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:

Difficulty: Self Selecting vs Self Defeating
by Christopher Gile on 06/07/13 11:25:00 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.


A big problem with having the user define what difficulty level they want at the start of the game is that players don’t exactly know how good they are at the game yet -- I mea, it is the start of the game, not the end. They can make assumptions based on other games in that genre or series but the point of making the game was you were making something new and so there shouldn’t be a game with a strict 1 to 1 comparison. So to compensate for the fact that players are basically just guessing games like Devil May Cry put in a feature where if the player was doing really poorly it would say “Dude, I see you're having problems. Want me to turn it down a bit?”

The problem with that idea is that the prompt only comes up after the player is almost certainly frustrated, can we do something about that problem before they get super frustrated? Indeed we can as evidenced by the fact that Starfox 64 does this exact thing. In Starfox 64 if you did something really well like shoot down a 100 enemies or kick in Star Wolf’s head the game would then make the next level the harder of the options available. The game was rewarding your knowledge of the game and your mechanical ability by making the game harder to sustain the challenge, or lessen it if you are just barely getting through levels.

This kind of dynamic difficulty has a couple of problems though, first of which is that going through the game will bring you to only 7 of 15 worlds. You can beat the game without ever seeing more than half of the game’s content which can be bad for the developer since it doesn’t translate into a great return on investment for creating levels. You could have a whole bunch of brilliant levels that the player never sees.

Starfox 64 though solves this problem by being in the arcade style of game which has no real saving and assumes the player will start from scratch over and over again with a relatively short amount of time taken to beat the game. Starfox 64 doesn’t actually take a super long time to beat and this short turnover time goes well with the fact that the game allows you to quickly distinguish yourself and get on a harder/different path.
Due to the fact that it causes players to lose out on a great deal of what a game has to offer in a single playthrough this style of difficulty doesn’t really go well with very long term games.

Games implementing choice like Mass Effect or KoTor don’t force players to make choices that prevent them from being able to experience whole levels in a game, they will give choices that will change the context of that level but not typically force you to choose between two levels for an entire play through as they want to show off all the cool content they made and don't expect you to play through the entire game all over again or expect you to replay any content at all.

But even given the arcade style restriction it forces on the game’s narrative form it is a rewarding way of doing difficulty as it allows players to rise to the task and makes getting to the harder levels an implicit reward for playing well as it allows them to see new content.

There is the opposite of self selecting difficulty and it is actually fairly common. In games like Metroid Fusion finding secrets make a game easier as the reward for finding the game’s secrets is additional power such as greater missile capacity or health. If you know the game well or are good at it, as evidenced by your finding secrets ability, the game gives you a power up and thus makes the game easier for you. It is a self defeating difficulty system where smart/good play is rewarded with an easier game and bad play is rewarded with greater challenges.

Now without raging on this system to much it is easy to see how such a pattern would develop. This system isn’t indeed to be a difficulty control but rather a reward system for players who explore the game. The game rewards the players with something they will value which in this case is in game power. Tragically this reward system has the undesirable effect of snowballing the game’s difficulty. Minor jumps in skill are amplified when that minor bump gets you an extra E-Tank in Megaman, but it also makes the game that much harder if you miss these items as the game is balanced on the assumption you will find at least some of them.

Okay, so how can we still keep secrets in a game rewarding to find without creating this undesirable effect on a game’s difficulty? 

The most obvious thing we can do is not make power the reward for finding secrets. Power is but one thing a game has to offer and the game could offer other things such as story/lore with a new ending unlocked if they find all of the game’s secrets. Other things that could be offered without balance concerns are things like skins/color pallets for the player. Ratchet and Clank games do this where finding secrets allows you to unlock costumes for the player to wear. It doesn’t effect gameplay but is a fun thing to play around with and incentivises players to explore with out making the game easier for them.

Final Fantasy 7 has numerous optional bosses that give the player stronger weapons and materia to fight with, Final Fantasy games have often have numerous optional bosses that provided such things. The interesting thing though is that these secret bosses are often much harder then the game’s mandatory bosses. I beat Final Fantasy 7 long before I beat both the Emerald and Ruby Weapons. So while these do make the final boss of the game easier as the game rewards you for beating them, it only does so after you have already surmounted a greater challenge then the final boss without those upgrades. In that way it allows players to challenge themselves, or not, without making the game easier for players who know about these optional bosses. Though this does have the side effect of making the final boss feel like a let down by comparison and I think it would be a good idea to make the final boss stronger if certain conditions had been meet.

