Alpha 4v68 Meta Analysis (update)

[RedParadize]

Preface:
First of, my goal is not to point out what the game should be or how historically accurate it is. This analysis only objective is determining what are the best way to built, fight and win in Alpha 4v68. I will split my analysis in defense and offense. Some of you may be aware that traditionally there is a third category added to this, namely speed. I will however not include it as its overlap with offense and defense is too great to be treated separately. I will touch cost effectiveness a little bit, but not maintenance cost. The main reason for this is that I feel that economics aspect are not very balanced in current version of the game.

(if you do not wish to read the wall of text, here is what it is all about)
The concept is to go for high speed, top armor and medium armament. It is possible to do so on a relatively small and inexpensive ship. A ship like I describe in my conclusion and trough the example I provide can easily go against multiple heavy battleship and their escort.

Defense:
Defense can be separated in 3 category: Hit avoidance, armor and damage mitigation.

Hit avoidance:
Ship size/target signature is the largest hit chance malus and is static, ranging from -20% for very small destroyer to more than +400% for largest super battleship. The other way to avoid hit is speed, malus range from 10% at 0kt to 90% at 37kt (credit to @Evil4Zerggin). Making speed the second largest hit chance malus. Combined together, small size/target signature and high speed can provide incredible level of protection against gun fire. As for torpedo, hit avoidance is achieved trough high speed and agility. At distance, a slight turn is enough to make torpedo wave miss, specially if going fast. For last second avoidance, Aux. engine and Shaft provide a nice turning bonus.

Armor:
Armor is the second line of defense. There is two important question: What to armor and how much armor ? It is important to note that turret have their ammunition located under the compartment they sit on, thus can be outside of main belt. Damaged compartment, even above water, affect top speed and its accuracy malus. For these reason it is unwise to opt a "all or nothing" approach.  Given how light and cheap armor is, immunity on a wide bracket of range is possible against the largest gun, even on relatively small ship. Therefore there is no reason to not opt for a all around and thick armor. However note that armor do not protect against torpedo.

Damage mitigation:
Damage mitigation is the last line of defense and can be divided into 5 types: Fire, flood, ammo explosion, engine damage and torpedo defense. Fire and flooding can be reasonably mitigated with "many bulkhead" plus 1st level of Anti-flood and Anti-fire for a minimal cost&weight. Penetrating damage is required damage engine or trigger ammo explosion. Given how massive Citadel and Barbette are, it is better to reinvest this mass into armor. Armor is simply lighter, cheaper and more effective than Barbette or Citadel.

As for torpedo defense, at first glance their damage reduction may seem attractive, but in any case a torpedo will destroy the compartment it hit. This is important because if one of the 3 mid bottom compartment is destroyed engine will be lost, regardless of citadel or torpedo protection. At best, Torpedo provide a damage reduction to compartment surrounding the place it hit. But does it matter? A ship that lost engine become a easy target from both torpedo and gun fire, most case scenario it will be lost. For these reasons I am of the opinion that torpedo defense is a extra at best. Specially considering how heavy and expensive it is.

Offence:
Weapons consist of Main gun, secondary gun and torpedo. Before covering them it is important to mention the effect of roll, pitch and balance have on accuracy. A heavily armed ship, with high pitch and roll, may end up landing less hit than a lightly armed one. There is no point in adding more gun if it impair the ability to hit of the rest of them.

Main gun:
Alpha 4 improved accuracy of lesser caliber, but bigger is better motto is still true. Mainly, it is a question of penetration. Some may argue that HE can compensate the lack of penetration, I am not of that opinion. When looking at damage done, most of it is from penetrating shots. Penetrating also mean risk of ammo explosion and engine damage. When you get those battle is most likely won already. Going for bigger caliber also allow to do the same amount of damage with a lesser number of turrets, thus reducing target signature, pitch, roll and ship mass.

Secondary gun:
Secondary is mainly useful against destroyer and lightly armored cruiser. Fire is not much of a treat, therefore secondary have almost no effect on heavy armored ship. Bigger is also better for secondary. A single triple 8" will kill a destroyer faster and further way than two triple 6" or six triple 5". It is unnecessary to have many secondary gun, a few will do, over doing it will just impair accuracy of the main battery. About casemate guns, don't use them! They have terrible effect on roll. Better have the secondary on turrets and close to center.

