Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
November 1, 2014
arrowPress Releases
November 1, 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:

In Defense of Surprising Gameplay
by Bart Stewart on 08/20/11 01:13:00 am   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.


Major game design studios these days seem to be dedicated to the utter and absolute elimination of all surprise from games.

On the one hand, many developers apear to be terrified that if a game allows for surprise -- if any unexpected emergent interactions between systems are permitted -- that some player somewhere will wonder for a microsecond what he is supposed to do next. At that point (the thinking appears to be), the player will conclude "this is boring!", quit playing the game, and insist to all his friends that the game is broken.

So these developers are now doing everything they can to make sure that every possible system interaction is 100 percent controlled. There must be no moment in which the player can see or do anything other than what the developer intended... even if that means limiting the number and span of gameplay systems to the point that games become nothing more than "walk" and "shoot enemy" with different textures.

But these developers aren't alone. Some players don't like surprises, either. And many (most?) developers have decided that these players -- and no others -- should be given what they want. So the developers obsessively design and playtest and focus group gameplay scenarios to insure that as long as players do the "right" steps in the "right" order (often substituting player skill for in-game character abilities, even in RPGs), they win. There's nothing wrong with having games like that... but why stop there?

What bothers me about the games being made based on these rejections of surprise is that not all gamers appreciate having every possible in-game behavior locked down. Some of us enjoy being surprised! In particular, surprise is a critical element of simulation play. If events can never vary between runs, then there is no opportunity for players to make interesting choices as inputs to produce outputs that are enjoyable for being unanticipated.

So why is it that it's almost exclusively the gamers who want gameplay they can be guaranteed of winning who get attention from developers?

Of course it's not much fun to lose because of randomness. But that's not the same thing as emergent surprise, which is what I'm really talking about here. Surprise doesn't have to have direct gameplay consequences; indirect effects may be enough. And of course games should be tested and modified to prevent game-breaking surprises, such as fires that spread unchecked across the entire gameworld. There can be surprises due to untestable interactions between many complex systems that do not result in a broken game, but instead contribute toward the presentation of gameworlds that feel alive in many small but meaningful ways. What is preventing developers from making games like this?

I believe the trend toward eliminating surprise from games needs to be countered in order to prevent games from becoming utterly banal snoozefests that nearly play themselves. Procedural content generation -- particularly in the area of object behavior -- is one possible path toward achieving that goal.

So I strongly endorse what developers like Andrew Doull and Miguel Cepero are doing with procedural content generation. And I hope their work will inspire other developers to embrace the idea that a little more surprise in their games, where appropriate, can actually be a good thing.

"A nice blend of prediction and surprise seem to be at the heart of the best art." --Wendy Carlos

Related Jobs

Rensselaer Polytechnic Institute
Rensselaer Polytechnic Institute — Troy, New York, United States

Assistant Professor in Music and Media
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Next Games
Next Games — Helsinki, Finland

Senior Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer


Michael Joseph
profile image
"many developers apear to be terrified that if a game allows for surprise -- if any unexpected emergent interactions between systems are permitted -- that some player somewhere will wonder for a microsecond what he is supposed to do next."

Good point. Do you think that a lot of game makers have a low opinion of their customer's intelligence? :) Maybe that's one of the downsides of too many focus groups. LOL.

Surprise I think that is also at least one difference between making a game you want to make (which may wind up being comparatively niche) versus making a game that appeals to a broad market. If you're just shooting for best of breed in a particular genre, you're not likely to surprise very much beyond the superficial. But if you're going for something fresh and original and surprising, you're assuming more risk but you gain the chance of producing something really special.

So another aspect of surprise is defying convention. Grand Theft Auto, Minecraft, Duke Nuke 3d and many other games that innovated and maybe even created new genres or sub-genres surprised audiences by their originality and were rewarded for it.

Fortune favors the bold.

Bart Stewart
profile image
Don't forget about Portal, which was surprising in more ways than one. "Deploying surprise in 3... 2... 1... wait -- that's not the surprise."

