how to balance my game?

Started by
9 comments, last by Shane C 9 years, 7 months ago

hi;

i didnt know really to put my post on which part in technical part or creative part but in the end i posted here.

as you know one of the most resonf of succes in games like titanfall, dota ,warcraft, and maybe l4d is balanced multiplayer in the meaning that characters behave diffrent but in the end player doesnt feel that other player is stronger than him.

for example in titanfall titans have better firepower and they have bigger amount health but they are slower and they are easy to aim in the other side pilots are faster with less fire power and health and are harder to aim. but in the end game make them in the end that in a same moment no one has advantage.

and there are alot of more examples as you know

im working on a tank project on unity that uses these kind of variaety as i said. but is that just playing with variables on program and script?

and how can i find they are compeletly balanced? i think varibles like speed and power are not rellated at all. i think i cant say

tank1: speed 20 power 80

tank2: speed 80 power 20

i think it doesnt work.

thank you for helping.

Advertisement
Play balancing requires play testing. Lots of it. Get people to playtest your game. That's how you balance it.

-- Tom Sloper -- sloperama.com

thank you mr sloper for answering but what is the touchstone of it? when can i say ok its balanced? who can say it is balanced?

A game is balanced when a player has a fair chance no matter which side he chooses to play.
When we were developing Star Trek DAC, we had 3 spaceship types. We knew it was balanced when there was no one spaceship type that everybody wanted to play (when the game was fair to players who used any spaceship type).

-- Tom Sloper -- sloperama.com

thank you. it was a very helpfull answer smile.png

Some reading material for play testing:

You might consider getting more data about your game by using a testing tool such as "Viewback", embedding a web server in your game, or by rolling your own analytics.

Hope that helps! :)

- Jason Astle-Adams

If you have implemented AI players already, then one option might be to let your AI do a big chunk of your testing for you.

Have them play thousands of matches overnight and see which tank type is winning the most games. Once you've got the stats balanced out for the AI, then you still need to do a ton of testing with humans, because your results from the AI may be biased due to quirks of the way the AI play, but it's still probably a decent starting point.


im working on a tank project on unity that uses these kind of variaety as i said. but is that just playing with variables on program and script?
and how can i find they are compeletly balanced? i think varibles like speed and power are not rellated at all. i think i cant say
tank1: speed 20 power 80
tank2: speed 80 power 20

i think it doesnt work.

I'm doing this mostly from memory, so please feel free to chime in anyone if this information is not accurate.

As far as game balancing comes, I am a firm believer of playtesting, but I believe there's a way to make most of the legwork upfront by creating a model. This also insures that your playtests have specific aims and are much closer to realization (because playtesting has a price).

Earth 2140 is a bad example of balancing: One faction is stronger, the other faster. Developers insured that the strongest faction was less mobile, disregarding how important mobility can be in an RTS.

The result is that, in multiplayer, everyone played the 'weakest' faction because its mobility allowed it to have the best concentration of force at all times.

Playtesting would've revealed this flaw, but someone with a keen understanding of an RTS or warfare would've also realized it.

To make a better estimate of proper balancing, the developers should've started with a smaller shift, bearing in mind the effect on economy and tactical positioning. It is not unreasonable to believe that a force with 25% less firepower and 10% extra speed can actually gain the upper hand through clever use of terrain, manoeuvres (kiting), etc.

A better approach to balancing would be as follows:

- Define your 'stats' and how they impact gameplay

- Define which core stats differentiate your factions (if any, as per Earth 2140 series)

- Establish a ratio that is sustainable for each faction

- Bring each other stats as an 'answer' to bring the balance closer together.

For example, assuming we have a 'fast' faction, and a 'strong' faction.

"Strong Faction"

- ('norm' = 100% for each stat)

"Fast Faction"

- Average unit Life 75% from 'norm'

- Average unit Armor 75% from 'norm'

- Average unit Firepower 75% from 'norm'

- Average unit speed 110% from 'norm'

Making theoretical fighting scenarios we might end up realizing that the 'strong' faction would lose most encounters as the enemy kites and moves into positions of strength, forcing concentration of force encounters and cutting reinforcements.

If this is the case, we can implement a new 'stat'.

For example, we might choose to grant the 'Strong Faction' a bonus to Range (or the fast faction a range malus). This would effectively give them the first shot in most encounters, also reducing the enemy's will to force kiting encounters (during which they'd end up taking significantly more damage).

As a general rule of thumb, you never want the fastest unit in the game to also have the longest range as this makes it effectively a kiter with no possible weakness.

Medieval-themed games remedy this by adding cavalry units which are essentially faster, and close on the archers at a speed that makes their otherwise somewhat mundane stats much more relevant (they might take 1 or 2 hits before closing on the archers and defeat them). The cavalry is a perfect example of a 'specialist' unit, something creating with a rock-paper-scissor mindset that serves a purpose.

So to answer your original question, simply 'adding' speed and firepower together and making sure they both equal 100 is largely insufficient.

My example of RTS might be far-fetched, especially given your current scope, but this is what I have the most experience in and I wasn't too sure what your game really is about without more details.

The model exemplified above can still be used, you may just have much fewer units to consider.

Balance does not necessarily have to be applied to the actual abilities of the tanks - you can also use external factors such as resource cost to manufacture or if in group warfare you might have an overall cost cap available for the initial combat setup, limited spawns and limited ammo can also be effective, terrain design can also be a decided factor on the types of tanks suitable.

Assuming you are making a FPS-like game with small battles between a few player-driven or AI tanks, you should consider how stats affect combat.

  • If a tank is slightly faster than normal, is it really better than a baseline tank? In which ways, and in which circumstances?
  • Can the environment or the kind of engagement make the advantage useless? For example, the fast tank could pass the whole battle in the same place in a sniper role or stick behind cover without serious movement.
  • Assuming being fast is good, what can you worsen to make the slightly fast tank fair? For example, slower turning can compensate faster straight movement. There's no reason to assume "power" (by the way, what is it?) is the right choice, only because it's the stat you have already designed
  • As the fast tank becomes faster, what are the qualitative tactical changes? Are they good variations, or boring easy strategies? For example, a ridiculously fast tank would be able to dodge artillery if there's enough room, which could be a good thing if obstacles make dodging challenging or a bad thing if nobody ever hits.

Omae Wa Mou Shindeiru

This topic is closed to new replies.

Advertisement