Jump to content
Sign in to follow this  
You need to play a total of 5 battles to post in this section.
MaxL_1023

Ricochet Mechanics and Spaced Armor

23 comments in this topic

Recommended Posts

422
[KRAB]
Members
840 posts
7,295 battles

I was looking at the armor models for the Missouri vs Iowa (along with a couple others) and had a thought. At intermediate impact angles between 45 and 60 degrees (for normal shells) the penetration/shatter vs ricochet calculation is done in a probabilistic manner - only some shells will be deflected. The question is:

Is this random check assessed each time the shell impacts an armor plate?

Basically, if I shoot a Missouri belt at a 50 degree angle, will my shell have to pass the check on the outer layer of 38mm plating, then again for the main belt? If so, the penetration probability curve is essentially squared for all hits on this armor layer.

This would make the Missouri armor significantly stronger than Iowa at intermediate angles, as the Iowa's 25mm plating would be overmatched by most BB AP it sees, leading to a higher penetration probability at these 45 to 60 degree angles (or 60-67.5 for SHS). 

If this behavior is intended (and balanced for) it is fine - I am just wondering if this is the case. It would affect my shot placement against ships with this armor scheme. 

Share this post


Link to post
Share on other sites
26,860
[HINON]
Supertester
21,350 posts
15,057 battles

A ricochet calculation is made for each armor plate the shell hits. This is what makes German turtleback so effective.

  • Cool 1

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles
Just now, Lert said:

A ricochet calculation is made for each armor plate the shell hits. This is what makes German turtleback so effective.

Missouri should probably be nerfed to 25mm torpedo plating then - it should not be THAT much better than the Iowa. I am all for economic bonuses in reward ships, but they shouldn't perform significantly better, which the reinforced citadel bulkhead and outer plating grants. 

Share this post


Link to post
Share on other sites
26,860
[HINON]
Supertester
21,350 posts
15,057 battles
1 minute ago, MaxL_1023 said:

Missouri should probably be nerfed to 25mm torpedo plating then

Any shell that hits the outer plating at such an angle that autobounce / autopen mechanics becomes an issue would not go through the main belt anyways.

Share this post


Link to post
Share on other sites
Alpha Tester
3,483 posts
1,556 battles

Personally I think the actual protection of ships in this game are pretty borked.   We just have to look at historical records to see how borked it is.  I mean the Bismarck took an estimate 400 hits from close range by shells ranging from 203mm to 406 mm (8 in to 16 in) and not one penetrated its armor belt.  I mean they tore the superstructure off the thing but not one shell penetrated its lower decks.

Also if you look at how spaced armor worked in armored vehicles.  The first plate generally deforms the penetrator head of the shell and/or causes the shell to change trajectory to some degree or another before it strikes the next plate which is angled to further deform and deflect the round and eat up even more energy.  On your typical BB, shells have to go through several layers of this sort of protection making it very, very unlikely that shells would reach anything vital before exploding.  Looking at your typical cross section of something like an Iowa, I have a hard time believing that even the Yamato's shells would consistently punch through to the citadel of an Iowa class unless it received multiple hits near the same area thus penetrating already buckled and cracked plates. 

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles
25 minutes ago, Lert said:

Any shell that hits the outer plating at such an angle that autobounce / autopen mechanics becomes an issue would not go through the main belt anyways.

At 45 degrees the Belt is only about 450mm effective (including the back slope) - inside of 10km another Iowa/MO would pen easily. 

Share this post


Link to post
Share on other sites
26,860
[HINON]
Supertester
21,350 posts
15,057 battles
1 minute ago, MaxL_1023 said:

At 45 degrees the Belt is only about 450mm effective (including the back slope) - inside of 10km another Iowa/MO would pen easily. 

At 45 degrees there's no functional difference between 25mm and 38mm either, and 0 ricochet chance.

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles
39 minutes ago, Lert said:

At 45 degrees there's no functional difference between 25mm and 38mm either, and 0 ricochet chance.

What I mean is that a 300mm plate will probably never shatter a 16/50 shell at close range - it will either penetrate or ricochet. Even at 60 degrees the effective armor is less than 600mm due to normalization. 

Share this post


Link to post
Share on other sites
26,860
[HINON]
Supertester
21,350 posts
15,057 battles
8 minutes ago, MaxL_1023 said:

What I mean is that a 300mm plate will probably never shatter a 16/50 shell at close range - it will either penetrate or ricochet. Even at 60 degrees the effective armor is less than 600mm due to normalization. 

I am aware.

