Challenge is one of those nefarious things that is subjective and even then difficult to get right. I switched Dragon Age Origins to easy after I got annoyed by the combat, but the sequel I switched to Hard. I complain that Puzzle Quest II is too easy. I got tired of Super Meat Boy. Sometimes I feel that my definition of challenge is all over the place. But after a wonderful video by a wonderful person called Jim Sterling on the Escapist, I realized my stance on game difficulty.
I hate frustration from forced repetition. Hard platformers where you need to perform flawlessly for 5 minutes between saves are pure hell for me. The fun part is to figure out what you need to do and which route to take to earn all the collectibles and beat the par time, but when you need to sit down and get 10 millisecond-accurate jumps in a row my blood begins to boil. Because repetition takes time, and I do not want to waste it. So the key is how much time a challenge takes.
Conversely, I love mental challenges and strategy. Figuring out efficient patterns, finding out routes, solving (non-arbitrary) puzzles, creating combos… Those things are fun. That’s why I tend to turn up the difficulty in my D&D combats: I want the players to feel challenged, but still survive. Puzzle Quest (as I remember) was fun because the enemies were quite merciless. Every move had to be thought out and you needed to have a grand strategy. But in Puzzle Quest II the enemies deal way too little damage and do not utilize their abilities as well as they could. In Dragon Age II the fights became a drudge when I realized that I could just bang my head on the keyboard and still win everyone. That’s when I switched to Hard. The fights still took an equal amount of time, but they needed more thought.
So those are my preferences towards difficulty. Now onto the main course.
In most video games, like first person shooters, action-adventures and such, there are three difficulty levels: Easy, Normal and Hard. In most cases, Easy difficulty lowers enemies’ HP and decreases their damage. Normal has no changes to those values, and on hard enemies have more HP and deal more damage. This is not a good way to do difficulty. Increasing enemy HP just means that it takes longer to kill them, and increasing their damage means you have to perform better mechanically. Your jumps need to be accurate and your shots need to all hit the head. This does not increase the need for strategy, it just means you need to play longer. This is why I usually avoid hard difficulty: It just wastes time without actually being more challenging.
So how to make games more or less difficult? Most of the time difficulty comes from multitasking. Navigating environments, figuring out enemy attack patterns, figuring out enemy weaknesses, managing resources… If you want less challenge, take some of those features away. Add infinite ammunition. Simplify enemy AI. Add pointers to the best place from where to shoot. And conversely, add more complexity if you want challenge. Add more different types of enemies. Add weapon jamming. Make the AI more cunning. Although, this will require more time and resources from the developer.
One way games have handled things better are achievements. Not the generic, “kill 1000 enemies” grinds. But those which require you to do stuff with a handicap. Half-Life Episode II and Puzzle Quest II have prime examples. Half-Life had and achievement that required the player to place a garden gnome into a space rocket. The only garden gnome in the game is found near the start, and the only space rocket near the end. In other words, you need to carry a garden gnome with you the whole game. Through all the dark tunnels, wide open spaces and car chases. Figuring out how to do that adds to the game difficulty by adding to complexity. Puzzle Quest II has an achievement that requires the player not to use any items except when dictated by the plot. It takes out an entire mechanic and valuable resource from the player. Many bosses have actually become challenging with this handicap. Beat Hazard had one achievement that required you to be hit by 50 or so mines in one song and not to die. It required a lot of planning to pull off. This is Jim Sterling’s point. Make the core game as easy as possible, and add additional challenges for those who desire them.
The other way is far more insidious. Fantasy Wars and Elven Legacy, strategy games from eastern Europe, both have extra objectives for every mission. These objectives range from completing the mission in X turns or less, or by having the player conquer an area somewhere, or such. The kicker is that completing these objectives award the player prizes, like experience, gold or items. These rewards are usually quite substantial. This actually makes difficulty a tactical choice: I can finish this level easily, but do I want to miss out on that wicked sword and 500 gold? In my case, I always went for gold. And boy was that hard. I remember thinking at one point “Well, this mission looks simple. There’s the main city that I need to conquer. Now what about the side objectives? Conquer that far-off village? Well, I guess that I can divide my forces… Finish the entire mission in seven turns or less?! It takes me at least 5 turns to move my first unit outside the main city! Madness!” I grit my teeth and played. I failed, learned, tried again and succeeded. Getting that sweet, sweet gold medal and all the rewards made the victory ever more satisfying.
I made a mathematical formula to represent this method. Imagine that the skill the game needs is 1, and the game gives 1 reward. The rewards make it easier for player to progress; A more efficient sword requires less attacks to fall an enemy. The difference between skill and reward is 0. If you want to decrease difficulty, make the game require 0.5 skill and give 0.75 reward. That means the difference between skill and reward changes to 0.25. So the player requires less skill to play the game, and the game gives the player more resources to succeed. And if you want the game to be harder, make the game require 2 skill and give 1.5 rewards. Now the game takes more skill to complete, but also rewards the player. If your game is a level-based roleplaying game, give the player more experience. Now he gets more (non-essential) content out of the game, and feels more challenged.
At least that is what I would do, were I a game designer. Oh wait.