That's not exactly the kind of surprise I mean, though. The "oops, you're not really getting cake" surprise is surprising the first time. But it never changes after that, and hence is no longer surprising. I'm thinking more of games that are designed to have bits that even the developers don't know for sure what might happen -- not just from doing a randomized table lookup, but through interactions of complex systems, which can be consistently surprising every time you play the game.

I understand developers being cautious about that... but not to the point of trying to eliminate it utterly from every game, which kills off a really interesting area of play.

Here's an example. Back in Star Wars Galaxies, the crafting process started with the collection of resources. There were different categories of resources; each resource had different qualities; and each quality varied numerically. The game then spawned new resources in different locations all the time. There were also crafting recipes calling for different resource categories, and the better the numerical values of the applicable qualities, the better the stats of the final crafted product could be.

That was great... but then the actual process of crafting was just a matter of cranking out many identical copies of the same product. There was no variation. And here's the salient point about that: for some people, that lack of variation in the output product was a good and desirable thing. For these players, variation -- surprise! -- was undesirable.

But for some other players, certain kinds of variation would have been pleasurable. The chance of a random "mutation" making one radically new kind of object would have given the crafting game a tremendously enjoyable feature for some players.

So I wouldn't attribute not wanting to offer that kind of gameplay to a low estimate of gamer intelligence, though. The fact is, there really are a lot of gamers who don't like surprise, and developers are just giving them what they want. My objection is that these aren't the only kinds of gamers out there. Those of us who are OK with some surprise in our games, who actually enjoy certain forms of unpredictability, just aren't getting this kind of content.

I see that as a missed opportunity.

Michael Joseph
profile image
"But for some other players, certain kinds of variation would have been pleasurable. The chance of a random "mutation" making one radically new kind of object would have given the crafting game a tremendously enjoyable feature for some players."

Oh... i see. I guess i understand why game developers would be "terrified" of implementing these types of emergent systems... they're not easy (Spore took how many years to make?) and sometimes the unpredictability can make things go horribly wrong. Somebody might find some game wrecking exploit in multiplayer. So you need a lot more testing and all of this of course translates directly to cost and maybe even hurts coding optimisations a bit if you can't make certain assumptions and that cuts into performance budgets... who knows.

beware of the unknown gnomes

EDIT: I think a good start is to revert to systems provided by pen & paper RPGs as opposed to rolling your own during production. They're insanely flexible and their rules have been play tested for years... Hrm, come to think of it it's somewhat surprising that systems like GURPS and such haven't been bought by big game publishers yet. A slogan like "Powered by GURPS" could come to be synonymous with the type of emergent gameplay experiences you describe. A company could release games in any number of genres and the "Powered by GURPS" logo could be used to generate a lot of cross over appeal. (eg. "I'm not a fan of martial arts fighting games but... this one is powered by GURPS and I love all the other PbG games")

Zenas Prime
profile image
"So why is it that it's almost exclusively the gamers who want gameplay they can be guaranteed of winning who get attention from developers?"

I tried to have this debate with my fellow WoW players and what it seems to come down to is that a very large proportion of them really do just want an "interactive movie" style experience that does't provide that much in the way of a challenge or suprise. The game industry seems to have shifted towards this market because thats where the money is coming from. Why waste your developing dollars on niche markets when you can appeal to the masses for far less investment and higher return?

Of course I don't agree with this paradigm but I can sure understand where it's coming from; big money enterprises.

Darren Tomlyn
profile image
(All my comments are based upon my blog - click my name).

The is, of course, ultimately based upon the relationship and difference between a game of chance and skill. Many of these issues are about games currently trying find some combination of the two.

One of the main reasons for this, is simply a matter of consistency in the behaviour of the game in relation and reaction to the players actions. But the reason why surprise is unwarranted in relation to the players behaviour, is precise because that's the only way it can BE a written story in the first place. How a game REACTS to the player, is of course, one of the main areas you're talking about - and that is fair - how games react to the written story of a player is one of the main areas they can currently improve upon.

