examine these topics and more!

Along the way we’ll also look at some features of the Civilization series, as it presents an excellent case study for how abstraction not only improves a game but can also frustrate a subset of players.


What is Abstraction?

Whether in games, graphic design or fine art, the use of abstraction is always a means to the same goal: emphasizing a few important elements over the larger, more complex picture.

As with most design knobs, abstraction isn’t black-and-white, where you either have it or you don’t. The spectrum ranges from offerings which try to model every factor, such as military sims (the “game” status of which is debatable) to the venerable Go, where the incredibly nuanced concept of warfare is simplified all the way down to two types of stones on a grid.

The amount of abstraction a game embraces is one of the key elements in determining what type of audience it will appeal to. As such, developers concerned even a little bit about sales numbers are wise to give more than just a passing thought to the degree of abstraction utilized in their products. Sales figures aside though, virtually everything is on the table – an opportunity which designers should find both exciting and daunting.

Now that we know what abstraction does, let’s look at some concrete examples of how it can actually be applied.



Abstraction in Action

The best opportunity to utilize abstraction is when dealing with detailed subject matter. Economics, present in some form in a large number of games, is a great example. Notabstracting economics isn’t really an option – hell, not a single person alive has a complete understanding of how it works. Even the models professional economists use are heavily simplified from reality. Needless to say, we have neither the brains nor the computing power to properly simulate the economic impact of every human alive.

Because economics is such a vast and detailed subject matter, designers can pull out any number of facets to emphasize. What is the key to a robust economy - having the most people? An advanced industry? A large trade network? Buying and selling at the most profitable opportunities? These concepts are just the tip of the iceberg.

In the Civ series, the ingredients for economic success always include one part good geography, one part technological advancement and a pinch of “population is power.” This approach is very much an extension of the overarching theme of Civ - human history as the story of continual upward progress. Even this is an abstraction, as countless kingdoms have not just risen but also fallen throughout the ages. Despite this fact, over the course of civilization the needle has generally points upwards and this was the feel Sid Meier was trying to emphasize in the first Civ. Most folks find it more fun to play a game where you’re building ever upward, rather than dealing with the risk of being toppled by your own people. It would have been equally valid for Civ to prominently feature internal strife and famine, but these were elements the designer simply didn’t care to represent.

Combat is another real-world concept that goes through a bit of a filter before showing up in games. Accurately simulating the nuances of single combat is far beyond what most titles want to offer, and those which present full-scale battles have no choice butabstraction – the only question is what form it should take. As with economics, the specifics mainly come down to the team’s preferences. Is success in battle primarily based on good timing? Proper positioning? Wielding the best weapons? It’s possible to incorporate all of these elements, but odds are the developers have a few pet ideas which they believe are especially interesting or pertinent.

Winning wars in Civ usually just comes down to bringing the biggest, most advanced army to the party. Tactics also factor in, but to a much lesser extent. This abstraction was no accident, and applied largely because Civ is an economic game at its core. Modeling the intricacies of battlefield strategy is no small challenge, and it’s very easy to simply “round down” and accurately claim that wars are typically won long before soldiers take to the field.



A Double-Edged Sword

Like all powerful tools, abstraction has the capacity to do as much damage as it does good. An abstracted icon can be hard to identify. Abstract art is only understood by a dozen people worldwide. Maybe. Too much abstraction in a game can cause the final product to bear very little resemblance to the source material. Abstract games that draw on topics only loosely can certainly be fun, but creating one means the designers are inviting a much greater challenge upon themselves.

In game design, the downside of too much abstraction is the potential for players to refuse to buy what you’re trying to sell. One famous (or infamous) example from the Civ series is the eternal struggle between tank and spearman. For those of unfamiliar with this “meme,” in the combat system there are rare times when a vastly superior unit can get extremely unlucky and lose a fight. Ultimately this is a consequence of abstracted form combat assumes in the series. For some people this is nothing more than a humorous quirk but for others it’s an unforgivable sin.