Torpedo:
Since Alpha 4, torpedo have become a primary weapon even on capital ship as soon as you can use deck launcher on them (1920). Torpedo bypass armor, ignore aiming penalty and do massive damage and crits. Pound for pound they are better than guns. On the down side, ship with strong torpedo defense can be almost impossible to finish with torpedo alone, so it can't be the only offensive weapon. Strong TDS do not negate torpedo, they will still destroy engines, making ships easy target for main battery. Deck torpedo launcher have to be placed far away from centerline, too much of them and it will have a big impact on roll. Therefore it is better to add just enough of them to be effective. Single launcher reload faster and are proportionally lighter than larger one, but they do not spread torpedo as good. For that reason, at range 5x torp launcher are preferable. Point black, multiple single launcher will do better.

Conclusion:
Taking it all in consideration, the ideal "meta" ship is as small it can be made to reduce target size, with the most mass effective tower available. To minimize target signature, only two main turrets, preferably of high pen big caliber. One or two secondary turrets of 6" or above, preferably on the rear. As Fast as it can be made, preferably 37kn or above. 2 to 4 deck torpedo launcher per side, preferably 5x. Enough armor to stop shell of the best gun of the era. Some aux engine and shaft, lvl 1 anti-fire and anti-flood. And lastly, no barbette, no torpedo protection, no citadel.

So little armament may look scary but remember, if you manage to reduce target size/signature to its minimum and dash at 37kt. You will have a hit chance of potentially more than 10 times what a conventional enemy would have. In other word, a fleet of conventional battleship would need 20 turrets to match your firepower. The only thing you have to care about is to stay in your armor immunity zone, stay far from torpedo armed ship and maneuver frequently.

Few example:
Excluding the deck torpedo, ship with that focus can be made in all era, even per-dreadnought. But it get far better once you have Battlecruiser hull available. To make it easier, chose hull with good "hull form". Here is some example:

1900: No secondary on that one, simply get rid of destroyer with main battery before engaging enemy. There is not much deck armor, but since accuracy is so low no one can shoot from far anyways. Immunity zone vs 12": 5km to 7km
Obsolete as soon as Dreadnought and battlecruiser arrive.


1910, More deck armor, but still no deck torpedo.
Immunity zone: 2.5km to 15km
Obsolete by +-1925


1925: At this point the concept ca be fully implemented.
Immunity zone: 7.5km to 25km
Remain competitive until the end of the game.

Edited February 15, 2020 by RedParadize

[RedParadize]

Same Meta, but without Speed:
I have been thinking allot about the meta I came up with. The basic concept was to avoid getting hit at all cost, and if hit survive it. One of the core element was speed, but what if I removed it from the equation? So what I did is take all the desired attribute of my previous Meta except speed and torpedo. Why removing torpedo? well if I am gonna be slow it doesn't make sense to expose myself to be torpedoed. I will simply try to stay at range, but can I do that without speed? To figure out I will add a bunch of destroyer into the mix.

Setup:
2 BB vs 4 SBB 8 Destroyer

The low signature steel brick:
Here what I came up with to test this. It basically have all the attribute of my meta but at 25 knots. I chose a Modernized hull because of their comparable size to my previous Meta Design, but also for its decent 90 resilience and best tower setup 50 Base Accuracy and 60 LR Accuracy for a very low weight. Two triple 18" and two triple 7" to fence of the DDs. And as the name suggest, a insane level of armor:


I had to restart the match a couple times to get the kind of enemy I desired. Namely a decently armored fast Super Battleship. Note the ridiculously high target size. As a result, I have twice more chance to hit him than it have to hit me (15.8% vs 7.3%). So while its fleet have a total of 16 triple 17" and I just have 4 triple 18", I am not that far behind in term of effective firepower. With armor factored in, I have the advantage.


Here is the DD advancing on me just prior to their death. I had no difficulty at avoiding the torpedo as grouped ship do not launch their torpedo simultaneously. In fact I think only two of them launched their torpedo. 


