>>>Alpha-3 General Feedback [HotFix v66]<<<

[RedParadize]

@arkhangelsk Screening DDs vs DDs only worked in Alpha 2 because target priority is pretty rudimentary at his point. Ship movement seem to be only dictated by target position and torpedo treat. Note that on player side a single DDs was enough to stop many AI DDs for that reason. DDs were so ineffective at fighting each other that single DD was able to effectively hold a full squadron on its own, diverting torpedo away form target sensitive to them. I called that a exploit and still think it was. Now, with last patch, there is still no target priority, but that lonely DD would get hammered that is a plus in my book, but it doesn't remove the need for better AI.

On the flip side, have you try to play a DD vs a modern CA? Or a CA vs a BC? Its pretty funny but totally ridiculous.

A DD can now own a CA just with its gun, just because the CA have such a hard time hitting it.
A CA can own a BC because AI place way less armor than player do on the CA.

So yeah, there is still plenty to work on besides accuracy.

Even if I think the patch over corrected a bit, I am of the opinion that Accuracy should be left as is for now and attention should be applied to armor. Simply put, trying to nail it in the current state of the game would be a waste of time. As Accuracy/Damage/Armor are all intertwine I expect multiple pass on this before nailing the balance.

[arkhangelsk]

Well, basically I only have time to play one game per day since Alpha 3 came out, and I'm not winning every time (due to the indestructible object problem I mentioned last post) so I still have to polish off one more Naval Academy mission (at least this hotfix should make it easier). I'm not sure I want to verify things like "A DD can now own a CA just with its gun, just because the CA have such a hard time hitting it." but it is consistent with what Tousansons is telling me. I'm glad we agree such results are problems.

I am aware that there is a tendency of them to do what you said in Para 1. However overall I don't find it as annoying as you do. Target fixation is realistic. For example, while it is a fighter battle, you might recall how much of the Japanese CAP Thach and his partner tied up with his Thach Weave. As Jon Parshall points out, the textbook solution there is to send a single flight after them while the rest continue CAP. And that's not what happened. Did Thach use an "exploit"?

Besides, I don't think Destroyers vs TB would have been winnable (in Alpha-2) anyway without this psychological realism. Imagine them using only 3-5 (since that's how many destroyers you can build with that budget) torpedo boats to delay your destroyers in one-v-ones while sending the other 5-7 to torpedo your battleship. You'd be F-ed. So I think it is fine, or close to it.

At least the solution isn't to paper it over by just making sure you can kill the lone challenger unrealistically fast. My suggestion is to revert accuracies back to pre-Hotfix levels, and try to play with the armor. I don't think trying to adjust off knowingly false values is a winning strategy.

[There was a submission] below - [about] how one destroyer absorbed the attention of a whole fleet, when they really should be hunting down the other survivors. Also note how many shells it took to finally put her down. Destroyers can be really hard to sink... (sure, you might insist she dodged, but overall in terms of number of shells to get results the conclusion stands).

 

 

Edited January 13, 2020 by arkhangelsk
Deletion of old media to conserve quota

[arkhangelsk]

OK, I finally killed that last Naval Academy mission. How did it go? OK... First, here is the winning ship.

So we enter the battle, and again my health starts going down stupidly fast. I was down to 87% health after taking one set of hits and 78% after the second. At this rate, I'd be dead in minutes. At least the computer seems to be feeling the pain too, since a mere seven minutes after the game started, he chooses to retreat.

Something nicked Geng, my leading battleship. Since it took only off 1% health I won't put a picture of it here. Below, we have enough ID of him to have its approximate health - he lost about half. At this rate, he should be gone in 10, 15 minutes tops...

I get some time to check out my little destroyers. What do they have? Oh I see - the accuracy drops by a factor of TEN over just 2500m, and a 3 inch gun has 6.4" equivalent penetration at 2500m. Muzzle velocity 1117 - what are they firing? APDS? By the way, the twin guns have 3% more accuracy.

