Arma 3 weapon damage

May 11, 2015

I decided to test some common Arma 3 weapons and their abilities to kill. This test was conducted using build 1.44.130654 without modifications.

Update (May 12, 2015):

With the help of Reddit’s Arma community, what has been found is that Arma 3’s handling of ballistic damage may actually be more sophisticated than its predecessors.

Unlike previous Arma games, Arma 3 appears to take into account angles and depths of penetration, increasing or decreasing the damage it issues characters based off of how much flesh a bullet should have torn through.

This would mean the results below — found in a test that assumed Arma 3’s simulation of damage was similar to Arma 2’s, where a foot is a foot — are fundamentally flawed.

If Arma 3 does indeed take angle and depth into account, I believe the change is positive, as it actively rewards players for aimed shots.

A game developer also commented that the test below may be faulty because the characters I specifically tested on, Virtual Reality characters, may handle wounds in different ways than actual characters.

Hypothesis

Weapon damage is dealt inconsistently in Arma 3.

Objective

Find peculiarities in how damage is received by game characters.

Method

Fatally administer bullets into as many human targets as possible. If a target is missed, do not count the miss toward test results. While testing specific body parts, if a body part other than the one being tested is mistakenly hit, ignore all results from that particular target.

Caveats

Targets were fired upon from three metres distance. The farther a bullet travels in Arma 3, the less damage it does upon impact. Damage decays differently for each weapon in Arma 3. This test only considers a select number of weapons and their immediate ability to produce damage.

Only one set of body armor was tested. Different pieces and combinations of body armor offer different protection and would provide different results.

Further gameplay considerations such as recoil, rate of fire, weapon stabilization, accuracy, player skill, and so on, were outside the scope of this test. Only damage was considered, not the ability to produce it.

The majority of weapons were not tested. I personally believe that Arma 3 has two kinds of weapons: ones that produce a lot of damage (anti-material rifles, sniper rifles, battle rifles, machine guns), and ones that produce little damage (assault rifles, submachine guns, pistols). I only tested what I consider to be Arma 3’s “line-infantry” weapons.

Disclaimer

All tests conducted and calculated by untrained amateurs with abundant free time. Attempt if you want, I guess.

Results

Weapon damage table

You may notice I tested the TRG-20 instead of the TRG-21. That was a mistake on my part. However, if weapon charts in the Virtual Arsenal are to be believed, there should not be an observable difference between the two within the confines of this test.

All results in the tables above were calculated into an arithmetic mean. As such, the results seem fairly normal. However, the objective of this test was to find peculiarities in how weapon damage is received, not the average damage produced by each weapon.

Peculiarities

In my test, headshots were only fatal 93% of the time. I briefly did some extra testing on the side and found that if I removed player helmets, headshots would be fatal 100% of the time with the rifles I tested. Further, with helmets removed, headshots are fatal 89% of the time with the Sting and 87% of the time with the P07.

Headshots

While testing headshots, it became immediately clear to me how inconsistent Arma 3’s handling of damage is. All headshots were administered through the paranasal sinuses, so no contact with the helmet should have been made.

The tested weapon with the most inconsistent ability to kill is the Mk20, which needs four to twelve footshots to kill a target. But while the Mk20 is the most egregious example of an inconsistent weapon, in practice, every weapon has an inconsistent range of effect.

Ranges

Weapon ranges table

Analysis

In general, weapon damage feels fairly arbitrary. I cannot think of many reasons why targets in identical conditions should have vastly different chances to survive wounds.

Without knowing the code at work behind Arma 3, the only possibility I can imagine involves a rudimentary calculation influenced by random chance.

I can think of at least six variables at play, maybe seven: the bullet’s calibre, the bullet’s velocity, the target’s health, which of the target’s hitboxes were struck, that hitbox’s armor, and an arbitrary dice roll or two for a critical hit potential or a chance to negate armor.

I can only assume chance was introduced because Arma 3, understandably, cannot simulate everything. How does the game simulate interesting ricochets on boring flat polygonal objects like the ground or a wall? It spins them, throws a little randomness in. Likewise, how does the game handle things like exit wounds or bullets changing directions while moving through something? Again, the game spins them with a little randomization.

The part I personally find frustrating is how Arma 3’s designers have added that random inconsistency to player damage as well.

For me personally, there is definitely an isolation effect at play. I was driven to conduct this test because of the many instances where I have hit people way too many times, sometimes only to have them turn and kill me instantly with their return fire. But after collecting the data, unless my copy of Arma 3 is fundamentally broken, I cannot imagine I am alone in my experiences.

Closing

In essence, I believe the results of my test adequately demonstrate that engagements in Arma 3 are a gamble. And I do not think they should be.

As I see it, there are three paths Bohemia Interactive can choose:

  1. Do nothing: “Works on my machine — WONTFIX”
  2. A configuration option or BIS_fnc module or something that users or mission makers may use to turn specific parts of the simulation off with ease, if they so desire, while still meaningfully retaining the benefit of equipping body armor to characters.
  3. Reconsider how weapons and armor interact with one another so that the range of potential outcomes is much smaller and more consistent.

Update (May 12, 2015):

With the help of Reddit’s Arma community, what has been found is that Arma 3’s handling of ballistic damage may actually be more sophisticated than its predecessors.

Unlike previous Arma games, Arma 3 appears to take into account angles and depths of penetration, increasing or decreasing the damage it issues characters based off of how much flesh a bullet should have torn through.

That would mean the results above — found in a test that assumed Arma 3’s simulation of damage was similar to Arma 2’s, where a foot is a foot — are fundamentally flawed. That would also mean my memories of sloppy gunfights are a result of me playing Arma 3 as if it were Arma 2.

If Arma 3 does indeed take angle and depth into account, I believe the change is positive, as it actively rewards players for aimed shots.

A game developer also commented that the test above may be faulty because the characters I specifically tested on, Virtual Reality characters, may handle wounds in different ways than actual characters.