Glitch Hunting: What is it and how do you do it?

Glitch Hunting is an important activity within speedrunning — but what is it, *exactly*? And how can you get started?

Glitch Hunting: What is it and how do you do it?
Pubblicato 4 years ago

For a lot of people, glitches are at the heart of speedrunning. For many of us, witnessing cool skips and tricks at GDQ or other speedrun marathons is what got us excited about speedrunning in the first place. But where do these glitches come from? Who finds them? And how?

Well, that’s where glitch hunting comes in.

What is glitch hunting?

In the most simple terms, glitch hunting is when someone attempts to find glitches and exploits within a given game. That said, there is a wide variety of approaches to glitch hunting that require different levels of skill and will often produce different results.

Whether intentionally or unintentionally, a lot of speedrunners will do a bit of glitch hunting at some point or another. Some people are so dedicated to the search for glitches that they will call themselves glitch hunters. There is plenty of room for crossover between speedrunners and glitch hunters but that does not mean the two are synonymous.

While the process of glitch hunting can take different shapes, we can more-or-less break it down into some steps that may help you get started. (You might even recognize these steps from science class.)

Step One: Observations

Before looking for glitches, you can set yourself up for success by getting to know the game you are hunting through.

This can take the form of repeated playthroughs but it can also include research into what discoveries other people have made about the game.

When it comes to repeated playthroughs it never hurts to keep recordings of everything so you can look back on any points of interest you stumble upon.

My own experience glitch hunting my primary speed game, Moomin’s Tale, has mostly included repeated playthroughs and discussion with the small Moomin speedrunning community.

As is often the case, community cooperation is one of the best paths forward. Whether someone plays the game casually, speedruns it, or glitch hunts it themself, you never know what discoveries can be made just by sharing experiences and working together.

This step can also include the use of additional tools to investigate a game’s code. My own coding skills are amateur at best and I’m just getting started using CheatEngine and other tools to understand the inner workings of different games.

For more technically complex examples of glitch hunting with advanced and custom tools, check out our spotlight on the Prince of Persia speedrun community.

Step Two: Questions

While learning about a game, follow your curious impulses and ask questions.

Sometimes the physics in this game do weird things. Is there a way to exploit that?

Someone else just found a cool glitch. Are there other places that glitch could be useful?

This eagle boss is a poorly-coded, late-game run-killer. Is there a way to skip the fight entirely?

The main point is turning the information you have gathered into questions.

For example, in Level 5 of Moomin’s Tale, there is a door on the other side of a small gap the player cannot usually jump over. Later in the level, one of the branching paths brings the player to that door presumably to trick new players into accidentally falling into the old room and forcing them to replay a portion of the level.

Additionally, there are fruit items in Moomin’s Tale, some of which give the player a speed boost until the next time they take damage. This boost also improves jumping distance.

Lastly, the player character’s sprite sometimes flickers between falling and landing when running into the edges of certain platforms.

With all this information combined, I asked: “If I have a speed boost and correctly time a jump across the gap then jump a second time while I am semi-colliding with the other platform, will that allow me to cross the gap and skip part of the level?”

Step Three: Hypothesis

Once you have your question, it can help to turn it into a hypothesis by coming up with a guess and a way to find out if that guess is true.

The physics in this game can be exploited by chaining these actions in this location. I will test it by attempting to demonstrate the trick.

That glitch someone else found can be used in this other part of the game. I wiIl test it by trying the glitch set up in the new location.

By getting out of bounds, I can exit the fight with the poorly-coded, late-game run-killer eagle boss and leave it behind. I will test this by editing the game code to put myself out of bounds to determine if it is possible before I start looking for a way to get out of bounds.

Of course, in some cases, your hypothesis might just be “If I jump into this wall something interesting will happen. I will test it by jumping into this wall.” You don’t need a specific goal in terms of what you find. It’s going to differ between glitch hunters and games and game sessions. Sometimes it’s good to have a clear goal but it's equally important to keep an open mind to the possibilities.

Returning to my example from Moomin’s Tale, my hypothesis was: “If I have a speed boost and correctly time a jump across the gap then jump a second time while I am semi-colliding with the other platform, I can cross the gap and skip part of the level. I will test this by setting up a savestate before making the jump so I can make multiple subsequent attempts.”

Step Four: Testing

It’s experimentation time. Put that hypothesis to the test and record what happens. You can even take notes if you’re testing something more complicated or you just like to be meticulous.

It’s up to you.

Demonstrate the trick, try the glitch, mess with the game code. There isn’t much else to say but to get to it.

Step Five: Analyzing and Interpreting Your Results

So how did it go? Were you right? Were you wrong? Something else?

Maybe you were right. You proved the physics in the game can be exploited by chaining those actions in that location.

Maybe you were half-right. You found that the glitch someone else found can be used in this other part of the game, but the setup is a bit different and it’s tougher to pull off.

A lot of times, you might have gotten it completely wrong. Getting out of bounds does not allow you to skip the poorly-coded, late-game run-killer eagle boss.

In the case of my jumping experiment in Moomin’s Tale, I was successful. The jump worked and I was able to skip a short segment of the level!

##Step Five: Sharing, Repeating, Further Testing

In the first two cases above, you should of course share your discovery with other members of the community but it is often just as important to share when your discovery went wrong.

At the very least you can save someone else from fruitlessly repeating your exact actions and maybe other people can contribute new adjustments on your experiment or completely new ideas for

solving the problem you’re facing.

In a way, every experiment becomes part of the “Observations” stage of the next experiment! Regardless of what you find, you now know a little bit more about the game and can take that information into account while you continue your hunt for glitches.

In the case of my Moomin’s Tale experiment, I went on to double check if this new route was in fact faster than the previous one by timing both routes and comparing.

One Last Thing

Of course, this cycle does not need to be as formal as these steps might make it sound. Some of these steps are unconscious or happen quickly as you try one thing after another.

But hopefully, this structure can help if you are new to glitch hunting or hitting some kind of roadblock in your hunt.

Whatever you do, good luck glitch hunting and wish me luck finding a way around the poorly-coded, late-game run-killer eagle boss in Moomin’s Tale because it is trying my soul.

Hai delle notizie da condividere?
Invia una storia per apparire su Speedrun.com!
Storie correlate

When games don't provide what speedrunners are looking for, sometimes it's up to the community to step up to make the changes for themselves.

The choice of whether to select a more competitive run as the default subcategory versus a more entertaining one is and should be community-driven.

A No Reset Run can shake up your typical routine and give you some valuable late-run experience.

Want to get started with speedrunning, but aren't sure what game to try first? We've got some advice to get you started!

This month we are spotlighting Juh0rse, aka Juh0, a Singaporean speedrunner with a love of horror games who founded Lion Speedrunners Assembly.

This month, we're spotlighting The League of Puzzle Speedrunners, a community focused on running and racing action puzzle games from the Puzzle League series.