 

One of the cruisers is identified as the Tateshina, let's see what she has. 9-inch triples, 2500m hit chance 71%. OK. Triple 5", 2500m hit chance ONE HUNDRED PERCENT, penetration close to 10 inch equivalent, ammunition speed 1123. Really?

3" triple ... HEY! Mine are 57% and his are ... NINETY-SIX percent. is it because of the hull? Well, at least poor RamJB is right then, the additional stability is really helping. We again have close to a 10 fold decrease moving to 5000m.

We have Kinugasa IDed finally. Let's see 7" triples 2500m, 85%, 4" 84% ... at least the accuracy did not invert themselves this time, but really? And I never knew 100mm class guns could penetrate close to 9" of armor at 2500m, velocity 857m/s...

 

And remember how you are supposed to use your smaller ships to at least deter the destroyers? At 7.4km, the enemy's hit rate is 0.2% on my destroyers. I laugh at the piddly hit rate and continue rushing after the battleship. The deterrence plan is not working.

We finally ID the battleship as Tosa. Ten minutes have passed since we last look at him, and his structure only dropped maybe another 10%. The "indestructible destroyed objects" are starting to show. Let me read out the accuracies of his secondaries. Triple 7" 2500m, 100%. Dual same but 5000m rises from 39 to 41%.
Triple 4" guns, 100% decreasing to 23% at 5000m, 3" guns 100% ... oh BTW 5000m hit chance has 27% on this gun. I'm guessing the 2500m hit chance is something like 250% by now but has been capped to 100%, and why are the guns getting more accurate as they get smaller?
2" guns ... 100% decreasing to 11% at 5000m. Oh, really, that wimpy little pom-pom ... it can shoot through 5.4 inches of steel armor!!!

 

OK, so this is how they are going to use the cruisers to defend against my destroyers ... torpedoes. I don't begrudge them, but I wasn't warned. Do you see the message box on the lower left? No torpedo warning! I thought that was Alpha-3's major feature!

 

Look at this. Now it swears it one of my destroyers warned me at 17:36. But look up there! It just isn't on the Report box! Heck, look down Even now!!!! Does this game simulate falsified reports by incompetent and corrupt captains? You have to grant, this is fast staffwork - only one minute after Datian has sunk, they had backtracked, found a plausible point to "notice" the torpedo, and are starting to work that falsification into the system

One of my destroyers got clipped by a 9" shell and lost 14% health. OK, I can roll with that, especially with the hit chance so low (it is 0.9%). By the way, my 3" at this point have about 3.3%.

Tosa lost about 2% health during the period from 3:11:37 to 3:07:38 - not quite as bad as last mission, but the rate of destruction is clearly stalling. And since we were pretty negative up to now, let's put up one positive image - I turned the damaged Jinyu away and it correctly recognizes my intent and launches torpedoes! We need subordinates that can read our intent, Good!

 

Around 3:03:57, I decided to try hitting the cruisers despite a discouraging 0.2% initial hit rate rather than continue to launch shells at the Tosa's Destroyed, thus Indestructible Compartments. The locked on percentage is 2.6%. For something like 7 minutes, I continue to get no hits - note that in a desperate attempt to squeeze out hit rate I have slowed to my optimal shooting speed. The story does kind of have a happy ending though. About 20 seconds after the below picture, I finally smacked Kinugasa and robbed 28% health! Though I must point out it was only doing a mere 28.1 knots.

 

Well, after that actually killing the cruisers wasn't so bad. The game ended after I killed both cruisers with 2 hours and 41 minutes remaining - it has apparently decided to spare me that multi hour chase to reduce Tosa to zero. The only other point of complaint is I swear I lost a destroyer despite having spotted torpedoes and evaded correctly because the auto anti-collision "pushed" me back into to the torpedo. After that, I just turned the AI on every time I read a torpedo warning and let it try to steer itself away from the torpedoes...

If you want to see the full set of photos, including those that did not make this selection, see if you can open this:

https://photos.app.goo.gl/4fXrqv1u2BgDm4NGA

Dinnertime!

