Thursday 7 February 2013

How to Balance a Game



How should you go about balancing a game?

There are a few schools of thought to this. The predominant answer that I have found is that it depends on the nature and genre of the game.

PVM

As a rule of thumb, when designing a PVM (player versus monster) game, the idea is to go big or go home. Does a certain boss feel too easy? Try doubling his stats. PVM focuses mainly on feel, and as such, minor refinements and tweaks tend not to make a huge difference. Many developers have ended up doubling those stats again.

Experiment with a wide variety of values, and see what 'feels' right. These types of games are mostly about the experience, rather than showcasing a particular mechanic. If something is not working exactly how you want it, consider just not emphasizing that element in the game; by this I mean if the jumping mechanic in your game is awkward, you have two options:
  1. Refine your jumping system until you have something smooth
OR
  1. Just don't include any jumping puzzles
Is jumping really important to your game? Do you really need to refine it, or can you simply ignore it? Most PVM problems can be solved by altering the CONTENT of your game rather than its MECHANICS.

PVP

If you are designing a PVP game, refinement becomes much more important. Monsters may not mind that a particular spell or combo is overpowered, but when game-breaking combos are unleashed on other players, they tend to take offence.

Generally speaking, many of your more vocal testers (or players) will complain about whatever beat them to the point that you have to take their words with a grain of salt. It is also important to consider both sides of an argument, sometimes powerful units are necessary and intrinsic to a balanced matchup. I personally don't like Marauders in Starcraft, but I acknowledge that the game would not be the same without them.
Rock is OP. Paper is fine.
-Scissors
This is comparing the balancing philosophies behind WoW, DotA, and LoL.

Depending on whether or not you are designing a video game (and indeed what system you are designing it for), you may or may not have the luxury of patching your games; that is to say, fixing anything after the initial release. Patching is usually not required, unless you are aiming to have an extremely high level of competitive play, although it is fairly standard among video games nowadays.

I maintain the opinion that what Blizzard is best at is Patching games, outstripping every other game producer  by far in that regard. Every single patch in Diablo 3 was a godsend, after a fairly poor initial reception. Starcraft is widely considered the best balanced RTS in history, after years of updates and feedback from their community. I can't personally speak for WoW, but I feel that its 10 million+ population speaks for itself.

If you do not have the ability to patch your game, then you had best make sure that you get it right the first time. Donald X. Vaccarino playtested for over a year and played thousands of games before the first release of Dominion.

You need to consider every element, and its every possible interaction with every other element in order to claim to have fully tested something. A daunting task, and it is usually best to have as many testers as possible for a project. Usually, you want smart/skilled players, who will notice possible interactions quickly, and therefore test much more efficiently than simply brute-force testing every possible interaction. You should generally balance for as high-level play as possible.

PVP games must have both well balanced MECHANICS as well as CONTENT. The only regard that it is more flexible than PVM is that human interaction can naturally balance out minor things.

Is one element really powerful? Have players bid on it; this helps to ensure that whoever gets it has to pay whatever everyone feels that it is worth.
Is there potential cooperation among the players? Alliances or cooperation allow underdogs to topple a dominant player.



No comments:

Post a Comment