Damage per second

From Liquipedia StarCraft 2 Wiki

Damage per second (DPS) is a term commonly used to describe the rate of damage inflicted over time. DPS is a Game Mechanic in Starcraft 2 and is a method for standardizing the amount of damage dealt by units. The term is an aspect of theorycrafting.

Explanation of DPS[edit]

  • Damage per second is the rate at which "damage" is dealt to an attackable object within the game over a unit amount of time (1 second). Every unit in the game has a different attack rate and damage dealt per attack. Attack rate is the time interval between every instance that damage is dealt. I.e. if a Zealot attacks a Stalker, the time until it next deals damage is 1.2 in game seconds. This time interval is a constant. This property is also referred to as the Attack Cooldown.
  • The amount of damage a unit deals is also constant for each kind of unit; however, abilities in the game allow this to be increased. An example is Attack upgrades. But in the instance that the damage is dealt it is a constant. I.e. if a Zealot attacks a Stalker, 16 damage is dealt with each attack (18 with the attack upgrade).
  • Since every unit has a different attack value and Cooldown, it can be difficult to directly compare their damage dealing potential. DPS is used as a way to resolve this issue by showing the damage a unit deals over a unit of time in order to simply compare two units damage dealing abilities. Therefore the higher the DPS value, the less time that one unit takes to defeat another unit.

Calculation[edit]

Damage per second is calculated by dividing a unit's Attack Value by the unit's Attack Cooldown.

Example: A Zealot has an Attack Value of 16 damage and an Attack Cooldown of 1.2 seconds. What is its Damage per second?
We use the equation Damage per second = (Attack value / Attack Cooldown). So Damage per second = (16 / 1.2) = 13.3 damage every second

Shortcomings[edit]

DPS has several key shortcomings.

First, DPS does not account for Splash damage. For example, the Archon's DPS is technically 14.3(+1.7), and this would be the case if the Archon were fighting a single solitary opponent. However, in the case which an Archon attacks an opponent closely surrounded by several opponent units; the Archon does damage to all of the opponent units. Depending on the number and spacing of the opponent units, the actual damage dealt may be many times what the DPS value of the Archon predicts.

Second, DPS does not take into account the attributes of the object being attacked. It is only a measure of the damage output of the unit doing the damage. DPS can be hindered by the opposing unit by their Armor Value, which decreases the Attack Value of the attacking unit by the same amount as the Armor Value of the unit being attacked, or the unit type, the tags given by Blizzard for each unit; if a unit has a bonus to Armoured and the attacked is not armoured then the DPS on that unit will be lower than the highest it could be.

Thirdly, because DPS is an average, it suffers from several key problems, [1] which can lead to some strange calculations when the attack value or cooldown of a unit is relatively high and the time frame being considered is relatively small. For example, consider a Siege Tank in Siege Mode. The tank does a base damage of 35 versus non-armored units and has a cooldown of 3 seconds, and therefore has a DPS value of 11.7 damage. Assuming an attack against a single non-armored opponent unit; the DPS value would suggest that 1 second after the tank began it's attack it would have dealt 11.7 damage. However in reality the tank would have dealt 35 damage 1 second after it's attack began. This is because the Siege Tank's attack occurred at T=0, and all damage is done at that moment.

Similar discrepancies will occur when using DPS to predict damage output at any interval greater than 3 seconds as well. Though as the amount of time the attack takes gets larger, the discrepancy between actual damage output and predicted damage output becomes smaller.

Finally, as some units and abilities deal one-time damage only, Damage per second cannot be calculated. These are the Baneling, Nuke, Yamato Cannon, and Seeker Missile.

Usefulness[edit]

As a measure of capability, it is more accurate than a simple damage value that does not reflect how often the damage can be dealt. As units have varying cooldown times between attacks, it is therefore a useful metric both to game players and game designers.

Damage and DPS by Unit[edit]

The following sortable table includes the DPS by unit with multiple permutations, DPS calculations are based on a unit's Damage output only.