I won, but not by a big margin.

Conclusion:
Speed is very important, but maybe not as much as target size. Specially if not the speed malus is not maxed out (37knot). Target signature and ship size play a significant role in hit probability. The part that I retain of my original meta, namely: low signature, no citadel, no torpedo defense and barbette make allot of sense. This allowed me to free up allot of displacement for armor, even more since I did not invest in speed.This meta work too, but its clearly not as fun as fast Battlecruiser.

I am now of the opinion that target size/signature and the lack of purpose of citadel and other damage mitigation buff are problem as serious as the speed malus.

[RedParadize]

Hahaha! Look at this:


The worst tower available and slower speed, yet I have better hit chance. With 45" of belt armor I only have to do what that BB name suggest.

[Tycondero]

The current speed malus should be revised. I agree that speed will make it difficult to hit a target, but it should depend on the movement vector e.g. when both ships are running next to each other in the same direction with matching speed it should be less difficult to hit eachother again. 

Edited February 17, 2020 by Tycondero

[arkhangelsk]

On 2/16/2020 at 7:02 AM, RedParadize said:

Hahaha! Look at this:


The worst tower available and slower speed, yet I have better hit chance. With 45" of belt armor I only have to do what that BB name suggest.

Not really (at least as per the read-out hit chances). You have 6 guns, so your chance of getting at least 1 hit is 1-(0.74^6)=about 84 percent. His chance of getting at least one hit is 1-(0.84^15)=about 92.7 percent. And his hits are worth more. If you are getting hurt less, it's likely because of your FORTY-FIVE inches of armor.

[Shaftoe]

Who the f@#k calls a battleship "FaceHugger"?

Good thread. Hopefully, next patch will make torpedoes more useable. 

Edited February 17, 2020 by Shaftoe

[RedParadize]

8 hours ago, arkhangelsk said:

Not really (at least as per the read-out hit chances). You have 6 guns, so your chance of getting at least 1 hit is 1-(0.74^6)=about 84 percent. His chance of getting at least one hit is 1-(0.84^15)=about 92.7 percent. And his hits are worth more. If you are getting hurt less, it's likely because of your FORTY-FIVE inches of armor.

Armor focus was the point of that design. I was just surprised that I had a better hit chance (per gun yes) than them. If it would have been only about that I would have lost. By with heavy armor it works, btw, I won that fight, vs two!


My latest challenge is to beat super battleship with the oldest and cheapest built possible. Agains't a decently protected SBB, best I could do is 1920 and is using the all armor meta. Speedy BC can't do it prior to 1925.


@Shaftoe Torpedo are usable, it fact it would have been much easier if I had allowd myself to used them.

Edited February 17, 2020 by RedParadize

[Zak MacKay]

On the speed part ship size also changes engine efficiency. So a smaller ship will have better engine efficiency. I use speed as one of my main anti torpedo methods. Just sailing away from the enemy destroyer and they won't be able to torpedo you. Battleships and especially battlecruisers can move extreamly fast meaning they cannot torpedo you at that range.

Torpedoes also cause the enemy to panic. You may be able to shut off half the enemy main guns when they turn to avoid torpedoes. It also prevents destroyers from approaching you and a lot of good things overall.

Citadel protection also increases resistance which reduces torpedo and gun damage while increasing armor's thickness. On the hull you are using All or nothing reduces gun damage by 10% and turtle back by 22%. It seems a bit heavy and costly compared to armor but its actually not the worst. I think its more due to armor overall being worlds ahead of everything else which makes them bad. But their not super far from being viable.

Some towers and funnels are definitely heavier then they first seem. Barrettes are also just way too heavy and big meaning I am shying away from them. Many towers act as barrettes for secondaries so kind of have the similar issue of your paying this high premium for placement. I changed a ship from having 2 turrets forward and 2 aft to having 3 forward and 1 aft and it just seemed a lot better. The arcs seemed better, I reduced weight, and even opened up some secondary mounts on the tower. Because the huge barrette was just really huge. The  secondary barrette I tried to find a use for but it just seems awful. Would I rather have better arcs in one place or place a separate gun somewhere else. Its a pretty obvious answer with more guns covering more area and being less weight.

