Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
January 18, 2017
arrowPress Releases

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Breaking Games Part II: Cheesing
by Joram Wolters on 05/02/13 12:06:00 pm

1 comments Share on Twitter Share on Facebook    RSS

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.



Abusing game mechanics or using meta game thinking to achieve an advantage over game enemies such as NPC's or other players.

Cheesing is something that requires less dexterity and luck than OTM, and more of a deeper understanding of the game's mechanics, and any opposing players. It's about finding weaknesses in the game's underlying economy, skill-tree, AI or other players and abusing these.

1. Meta Game Cheese

Most popularly known in the Star Craft games, where most cheese is a form of meta-game thinking. Meta game thinking is using knowledge from outside the game to influence the game. Most cheeses in Star Craft work on the assumption that the opposing player makes assumptions. For instance, a lot of the tactics in SC revolve around getting reinforcements at the location where you need them. This is all based on the normal model of play where players build a base and they build all their buildings and units at that base. Inexperienced players will always expect their opponent to display this typical behavior. That's why most cheese in Star Craft revolves around placing buildings outside your own base, or even inside an enemy's base.

Meta game cheese is the art of reading your opponents mind, or of being very aware of large trends in the game's multiplayer (also commonly referred to as 'the meta game'). This isn't to say that meta-game cheesing won't help you break single player games though. Breaking single player games by meta game thinking revolves a lot around being aware of trends again. These can be trends either within the game itself or within games in general. For instance, most experienced players will know that if they've just been given a powerful weapon for the first time in a game, a very difficult challenge is probably just around the next bend. Try to be aware of similar situations and narrative devices being used in games so you can become aware of these trends and then abuse them to your benefit.

2. Camping

Another form of cheesing which is very popular is camping, this well-known variety is when players in competitive multiplayer games stay in one place and kill anyone who comes close to them. As camping is a pretty familiar term, I will not describe it further here.

3. Eco-cheese

A fairly extensive subject which I think might warrant it's own article, but I'll try and keep it short.

A more complex form of cheesing is ‘eco-cheese’, where you upset the game’s balance by acquiring an absurd amount of in-game money as soon in the game as possible. Doing the same with XP boils down to pretty much the same thing, but it still gets its own segment further down.

If you want to break a game through eco-cheese, be mindful of anything that generates a regular income. Always go for upgrades or items which give a percentage bonus on your income. Eco-cheese of this or similar kind effectively ruined Far Cry 3, Saints Row: the third and XCOM: Enemy Unknown for me. In all three games I was maxed out on available gear and skills well before the games’ ends. Game designers have become pretty effective  in limiting eco-cheese by linking the weapons you can buy or skills you can use directly to how far you’ve progressed in the narrative. What they don’t understand is that these limitations are usually not strong enough. If you can buy all the weapons and skills you want as soon as they become available, you’ll never be as delighted to have finally unlocked them. 

A less effective form of eco-cheese – one that has lost nearly all of its effectiveness in recent games – is playing the market game. There are two forms of this, one of which requires less effort but more time and the latter requiring more effort but less time.

The first method is to simply pick up EVERYTHING  you encounter, and sell it whenever you can. Even though the objects you’ll be picking up are worth nearly nothing, they usually nearly have no weight. This works best in games where you get to teleport back to towns, like Diablo or Divine Divinity. Not very elegant, but you'd be surprised how effective it can be. If you don't mind walking around at a slow pace because of over encumbrance in games like Fallout or Skyrim, you'll get pretty decent results.

The more efficient method of playing the market is to actively find out which items are more valuable at one store than the other, and then find a way to abuse this mechanic. For instance in Skyrim, there’s a vendor where iron is really cheap, and a vendor close by where you can sell iron gloves at a reasonably high price. You can buy the iron cheap, craft iron gloves, sell them, walk back, buy iron with the profits and repeat. There is often a system in place which is supposed to limit this behavior. For instance in Skyrim, the shops have limited stock and restock only once a day. I love how these games usually also have a sleeping feature *hint hint*.  Doing this run serves a double purpose, as I’ll be discussing in the next segment. Something akin to eco-cheese is the abuse of any type of crafting mechanism. Games like the witcher, every Elder Scrolls game and even Far Cry 3 (again) where you are allowed to craft potions yourself are prone to be vulnerable to this. Again, this approach takes a lot of patience, as you’ll be spending a lot of time collecting pretty much every craftable resource, or very many of a specific kind.