Unit Name Secondary Attribute(s) Race Damage vs Ground Damage vs Air Bonus Damage DPS vs Ground DPS vs Air Bonus DPS
Zergling Adrenal Glands Zerg 5(+1) 7.2(+1.4) 8.5(+1.7)
Drone Zerg 5 3.3
Queen Zerg 4(+1) (x2) 9(+1) 8(+2) 9(+1)
Roach Zerg 16(+2) 8(+1)
Hydralisk Zerg 12(+1) 12(+1) 14.5(+1.2) 14.5(+1.2)
Infestor Fungal Growth, splash damage Zerg 30 30 40 vs Armored 7.5 7.5 10 vs Armored
Infested Terran spawn Zerg 8(+1) 8(+1) 9.3(+1.2) 9.3(+1.2)
Ultralisk splash damage Zerg 15(+2) 35(+4) vs Armored 17.4(+2.3) 40.6(+4.6) vs Armored
Brood Lord Zerg 20(+2) 8(+0.8)
Broodling spawn Zerg 4(+1) 6.2(+1.5)
Mutalisk splash damage Zerg 9(+1) 9(+1) 9(+1.000), 3(+0.333), 1(+0.111) 5.9(+0.7) 5.9(+0.7) 5.9(+0.7), 2(+0.2), 0.7(+0.1)
Corruptor Zerg 14(+1) 20(+2) vs Massive 7.4(+0.5) 10.6(+1) vs Massive
Spine Crawler Zerg 25 30 vs Armored 13.5 16.2 vs Armored
Spore Crawler Zerg 15 17.4
Baneling Attack Structure, splash damage Zerg 20(+2) 80(+5) vs Structure, 35(+4) vs Light
Probe Protoss 5 3.3
Zealot Protoss 8(+1) (x2) 13.3(+1.7)
Stalker Protoss 10(+1) 10(+1) 14(+1) vs Armored 6.9(+0.7) 6.9(+0.7) 9.7(+0.7) vs Armored
Sentry Protoss 6(+1) 6(+1) 6(+1) 6(+1)
High Templar Psionic Storm Protoss 80 80 20 20
Dark Templar Protoss 45(+5) 26.6(+3)
Immortal Protoss 20(+2) 50(+5) vs Armored 13.8(+1.4) 34.5(+3.5) vs Armored
Colossus splash damage Protoss 15(+2) (x2) 18.2(+2.4)
Phoenix Protoss 5(+1) (x2) 10(+1) (x2) vs Light 9(+1.8) 18(+1.8) vs Light
Void Ray Prismatic Beams Protoss 6(+1) 6(+1) 10(+1) vs Armored, +20% vs Massive 10(+1.7) 10(+1.7) 16.7(+1.7) vs Armored, +20% vs Massive
Void Ray, Charged Prismatic Beams Protoss 8(+1) 8(+1) 16(+2) vs Armored, +20% vs Massive 13.3(+1.7) 13.3(+1.7) 26.6(+3.4) vs Armored, +20% vs Massive
Carrier with 8 Interceptors Protoss 80(+16) 80(+16) 26.7(+5.3) 26.7(+5.3)
Interceptor spawn Protoss 5(+1) (x2) 5(+1) (x2) 3.3(+0.7) 3.3(+0.7)
Mothership Protoss 6(+1) (x6) 6(+1) (x6) 16.3(+2.7) 16.3(+2.7)
Photon Cannon Protoss 20 20 16 16
SCV Terran 5 3.3
Marine Stimpack Terran 6(+1) 6(+1) 7(+1.2) 7(+1.2) 10.5(+1.7)
Marauder Stimpack Terran 10(+1) 20(+2) vs Armored 6.7(+0.7) 13.4(+1.4) vs Armored, 10(+1) Stim, 20(+2) Stim vs Armored
Reaper D-8 Charges (buildings) Terran 4(+1) (x2) 30(+3) vs Structure, 9(+1) (x2) vs Light 7.3(+1.8) 16.7(+1.7) vs Structure, 16.4(+1.8) vs Light
Ghost Terran 10(+1) 10(+1) 20(+2) vs Light 6.7(+0.7) 6.7(+0.7) 13.4(+1.4) vs Light
Ghost Sniper Round Terran 25 vs Biological 25 vs Biological 50 vs Psionic 50 vs Biological 50 vs Biological 100 vs Psionic
Ghost Tactical Nuke, splash damage Terran 300 300 500 vs Structure
Hellion Infernal Pre-Igniter, splash damage Terran 8(+1) 19(+2) vs Light (Infernal Pre-Igniter), 14(+2) vs Light 3.2(+0.4) 7.6(+0.8) vs Light (Infernal Pre-Igniter), 5.6(+0.8) vs Light
Siege Tank Tank Mode Terran 15(+2) 25(+3) vs Armored 14.4(+1.9) 24(+2.9) vs Armored
Siege Tank Siege Mode, splash damage Terran 35(+3) 50(+5) vs Armored 11.7(+1) 16.7(+1.7) vs Armored
Thor splash damage Terran 30(+3) (x2) 6(+1) (x4) (splash) 12(+2) (x4) (splash) vs Light Air 46.9(+4.7) 8(+1.3) 16(+2.6) vs Light Air
Thor 250mm Strike Cannons Terran 500 83.3
Auto-Turret spawn Terran 8 8 10 10
Viking Terran 12(+1) 10(+1) (x2) 14(+1) (x2) vs Armored Air 12(+1) 10(+1) 14(+1) vs Armored Air
Raven Seeker Missile, splash damage Terran 100 100
Banshee Terran 12(+1) (x2) 19.2(+1.6)
Battlecruiser Yamato Cannon Terran 8(+1) 6(+1) 300 35.6(+4.4) 26.7(+4.4)
Missile Turret Terran 12 (x2) 27.9
Planetary Fortress splash damage Terran 40 20

References[edit]