What I'm saying is that in most cases that additional armor isn't going to make a difference. Only when under fire under very narrow angle deltas. The belt and the outer layer are under the same angle, and only across a delta of 15 degrees and for 16" and larger AP does it trigger an additional ricochet check.

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles

I disagree - if you are using your back turret at all you will spend a lot of time in this angle range. 

Share this post


Link to post
Share on other sites
1,042
[Y0L0]
Members
9,872 posts
13,704 battles

you also have to account for AP  normalization.  US AP has fairly high normalization  which will reduced perceived angle.   

Share this post


Link to post
Share on other sites
3,017
[DAKI]
Privateers, Members
8,717 posts
7,708 battles
13 minutes ago, MaxL_1023 said:

I disagree - if you are using your back turret at all you will spend a lot of time in this angle range. 

Doubt

T4U0JdR.png

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles

If you back turret has 31 degrees at ~10 km, the A turret will see closer to 34-35 degrees, well within the probability range. Also, angling even 5 degrees more than the minimum can easily occur without it being a mistake if multiple targets are present and you want to present a worse angle to something else. 

Share this post


Link to post
Share on other sites
3,017
[DAKI]
Privateers, Members
8,717 posts
7,708 battles
14 minutes ago, MaxL_1023 said:

If you back turret has 31 degrees at ~10 km, the A turret will see closer to 34-35 degrees, well within the probability range. Also, angling even 5 degrees more than the minimum can easily occur without it being a mistake if multiple targets are present and you want to present a worse angle to something else. 

That assumes that you are unable to make use of your rudder. A somewhat smart player will not stay for the full 30 seconds explosed, but rather angle in after firing the turret, or will rather not fire the turret at all.

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles

The point is that the MO can get away with maneuvers which would statistically hurt an Iowa more. Battles are fluid - you are not bow-tanking all the time, and MO is stronger at that as well due to the thicker transverse bulkhead. 

Share this post


Link to post
Share on other sites
7,272
[WOLF3]
[WOLF3]
Members
21,246 posts
19,617 battles
1 hour ago, Midnitewolf said:

Personally I think the actual protection of ships in this game are pretty borked.   We just have to look at historical records to see how borked it is.  I mean the Bismarck took an estimate 400 hits from close range by shells ranging from 203mm to 406 mm (8 in to 16 in) and not one penetrated its armor belt.  I mean they tore the superstructure off the thing but not one shell penetrated its lower decks.

 

The damage she took may not have broken the belt but it crippled everything else.  She was a dead, floating hulk rendered defenseless, inoperable before the torps finished her off.

Share this post


Link to post
Share on other sites
2,759
[HINON]
Privateers, In AlfaTesters
7,669 posts
2,114 battles
12 minutes ago, MaxL_1023 said:

The point is that the MO can get away with maneuvers which would statistically hurt an Iowa more. Battles are fluid - you are not bow-tanking all the time, and MO is stronger at that as well due to the thicker transverse bulkhead. 

Nope... On Iowa, at this angle the shell would just overmatch the side armor, hit the main belt, and then either defeat it or not defeat it.

On Missouri, if the angle is enough for it to auto bounce, then regardess of the outer plating, it's going to bounce off the 307mm plating. If it's not at autobounce angle.. well, 38mm plating doesn't do jack to slow down BB-caliber shells, so again, no difference.

Statistically, there simple is no difference in armor protection against major-caliber AP shells. Any time the 38mm armor would be useful, the main belt would not ben penetrated anyways.

 

As for the transverse bulkhead, it's irrelevant in-game. If a shell overmatches your bow, it's Yamato's 18.1" AP. And that's going through your frontal bulkhead regardless of if you're Iowa or Missouri.

Share this post


Link to post
Share on other sites
422
[KRAB]
Members
840 posts
7,295 battles
Just now, Phoenix_jz said:

Nope... On Iowa, at this angle the shell would just overmatch the side armor, hit the main belt, and then either defeat it or not defeat it.

On Missouri, if the angle is enough for it to auto bounce, then regardess of the outer plating, it's going to bounce off the 307mm plating. If it's not at autobounce angle.. well, 38mm plating doesn't do jack to slow down BB-caliber shells, so again, no difference.

Statistically, there simple is no difference in armor protection against major-caliber AP shells. Any time the 38mm armor would be useful, the main belt would not ben penetrated anyways.

 

As for the transverse bulkhead, it's irrelevant in-game. If a shell overmatches your bow, it's Yamato's 18.1" AP. And that's going through your frontal bulkhead regardless of if you're Iowa or Missouri.