The main reason for this, however, is because (computer) games are generally being made for the opposite reason - to tell a story to the player, rather than enable a story to be written, and then merely react to such behaviour in many different ways (surprise!) instead.

There are, however, other aspects of this problem, such as the static (and persistent) worlds many games - especially MMO's - take place in. One of the related issues here, is also about just how much information is given tot he player in the first place, about the setting and other stories taking place within - such as missions etc..

Dynamically generated content should be one of the greatest ingredients developers have when creating games for computers, but such a concept has yet to come anywhere near reaching its full potential.

Dave Endresak
profile image
You had me interested until you got to the part where you claimed that RPGs have changed to where they eliminate relying on player skill. As we just clarified recently in the article about the problems with RPGs, this is the actual problem with so-called "RPGs" - that is, for the past decade or so, RPGs have changed to where they focus on real time actions which remove any ability to actually play the role of any character, instead relying on player skills, not character skills. Role playing is not about "twitch" gameplay of any kind, but today's developers (and yes, many players, too) seem to have lost sight of that fact. Instead, they just complain about products.

By the way, Bart, Record of Agarest War has some of the concept that you suggest as far as crafting is concerned. You craft weapons, but the smith will sometimes "goof up" and wind up with some random result, even results that are far more potent than what was originally intended to be crafted. Unfortunately, this leads to simply attempting the crafting over and over in order to get some sort of randomized item that offers advantages to certain characters, or even simply doing this until all characters have been provided with various random powerful but random items. Granted, this is player choice, and I am a very strong proponent of player choice. I just think that this type of result could be graduated a bit better. Still, I thought it was worth pointing out an example of what you suggest.

Also, even more importantly, user created content is queen, but many companies continue to fail to offer such empowerment to players. This could be anything from new character models, maps, dungeons, adventures, clothing, armor, weapons, etc, to entirely new races, objects, and expansions. Bethesda has perhaps the most active modding community, but other companies like Firaxis are certainly supportive of their players. Other examples would be new maps for strategy games or new circuits for racing games.

Obviously, there is no way to know exactly what users might come up with next. However, supporting consumer generated media content is why movements such as Vocaloid music and games have grown to worldwide popullarity in such a short space of time. Part of it isn't even the enjoyment of the content. Part of it is simply being empowered and feeling that you are actually part of the creation process, even if your part is simply feedback or using the content created by others. This is somewhat analogous to the free to play model and the 80/20 rule where only 20% of users actually pay, but the other 80% are important because they are the majority of the community and make the community viable.

Jason Schwenn
profile image
"I believe the trend toward eliminating surprise from games needs to be countered in order to prevent games from becoming utterly banal snoozefests that nearly play themselves."

Aren't we already there?

I am personally concerned as even the "Games of the Year" that garner the 9s and 10s are usually so predictable and boring to me that I'm not even sure where I'm supposed to place hope in the current mainstream game industry.

Mass Effect 1/2, Batman: AA, nearly any AAA FPS, Assassin's Creed and such are all hand-holding snoozefests for me. Hell, what has GTA done that's exciting since the first GTA III?

I know I'm a bit too hardcore about my standards and views, even for the supposed hardcores, but it wasn't that long ago that games like GTA III, KOTOR, Morrowind gave me such huge hope for the 'games of the future'.

Here I am in that near- future and the only thing I'd tell myself 5-7 years ago is "prepare to be greatly disappointed."

Seems to me too that the big expensive AAA games are just going to continue on the same path of higher costs and monetizing gameplay mercilessly and the indie scene will continue to exist mainly to audition for the AAA companies or to make the same artsy 2d platformers or flock to portables/smartphones (while still maintaining their condescending indie snobbery of course).

The video game industry is the newest creative kid on the block, and it's alarming just how fast it became assimilated into corporate big-business and lazy entitled consumer culture.