Edited January 13, 2020 by arkhangelsk
Deletion of old media to conserve quota

[Nick Thomadis]

12 hours ago, arkhangelsk said:

First, let me thank the administrator/moderator and gamemaker for providing concrete examples of what he considers unacceptable. I've had my posts removed or edited before in other boards, and often it is unclear exactly what's wrong with them, where the borderline is.

I am sorry if I sounded harsh in the forum to RamJB. The detailed feedback is much appreciated but for very long debates and historical analysis is better to create another thread, not use this central feedback thread. The next update is going to be improved further according to RamJB and everyone's feedback.

 

[akd]

1 hour ago, arkhangelsk said:

Oh, really, that wimpy little pom-pom ... it can shoot through 5.4 inches of steel armor!!!

To be fair, I believe the listed penetration is not steel armor, but wrought iron equivalent.  Still, it is very, very high.

I disagree that these weapons should be considered “Pom-poms” (i.e. 37-40mm automatic guns).   2-inch guns would probably be 6pdr.  Of course these fell out of favor rapidly in the 21st century, so we don’t really have good comparables for late-mark naval 6pdrs.  An automatic one (with semi-automatic firing only in the naval mount) was used on Brit MTBs during WWII, and I think AP performance was the same as the 6pdrs mounted in tanks, which would have at most about 3.5” of penetration of steel armor at 1,000m, so we might conclude these are overperforming by 100% (14 inches iron being equivalent to about 7 inches of Krupp steel).

Edited December 21, 2019 by akd

[arkhangelsk]

All penetration values were taken at 2500m and I already divided everything read off the screen by two in the text to compensate for the difference between iron and armor steel.

12 hours ago, RedParadize said:

@arkhangelsk 

A DD can now own a CA just with its gun, just because the CA have such a hard time hitting it.

I am happy to report that at least I have not yet been able to reproduce this worst case scenario. Photos from the attempt - I only killed about 10% health before he got me. Shots from the attempt below:

https://photos.app.goo.gl/qLxgukfNe155fAps6

Edited December 21, 2019 by arkhangelsk
Add a battle report.

[Diamondhero253]

I wrote a bit of a long suggestion in the bug reporter, that I think would also be relevant here, since I think some of my suggestions are worth noting, but I am very grateful for the quick responses to suggestions from the devs. This was my general response to the pre-hotfix version of the game.

"

I have noted several things that could be improved about, or added to, the custom battles:

-Allowing ships to be saved and loaded

-Allowing ships of multiple time periods, to more accurately represent fleets

-The option to allow all hulls should expand to any hull that nation has, not just hulls of that year

-Give certain tech bonuses, or debuffs, to sides (having modern ship that can only use iron armour?)

-Ability to save battles and make your own custom missions (once the game is released, you could tie this feature in with steam workshop)

-Allow player to design multiple types of ships in the same battle by allowing them to load saved ships

-Having mixed fleets of different ships from the same class, such as a super battleship supported by smaller battleships, or a fleet with pre-dreadnoughts and semi-dreadnoughts.

 

Additionally, I think there should also be more ship types in the game, such as Coastal Defense Ships, Monitors, Frigates, Patrol Boats, Corvettes, Gunboats, Minelayers, and potentially even Submarines and Carriers, but these would significantly change the balance of the game, and should be considered carefully.

 

As I have stated previously, more nations, particularly the Ottoman Empire, should be represented in the game.

 

Finally, it would be interesting to expand the timeframe of the game, adding anything from ironclads to more modern ships from the 50s and 60s. It would be nice to have, even if outside the scope of the game. Im not suggesting missile cruisers or wooden ships-of-the-line, just things that are within reason. As many features as possible would be preferred, even if it has to come in DLC.

"

[RedParadize]

@Nick Thomadis Does target signature contribute to hit probability?

Edited December 22, 2019 by RedParadize
typo

[Nick Thomadis]

15 minutes ago, RedParadize said:

@Nick Thomadis Does target signature contribute to hip probability?

Yes, it does. It increases the accuracy parameter related to size of target.