While I certainly wasn’t one of this “feature’s” largest detractors, I did find such occurrences to be a little out of place in a game that, while not a simulation by any means, at least tries to tie itself to world history. As a result, I modified the combat resolution model in Civ 5 so that the possible outcomes for a battle live inside a much narrower band.

One of the other design changes I made in Civ 5 was to abstract naval transport. Instead of needing to ferry armies around on actual boat units, the idea was that once your civilization is capable of launching seaworthy vessels, transport ships are simply “available” for your forces as soon as they enter the water. This made crossing the seas much less of a hassle, and as a result there are many players which vastly prefer this model. But it wasn’t universally beloved. A number of other people found this design tooabstract, believing it ridiculous that a group of spearmen could “magically transform” into a boat.

Although it’s second-nature for many of us at this point, even the basic abstraction of “turns” (when some agents can act and others cannot) out of our real-time experience of living is a fairly unnatural translation. Outside of artificially-created constructs, how much of the world could really be considered “turn-based?” Turns are a crucial component of many games, and they help focus attention in ways real-time titles struggle to or simply cannot offer. The way turns “freeze” time in Civ opens up a better opportunity to craft both short-term and long-term strategies compared with a real-time cousin with similar subject matter, such as Age of Empires.

Civ also applies another abstraction relating to turns. In order to inject a sense of progressing through history the turns are labeled with very specific dates, such as 3950 BC or 1492 AD. Few people have taken issue with this, but there are those who dislike the unrealistic holes this opens up – after all, there’s no way it would take 50 years to get from New York to Boston no matter how slow you walked.




Fans of Realism

While most people are fine with a fair helping of abstraction in the games they play, some are not quite as forgiving. Given that all games are entertainment, there’s no objective “good” or “bad,” and as such their opinions are just as valid, even if their numbers are few.

But even hardcore realism fans are looking for some form of abstraction. Civ players who ask for more realism simply want to slide over a few notches on the abstraction spectrum – they don’t actually want to live out the life of a historical leader with all that entails. Even sports sims aren’t perfect replications, if only because you’re giving commands through a controller or mouse. I can’t imagine anyone really wanting their gaming experience to include bundling up to ward of sub-freezing temperatures while dealing with insubordinate players who refuse to follow through on the play you called! Anyone interested in this would actually, you know, coach football in some capacity.

The only time I can see the goal including a perfect duplication of reality is when the target audience is comprised of individuals who simply lacks the capacity to perform the represented activities outside of a game. Projects of this type are part of a very rare breed.

As I noted earlier in this article, abstraction is simply one part of the designer’s toolbox. While true realism is never the objective, nearly everyone at least expects the games they play to be believable. As long as a title is consistent with itself and its subject matter then you as a designer have done your job.

- Jon

Related Jobs

DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer
Bohemia Interactive Simulations
Bohemia Interactive Simulations — ORLANDO, Florida, United States

Game Designer


Steven Christian
profile image
Everything is abstract, from the fact that a worker is larger than a city, a marine is larger than a battleship, or the fact that they all have 100hp. Hp itself is an abstraction. UI is an abstraction. Everything is.

Bart Stewart
profile image
I'd go so far as to say that being able to judge the right level of abstraction for the dynamic feature of a particular game is one of the two or three core competences of a game designer.

It's tougher for gamers who like simulationist fun to see that adding realism isn't always the best solution. So I've started using the word "plausibility" to describe the right metric for assessing design ideas. Plausibility works better because it stays fully inside the magic circle instead of importing real-world behaviors that may not entirely support the point of the game. And as a practical matter, plausibility is also cheaper to implement than full-up simulated realism.

Chris Hendricks
profile image
Incidentally, thank you for the whole "armies magically turn into boats" thing. My initial frustration at not being able to find transport ships immediately turned into happiness when I realized that I didn't even need them. That was a great simplification.

Toby Grierson
profile image
I think my favorite civ abstraction was the air missions in Civilization 3.

Before, you would move your air units around and just fight, but with a fuel limit, which is IMHO bonkers.

In 3, your air unit is tied to a base and has actions like "rebase", "strike", "bombard", "air superiority" (intercept others) and so on. Rebasing is unlimited range, while others have finite range.

