EPTThis will not be a blog directly regarding Starcraft 2. This will not even be a blog directly regarding RTS. Discussion of HotS has reminded me, however, of some thoughts I've had before about platforming games (and how this might apply to games in general) that I thought I would share. First, some history.
In 1983, Nintendo produced an arcade game called Mario Bros. Most of you have probably seen it; it's been released as a mini-game a few times on other Nintendo games, and Super Smash Brothers Brawl even had it as a level.
![[image loading]](http://i.imgur.com/3TFsh.gif)
Remember me?
Controls were very simple. Run. Jump. While jumping, you have basically no control over your character until you land. And there aren't really any walls around, so you don't have to worry about being able to jump off of them. Yet with such simple gameplay, the game was a hit.
In fact, was a big enough hit that a sequel was even made, called Super Mario Bros. This featured the now well-known side-scrolling gameplay, complete with blocks that give coins, power-ups, and lives, and enemies whose heads can be jumped on to defeat them and bounce into the air.
![[image loading]](http://i.imgur.com/cDik9.gif)
Back when "2 Player Game" meant "identical to the 1 Player Game, but the game tells you when to hand off the controller."
At this point it is necessary to mention difficulty curves. Most games start off relatively easy; if not, many players immediately get discouraged and quit. On the other hand, a game can't stay that easy, or else players quickly get bored and quit. There is a healthy range in which a game's difficulty has to stay in order to keep its players involved. While a difficulty curve (a graph of game difficulty against time spent playing) is not identical to a learning curve (a graph of player skill against time spent playing), a well-designed difficulty curve should stay relatively close to the learning curve to keep the player interested without making them discouraged.
![[image loading]](http://i.imgur.com/2HfJZ.jpg)
Not to nitpick, but this picture describes DIFFICULTY curves, not LEARNING curves.
In Mario Bros. the difficulty curve drew itself; the arcade game kept going as long as the player could keep up, and when they couldn't keep up any more, they died and the game asked for more money. But in Super Mario Bros. the designers had to specifically set a difficulty curve. Nintendo has historically assumed Americans were bad at video games, so they generally release Mario games with fairly easy difficulty curves.
Super Mario Bros. 2 was designed as a very difficult game, but it was only released in Japan; they actually made a completely different (easier) game for the American release of Super Mario Bros. 2. They later gave American hardcore gamers a chance to play the original, hard version when they released Super Mario Bros.: The Lost Levels. This game is one of the hardest platformers in existence. The control scheme had not yet changed; in-air control was still basically non-existent, and there were no wall-jumps. So the difficulty of the game came from its completely unforgiving design; many of the challenges consisted simply of jumping at precisely the right time to land on the proper platform, because anything less than perfect timing of the jump, and you would simply watch Mario arc through the air into a pit while the player watched helplessly. Lives are relatively scarce, and punishments for running out of lives are brutal. American gamers found the game too hard, and Japan promised to never again release a challenging Mario game in America.
It's hard to play this game and not feel like the Nintendo developers have a deep, irrational hatred for you.
Mario side-scrollers were pretty easy from that point on. Gameplay additions made the controls less unforgiving; in-air control was increased, and wall jumps were added. I like to imagine somewhere there was a Mario-themed forum much like TL for Starcraft, where the older Lost Levels veterans made fun of the newer Super Mario World forumgoers for their sloppy mechanics and discussed how in-air control and wall jumps were lowering the skill ceiling.
Now lets change tracks completely, to the indie title Super Meat Boy (a game which delights in its initials). The UI for Super Meat Boy is far more forgiving than the old Mario games; your character can run much faster than Mario ever did, and you have so much wall jump and in-air control that you can literally jump off a wall, in-air control back to that wall, and wall jump again and climb that wall. Lives are not scarce; in fact, the game doesn't even count deaths, has very short load times between dying and starting again, and offers replays at the end of each level showing all your lives simultaneously, as if they are inviting you to die a lot. If removing limitations on player control is easy-moding the game, then Super Meat Boy should be the easiest side-scroller on the planet.
![[image loading]](http://i.imgur.com/qD4fb.png)
What, you don't have meat curtains?
Yet Super Meat Boy is one of the hardest games you'll ever play, almost certainly harder than Lost Levels. Why? Because Meat Boy didn't just add easier player controls, they also added a HELL of a lot more difficulty. You're surrounded by spinning saw blades, homing missiles, lava pits, and all manner of destructive circumstances that Mario was way too much of a sissy to handle. The PC version can technically be played with keyboard controls, but it's virtually required to use a gamepad.
![[image loading]](http://i.imgur.com/bYYcK.png)
Alternate Warning Text: "A gamepad isn't required, but neither is bathing. Think about it."
The takeaway lesson is not just that Team Meat found an alternate way to make a difficult 2D platformer. They found a better way. When I walk away from a session of Super Meat Boy, I feel discouraged and upset about the challenge I just failed repeatedly, or swelled with pride at the incredible obstacles I just thwarted, but in either case, I'm excited by the awesome feats they let me pull off so easily. As I improve, I learn to leap off a conveyor belt pulling me upward at just the right second to let me grab the collectible between two sawblades without going too high or too low and dying, where in Lost Levels the same level of practice allows you to, for instance, jump to a small platform without falling. Despite the difficulty involved, I am hoping to 100% complete Super Meat Boy some day, including all A+ time trials and all collectibles (and I've already done so up to the dark world of World 4). If I play Lost Levels, on the other hand, I walk away discouraged about challenges I've failed, or somewhat happy at having succeeded in such an unforgiving environment, but ultimately, I just feel like my character doesn't do what I tell them to, and I feel like playing a different game in the future.
The most aptly named of all the Super Meat Boy worlds.
Conclusions
-Games need ways to make the gameplay hard.
In early arcade games, this problem was pretty easy; technology limited control systems enough that just playing a game properly was, in most cases, pretty hard. Arcade games built their own difficulty curves pretty easily, and early Mario games didn't have to do anything that special to be really hard, because the controls were basic and unforgiving. As technology improves, it becomes easier and easier to make control systems intuitive, but that means developers must come up with other ways to ensure their games remain challenging.
-As technology evolves, game design must evolve with it.
Adding wall-jumps may have made Mario games easier, but it was a good change; it made gameplay more exciting, more fun, and more dynamic. Maybe those games were too easy, but if so, the solution is not to revert to the former era of more difficult controls. You must find new solutions to the problems that new technology presents, not try to apply the same old formula that worked in the '80s.
-Improve gameplay UI where you can, and worry about difficulty curves after.
It's not nearly as fun to set arbitrary obstacles that raise skill ceiling; ideally, challenges will be fun and exciting. In my opinion (and remember, this is just my opinion), going back to my Command Center every few seconds to send a worker to mine is not an example of a fun and exciting challenge.
-Go play Super Meat Boy, damnit!
It's a really good game. It makes you want to ragequit every thirty seconds, but that just makes it all the more glorious when you conquer Doctor Fetus's infernal obstacles.