Bart Stewart
profile image
Thanks for the comments, everyone!

There are some good counterpoints that have been made.

1. It's very easy to neglect to spell out large-scale assumptions when focusing on a specific concern. In this case, I failed to note that what I'm really thinking of here are single-player games. Thanks for helping me see that, Danny.

Multiplayer games of any scope are almost always surprising because, by definition, they include other human beings... and it's fair to regard people as "unpredictable systems." As I've said before, this is one of the reasons why multiplayer games (including MMORPGs) are freakishly difficult to do well. In addition to just getting the game itself to work right (player-to-world interactions), you have to spend a lot of time and effort on keeping players from cheating or griefing each other (player-to-player interactions). The kinds of things that creative human players can do to each other are actually *too much* surprise unless you really nail down all such interactions. (And even they'll they'll find ways to surprise you.)

So what I was speaking to in my original post was only single-player games. How can these games have some of the same surprising fun as multiplayer games?

2. Dave, I may have phrased it badly, but I was actually agreeing with you. I believe developers of MMORPGs are indeed shifting gameplay more toward player skill at following specific situational rules -- think about the highly coordinated player actions required to defeat a high-level boss.

However, I do give developers some credit here for simply giving gamers what they say they want. As I noted in the SWG crafting example, it's not just developers who dislike surprise. Some gamers regard surprising content as an annoyance because, to them, it appears to be randomized interference deliberately injected into a game to block them (the players) from reaching the defined goal as quickly as possible. For these gamers, a "surprise" like unexpected instances in a bunch of mass-produced widgets is not a fun new bit of content to be enjoyed as part of a creative game; it is one less widget that can be sold to "win" an economic game.

I want to point out here (because it can be hard to hear tone of voice in a written mesage) that I'm not criticizing or looking down on these gamers. People like what they like. In catering to these gamers by locking down systems to minimize surprise, developers are giving these gamers what they want. That's not a bad thing.

What I do think is a bad thing is catering *only* to these gamers. For all those who dislike surprise because it gets in the way of blasting to the "win" state, there are other gamers for whom the fun is in the journey, not just reaching the destination. For these gamers, surprise is a crucial part of the journey. (In a larger sense, this is why people travel. Surprise is what exists "beyond the fields we know.")

A crafting system that allowed the occasional oddball item to be produced by simply clicking the same thing over and over again, as you say, would just encourage grinding by the destination-oriented gamers. But it doesn't have to be designed that way; it would have been relatively simple to take a deep crafting system like SWG's and tweak it so that only thoughtful experimentation could produce surprising new things. That's a creative production system that would appeal to lovers of surprise.

3. Finally, I should emphasize the Wendy Carlos quote: too much surprise is no better than too little. The interesting stuff seems to happen somewhere in the intersection of order (rules) and chaos (freedom).

That is, anything too surprising just looks random. To really engage hearts and minds, art should show us something familiar in a new way -- surprise needs to be constrained to be enjoyable.

In game design terms, that's the kind of thing I was describing in talking about fire systems like the one in Far Cry 2. An always-predictable exploding barrel is minimal fun; a fire that spreads uncontrolled (as was nearly the case in Minecraft until the 1.6 patch) is too much fun. A fire system like that of Far Cry 2, which while unpredictable was damped (although maybe a little too damped!), is a system of constrained unpredictability that maximizes fun over repeated play sessions. You never know how exactly how a fire will spread, but you can be pretty sure that this constrained unpredictability will make every encounter more fun.

And that's what I'm suggesting in this thread. Let there be single-player games that always tell players what is expected of them, challenging their memory and reflexes. It's good to have such games.

But also let there be games that, in a well-constrained way, offer worlds full of unexpected events and encounters that challenge the perceptiveness and creativity of players. Those are fun, too!

