Literature and cinema have the whole difficulty conundrum figured out.
The movies know better than to have Luke fight Vader before he's even learned about the Force. Why? Because that way, the Emperor wins in 20 minutes flat and there is no restarting at the last checkpoint for the Rebellion. It makes sense for Luke to grow in confidence and power, and face progressively greater challenges before the final confrontations occur.
I suppose you could say that some books have a checkpoint system, like those bad-ass Choose Your Own Adventure books. For those, every finger or bookmark is a save slot. On the whole, however, books also only have one shot to get it right.
This sense of "challenge either matching or driving capability" is a given in those two mediums, seeing as how people throwing hardcovers or theatre seats around when the protagonist dies an untimely death is hardly practical. Controllers are more aerodynamic and are properly weighted for the endeavors of fiery tantrum throwing.
The problem with applying this narrative arc to games is that the player who guides the protagonist is a large set of variable where each variable may cover a huge range. Developers must take many dimentions of player ability into account : twitch reflexes, strategic thinking, problem-solving, and more all must be considered.
Let's take a look at some of the strategies that developers commonly apply to difficulty in games and their effectiveness.
"No! That's not true! That level's IMPOSSIBLE!" Strategy #1 : The Texas Toast Approach.
Texas Toast was created with the following logic in mind : "If toast is great, then if we make it bigger, it will therefore be extra great."
Applied to games, this simply means that if you want to increase the difficulty of a game, simply increase the value of an attribute that contributes to difficulty. This often boils down to 1 or more of 3 usual suspects; enemy damage, enemy hitpoints/health, or number of enemies.
This is the most commonly used strategy for difficulty scaling in my experience, and in most cases it reeks of either laziness or lack of imagination. The FPS and TPS genres are the most blatant abusers of this technique. The difference between Easy, Normal, and Hard modes is usually only EnemyHP x 1, EnemyHP x 2, or EnemyHP x 3.
Uncharted : Drake's Fortune is one of my favorite games of this console generation, and yet even it falls into this category. The damage-soaking ability of the enemies in that game were well documented as a point of contention for many gamers. When a game's challenge is defined solely as the number of times you must shoot a baddie or can be shot by one, the difficulty curve becomes annoyingly one-dimentional.
Now, I don't want to throw out the baby with the bathwater, so it must be noted that there are some games and genres where this is appropriate. Shmups come to mind here. From R-Type to Ikaruga, spamming enemies, bullets, etc. . . is an effective means of difficulty scaling, because twitch reflexes are the foundation and sole focus of the gameplay.
Butter & Syrup Soaking > Damage Soaking Strategy #2 : The Populist Approach.
Just as in politics, the populist approach dictates that all enemies in a game's eco-system should have an equal chance of eating your lunch. In this scenario, the system provides weaker enemies with a boost in power as you advance so that they can still provide you with a challenge. It's a nice idea in theory, but in practice leaves much to be desired.
There is a sense of accomplishment that arises from just decimating enemies you fought on even footing earlier in the game. There's nothing quite like casting a Tornado of Instantaneous Disintegration at a lone kobold with a wooden club to pay them back for all the times they caught you by surprise and made you reload a save. Scaling enemies robs you of this feeling. This was a major issue for many who played TES : Oblivion; an unnecessary bleminsh on an otherwise excellent and enjoyable title.
It looks like developers are beginning to recognize the issues inherent with this approach. Discussing difficulty scaling in Oblivion versus their strategy for the upcoming Borderlands, Randy Pitchford of Gearbox was recently quoted as saying
, "We don't have that approach at all. If you're a level 1 and you're in an area that's tuned for level 1, that might be an even challenge. Later if you come back when you're level 15 you're just going to own that place and that's cool, that feels really good, that proves to you the power you built."
I have yet to play a game that I felt benefitted from the scaling enemies approach.
That's not quite what I meant by a smooth difficulty curve, Randy. Strategy #3 : The "Bring Out The Gimp" Approach.
Limiting what a player can do is another common means of addressing difficulty. For example, winning the heart of the woman of your dreams is a satisfying challenge. In this approach, a game developer would look at that scenario and say, "OK, then when they switch to Hard mode, they'll have to do it, but with no job and no car."
The survival horror genre uses generous helpings of this technique. By limiting what the character can do, it increases the both the difficulty and the tension present in the game. This works with varying degrees of success. For example, RE5 in particular received an ass-ton of criticism for limiting the player's ability to move while firing.
Now, while the success of this mechanic is debatable, it's interesting to note that almost no one complains about the equally gimping mechanic of limiting ammunition/healing items that is a mainstay of this genre.
It's my opinion that limiting ammunition is a much more organic means of nerfing a player, which lends itself well to the necessary suspension of disbelief that makes a good horror game fun. I also found the blackout zones in Infamous to be an excellent twist on this technique, providing nice variation in difficulty without being overtly frustrating. This is how Capcom wants you to feel. Strategy #4 : The Cheap Shot Approach.
Oh, this one drives me crazy. It is usually found in JRPGs, and is when a boss or enemy has an unblockable, undodgeable, unpredicatable attack that instantly atomizes you and your crew regardless of how much you've power levelled. This is a really annoying way to keep players who grind before advancing from waltzing through certain areas or confrontations.
If I've put in the work prior to defeating an enemy to build my party to the point where they are an ultimate fighting force, as a developer, you're basically telling me that I wasted hours of my life if that super-party can be neutralized at some random moment in a boss fight, when they wake up and realize that they conveniently know a "Death To Party" spell.
This is really just a horrible way to elongate the length of a game by adding an element of arbitrary failure. You studied really, really hard for that final exam? Well that's great! 15 minutes before the exam is over I'm going to rip up your essay and have you take the test over again! Have fun!
Unspottable snipers in shooters also fall into this category, among other examples.
"For death awaits you all! . . .with nasty, big, pointy teeth!!!!" Approach # 5 : The "No, Really, Combat Evolved" Approach
I wanted to close with what I find to be the most satisfying means of creating a difficulty curve, and that is by scaling the level of strategy required to be successful. It doesn't matter whether a game is focused on combat, puzzle-solving, platforming, or tactics. Forcing the player to adapt their playstyle or use the tools given to them in a new way is a rewarding means of creating a smooth difficulty curve.
Diablo I and II start by sending mobs of a single type of creature at you. At first, you can just spam your way senselessly through them. Later, they mix up the enemy types so that some are resistant to physical attacks, while others are resistance to magic, which forces you to use different types of attacks. Or, they add in an enemy shaman that can resurrect it's brethren, forcing you to rethink the order in which you attack your foes.
Good RTSs also use this approach by varying the different tactics available. For example, I'm a turtler by nature. If I can, I will set up a defense perimeter and not venture into enemy territory until I have a vast host of destruction at my fingertips. While it can be a successful (if somewhat boring) strategy, some factions and tactics can overcome the turtle easily by focusing on outproducing it. Unless the player adapts and realizes that a quick strike to disable enemy production is the only way to ensure long-term victory, they condemn themselves to a slow and painful death inside their shell.
There are plenty of other examples of this approach being used well for a wide variety of games, so feel free to share your favorites in the comments!
He never did figure out that anti-Semitism was a losing strategy.
I personally feel that the last approach has the most room for application in game design. It is flexible, fair, and easy enough to implement once it has been properly planned. The downside is that it does take a good amount of forethought and imagination to come up with new ways to stretch the minds of players to make the difficulty curve an enjoyable experience.
I'm always looking to support games and developers who force me to stretch my mind as I play. How about you all? What approaches to difficulty are the most rewarding to you? Are there any approaches I've left out?