Last winter, the internet watched as tens of thousands of Twitch users "cooperated" to play Pokemon Red. Chatters typed out game commands which were executed one at a time, and while that mostly sent the game's protagonist in circles, they did eventually lead the rookie Pokemon trainer to victory. Now, thousands of Twitch users are trying to do the same thing with Dark Souls... but things aren't going so great this time. In the first 22 hours of play, Twitch has yet to leave the first area of the game, the Undead Asylum.
In some ways, this feels inevitable. There are so many gimmicky (and amazing) Dark Souls runs out there. There are players who beat the game without ever leveling up, and those who play the game blindfolded. Most recently, one player beat it using only his voice. Dark Souls has a long history of its devoted fans performing incredible feats. So this is only the latest in a long line of unbelievable runs. On the other hand... this run feels impossible.
There is a line of thinking that says that crowds have a sort of wisdom--take a group of people, let them all submit a solution for a question, and there's a good chance that the average of all of that input will be close to the solution. Plenty of experiments have shown that there's definitely something to that theory. But sometimes, this idea taken to a utopian extreme that forgets (or neglects to consider) that there are limits to what a crowd can do on its own, and that structures, technologies, and all sorts of mechanisms limit and direct what we do. (For more on this, check out All Watched Over By Machines Of Loving Grace, or at least watch this clip!)
I really wanna believe that Twitch can beat Dark Souls. But the thing is, Twitch Plays Pokemon had a lot of advantages over Twitch Plays Dark Souls:
- Pokemon has only a handful of inputs, and "good play" only requires one button press at a time. Even basic combat in Dark Souls needs players to do some combination of moving, blocking, attacking, and occasional item use.
- In Pokemon movement is tile based. During Twitch Plays Pokemon, if the trainer needed to go northeast, a bunch of players typed "up" and a bunch typed "right", and the average guided them fairly well. In Twitch Plays Dark Souls, even moving down a hallway seems to be a colossal challenge.
- Plus, both movement and combat are turn based in Pokemon. This limited the amount of incorrect options that Twitch could input, and it meant the chat wasn't required the chat to execute (or luck into) masterful timing. There were still plenty of bad inputs, but Pokemon's limited options meant that chat could push towards the good options together. Dark Souls is real-time, and that introduces a ton of additional difficulty.
- Also limiting the damage that the chaotic chat could do in Pokemon was the fact that there many of the bad inputs simply paused the game or brought up a submenu. (And menus in Dark Souls do not pause.)
- Maybe most important of all: Failing to "progress" in Pokemon mostly meant getting stuck in combat over and over, and that meant that the trainer's Pokemon gained experienced. Which in turn meant that by the time Twitch lead the trainer to the next story challenge, they were prepared to win those fights. If Dark Souls worked this way, I'd think there was still hope: Players would run their heads into a wall until they were buff and tough enough to win by brute strength alone.
And let me be clear, none of this means that I don't think that Twitch Plays Pokemon wasn't a huge, ridiculous accomplishment. It was! But it was also just a lot simpler.
I'd love to see Twitch beat Dark Souls. I really would. As it stands, even the tiniest victories feel amazing. I'm just not sure it's gonna be the storybook ending we're hoping for this time. But who knows. Maybe a week from now I'll be writing an article about how Twitch users managed to do the unthinkable. It wouldn't be the first time.