Will Ooi
profile image
One really fantastic piece of obfuscation which was risky and divisive but, I think, ultimately worthwhile, was Metal Gear Solid 2 and how the player was teased with controlling Solid Snake for the first level of the game, only for the title to then make them play the remainder as Raiden - who was in several ways the antithesis of Snake. It was such a brave decision to disconnect the player from direct control, but in doing so helped strengthen that aura of the character - which created in the sequels that followed an even greater yearning for a return to 'normality'.

And unlike other narrative techniques where a surprise element or twist was incorporated into the story itself, MGS2 didn't suffer from - for want of a better term - "second playthrough twist already-known syndrome" (there must be an official name for this, somewhere =p ).

Funnily enough the series ended up as quite a letdown simply because of all the twists that Kojima tried to insert into the story in MGS3 & 4 - when the most effective surprise was the one that broke the fourth wall.

Jonathan Lawn
profile image
I'd like to support Bart's article. Some players like determinism, some like to think on their feet. I'm in the latter camp, and I think the former is getting a lot more attention.

Perhaps there's again an issues with labelling here though - we (gamers in general) lack a common language so that we know what we're getting. "Randomness" ought to be an easy concept to convey in the tagline, but perhaps it's difficult to make it sound marketable.

Tora Teig
profile image
Well said. We are much puppeteered through modern games. The experience is breathtaking and fascinating, but it wont be so the second time. And with so little freedom I think a lot of story and world credibility is lost.

Thank you for posting this :)

Luis Guimaraes
profile image
Yet the most successful titles today are multiplayer, which means unpredictability is important for players. But the kind of games we want is more ambitious to develop anyway.

Perception of fairness is also important in this case. The way to develop perception of fairness is evolving and it's natural that some attempts will go too much beyond the sweet spot (i.e. player pampering), but things will get better.

Bart Stewart
profile image
I should have known that soon after I posted this, someone would come along and provide a counter-example of a game that creates complex systems, lets them interact, and trusts the player to enjoy being creative using them to solve problems. And sure enough....

Here are some quotes from an interview by Rock, Paper, Shotgun with Harvey Smith, designer on the original Deus Ex:

"[T]he team, all of us, had to be reminded constantly about what our goal was – to over-connect the systems, the rules and the mechanics."

"There’s all these crazy combinations that we had no plans for, but we trust that if we make things general purpose, the player can get creative."

"It is exhilarating, because so few times in most games can something surprising happen."

What he's talking about in this interview ( is the game Dishonored, currently in development by Bethesda's Arkane Studios, fans of Looking Glass and makers of Arx Fatalis and Dark Messiah of Might & Magic.

If anyone can pull off this kind of trick, these are the folks. There's always the chance it could wind up failing on some important level, but the odds seem pretty good here. This is a veteran team with experience in this sort of emergent-solution game and Bethesda's resources behind it.

I hope that future-me will be able to use Dishonored as my prime evidence that surprise in games *can* be brilliant fun. We'll see.

Bart Stewart
profile image
Well, hello, past-me!

Dishonored turned out to be a very good game, but not quite as open to player invention as, say, Deus Ex.

Gameplay in Deus Ex felt more open, I think, because the breadth of abilities was much wider, and because environmental puzzles were designed to be solvable by several of those ability modes. This created more opportunities for tactical creativity than in Dishonored, where virtually every ability and tool was for inflicting mayhem and Blink could solve most environmental challenges.

Dishonored was not a bad game. Movement was extremely satisfying, and using Blink to maneuver so as to drop all the opponents in a level was interesting fun. The "feel" of the city of Dunwall was also satisfying.

But it felt more constrained, more directed. Maybe Dishonored 2, then....

Walker Hardin
profile image
I've been looking forward to Dishonored since it was announced for exactly these reasons. The successful execution of random elements also kept me playing Spelunky every day for a year and a half.

I agree wholeheartedly with your point and believe the integration of procedural systems to create an ever-fresh experience is one of the most promising design spaces to explore. I'd love to see more games that champion this.

You might like this post I wrote on the subject...