Interview: Twitch Plays Pokemon – Collaborative Gaming Chaos
There’s a famous theory that if you give thousands of monkeys an equal number of typewriters, they will eventually write the complete works of Shakespeare. But what if you took thousands of gamers, and gave them each a controller connected to the same game? Would they be able to complete it?
This is the most apt summary of the viral hit known as “Twitch Plays Pokémon”, a livestream of a Pokémon playthrough that has a ridiculous twist- the game is an emulation controlled by the stream chat. For example, if someone types ‘up’, the emulator will read that button input and move the player character upwards. It sounds like It would be an absolute mess (it is), but it’s the most fascinating phenomena I’ve seen in a long while, and I’ve found myself unable to tear myself away from the screen.
In this version of Pokémon, the smallest of actions becomes an ordeal. Walking into a store and buying Pokéballs is simply a matter of pressing the D-Pad and A button a few times, but when you have thousands of people pressing buttons at once (the most I’ve seen participating simultaneously was just over 13 000 people), you’ll find that the game doesn’t often do what you want it to.
Adding to the hilarity is the delay between the chat and the livestream- while you may be seeing one thing on the stream, what you’re seeing may be something that has already happened in the past due to the time it takes the data to be downloaded to your computer. So when you enter ‘start’ into the chat to close the pause menu, your command may be read later when the player character is walking around the game world, causing the pause menu to open back up- and many of the participants to yell obscenities at you in the chat.
It’s such a fascinating experience watching the stream, there’s a constant tug-of-war going on between participants as their commands stack and counteract each other. Instead of walking in a simple path to their destination, the Pokémon Trainer bumbles around in a stupor, going everywhere except where they need to. Oak’s wise words constantly ring in his head as he tries to use random items within battle- “RED! This isn’t the time to use that!” Deep down the Trainer knows he has to make his way to the Gym and get his next Badge, but instead he finds himself staring at a map for reasons he can’t explain. There was a time I was watching where the audience was trying to pick up an item on a ledge. It could only be approached from one entrance and moving too far would make the Trainer jump off the edge, forcing you to come around again. It’s such a menial, rote task in Pokémon games that requires no extensive thought or effort, and yet it had been turned into a gripping spectator sport. The chat would cry out in frustration when too many people moved the Trainer to the right, pushing them off the ledge. We would come so close and then, due to forces beyond our control, have to restart.
But as frustrating as it is to watch and participate in, there’s a huge payoff when it works. Somehow the collective have managed to navigate dungeons, find hidden items, catch and train Pokémon, and at the time of writing are on their way to tackle the game’s third Gym Battle (albeit after more than 50 hours, but it’s still impressive). It’s bizarrely fulfilling when you watch (or help) the other players overcome these challenges- when you see your command pop up on the screen it’s like being acknowledged by a celebrity. The communal effort makes the participants form a strange bond with each other.
The chat screams when someone screws up the process with a poorly timed button press, and cheers when they succeed at long last. The hype in moments like the beginning of a Gym Leader battle is incredible. Everyone gets pumped up ready to win and, surprisingly, most people have progress as their goal, with the trolls and detractors in the minority. Everybody cheers each other on as the battle progresses, and gets attached to their Pokémon despite sharing them with thousands of other people. Everybody is excited for “ABBBBBBK O” the Charmeleon to evolve into a Charizard, and was ecstatic when they successfully managed to catch a Spearow and trade it for DUX the Farfetch’d. When Misty took out their Pidgey the chat went off, and the outcry even spread to Twitter.
— Cody (@Moomoomoo1) February 14, 2014
PIDGEY NOOOOO — phil kollar probably (@pkollar) February 14, 2014
The internet killed Pidgey 🙁 http://t.co/f1q65TGPab — Wario64 (@Wario64) February 14, 2014
— Jason Lacour (@Nanashrew) February 14, 2014
It was odd seeing so many people discussing the game on Twitter, and finding out that other people were as engrossed in it as you are. You would turn off the stream and go to bed, later waking up to see people tweeting about all the achievements the players made while you were sleeping. The participants start making in-jokes as they share common experiences with the others. They began worshipping the Moon Stone that was always selected in battles instead of an attack, and were heartbroken when they thought it had been discarded by accident. Some people have even tried forming some semblance of democracy and get ‘leaders’ nominated who will be the ones to input commands to the game. There’s so much going on at once, it’s chaotic. But the thing that has people so mesmerised is that the players are actually succeeding. Despite the stream delays, despite the trolls, despite the miscommunication… the game is actually progressing.