04. XP and Skill-Cheese.

When it comes to using skills or items to further benefit your eco-cheese, always strive for synergy! Try and find items and skills which influence each other, or better yet, influence each others influence over each other. Having a sword which gives you HP every time you hit something suddenly becomes very effective when you have a skill which increases the attack speed every time you gain health. Players of Magic: the gathering or Diablo III should be well-familiar with the concept.

The cool part about games where crafting is a skill is that by any type of repeated crafting you're doing (such as the market example above, or crafting potions in Elder Scrolls) will train your crafting skill, further upsetting the games' balance.

The last tip I'll leave you with is a brief explanation on 'solo-statting' and 'dump-statting'. Solo-statting is expending all your XP points, skill slots or whathaveyou on a single stat or skill. Usually this will go horribly wrong right in the beginning, kind of even out after a while and near the game's end you'll be completely overpowered. Dump-statting is the opposite practice, where you completely ignore a single stat or skill-tree. The latter is nearly always a good practice.

05. Tricking the AI

The trick to tricking the AI in games is to forget everything you know about real-people behavior and to exclusively focus on the behavior you perceive in the AI enemies. Take notes of how AI agents respond to different situations and try to apply your findings in other situations.

The main problems for game AI's are Detection, Pathing and Remembering, I'll cover the three briefly.

Detection is something which is usually kind of weird in games. One time it could be that you are right in front of agents without being spotted, the next moment they'll find you instantly when you sniper one from 1200 meters with a silenced sniper (playing FC3 on 'warrior' is great fun). Try to remember that the AI is governed by game rules, not by anything real. The usual breakdown of an AI agents perception is something like this: The agent has two modes of perception, a sphere for sound and a raycast (straight lines) for vision. The sphere and raycast start out very small. The AI agent can hold a number of states, for instance Idle, Curious, Alerted and In Combat. The sphere and raycast will change size dramatically depending on which state the agent currently occupies. Most AI agents will default back to Idle after having had a certain state for a while without anything new happening. This allows for the infamous cheese where you kill off a guard, lay low for a while, kill the next one, etc.

Pathing is the bane of all AI programmers. I have never seen anything go wrong in games more often than pathing. Here's an example of me breaking Diablo III's pathing: Most pathing cheeses involve getting your character in a spot where the AI can see you, but the road to getting to you is far away. Poor pathing solvers will ensure the agent can't get to you. Often there are spots where you can jump on something in a manner which the AI can't follow. The AI agent will try an alternative (longer) route to get to you. You jump down, the AI takes the long way around again, you jump on the thing again etc. This will help in any case where you have a ranged weapon against melee opponents. This trick got me many achievements in Left4Dead(2) and allowed me to beat mammoths and giants way too early in Skyrim.

Remembering is something which still causes a little cognitive dissonance with players. I've heard many upon many players express their amazement at how they had been completely forgotten by the AI. I've covered the core of the problem at the 'Detection' section. An easy fix for it would be to have AI agents default back to the 'Alerted' state, instead of them going full on Idle.

To compensate for the lenght of this article I'll conclude the series with a shorter article next time: Using hacks, mods and console commands      

Related Jobs

Digital Extremes
Digital Extremes — London, Ontario, Canada

Weapons Artist
SYBO Games
SYBO Games — Copenhagen, Denmark

Senior Mobile Engineer(s) and Game Designer(s)
Tesla Motors, Inc.
Tesla Motors, Inc. — Los Angeles, California, United States

Developer, Digital Design Technologies
DreamSail Games
DreamSail Games — New York, New York, United States

Game Designer

Loading Comments

loader image