Originally posted 8/16/16 on Kickstarter. The problems described here have been fixed in the current version of the demo.
Hi! Today I'm gonna talk about combat balance and strategy, and give you some more insight into what will be changing with this next update of the demo.
I think one of the main strengths of Lonely Star is the fast, kinetic, complex combat system. It's something I've poured a ton of effort into over the past few years. I'm a big fan of games that offer something more than novelty, games that still provide a dynamic challenge even when the player knows what to expect.
So it's kind of a problem that the winning strategy, in any melee fight, is to run away, crouch in place, and hammer the attack button over and over.
OK-- this is a constant struggle in game development. The developer knows how everything works, knows all the strategies and angles, so, to them, the challenge quickly becomes trivial. Then there's a temptation to keep making the game more and more difficult, and therefore more and more inaccessible to the new player. It's something I have to be careful of.
Plus, encounters in Lonely Star are usually designed to pose more than one type of threat to the player. In the scenario you see above, if one of the swordsmen's weapons was replaced with a crossbow, they'd just stand back and shoot the player.
But still! This isn't a problem I can afford to ignore. It looks stupid, and more importantly, it's "degenerate"-- a strategy so effective that it destroys the complexity of the game. Go ahead and try it; once you see how well it works, you're probably not going to go back to ducking and dodging.
There's a certain amount of smoke and mirrors to any game like this; by throwing a ton of complexity at the player, it tricks them into thinking that they have to engage with it. Taekwan Kim writes:
What a degenerate strategy does, then, is cut through all the obfuscation to expose the underlying artifice in its starkest form. It reveals the implicit as only being implicit, that the rules only have meaning because the player chooses to submit to their arbitrariness, and chooses to believe in a context in which they aren’t arbitrary.
Especially in a difficult game, the player will be thrilled to find an easy, successful approach, so they'll probably keep using it even at the cost of making the game boring to themselves. That destroys the game's ability to offer anything but novelty; pretty soon, the player feels like they've solved it, and it's about as compelling as tic-tac-toe.
So the first thing I need to figure out is, what exactly is the problem? Look, it doesn't work if the player is standing instead of crouching:
A crouching attack has less range, less power, and a longer recovery than a standing attack, and the player can't dodge during it, all of which is balanced by a single very powerful property: it can sweep the enemy's legs and stun them.
When you play a good singleplayer action game, at least one based on dynamic close-range combat, you'll notice that (at least after the first level or two) the game often pits the player against enemies whose attacks are not interrupted when they are hit by the player. Armored enemies in Lonely Star work the same way, but in the demo, there are no enemies with armored legs, only armored torsos.
I could fix my problem here by taking out the special properties of the crouch attack. That's not what I want to do, though. Sacrificing a lot of complexity to save whatever's left over doesn't seem like a good approach. The crouching attack works great as a countermove, especially against an enemy with an armored upper body:
But it's not foolproof. The player has to make sure they're not dodging right into another enemy's attack range, or they'll be helpless during their attack recovery:
I want to preserve this stuff. So here's what I'm working on instead:
Two things changed here, and they're both in the enemies' behavior. The crouching attack itself hasn't been changed at all.
1) I made the enemies aware of their reach advantage. Outside of a few early-game, easy enemy types, there's no reason for them to blunder straight into an attack when they can strike safely from outside that range.
2) I made the enemies aware that, when the player misses an attack, that's an opportunity to counterattack. If the player misses a standing attack, they can still dodge afterwards; if they miss a crouching attack, they might be out of luck. That keeps the risk-reward balance of the different attacks in check.
So even if the player has the range advantage, not the enemy...
...the old crouch attack spam strategy comes out neutral, not in the player's favor, because the enemies are at least rudimentarily aware of the rhythm of the battle.
Now this fix makes the game harder, so I've modified a couple of other things to balance it out:
1) Enemies now may hesitate when they're within striking distance of the player
2) Enemies will hesitate in their approach when the player is swinging a weapon around, even if they're outside of its range. This instinctively makes sense, right? I like that there's a flavor of intimidation to it, a little suggestion of a thinking, feeling consciousness inside those 5x5 pixel skulls, not just an insectile snarl of code. In general, enemies are more prone to hesitation when the player gets aggressive, and more aggressive when the player runs away.