Breaking Games Part 1 - Off the Map

"Ik ben Joram, en het moet kapot" - a phrase coined by a couple of friends of mine - roughly translates to "My name is Joram, and it has to break". My friends and classmates started using this phrase somewhere in my second year studying game design at the Utrecht school of art. The reason they've been using it is because well.. I break stuff. All the time. This includes electronics, appliances, people, furniture and most importantly, games. I am usually not aware of it but apparently I display behavior in games which designers had not anticipated upon. Which causes me to frequently fall through the geometry, get stuck or suddenly fly up into the air.

At the root of this is when me and my friends had been playing Halo for a long time, we had gotten bored with finishing the game for the umpteenth time and we were quite done with battling each other as well. We did however; find great joy in breaking the game, something for which Halo is an excellent game, as you can see in this vid.

After doing 'halo tricking' for a couple of years, the mentality of playing a game to break it started seeping into my regular playing style. I naturally played games in a way which was sure to break them.

After digging deep and figuring out what it was about my playstyle that caused me to break games, I finally distilled a list of valuable tips to help you start breaking games for yourselves. This was going to be one long post, but as I’ve started writing, I found there was more and more to write about, so this is now part one of a three part article. This - the first part - will give some general guidelines to breaking games, and will cover the first type of game-breaking. So without further ado, here it is.

I. General tips

1. Have a goal.

To start breaking games, we must first understand what it means to break a game. As shown quite competently in the Halo example above, the most obvious ways are to escape the game’s intended level geometry, called OTM (Off The Map). But there are many more ways, of which I will discuss the following in these posts: OTM, Cheesing, min/max’ing, hacking/glitching. We’ll get back to the definitions later as we discuss each method.  The point of knowing what type of breaking you’ll be doing is because without a clear goal, your chances of finding a glitch or bug diminish substantially.

2. Understand the design(er)

Secondly, you need to have some understanding of the underlying structures of a game, and its design. Not only of the game’s engine and programming, but maybe more importantly of the game’s design. If you understand what behavior the designer expected to see in the players, it is easier to come up with ways to play the game in a way that it won’t be able to handle correctly. For instance, in most shooters, the designers don’t expect a player to throw a grenade under themselves. This is why in most shooters, if a grenade causes knockback, and you time the grenade detonation well, you can probably get OTM. - At least, as long as the blast doesn't kill you.

3. Good things to those who are patient.

Thirdly, you need to be really patient. Like, really – really -  patient. It varies per type of breaking you want to do, but breaking games is roughly 90% luck, or 90% repetition. It took me and a friend roughly thirty minutes to melee a warthog through a corridor in Halo. When we got it through, we tried to do a jump, which failed, so we had to start all over. I’m guessing it took us about twenty tries before we made the jump. Another example is in Fable, where you can ruin the game’s economy by repetitively buying, upgrading and selling houses for a minor profit. Repeat the process often enough, and you’ll be swimming in gold.  Many other RPG’s will not expect someone to literally pick up every herb they see and craft an insane amount of performance enhancing potions (the witcher, skyrim).

4. Location, timing, timing.

Pretty much everything in life is about being in the right place at the right time. Breaking games exemplifies this, it's about being at the right pixel at the right frame. This reiterates why the above (being patient) is so important, you simply can not get the level of precision you need at the first try. It's easier to just try the jump/run/glitch a million times than to try and plan it perfectly. This doesn't let you as a player off the hook though, you need to have a clear understanding of what you're trying to achieve, and carefully choose the place and time at which you're going to do it. Make sure you're as comfortable as possible with the game's controls, a lot of this is about not panicking when it counts and nailing jumps.

II. Different types of breaking games

To escape the game’s intended play area. Fairly straightforward, most common in shooters and (3D) platform games. The best way to start is to get a good grip on the game’s allowances and restrictions. By this I mean what will the game permit you to do and where it is trying to restrict you. Do explosions cause knockback, or do they kill you instantly? Is there fall damage? Where are the real level boundaries? Key here is to adopt a kind of ‘matrix vision’ when looking at level geometry. Try to see the collision boxes of objects instead of their visual counterparts.

Other important things to look for are things that generate force (explosions, moving level geometry, vehicles, physics objects such as boxes and barrels, etc.). Especially things that will force the game engine to calculate complex collision such as a stack of barrels rolling together. Adding a game character here will nearly always upset the mix. Another common way to generate force is to crouch under a ledge and repeatedly spam the jump button. Many game engines will employ an 'addForce' parameter, as long as you're stuck under the ledge, the force will keep building up, allowing for one 'super jump' when you leave the ledge.

Something which is a blessing in situations like these is if there is anything that will affect player health. This is because most game characters will take damage if they get hit by game objects or if they take a large fall. If there are medkits, shield boosters or other effectors on player health, try to time picking them up with the exact moment of taking the damage, the game code will usually just contain something like ‘fill up the health/shield to full’, regardless of how much damage the character took. Having this one frame of immortality is pretty vital in breaking games.   When trying to get to low places, try and find something which will slow down your descent. Sloped hills and the usual collision-free scenery props won’t help you here. Try to find a physics object like a box or barrel which you can throw down with you, and try to have it halt your descent half way down by jumping from it. In games like GTA or Saints row it helps to first dive down a cliff in a car, jump out at the right time and try to jump forward off of the car, it’ll translate your momentum in a different direction.

The last thing to remember when going OTM is to try and find places where different collision meshes come together. Many level designers get pretty lazy when defining the outer boundaries of their maps, and because the collision meshes for these boundaries are usually invisible they’re usually not put together too neatly. It is in these places where you can try and glitch part of your game character’s collision in between the two boundary boxes. Especially games that don’t use box collision but more complex mesh-based collision tend to have trouble here. This concludes pretty much all the general stuff I have to say on OTM, I'll leave you with a short vid from my playtesting article in which I employ some of the principles described above. The next article will concern the art of cheesing. -No not sniffing cat's urine, you'll see.