Speedrunning is a hobby with a simple goal: beating video games as quickly as possible. Beneath the surface, though, things are more complicated—rather than just sprinting from the beginning to the end, runners often use tricks and exploit glitches that push game engines to their breaking points. While some of these techniques are uncovered by the runners, many are found by dedicated glitch hunters, people who investigate the mechanics of games in search of ways to exploit them. Looking at the work of glitch hunters reveals a fascinating subculture that can tell us something, surprisingly enough, about the process of scientific research.
Super Mario 64 was one of the first games I ever played, and it still holds a special place in my heart for teaching me that there could be more to a game than just playing it through to completion. After collecting all of the game’s stars, I stumbled on a website called Beyond 120 Stars devoted to cataloging user-submitted tricks and challenges. It’s a charming monument to exploring every corner of a game in the days of a decentralized internet, drenched in the online aesthetic of the early ‘00s. It’s also where I found my first speedrun.
But something drew me to the website beyond the promise of going fast. I loved the idea of the subtler challenges, like trying to get as many stars in a course as possible without jumping. Years later, I’d discover the YouTube channel of pannenkoek2012 and learn I wasn’t alone.
Pannenkoek2012’s first video is dated August 16th, 2010. Many of his early efforts focused on maximizing his coin count in each of the game’s stages, often using glitches like cloning objects that go well beyond a typical player’s knowledge of the game. After dozens of videos documenting glitches and collecting coins, the majority of pannenkoek2012’s efforts centred around the A Button Challenge: how much progress can you make without pressing A?
In Super Mario 64, as in most Mario games, the A Button makes Mario jump—the simple mechanic that drives the entire genre of platformers. But Super Mario 64 is a complex game with plenty of movement options. Rolling out of a dive—which requires only enough running speed and the B button—gives Mario enough height for most practical purposes. Other techniques for gaining height and distance are more ingenious, like using the slight height boost from being hit by fire or collecting a star to inch up to an out-of-reach spot. Sometimes glitches like cloning are employed—in one delightful video, pannenkoek2012 clones a staircase of Goombas to reach some otherwise inaccessible coins.
Pannenkoek2012 has achieved memetic status online largely thanks to an A Button Challenge video for the Watch for Rolling Rocks star that captivated audiences—even those unfamiliar with or uninterested in the mechanics of videogame glitches—with its detailed, easy-to-understand explanations and ludicrous-sounding ideas like exploring alternate universes. But the beauty of these videos is not just in their utility for this and other challenges. It’s in the loving attention to detail that pannenkoek2012—and many others—lavish on every aspect of the game. It’s in the fact that a nine-minute video on how characters blink, which has no application to the game other than learning how it works, has over 500,000 views, and even finds a little romance in the synced-up blinking animations of Mario and Princess Peach. It’s in the nearly two hours of video documenting every imaginable detail of walls, floors, and ceilings in the game, just because this is something that someone decided was worth knowing about. Even at his most obtuse, pannenkoek2012 shows that there is some value in knowledge for its own sake. More on that later.
Glitch Hunting in Hyrule
Other glitch hunters tackle different types of games. Glitch hunter Stryder7x applies a similar depth of analysis to the Paper Mario series, often finding humorous and unexpected ways of crashing the game. YouTube channel TheZZAZZGlitch dives into the fascinating programming of early Pokémon games, and has even worked on an A Button Challenge for Pokémon’s first generation of games.
But classic games are not the only magnets for deep investigation. As Nintendo’s first open world Legend of Zelda game, Breath of the Wild turned a lot of heads when it was first announced. While its wide-open environments lead to a great variety of possible routes, equally important is the game’s robust physics and the interactions between its different systems. The technique of Stasis Launching, for example, turns trees and rocks into rockets and catapults and allows for speedy movement around the vast game world.
Glitch hunters are still at work on cracking open all that Breath of the Wild has to offer, and they’re making some incredible progress. On September 3rd, speedrunner PuppetMaster9 posted a Reddit thread describing the result of a huge community of Breath of the Wild glitch hunters—a discovery called the Shield Skew Clip. The trick relies on the unusual properties associated with the game’s shield-surfing mechanic and can be used to take Link outside the boundaries of the world. This trick skips some of the slowest cutscenes and longest detours in the game—a very valuable technique for speedrunners. New progress is being made almost daily on finding new glitches and building on the results of previous discoveries.
New glitch-hunting progress is being made almost daily. Just days ago, a new super-fast movement glitch was discovered that looks to upend speedrunning strategies all over again. For researchers, runners, and fans of the game, there’s a lot to look forward to.
Basic and Applied Research
Since so much of glitch hunting is about investigating the physics and programming of game worlds, it’s easy to draw a comparison to the scientific study of the physical world. Basic research is scientific research done with the purpose of exploring fundamental concepts and understanding how things work, and includes disciplines like pure mathematics and theoretical physics. In many ways, glitch hunters provide a model for understanding the development of a basic research discipline.
Speedrunners develop their own language and terminology just like scientists do—in Super Mario 64, for example, it’s common to see acronyms like the HOLP (Held Object’s Last Position) or BLJ (Backwards Long Jump) thrown around. Sometimes this jargon follows the tradition of naming a discovery after the person who found it. Super Mario 64 glitch hunters refer to the Pedro Spot—a valuable bit of level geometry named after “a guy named Pedro”—the same way a mathematician might refer to a Banach Space or a Fourier Series.
We can also see the role that glitch hunting plays for speedrunners in the relationship between basic and applied research. Just as runners do things quickly with the help of peculiarities in game engines discovered by glitch hunters, applied researchers use the concepts and techniques developed by basic researchers to create real-world technologies.
At this point, speedrunning is a hobby with something close to mainstream popularity and an entertainment value that’s well-documented—every year, the semi-annual Games Done Quick speedrunning marathons raise millions of dollars for charity. But, outside of major exceptions like pannenkoek2012, glitch hunting doesn’t quite reach the same level of popularity.
There is something quietly difficult about accepting the inherent value of basic research, about really believing that studying physical space or how we organize numbers is valuable even if we get nothing material directly from it. With the world as gone-to-heck as it is, the pull of doing research with a real impact is strong—I’m a grad student in mathematics who tends towards applied research for exactly that reason. But I’ve spent some time researching some pretty abstract pure mathematics, and the experience taught me a lot about the value of basic research. By and large, the basic researchers I’ve met do what they do because there is something interesting, enlightening, and fun about discovering something new, and that’s the same fun I’ve seen reflected in the work of glitch hunters.
Sometimes the going is good. The current breakthroughs in Breath of the Wild show a research landscape blown wide open by new discoveries—the timing is right for speedrunners to shave serious time off their records and for glitch hunters to gain new insight into how the game works.
Sometimes the going is less good. A lot of my struggles with pure math research revolved around not just failure, but stagnation—the fear that I could stare at a page for hours and not move the problem forward. When I look at a glitch hunting video with a title like “Disappointing Observations”, I see a lot of these frustrations replicated on the screen. On a different note, pannenkoek2012 has talked openly about his struggles with the popularity of his work, casting a harsh light on both the difficulty of communicating research and the human cost of being a meme online.
But some people will never be kept away by these difficulties. Mathematicians will keep putting out new papers, glitch hunters will keep posting new videos, and devoted fans will keep watching along with bated breath, waiting for pannenkoek2012 and other A Button Challengers to shave off the next A press. Basic research is rewarding, both for the people who benefit from its applications and for those who are just curious about the world’s mysteries—no matter which world that is.