Thursday
Feb282013
Please, don't make math a core mechanic of gameplay.
Thursday, February 28, 2013 at 10:12PM We've established a few things in the past few weeks... like, one a week. Number one most important, though: meaningful choices in games are important. But choices aren't necessarily meaningful if you don't understand what you've decided to do. Sure, games require a certain amount of mystery over how things will turn out--that's a matter of future meaningful choices playing out meaningfully--but players have to understand how the immediate factors about their current choice are different from the immediate factors of another choice. You can't make meaningful choices without having all of the information presented in a human-friendly way.
Yep, human-friendly. Video games are inherently more math-based than board games, because computers are excellent at doing computations--so smart designers take advantage of that strength and allow for a large quantity of complicated equations to take place simultaneously, giving players high-speed action. While the computer handles the math, the (human) player still needs to know what math is taking place--for example, in real-time strategy games, you clash huge armies against each other. The computer can tell you who wins, but players want to know which troop types are going to beat each other so that they can create deadlier armies. All troops are good at killing some opposing troops, and not so good at killing other opposing troops. Players can realize these strengths and weaknesses through play, but they can also look at the armor and damage of given troops to get a sense of what works best.
Much of my RTS experience comes from Blizzard's games, so we'll look at them for some examples. We'll start with Starcraft: a dragoon shoots a high-powered ball of energy at opponents, meaning it's very effective against armored troops. However, a zergling swings its claws rapidly, butchering troops with little armor in seconds. The weapon damage and armor of a troop is given when the player clicks on it--the dragoon deals 20 damage, and the zergling deals 5 damage. Dragoons have 180 hit points, and zerglings have 35 hit points--so two dragoon shots will kill a zergling, and 36 zergling attacks will kill the dragoon.
...right?
Oh, we forgot about armor! Every point of armor your target has decreases damage dealt by 1. So, 20 - 1 = 19, and 5 - 1 = 4. That means 1 armor reduces a dragoon's damage by 5%, and a zergling's by 20%. That's such a huge difference! And dragoons have 1 armor, and zerglings have none, so you'll need a whole bunch of zerglings to take down each dragoon, right?
HAHAHAHAHA no.
Starcraft featured three damage types (concussive, normal, and explosive), and three troop sizes (small, medium, and large), which all modify how much damage is actually dealt to your target.
Did the game mention this ever? Anywhere?
Hell no! You just had to play the game to realize that dragoons, which by all appearances should roast zerglings... got sploded hardcore instead. Dragoons deal explosive damage, and small troops (including zerglings) take 50% damage from explosive attacks. So, it's not two hits from a dragoon--it's four. Add to this the fact that zerglings attack significantly faster than dragoons, and your math on how many zerglings will outmatch a dragoon gets turned entirely on its head.
While Starcraft's designers did a great job of giving players easy numbers to crunch, they hid other numbers from players (damage types and sizes), so players had to do research, and still had to do some math to realize how effective their troops would actually be.
The next Blizzard RTS was Warcraft III, which had more features--there were four playable races (vs. Starcraft's three), and utilized hero troops that players could level up to cast powerful spells--and it ramped up the un-friendly math by many whole a lots.
It featured six armor types and SEVEN weapon types. There are as many armor types alone as damage and size modifiers in Starcraft! This makes understanding counters significantly more difficult for players; however, Blizzard created a unique icon for each weapon and armor type, which made recognizing patterns slightly easier. That's about all they did to improve understanding choices, however, because they used the worst numbers EVER:
Much of my RTS experience comes from Blizzard's games, so we'll look at them for some examples. We'll start with Starcraft: a dragoon shoots a high-powered ball of energy at opponents, meaning it's very effective against armored troops. However, a zergling swings its claws rapidly, butchering troops with little armor in seconds. The weapon damage and armor of a troop is given when the player clicks on it--the dragoon deals 20 damage, and the zergling deals 5 damage. Dragoons have 180 hit points, and zerglings have 35 hit points--so two dragoon shots will kill a zergling, and 36 zergling attacks will kill the dragoon.
...right?
Oh, we forgot about armor! Every point of armor your target has decreases damage dealt by 1. So, 20 - 1 = 19, and 5 - 1 = 4. That means 1 armor reduces a dragoon's damage by 5%, and a zergling's by 20%. That's such a huge difference! And dragoons have 1 armor, and zerglings have none, so you'll need a whole bunch of zerglings to take down each dragoon, right?
