Ship Stats & Guides
Player Leaderboard
Captain Skill Calculator
Fun and Engaging Science Part Fivish7/24/2017 2:41:00 PM
Published By: LittleWhiteMouse
Previous Articles in this Series: One, Two, Three, Four
Thank you for joining me on another episode of Fun and Engaging Science -- the series where I explore detonation mechanics in World of Warships.  Like my last episode, we're going to be looking at module damage today, so this applies not only to making ships explode, but also knocking out modules like AA-guns, Main Batteries and Steering Gears.
In my last article, I discussed the blast radius of various HE attacks.  This allowed us to know when specific modules were being struck.  The next logical step was to try and isolate how much damage these attacks were doing.  From this, I could extrapolate how many hits would be needed to fully deplete magazines of various ships and thus properly calculate the odds of individual strikes causing detonations.  Internal modules such as steering gears, magazines and engines provide little in the way of feedback when they're taking damage.  To this end, I elected to target surface modules -- specifically the main battery and torpedo armaments of various destroyers and hammer them with a large HE shell and count the number of hits needed to knock them out.
I immediately ran into a very obvious problem:  The number of hits needed to knock out a turret was highly inconsistent.
It didn't matter what ship I was using or what the target was.  There was always a variance to the number of hits needed to knock out a main battery turret.  To really try and isolate this, I elected to use Furutuka with her single-gun 203mm mounts and fire point blank into the side of South Carolina's turrets.  This involved parking within 1km of the Battleship and slamming AP shells through them over and over and over again. 
  • A South Carolina's turrets have 15,000hp
  • Furutaka's AP shells do 4,700 damage.
  • My theory was that it should take about 4 hits to knock out one of South Carolina's guns barring any internal damage mitigation.
I collected hours and hours worth of this data, discovering maximums and minimums.  However, I rarely saw consistent results.  For example, it might take as few as six hits to knock out one turret while it might take as many as fourteen to knock out another.  This meant that there was something else at play here.  If the internal damage mitigation provided by the ship's guns was a flat value, then while it might not take four hits to knock out a South Carolina's turrets, it should be the same value over and over.  This simply wasn't the case, though.  The disparity wasn't just one or two hits, but easily more than double.  The specifics of this data was pretty telling.  As best I could figure, I was looking at a formula like this:
Module Damage = { Shell Damage } * { Module Protection Coefficient } * { X? }
I thought I was getting close to isolating the damage coefficient through a large sample size (my goal was 400 turrets destroyed). I didn't know if the module protection coefficient (whose existence Sub_Octavian had confirmed) was a fluctuating variable or if there was another variable interacting on top of this.  Maybe both were dictated by RNG.  However, I lost all of this hard work when an installation error on a patch update wiped out my client and all of my saved files. 
This was pretty heartbreaking.  However, Sub_Octavian came to my rescue the same day the client-wipe occured.  We had a nice chat about module mechanics and I got to ask him directly about what my experiments were uncovering:
There is an additional layer of RNG protecting modules from taking damage.  X does exist.
Our conversation summarizes to this:  Whenever a module is attacked, the base damage the attack inflicts is modified by RNG.  This can increase or decrease the amount of damage done.  So if an attack would do, say, 1,000 damage base, this could be modified up or down by a set percentage determined by an RNG parameter.  This is a similar system found in World of Tanks with things like shell damage, which varies as much as +/- 25%.  If we use my 1000 damage example above, RNG would mean that these attacks could do as little as 750 damage or as much as 1,250. The exact value World of Warships uses isn't being given, however it was made clear that the value is less than +/-100% (so no zero-damage or double-damage strikes).  This new damage value then interacts with module protection and armour protection.
Note that this RNG variance is rolled for each module individually when struck.  To quote Sub_Octavian:
"RNG is added to exclude the scenarios when, for example, a single shell splash-destroys all AA mounts in area at once. With it, only some will be destroyed, and some will survive for a few more hits. Or, for example, to exclude the type of thinking 'okay, if I shoot North Carolina turret with 3 shells, it's a guarantee to destroy it' - such scenarios are not desirable gameplay-wise."
 This furthers our understanding of how module damage (and by extension, detonations) operate in World of Warships.  Let's recap HE module damage:
  • When an HE shell strikes near or on a ship, an AOE effect is plotted.  HE shells never penetrate inside the ship and all explosion radii are calculated on the surface where they land.
  • All modules within this AOE effect are struck, regardless if the HE shell does damage to the ship or not.
  • All modules within the AOE effect individually roll to see how much damage they take based on the alpha-damage of the attack +/- an undisclosed RNG value.
  • The amount of armour between the point of origin can mitigate the amount of damage done.
  • Some modules have additional damage mitigation baked into their profiles.
  • The ability for HE attacks to punch through this armour comes down to a separate (and hidden) statistic Sub_Octavian refers to as "splash penetration".  This means that HE attacks that fail to do damage to the ship's hit points can still damage internal modules if this "splash penetration" is sufficient to reach them.
  • Main Battery, Secondary Battery, AA Guns and Torpedoes are destroyed when their hit point pools reach zero.
  • All modules have a chance of receiving a critical hit which increases as they take more damage.  Magazines have a base 1% chance to detonate with up to 70% of their HP remaining and this increases to 10% to detonate when their hit point pool reaches zero.  All critical hits are calculated after damage has been applied.

AP shells follow similar principles except the AP shell must stop inside a given module when it explodes. This will ignore any mitigation caused by armour, but the module's own protection may mitigate AP damage.  AP shells have no AOE blast and should only affect a single module at a time.  An AP shell that passes through a module in the form of an overpenetration causes no damage to the module in question.  This applies to AP bombs as well.

 This discovery means two things for detonations.
  1. It is more difficult to predict when a detonation may occur from individual attacks.
  2. It is increasingly important to prevent taking strikes to your own magazines whenever possible.
Until we can isolate exactly how much damage is being done per strike, players should take this new layer of RNG to heart.  Individual damaging hits to their magazines can potentially be much more of a risk than was initially thought should the roll be higher than expected.  We can easily calculate where a ship needs to be struck to put their magazines at risk and we're gradually chipping away at learning the exact numbers behind module damage.