You are wrong about the second point - if the angle is in the intermediate range, a battleship shell can and will penetrate the 38mm plate and then defeat the 307mm belt if it has more than ~600mm of base penetration. However, it could fail either one of the autobounce checks. 

The transverse bulkhead helps against medium to long range BB fire at angles near 30 degrees, where the bow's sloping can make the effective impact angle large enough for the shell to potentially penetrate, then hit the bulkhead. It also will stop USN 8'' SHS at a 30 degree angle - they will be able to get through the sloped 32mm at an angle of 60 degrees, but usually not have enough energy left to get through the bulkhead at a 30 degree angle when they hit it before detonating. 

Share this post


Link to post
Share on other sites
1,042
[Y0L0]
Members
9,872 posts
13,704 battles

shell normalization table    from 5.6.1  not sure if it has changed much, but was quick to find.      typically 5-6 deg shell normalization for  BBs.   that means  45 deg is treated as 39-40.        with rear turret,   you are looking at about 55deg,    with normalization, it is 50 deg.          auto ricochet of 60deg    unless overmatch. 

 

0.5.6.1 LIVE AP normalisation + autobounce values

 

Definitions:

  • Normalisation is how much a shell ''digs into'' the armour when striking it. For the armour calculation, it effectively shaves a few degrees off the slope, thus reducing the effective armour thickness.
  • Autobounce angle: if a shell strikes at an angle steeper than its autobounce/auto-ricochet armour, it'll automatically ricochet.

Everything in degrees. Premium ships in bold, special ships in italics.

The reason some shells are split is because they're split in the files, and I'll probably end up reusing these tables for stuff like krupp and mass values whenever I get around to datamining those so I wanna keep them split in these tables, too.

If I forgot to list a ship, poke me. I'm not perfect - close to it, but just not quite there yet. I'll turn these into proper spreadsheets at some point.

Obviously these are just the AP shells.

USN

Cruiser shells

Caliber Name Auto-ricochet Normalisation Ships used on
5"/127mm Mark 15 60.0 8.0 Erie
5"/127mm AP 60.0 8.0 Chester
5"/127mm Mark 38 60.0 8.0 Atlanta
6"/152mm Mark 28 60.0 8.0 Erie
6"/152mm AP, 105lbs 60.0 8.0 Chester, Albany, St. Louis, Phoenix (stock)
6"/152mm Mark 27, mod. 5 60.0 8.0 Phoenix (top), Omaha, Marblehead
6"/152mm Mark 35 60.0 6.8 Cleveland
8"/203mm AP, 260lbs 67.5 5.6 Pensacola (stock)
8"/203mm Mark 19 67.5 5.6 Pensacola (top), Indianapolis, New Orleans, Baltimore (stock)
8"/203mm Mark 21 67.5 4.4 Baltimore (top), Des Moines, Buffalo

Battleship shells

Caliber Name Auto-ricochet Normalisation Ships used on
12"/305mm AP 60.0 4.0 South Carolina
12"/305mm AP, 870lbs 60.0 5.0 Wyoming (stock), Arkansas Beta
12"/305mm Mark 15 60.0 5.0 Wyoming (top)
14"/356mm AP, 1400lbs 60.0 5.0 New York (stock), New Mexico (stock)
14"/356mm Mark 16 60.0 5.0 New York (top), New Mexico (top)
14"/356mm Mark 16 (Texas) 60.0 5.0 Texas
16"/406mm Mark 3 60.0 4.0 Colorado (stock)
16"/406mm Mark 5 60.0 4.0 Colorado (top)
16"/406mm Mark 8 (NC) 60.0 6.8 North Carolina
16"/406mm Mark 8 60.0 6.8 Iowa, Montana

Destroyer shells

Caliber Name Auto-ricochet Normalisation Ships used on
3"/76mm Mark 29 60.0 8.0 Smith
4"/102mm Mark 16 60.0 8.0 Sampson, Wickes, Clemson
5"/127mm Mark 15 60.0 8.0 Nicholas (stock)
5"/127mm Mark 38 60.0 8.0 Nicholas (top), Farragut, Mahan, Sims, Benson, Fletcher, Gearing

IJN

Cruiser shells

Caliber Name Auto-ricochet Normalisation Ships used on
4.7"/120mm Type 0 60.0 8.0 Hashidate
5.5"/140mm Common 60.0 8.0 Tenryu, Kuma, Katori, Yubari, Iwaki Alpha, Kitakami
6"/152mm Type 4 60.0 8.0 Chikuma
6.1"/155mm Type 91 60.0 8.6 Mogami (stock)
7.9"/200mm Type 5 60.0 8.2 Furutaka (stock)
8"/203mm Type 91 60.0 8.2 Furutaka (top), Aoba, Myoko, Tone, Mogami (top), Atago, Ibuki
8"/203mm Type 92 60.0 8.2 Zao