[Teckelmaster]

Concerning the penetration calculator, when i follow an enemy ship on the screen, while i have my ship selected and i hover over the enemy ship, i can not see the calculation for penetration, once i move away from following, e.g. by simply moving the screen a bit in any direction with the mouse, thus "unfollow" the enemy ship, i can hover over and see the penetration values. Not sure, if that is intended, took me a while to figure that out, so i thought i let you know.

[RedParadize]

1 hour ago, Nick Thomadis said:

Yes, it does. It increases the accuracy parameter related to size of target.

I should have asked this long time ago then. I the way it currently works disproportionally favor bigger ship with fewer gun of larger caliber and no secondary. Some example where it looks particularly out of balance:

- 2 3x18" add 88 signature. 3 3x14" add 94 and 4 3x11" add 88.
- Iowa front+back tower add 50 signature, Full dress secondary will add 47 to this. Its biggest funnel only 13.
- Hull do and hull size do not add any signature.

Here is demonstrative example of how much disparity it can create:

A moderately dressed BB, 29000t and 187 target signature:

A low signature Super BB, 56000t, 151ts:


Now, AI generated ship are much worst than the 29000t ship I posted, often dashing at 250-300 target signature. This make me realize how much I was gaming the system compare to AI. If you add to this high speed, the superior armor, very low roll/pitch and no smoke penalty, it is easy to win a 1 vs 3 against AI.

As a comparative, here is a not so badly designed AI BB:

I have seen much worst than this. I do not know how much malus Target signature give. But just the difference in roll, pitch, smoke give a +25% base accuracy bonus to the stripped down and balanced design I posted above. AI generated ship deserve a post on its own, so I wont elaborate on that here.

Could we get a bar in stats that combine all base accuracy bonus and malus? it would be great for clarity and testing purpose.

Edited December 21, 2019 by RedParadize

[RedParadize]

Humm. isn't 42" penetration a bit high for a 14" mk 4 at 10km?

Edited December 21, 2019 by RedParadize

[arkhangelsk]

3 hours ago, RedParadize said:

I have seen much worst than this. I do not know how much malus Target signature give. But just the difference in roll, pitch, smoke give a +25% base accuracy bonus to the stripped down and balanced design I posted above. AI generated ship deserve a post on its own, so I wont elaborate on that here.

Could we get a bar in stats that combine all base accuracy bonus and malus? it would be great for clarity and testing purpose.

To be fair, so far in game this "target signature" does not seem to hurt that much. I've actually been shot at by the enemy first - even though it is physically bigger, can have no better gear than me (because I choose the best towers and radars ... etc), and since they always put secondaries on they are bigger than me in signature ... they still shoot first.

P.S. Not related to the above, but here is attempt 2 to sink a CA with a destroyer. Result, main guns still the best way to defeat destroyer

https://photos.app.goo.gl/jvuurLRF6bH5xVe69

Edited December 22, 2019 by arkhangelsk
Post Script

[RedParadize]

1 hour ago, arkhangelsk said:

To be fair, so far in game this "target signature" does not seem to hurt that much. I've actually been shot at by the enemy first - even though it is physically bigger, can have no better gear than me (because I choose the best towers and radars ... etc), and since they always put secondaries on they are bigger than me in signature ... they still shoot first.

P.S. Not related to the above, but here is attempt 2 to sink a CA with a destroyer. Result, main guns still the best way to defeat destroyer

https://photos.app.goo.gl/jvuurLRF6bH5xVe69

Well, my testing seem to suggest otherwise. I would be curious to see how much exactly trough.

Btw, it was in 1940s era that I easily sunk CA with DDs. I am fairly certain you can do it in earlier era too. I almost soloed a BC yesterday, but could not reproduce it, call it a lucky event. Here is the one of the ship I used to do it (the most successful was smaller that that. +-24TS):

You have to make it as quick as possible stay at range where enemy get very low hit chance.
Gonna try in earier era, should be fun.

Edit: its actually easier in1914. You can keep the enemy under 0.6% hit chance while have yours at around 70%. Time to try again against BC!