Small guns seem like they should be a lot lighter making it useful. A 4" gun is just pretty heavy and spamming them across your ship can add hundreds of tonnes to its weight without really adding anything. Many high tier towers benefit is being able to spam a bunch of 3-4" guns which are just worthless later on. It seems like single mounts should be different somehow to make spamming them on towers more effective. But right now I would heavily consider if the secondary gun will actually be worth its weight.

[Evil4Zerggin]

Edit: Actually, needs more investigation.

Edited February 18, 2020 by Evil4Zerggin

[Cptbarney]

obviously things will need to be tuned down for the campaign would be pretty mental to see a 175k tonne Super-heavy battleship going 65knots with quintuple 24inch guns and 70inchs of armour lol.

[Evil4Zerggin]

Research suggests:

• HP requirement seems to be of the form displacement^D * speed^S * C.
• D, S, and C are constant per-hull, but can be different for every hull. These cannot be fully explained by Hull Form alone.
• When I first looked (Alpha 3?), D was usually about 0.6 and S was usually about 3.
• In Alpha 4, D seems to have increased to 0.7-1.01 for the hulls I looked at. This relatively benefits smaller ships.
• In Alpha 4, I've seen S as low as 2.51 for some hulls though others stayed at 3. IMO 3 is a good starting point; the lower it is, the cheaper fast speed is relative to slow speed which a lot of people wanted to avoid.
• Under an ideal quadratic drag + uniform ship scaling model, D would be 2/3 and S would be 3.
• HP per engine ton is also a constant per hull, and while it has some correlation with Hull Form, that cannot be a full explanation. For example, early TBs get 8x the HP/ton of CLs with the same Hull Form.

[arkhangelsk]

8 hours ago, Evil4Zerggin said:

Research suggests:

• HP requirement seems to be of the form displacement^D * speed^S * C.
• D, S, and C are constant per-hull, but can be different for every hull. These cannot be fully explained by Hull Form alone.

Actually, it seems that the HP requirement is by type (BB, BC) rather than by Hull Form. If you ask for a certain speed with a certain displacement, the computed HP (and HP required per ton of ship) is always the same as long as you remain within the same type. Hull Form only changes the weight of the engines required for that purported HP.

[Evil4Zerggin]

Great info! Will reverse-engineer further.

[Bluishdoor76]

hu who would of thought 1 meter of steel would be so effective :3

 

sorry couldnt resist, cool thread tho

[Evil4Zerggin]

Looked more closely at Hull Form:

• Only the base Hull Form matters for HP per engine/boiler weight. Tech and component modifiers to Hull Form say they change the engine/boiler weight modifier, but they do not.
• Each point of Hull Form away from ~82 gives a 1.1% bonus/penalty to HP per engine/boiler weight.
• This stacks multiplicatively with other modifiers.
• Ship types seem to have an extra multiplier to HP per engine/boiler weight. For most ship types this is only around 1.13, but for DDs it is nearly 5, and for TBs it is over 9.

[Evil4Zerggin]

And the displacement and speed powers:

• BB: 1.01 displacement, 2.3 speed, in other words HP requirement is proportional to displacement^1.01 * speed^2.3. 1.01 displacement power slightly favors smaller ships; since Target Ship Size heavily favors smaller ships, this means small BBs are favored overall. 2.3 speed power is lower than I seem to remember in Alpha 3, which would be a relative boost to faster ships or a relative penalty to slower ships.
• BC: 1.01 displacement, 2.51 speed. Same story for displacement as BBs. The speed power is a bit higher than for BBs but high speed is still a bargain considering the way Target Fast Speed currently works.
• CA: 0.28 displacement, 4.06 speed. Way out of line. With only 0.28 displacement power, bigger CAs are heavily favored by engine weight. But going fast is hugely penalized.
• CL: 0.6 0.603 displacement, 3 speed.
• DD: 0.6 displacement, 2.8 speed.
• TB: 0.7 displacement, 2.8 speed. These three are pretty close to the simple quadratic drag + uniform ship scaling model which would give 0.67 displacement power and 3 speed power.