Battleship shells

Caliber Name Auto-ricochet Normalisation Ships used on
12"/305mm AP, 850lbs 60.0 8.0 Mikasa
12"/305mm 41st Year Type 60.0 8.0 Kawachi (stock)
12"/305mm 41st Year Type 60.0 8.0 Kawachi (top)
12"/305mm Mark VI 60.0 8.0 Ishizuchi
14"/356mm Type 5 60.0 5.25 Myogi, Kongo (stock)
14"/356mm Type 91 60.0 5.25 Kongo (top), Fuso
16.1"/410mm Type 88 60.0 5.05 Nagato (stock), Amagi (stock)
16.1"/410mm Type 91 60.0 5.05 Nagato (top), Amagi (top)
16.1"/410mm Type 91 (Izumo) 60.0 5.05 Izumo
18.1"/460mm Type 91 60.0 5.25 Yamato

Battleship secondary shells

Caliber Name Auto-ricochet Normalisation Ships used on
5.5"/140mm Type 2 60.0 8.0 Nagato, Amagi
6"/152mm Type 4 60.0 8.0 Myogi, Ishizuchi, Kongo, Fuso
6.1"/155mm Type 91 60.0 8.6 Izumo, Yamato

Destroyer shells

Caliber Name Auto-ricochet Normalisation Ships used on
3"/76mm Type 0 60.0 68.0 Tachibana
4.7"/120mm APC 60.0 8.0 Umikaze, Isokaze
4.7"/120mm Type 0 60.0 8.0 Wakatake, Minekaze, Kamikaze, Mutsuki
5"/127mm Type 0 60.0 8.0 Hatsuharu, Fubuki, Kagero, Shimakaze

Russian navy

Cruiser shells

Caliber Name Auto-ricochet Normalisation Ships used on
4.7"/120mm Model 1905 60.0 8.0 Novik
5.1"/130mm SAP-46 (Orlan) 60.0 8.0 Orlan
5.1"/130mm Model 1911 60.0 8.0 Bogatyr (top), Svetlana (stock)
5.1"/130mm Model 1917 60.0 8.0 Svetlana (top)
5.1"/130mm SAP-46 (Profintern) 60.0 8.0 Profintern
6"/152mm AP, 91lb (Bogatyr) 60.0 8.0 Bogatyr (stock)
6"/152mm AP, 91lb 60.0 8.0 Diana, Aurora
6"/152mm Mark 27, mod 8 60.0 8.0 Murmansk
6"/152mm B-35 60.0 8.0 Budyonny, Shchors, Chapayev
6"/152mm B-35 (Kutuzov) 60.0 8.0 Mikhail Kutuzov
7"/180mm AP-32 60.0 6.0 Kirov
7"/180mm AP-32 (Dankskoi) 60.0 6.0 Dmitri Donskoi
7"/180mm AP-32 (Molotov) 60.0 6.0 Molotov
9"/220mm AP-23 60.0 6.0 Moskva

Destroyer shells

Caliber Name Auto-ricochet Normalisation Ships used on
4"/102mm Model 1915 60.0 8.0 Storozhevoi, Derzki, Izyaslav
5.1"/130mm SAP-46 60.0 8.0 Gnevny, Ognevoi, Kiev, Tashkent, Udaloi (stock)
5.1"/130mm SAP-46 (Gremyaschy) 60.0 8.0 Gremyaschy
5.1/130mm SAP-46 (B-2-U) 60.0 8.0 Udaloi (top), Khabarovsk

Battleship shells

Caliber Name Auto-ricochet Normalisation Ships used on
12"/305mm Model 1911 60.0 5.0 Imperator Nikolai I

German navy

Cruiser shells

Caliber Name Auto-ricochet Normalisation Ships used on
4.1"/105mm SAP Pz.Gr. 60.0 8.0 Dresden
4.1"/105mm Pz.Gr. L/3.9 60.0 8.0 Emden
4.1"/105mm P.S.Gr L/3.8 60.0 8.0 Kolberg
5"/128mm P.S.Gr. L/4.5 60.0 8.0 Hermelin
5.9"/150mm Psgr. L/3.7 60.0 8.0 Karlsruhe
5.9"/150mm Psgr. L/3.7 (C/25) 60.0 8.0 Königsberg, Nürnberg
8"/203mm Pz. Spr. Ggr. L/4.4 60.0 5.0 Admiral Hipper, Roon, Hindenburg
8.3"/210mm Psgr. L/3.1 60.0 5.0 Yorck