Edit2: Yep, you can do it against BC:

I did manage to do this vs a BB of the same era too. But you have to roll a battle were armor is at least under 12". I did run out of ammo trough.
 

Edited December 22, 2019 by RedParadize

[captinjoehenry]

@Nick Thomadis I feel you were a bit excessively harsh on @RAMJB as while he does repeat himself a bit he's stayed rather civil about the whole discussion and provides plenty of sources and other information to back up his point and has largely engaged in civil and productive discussions.

[arkhangelsk]

Since some people are defending him, I will as well. While I respect the administrator's discretion to decide what kind of language is acceptable on his board, I plead that he consider the following:

To the extent RamJB diverted from fully civil discourse, you have to admit he was quite heavily provoked. One can only stay so cool when they provide a page of text that they believe fully answers their opponent's question, only to be met with a single line that's best summarized as "No count" without even an explanation.

BTW, speaking as a third-party (albeit one that generally agrees with his viewpoint), he does answer the question and meets the "burden of proof" insofar as it should be ascribed to him.

Finally, let me be brutally honest here, this board needs him. More than it needs 3 or 5 of us (and I include myself in the "us"). If UAD is to develop into a relatively realistic naval simulator, it does need to give weight to those with the knowledge (which numerically will be in the minority) to say whether what's happening matches up with reality, and as far as I can see in this forum none can beat RamJB in the naval knowledge area. If he's gone for good it'll be very much to this board, and likely game's loss.

Edited December 22, 2019 by arkhangelsk
Best to add this.

[RedParadize]

I want to make it clear that I did not ask for a admin intervention nor complained about RamJB. I did not want to address this openly here but I feel I now have to.

Do I think RamJB is knowledgeable when it come to naval warfare? Yes I do. Having said that, it was indeed pretty unpleasant to voice opinion here, backed by source or not. Some may say that he was provoked, I would reply to that I do not like to have one of sentence nip picked, quoted out of context and twisted into something I did not say. Only to be presented with wall of text that do not address the argument I made in the first place.

I do no think Admin were harsher than himself have been. They just told him to tone it down. Is it that harsh?

Edited December 22, 2019 by RedParadize
Edited for clarity

[arkhangelsk]

2 hours ago, RedParadize said:

I want to make it clear that I did not ask for a admin intervention nor complained about RamJB. I did not want to address this openly here but I feel I now have to.

I believe that you did not ask for an admin intervention. For the rest, since it would be an off-topic discussion and further I don't want to start any flame wars. I shall assemble a PM and send to you ASAP..

Update: PM has been sent

Edited December 22, 2019 by arkhangelsk
Update

[Steeltrap]

 

On 12/21/2019 at 1:26 PM, RedParadize said:

Even if I think the patch over corrected a bit, I am of the opinion that Accuracy should be left as is for now and attention should be applied to armor. Simply put, trying to nail it in the current state of the game would be a waste of time. As Accuracy/Damage/Armor are all intertwine I expect multiple pass on this before nailing the balance.

From a design point of view, accuracy/damage/armour really aren't intertwined other than in the sense of requirements and sequencing. I'm not saying this to nit-pick, lol, more as a general discussion about the challenges the devs face and how it affects us.

If I were to approach the tactical battlefield as a total process, the common method would be to cut it into the sub-processes and then farm those out to individual teams (assuming I had the resources).

If we take the 3 processes mentioned, although I'll call them gunnery, hit resolution and damage, each can be developed, independently and indeed OUGHT to be. The interrelations from a process design perspective are correct sequencing, requirements, inputs and outputs.

In this case the sequencing is as I listed it: gunnery ---> hit resolution -----> damage.