HAHAHAHAHA no.
Starcraft featured three damage types (concussive, normal, and explosive), and three troop sizes (small, medium, and large), which all modify how much damage is actually dealt to your target.
Did the game mention this ever? Anywhere?
Hell no! You just had to play the game to realize that dragoons, which by all appearances should roast zerglings... got sploded hardcore instead. Dragoons deal explosive damage, and small troops (including zerglings) take 50% damage from explosive attacks. So, it's not two hits from a dragoon--it's four. Add to this the fact that zerglings attack significantly faster than dragoons, and your math on how many zerglings will outmatch a dragoon gets turned entirely on its head.
While Starcraft's designers did a great job of giving players easy numbers to crunch, they hid other numbers from players (damage types and sizes), so players had to do research, and still had to do some math to realize how effective their troops would actually be.
The next Blizzard RTS was Warcraft III, which had more features--there were four playable races (vs. Starcraft's three), and utilized hero troops that players could level up to cast powerful spells--and it ramped up the un-friendly math by many whole a lots.
It featured six armor types and SEVEN weapon types. There are as many armor types alone as damage and size modifiers in Starcraft! This makes understanding counters significantly more difficult for players; however, Blizzard created a unique icon for each weapon and armor type, which made recognizing patterns slightly easier. That's about all they did to improve understanding choices, however, because they used the worst numbers EVER:
| Light | Medium | Heavy | Fort | Hero | Unarmored | |
| Normal | 150% | 100% | 100% | 50% | 100% | 100% |
| Pierce | 75% | 100% | 150% | 35% | 50% | 150% |
| Siege | 50% | 100% | 100% | 150% | 50% | 150% |
| Magic | 100% | 200% | 100% | 50% | 50% | 75% |
| Chaos | 100% | 100% | 100% | 100% | 100% | 100% |
| Spells | 100% | 100% | 100% | 100% | 70% | 100% |
| Hero | 100% | 100% | 100% | 50% | 100% | 100% |
Mmmkay, they're not quite as bad as I remember--but Starcraft utilized multiples of 25%. Four options total, and nothing ever dealt more damage than was listed in the tool tip. Here, we have 35%, 50%, 70%, 75%, 100%, 150%, and 200%. Seven different values, without any common denominator, with each type modifying damage in a seemingly arbitrary fashion across armor types. And did I mention that units did a range of damage? An archer might do 13-17 damage, or a ballista might do 23-117. So players also need to find the average damage dealt before they can do any other calculations... and they need to do a lot of calculations.
Because here's Blizzard's explanation for how armor affects damage:
"For positive Armor, damage reduction =((armor)*0.06)/(1+0.06*(armor))
For negative Armor, it is damage increase = 2-0.94^(-armor) since you take more damage for negative armor scores."
Uh-huh. Totally friendly.
Fortunately, Blizzard took a huge step forward with Starcraft II. All troops deal the damage listed in the tool tip, no matter what they are attacking*. It's that simple.
Asterisk? What?
All right, fine. It's not utterly simple--but it's still easy to understand. Troops do have modified damage, but those are limited strictly to bonuses against one type of target. For example, an Ultralisk deals 15 damage with every swing against anything, except that it deals 20 damage to armored troops. There's not even any addition involved; it deals 15 damage to everything that isn't armored, and 20 damage to armored targets. No 15 (+5 vs. armored). Just 15, or 20 vs. armored. Nice, straightforward, pre-mathed integers. Treating numbers in this way makes it easier for players to understand the variables in a game, and allows them to make better-informed decisions. Sure, it has some setbacks--Warcraft 3 used such a convoluted system because that system allowed it to fine-tune how effective troops were against a wide array of opposing forces, and that isn't possible with Starcraft 2's system--but I played Warcraft 3 for longer than Starcraft 2, and I definitely never knew the War 3 system by heart. Not even close. I would feel much more comfortable stepping back into SC2 today than I would War3, largely because I could re-learn the game more easily.
All of this boils down to each game's barrier to entry. When a game has a low barrier to entry, players understand the basics of the game more easily, and actually know how to win or why they lost. Easy numbers makes the barrier to entry lower, which is important in a genre like real-time strategies, which generally have a high barrier to entry due to their complexity.
Reader Comments