So you don't have units just sitting out over the ocean for a 2-year turn; no fighter planes vs. phallanx. It's a much better abstraction of actual air operations.

Ramin Shokrizade
profile image
I really liked your point that even real world economies are abstractions, this is very true. "Real" economies are so close to being "virtual" economies now that sometimes it is hard to see or explain the difference. "Real" economics usually abstract or even totally eliminate environmental factors, assuming all environmental inputs are infinite and all outputs to the environment are zero. This is, imo, the greatest source of error in trend analysis today in "real" economics. Sid Meier abstracted the environment with the first four iterations of Civ by including global warming. It may have been an abstraction, but at least the topic was not ignored. In 1990 (the time of the first Civ) very few people even knew what that was.

Much of what I know about both real and virtual economics I learned from playing all of the Civ iterations over the last 22 years. I found it interesting that the problem with battleships vs. phalanxes was caused because units all had 1 hp in the beginning. So they either completely lived or died after every battle. In later versions they were given 10 hp and this removed much of the weirdness. Now in the latest Civ5 expansion they have 100 hp. Again just by adding one more decimal point the game gets a tad more complex and realistic at the same time. EVE Online suffered initially from the same problem until they added one decimal point to the price of everything. This solved some huge exploits. I guess you could say that by not using enough digits for some values you were overly abstracting the quantification :)

Peter Ebbesen
profile image
I've got to admit that I'm one of those who hated the 'everybody has a boat' abstraction. It worked, certainly, but to me it felt like "yet another developer, who gives up on the importance of navies in order to turn the map into flatland for armies to walk on". From ancient times to this very day, deploying and supporting armies across seas have been at least an order of magnitude more difficult and expensive than across land and realistically only an option for those who dedicated a significant portion of their economy to do so.

Now, the Civilization series has always been very lightweight in terms of dealing with logistics, and understandably so given its wide target audience as logistics is hard, but this particular abstraction I could well have done without. The only simplification/choice-of-a-different-level-of-abstraction that saddened this strategy-grognard more was the introduction of empire-wide happiness in place of city-specific happiness and individual grumpy citizen-wannabe-revolters to cuddle/suppress.

And yet... all of this would have been forgiven if Civ5 had just shipped with an AI that was capable of dealing with 1UPT. A reduction in strategic and logistical complexity would be a fair trade off for the increase in tactical complexity you get when you turn the strategic map into a tactical map via 1UPT

Be that as it may, thanks for a nice article on the issue of abstractions and the thoughts that go into choosing the right level of abstraction, Jon.

Brandon Van Every
profile image
The Civ series has so many forms of "task fatigue," things that take far too many mouseclicks to get through, that I welcome any form of injunctive relief.

Reemi Pedersen
profile image
This was great reading, Thank you!

Ian Welsh
profile image
Technology isn't always as big a deal as people make out. The Taliban, with vastly inferior tech, is beating the coalition as we speak. Ethiopians did quite a bit of damage to Italian tanks. Hezbollah defeated Israel. There are other examples. Civ has always had strange choices about what is superior, anyway: for example horse archers (superior to knights in most situations) have always been badly handled. The outside chance of a technologically inferior force winning a battle is more realistic than people make out. It's a case of the real world being far stranger than fiction.

Brandon Van Every
profile image
Civ has no model of guerrilla / asymmetric warfare. All units are brought into combat on a tile. The only comparison is their economic value, as the article stated.

Joshua Darlington
profile image
It would be impossible to simulate life on a 1 to 1 ratio as we dont really understand life and the universe at that level. Plus our brain dumbs down all our perceptions so they can be processed at a minimum functional rate of speed. Unsurprisingly, computer games have similar contours of resource management. If you are balancing representation versus speed in entertainment media, the key dynamic to track is entertainment. In the entertainment business the key dynamic is business. So most game worlds are schematic representations designed for max money followed by max entertainment and "simulation" is complexity residue.

Brandon Van Every
profile image
For sake of argument, what's the "business dynamic" of a film?