Balancing stuff

It's something that I was always going to have to do, but it's time to talk about balance.

I have just about finished all the coding associated with gameplay mechanics, and am about to start, in earnest, the aesthetic feel of the mod (which I have talked about before, although some things have changed, ye shall see!).

It's been apparent for a while that balancing this mod will be quite difficult, and there most likely never be a possible configuration that something this diverse can be completely balanced. But damn it, I'm going to try.

And in this post, I'll just dump down ideas I have for that.

The first sort of balance method is using what I shall call "Move to Range Calculations". This is very simple, and most likely inaccurate in most cases. However, it can work for some cases, particular with specific unit counters. If a unit can beat an equivalent cost of its intended counter, then something should probably be done.

The way it works is you have an imaginary battlefield, with the equivalent cost of two units arranged on either side. The battlefield is as wide as the longest ranged weapon's range. The shorter ranged unit (or units) move towards the longer ranged the unit, with the longer ranged unit firing as fast as it can.

This is obviously very simple, and doesn't work for things like artillery, which miss moving targets, and doesn't account for anything like novel maneuvers or anything like that. It also doesn't work for support units that do no damage. It becomes more complex, although not impossible, for weapons that scatter and have a blast radius.

It does work for very simple things, though, for instance HAT vs Rush AT Infantry. The calculation shows that the HAT would kill one Rush AT Infantry before being killed by the other five. Definitely, a hard counter. It also shows that in HAT vs Rush Light Tank, the HAT would win after being reduced to 330 HP.

Obvious disreprencies occur if, say, Rush AT Infantry were found to be beating equivalent costs of Turtle Anti-Infantry Infantry, or Stealth Advanced Anti-Infantry Infantry.

Anyway, this method's major weaknesses are obvious, and its major strength is that it can be done entirely in the head, without the need for firing up the game.

Of course, unit match-ups are only a small part of balance. All the unit match-ups could be balanced, and yet some team would be better than another, as things like ease of resource collection, ramping up eco, strategic concerns etc.

For instance, while a HAT will win most armoured disputes, it is strategically weak as it cannot relocate to deploy its considerable firepower. Hence the best way to determine balance, if the most costly, is by actually playing lots of games, and then examining how various units and abilities change the game, probably while watching replays. You don't want to mistake one player's incompetence for their units being underpowered.

Once you have identified the unit that is balanced or unbalanced, you can go about the problem is several ways. It's obviously not going to be every aspect of the unit that is unbalanced, so you have to identify what it is causing it. The unit will be too fast, too powerful, too tough, or maybe too cheap and plentiful.

The various was of going about that are quite varied. If a weapon is too powerful drop its power, or increase its reload time. Maybe give it a pre-attack delay. Shortening the range could also work.

One of the things that will most likely crop up during playtesting is the three team's income will probably not be balanced. So it needs changing. Turtle is very easy, as I can change the number of collectors spawned, the speed at which they travel, how much money they pick up per trip... That's about it. But quite a lot can be changed. With Rush, it is slightly more complex, as I can simply change the amount of money each building gives you. However, a significant portion of the Rush economy is map dependant.

So, once I've made a map, with that, I can control how many buildings in total are on the map, how many buildings are in a cluster (important, as 10 buildings in two clusters are more defensible than ten buildings in 4), where the buildings are located in relation to choke points etc. All of this, I'll figure out when I go over map design.

With Stealth, it goes a little backwards. Normally, one would be able to change how much of a unit's price they get in return. However, for this, it means that a Stealth vs Stealth game would occasionally run down to almost nothing (scratch occasionally). So with Stealth, to make their economy "worse", you make everything on the Stealth Faction more expensive, and to make their economy better, you make everything on the Stealth Faction less expensive.