Or why not combine all of these? Optional objectives that make a game harder even as it provides new tools for you to work with that unlocks an alternate ending. By the way, that last sentence counts as gushing about Cave Story.

This was a combination of two of my posts here. 

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


Christian Nutt
profile image
Kid Icarus for 3DS has a great implementation of a system similar to the Starfox 64 system you describe -- every time you complete a level the game automatically nudges up the difficulty for the next one from the difficulty you just beat, and you get better rewards for doing it. But if you're not confident you can slide the difficulty down a bit to keep it in place. This is on a scale from 1 to 10 with tenths (so, 6.1 is a possibility, so is 6.2, etc.)

Adam Rebika
profile image
"In games like Metroid Fusion finding secrets make a game easier as the reward for finding the game’s secrets is additional power such as greater missile capacity or health. If you know the game well or are good at it, as evidenced by your finding secrets ability, the game gives you a power up and thus makes the game easier for you. It is a self defeating difficulty system where smart/good play is rewarded with an easier game and bad play is rewarded with greater challenges."

I actually see this working the other way around. This can be seen as a feature to prevent people from getting stuck in the game and having nothing to do.
A boss is too hard for you? Instead of frustrating yourself by trying to defeat it again and again and again, just go back to the easier portions of the game and have fun exploring them. You'll find some power ups that will then allow you to defeat said boss.

Dane MacMahon
profile image
I kind of liked how in older RPG games which class you picked was sort of picking a difficulty level. Warriors tended to be made for simple clicking, for people who wanted the story and not to worry much about combat. Rogues, wizards and more tended to be harder and harder to win with, requiring more tactical thought.

Stephen Chin
profile image
Desktop Dungeons so something akin to what the article writer suggests. Initially, most of the classes, monsters, and dungeons are locked. As you beat the game (it's a hour long rogue-like) with different classes in different dungeons and against different bosses, more of the game is unlocked. Thus if you're having difficulty with the game, the game doesn't just drop more monsters with more complex mechanics on you. But as you master the game, the game offers you both more challenge but also more complex mechanics of you own to master.

Adam Bishop
profile image
"It is a self defeating difficulty system where smart/good play is rewarded with an easier game and bad play is rewarded with greater challenges."

This is a problem I've spent a fair amount of time thinking about. You mention the extra-powerful materia available in Final Fantasy 7, and I think that's sort of a problem with RPGs in general: the players who complete the most demanding tasks are rewarded with items that make further tasks (including bosses who are part of the narrative) easier. I'm actually working on a game at the moment that tries to address this issue, I'm hoping to have it ready to show off later this summer.

Joshua McDonald
profile image
I've spent some time on this one, too, but first, it should be pointed out that the snowball effect is all over in games, though often more subtle than this. In Mario, doing well enough to get and keep fire-power makes the area ahead of you easier. In a strategy-rpg, the better the job you do building your team, the less clever tactics you need to use in the actual battles.

Though there are ways to mitigate this (some given in the article), effect, most of them have two major drawbacks:

1. Players should feel more powerful when they become more powerful: If getting the super laser in a space shooter makes your shots do five times the damage, but you give all the enemies five times more hit points, your players leave unsatisfied. Though that's an overly simplified example, the principle is fully applicable. I loved Baldur's Gate 2 until I realized that it was dynamically adjusting the power of enemies, at which point, all the cool things I thought I had done or the power I had built up felt meaningless.

2. Many players want functional rewards for their accomplishments: Sure, there are a lot of people who are satisfied with an achievement, extra story, or unlocking a costume, but there are a lot of others who aren't. If killing Big Insane Optional Super-Boss doesn't give them a great equipment upgrade, some will leave unsatisfied.

It's a complex problem, because I essentially agree with the concerns in the article and many of the comments. I just find that many solutions can be worse than the problem, depending on the type of game and your goals with it.