Edited February 22, 2020 by Evil4Zerggin

[BuckleUpBones]

3 hours ago, Evil4Zerggin said:

And the displacement and speed powers:

• BB: 1.01 displacement, 2.3 speed, in other words HP requirement is proportional to displacement^1.01 * speed^2.3. 1.01 displacement power slightly favors smaller ships; since Target Ship Size heavily favors smaller ships, this means small BBs are favored overall. 2.3 speed power is lower than I seem to remember in Alpha 3, which would be a relative boost to faster ships or a relative penalty to slower ships.
• BC: 1.01 displacement, 2.51 speed. Same story for displacement as BBs. The speed power is a bit higher than for BBs but high speed is still a bargain considering the way Target Fast Speed currently works.
• CA: 0.28 displacement, 4.06 speed. Way out of line. With only 0.28 displacement power, bigger CAs are heavily favored by engine weight. But going fast is hugely penalized.
• CL: 0.6 displacement, 3 speed.
• DD: 0.6 displacement, 2.8 speed.
• TB: 0.7 displacement, 2.8 speed. These three are pretty close to the simple quadratic drag + uniform ship scaling model which would give 0.67 displacement power and 3 speed power.

Why are you doing this, decoding the game, isn't this all abit meaningless to the general playerbase, it seems like a personal diversion and one that is not going to help the playerbase, so I ask you...

How is the playerbase going use all these 'possible' decoded calculations and algorithms?  Also as alpha users, how does this assist the Dev's?

 

Edited February 20, 2020 by BuckleUpBones

[Shaftoe]

4 hours ago, BuckleUpBones said:

Why are you doing this, decoding the game, isn't this all abit meaningless to the general playerbase, it seems like a personal diversion and one that is not going to help the playerbase, so I ask you...

How is the playerbase going use all these 'possible' decoded calculations and algorithms?  Also as alpha users, how does this assist the Dev's?

 

The man just did this in his free time and shared results of the work. Nobody is obligated to use or read it. But I gotta say, what he did is interesting. You know, sometimes people just do stuff like that, the "meaningless work". A lot of hobbys can be described as such. And yet people still do it. It does not warrant public scrutiny, and it does not need to have some high purpose to be valid knowledge or at least an interesting read for those who'd care to check it out. 

In case you're unaware, jupming into thread out of nowhere and questioning people's hobbies (or other seemingly useless activities) does not make you look good, nor smart. Especially when something you question is well-liked by others (as in this case, judging by amount of likes he received from active part of this community).

So, in light of questions you asked him, I would like to ask one of you: what is the purpose of appealing to playerbase's interests here (aside from embarrassing yourself)?

[BuckleUpBones]

10 hours ago, Shaftoe said:

what is the purpose of appealing to playerbase's interests here

Forum is for relaying bugs, problems and feedback, what we see is what we report on and discuss, underlying figures is Dev’s domain.

10 hours ago, Shaftoe said:

In case you're unaware, jupming into thread out of nowhere and questioning people's hobbies (or other seemingly useless activities) does not make you look good, nor smart. Especially when something you question is well-liked by others (as in this case, judging by amount of likes he received from active part of this community).

It’s alittle short-sighted to think forum users who aren't posting within a topic are not reading them, “jumping in” from any user/readers on an open forum should not be directly discouraged.

Edited February 20, 2020 by BuckleUpBones

[Hethwill]

Looking at numbers - for example odds of combat, or percentile accuracy, and so on and forth - and correlating them to historical data is many player's hobby inside the hobby and when done beforehand during development can prove invaluable for the devs.

Thing is, many players of games can't help it. They will naturally dissect the numbers no matter what.

[BuckleUpBones]

2 minutes ago, Hethwill said:

Looking at numbers - for example odds of combat, or percentile accuracy, and so on and forth - and correlating them to historical data is many player's hobby inside the hobby and when done beforehand during development can prove invaluable for the devs.

Thing is, many players of games can't help it. They will naturally dissect the numbers no matter what.

Acknowledge.

[RedParadize]

If the goal is not to find the best way to win then its not really a game isn't it?