Battleship shells

Caliber Name Auto-ricochet Normalisation Ships used on
15"/380mm P.S.Gr L/4.4 60.0 5.05 Tirpitz

Royal Navy

Battleship shells

Caliber Name Auto-ricochet Normalisation Ships used on
15"/381mm Mark XXIIb 60.0 5.05 Warspite

Destroyer shells

Caliber Name Auto-ricochet Normalisation Ships used on
4"/102mm Mark XVI 60.0 8.0 Campbeltown

Poland

Destroyer shells

Caliber Name Auto-ricochet Normalisation Ships used on
4.7"/120mm AP 60.0 8.0 Blyskawica

Pan-Asia

Destroyer shells

Caliber Name Auto-ricochet Normalisation Ships used on
5"/127mm Mark 38 60.0 8.0 Lo Yang
5.1"/130mm SAP A 60.0 8.0 Anshan
Edited by centarina

Share this post


Link to post
Share on other sites
306
[RUST]
Beta Testers
994 posts
11,220 battles

@MaxL_1023 I honestly think you are making a mountain out of a mole hill. As others have indicated, it's a pretty narrow set of circumstances for this to occur. This isn't tanks where players can control the expected angle of incoming fire to the precision that is required for this to matter. RNG on the dispersion alone will factor way more into whether you will or will not take damage for a given salvo received.  If you find yourself regularly within 10km  to where you can control your angle with that much precision, you have bigger problems with respect to map awareness and general decision on how you should deploy your ship. 

Also anyone that is worth a damn will aim at your super structure to land penetrating hits and not be baited into shooting your side armor at the angles you are talking about. 

Share this post


Link to post
Share on other sites
Alpha Tester
3,832 posts
4,687 battles
3 hours ago, Lert said:

A ricochet calculation is made for each armor plate the shell hits. This is what makes German turtleback so effective.

Actually, it's normalization that is the real culprit. Shells normalize upwards on impact due to angle of fall, that makes the critical ranges for German BB turtlebacks significantly further than they otherwise would be. German BB turtlebacks are around 65 degrees as an average(they can vary from ~50 to up to 68.5 over the length of the ship, but generally tend towards the higher values except at extremities). But the upward normalization means that you need to generate an additional 6 degrees AoF in order to hit the critical ranges for penetration, which represents ~5km of range depending on the gun involved. This also has majorly important effects when it comes to low citadel ships.

 

Including multiple angle effects where the shell has both AoF and angle off the bow is pretty complicated, but I might run the numbers sometime to see how it works in practice.

2 hours ago, centarina said:

you also have to account for AP  normalization.  US AP has fairly high normalization  which will reduced perceived angle.   

All normalization has been normalized(heh) to the same values for caliber ranges. Nobody has different normalization. Normalization occurs after the ricochet check, not before it. However on a ship with multiple plates this does mean that the outer plate will normalize the shell in, which will usually reduce the chance of the second check failing.

Share this post


Link to post
Share on other sites
26,860
[HINON]
Supertester
21,350 posts
15,057 battles
4 minutes ago, Aetreus said:

Actually, it's normalization that is the real culprit.

It's a combination of auto ricochet angles, normalization and the fact that each plate triggers a check.

Share this post


Link to post
Share on other sites
Members
0 posts
3 hours ago, Midnitewolf said:

Personally I think the actual protection of ships in this game are pretty borked.   We just have to look at historical records to see how borked it is.  I mean the Bismarck took an estimate 400 hits from close range by shells ranging from 203mm to 406 mm (8 in to 16 in) and not one penetrated its armor belt.  I mean they tore the superstructure off the thing but not one shell penetrated its lower decks.

Also if you look at how spaced armor worked in armored vehicles.  The first plate generally deforms the penetrator head of the shell and/or causes the shell to change trajectory to some degree or another before it strikes the next plate which is angled to further deform and deflect the round and eat up even more energy.  On your typical BB, shells have to go through several layers of this sort of protection making it very, very unlikely that shells would reach anything vital before exploding.  Looking at your typical cross section of something like an Iowa, I have a hard time believing that even the Yamato's shells would consistently punch through to the citadel of an Iowa class unless it received multiple hits near the same area thus penetrating already buckled and cracked plates. 

I think WoWS is the only naval game ever made where a ship's citadel is more of a liability than insurance.  In this game world, ships are more survivable without citadel protection, lol.

Yeah, it's silly, but it is what it is.  Still enjoy the game, tho.

 

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×