This peculiar experience is the work of a self-taught professional programmer, who we’ll call Red due to them wanting to remain anonymous. Red’s primary programming language is Python, but they know some others as well. Their previous project was similar to SaltyBet, another Twitch stream with audience participation that allowed viewers to place (fake) bets on matches in the fighting game MUGEN, earning points if they bet on the winner. After the interest in such events died down, Red decided they wanted to do something different, but of a similar nature. They were intrigued by the idea of a communal playthrough of a game, and whether the players could actually complete a game.
So why Pokémon? “Pokémon seems like a perfect choice because it’s highly turn-based and predictable,” says Red, who explains that due to the latency of the stream a game that requires quick reactions won’t work. “Pokémon allows the player to express themselves a lot more than other games so it’s easier for the viewers to influence the game”. According to Red, biggest factor is that Pokémon is a very forgiving game- there’s no permanent fail state, losing a battle simply subtracts some of your money and returns you to the nearest Pokémon Centre.
Red believes that the reason for the stream’s success, despite how frustrating it can be at times, is because people find the idea of contributing to a group effort (especially one that is publicly visible) to be appealing. “Pokémon is a simple game that many viewers are familiar with, most viewers will understand the ‘correct’ action to take at any particular time,” Red explains, “It can be exciting and suspenseful seeing if that expected action happens or not. I think the frustration provides an emotional contrast to when things randomly go smoothly, walking into a Pokémon Center feels like a huge group accomplishment due to its difficulty”.
Red works from home, allowing them to always have the stream open in a second monitor so they can keep an eye on the action at all times. Over the past few days there’s been some moments that they’ve found to be highlights. “The capturing and naming of Pokémon are easily the best moments for me, given the difficulty of using Pokéballs and the unpredictability and permanence of assigning a nickname”. Red also explains one of their favourite moments where a group of players worked together to locate and pick up a hidden Rare Candy item. “It showed the usefulness of collective knowledge,” Red says, “It was also impressive how quickly it was located, picked up and used on a Pokémon”.
Despite all odds players are making progress in the game, and it’s looking like they might be on their way to finish it. It may take hours just to teach Cut to a Farfetch’d and cut a tree down, but they’re getting there. Red thinks it’s possible they may eventually finish it, but there’s a few notable areas where they may get stuck. “I think the Safari Zone is going to be a significant hurdle, I would like to interfere as little as possible but I might need to make an exception for that part of the game.” The Safari Zone is an area of the game where players can only take a limited number of steps within it before being sent back to the beginning- it’s not hard to imagine what problems that would cause for this stream. Red also notes that HMs (Hidden Machines) like Cut and Surf could pose a problem, and at the time of writing that’s definitely ringing true- the players’ struggle to teach Cut to DUX the Farfetch’d is hilarious.
So what about the future? When Pokémon is either completed or abandoned, does Red have any other plans in mind? “It’s something I’m still thinking about, I would like to do something similar, but with a different game, but I think the kinds of games that are compatible with this sort of format aren’t very common. I’m trying to get around this lack of compatibility by thinking of different ways to accept button input (such as choosing the most commonly suggested button input) and using the emulator’s ability to pause or slow down in-between button inputs”. For now, we’ll just have to hope that all these people mashing on their keyboards jointly become Pokémon Masters, and that Pidgey (or rather, Pidgeotto now) doesn’t get knocked out again on the way.
You can watch the continuing adventure here!
ALL HAIL THE HELIX FOSSIL
It’s an awesome concept but sadly wherever an awesome concept exists someone will clone it and now there are a whole heap of TPP clones running on Twitch trying to outdo each other.
Must say the TPP by RNG is pretty cool in concept though