I think everyone have to take Evil4Zergling or my conclusion with a grain of salt. We do not have access to UA:D internal mechanics, we simply test things, but maybe more than others.  

@Evil4Zerggin I like your analysis. But I believe your mathematics formula are not the ones used. At best, they only represent the result of your test, which unavoidably have a limited sampling. There is more than likely more variable than what we can see. We can doubt the accuracy of value like HP as they are most likely calculated on top of the rest. Having said that, its still useful to test and reflect on these result.

[Shaftoe]

1 hour ago, RedParadize said:

If the goal is not to find the best way to win then its not really a game isn't it?

That's quite a statement. What if the goal is to re-create historical ships and try to beat certain scenarios with them? What if for somebody it's all about the looks of 2 fleets mauling each other? Tbh, personally, I don't care whether or not my ships are "meta". So long as they get the job done and I like them, the game is fine by me.

[RedParadize]

@Shaftoe "Meta" is just a gamer word for sound design and strategy really. If your ideal is to beat enemy using historical ship then ideally the "meta" would have to match real life. In theory that sound great, but what if they designer of the era were wrong? Think of how a ship like the Dreadnought changed everything... The thing is that if the game were to perfectly match history then player would effectively be prescient and design ship that are always ahead of their time. On the other hand, if the player really wish to stick with the era way of doing things then designing ship is more or less unnecessary. Btw its the one thing I do not like about the new hull they introduced, they are too limited. One or two choice of tower and you pretty much have to design it like the ship they are based on.

 

[Steeltrap]

6 hours ago, Hethwill said:

Looking at numbers - for example odds of combat, or percentile accuracy, and so on and forth - and correlating them to historical data is many player's hobby inside the hobby and when done beforehand during development can prove invaluable for the devs.

Thing is, many players of games can't help it. They will naturally dissect the numbers no matter what.

Exactly why I bang on and on about sharing things with us so we can understand them.

"Testing" requires knowing what is correct v incorrect, acceptable v unacceptable.

We're experimenting and guessing in most cases, not testing.

Mind you, I don't know what the "Anonymous Analytics" within the game's options menu means or does, so perhaps we're 'testing' through generating data for the devs without realising it. As an aside, as I couldn't find any information about it, I turned it off. Anyone know what it is and does?

As to the suggestion people ought NOT do this sort of investigation and analysis, I find that bizarre.  People are welcome to think that, but trying to tell others their behaviour is "unacceptable" somehow seems remarkably presumptuous from a fellow player.

If the DEVS tell us not to do things, that's different. They won't, generally, because the info is in the game and thus people can piece it together.

Some people learn best through observation/action, others through absorbing information first. For the second group, the more the better. They're the people who will read a game's manual cover to cover before trying to play.

If those people aren't given a 'comprehensive' manual, they're going to try to find that which they feel they need to know to understand how to play the game "effectively".

Others, of course, will dive right in.

My closest friend and I are polar opposites. In 40 years of playing games he's the "dive in and see what happens" while I'm the "read the manual cover to cover, think about it, go back and read certain bits again, now load the game and see if my understanding of the manual and its implications are correct". We've played games where years later I've found he didn't realise you could do certain things because he never read the manual or, with more modern games, there isn't a manual or it sucks in terms of thoroughness and quality. I have suffered a lot over the years as beautiful, accurate manuals vanished and game devs relied on fans building their own wiki etc. Yes, I'm a "RTFM" type, LOL.

I've always thought keeping players ignorant, especially during testing, is less than ideal.

Impressions matter. Uninformed people are more likely to draw incorrect conclusions through no fault of their own. Doesn't matter now, as those who paid to play are pretty much all enthusiastic supporters, but eventually the question of whether people being kept ignorant of mechanics and thus blurting incorrect and unflattering conclusions into the public domain will arise. We all know about Steam Reviews.

Of course these are simply my opinions. Others are welcome to disagree, although unless it comes to a question of fact, don't tell me I'm "incorrect", and certainly NEVER use the word "wrong" if you mean "incorrect". If you don't know the difference, RTFM; (which in this case is a dictionary). 😎

Cheers

Edited February 21, 2020 by Steeltrap