1. GUNNERY

The basic requirement is to delivers a hit/miss result for each shot fired. Any hit triggers the Hit Resolution process. Misses do not, obviously. The process is at its highest level very simple:

(a) Shot fired ------> Was a hit scored? Yes --------> (b) pass required info to 2. HIT RESOLUTION ---------> (c) END

                                                                   No -------------------------------------------------------------------------------------------^

The real meat of the process of course is how "Was a hit scored?" is answered. We know (from a thread on the gunnery model) the devs have abstracted this by means of a cascade of factors that take a starting point (base hit chance) then modify it to produce a final hit chance where, presumably, the perform RNG to result in a 'Yes' or 'No'. And it's THIS choice that drives a bunch of requirements. I might add, it appears it could be even more complicated because the detailed stats display "any 1 of n salvo" plus "50% of n salvo", each of which shows how many guns you can fire. In the case of Iowa or Yamato main battery, for example, it would be (3x3=9, any 1 of 9).

Those requirements are whatever is needed to generate each and every potential modifier.

Some of them are environmental (time of day, overhead condition, wind, wave state), so they need a way of generating those and applying values to them. Furthermore, we know THOSE factors are themselves modified by various characteristics of EACH INDIVIDUAL SHIP, such as stability, pitch, roll, and weight offset. If you hold your mouse over any of those ship factors in the ship designer you can see just how they in fact reduce or increase the effects of waves, turning etc.

Others are related to "tower and tech". Plus there's the matter of the gun whose shot is being addressed having its own characteristics, both accuracy (varies with range) and mount (penalties can apply for dual or triple mounts).

So we can see there's a hell of a lot of moving parts, some of them within the players' control (how we choose to design our ship) and others not (environmental, certain base hull characteristics etc).

The point of all this is it highlights how many different ways there are for the devs to alter the "final hit chance" on which everything depends. You have been questioning the whole "signature" system, which translates to "target ship size" modifier (I already knew that) and the implications. Well, the devs could tweak whatever table exists that translates "signature" to "target ship size". Or they could play with tower bonuses. Or the hit chances of individual guns. Or any of them.

2. HIT RESOLUTION

The basic requirement is to determine if a hit is to do damage and where. Nothing else. Any hit triggers this process. There are a few more possible results, so it's a slightly more complicated high level process, and I'm going to cheat a bit anyway:

(a) Hit Scored ---> (b) determine bearing of incoming fire ---> (c) determine hit location -> Is hit AP or HE?  AP --> (d) Ricochet? Yes --->. No damage ---------> END

                                                                                                                                                                                                                               No -----> (e) Penetrate? No -----^

                                                                                                                                                                                                                                                                      Yes -----> (g) Pass to 3. DAMAGE

                                                                                                                                                                                                                                                                 Partial -------"

                                                                                                                                                                                               HE  ------------------------------> (f) Penetrate? Yes/No/Partial ----^

You can see there are several things we're going to need for all the steps (b) through (g).

A system to work out (b) bearing (easy enough) and thus (c) hit location, which is a lot more complicated because it means being able to reference each and every ship to determine what CAN be hit from that angle, assign a probability to each potential location and then resolve what IS hit (more RNG).

[note I did NOT include the question of plunging fire. Clearly it's there somewhere, as the 'penetration popup' assigns a chance of a vertical/horizontal strike. Were I doing this professionally I would list that calculation as another step even though it could be done within (c) because in an initial pass at least I want to be able to see every significant step]

I assumed HE can't ricochet (believe that's true) so I asked the AP/HE question first (more efficient, and you want your process efficient BEFORE you automate it). We still need to work out ricochet mechanics (d) and penetration/partial mechanics and any differences that might apply within these between AP and HE.

Ultimately we end up with the following: Ricochet, Non-Pen, Partial Pen, Full Pen. The last 3, along with the Type of Shell AND hit location, are passed to 3. DAMAGE.

Anyway, I won't cover that. Indeed, people might wondered why I've done any of this, LOL. There are 2 reasons.

The first is to demonstrate WHY it is they are and ought to be developed as independent processes, which is NOT the same as saying "blind to what is going on up and down stream in the whole combat process". As with any process, they have inputs and generate outputs. Anything required to generate the required output is an input. Any input needs to be listed and sourced. We know, for example, we need hit chances for each gun; when designing the gunnery stats an accuracy curve against range is required (it's probably a function, but whatever lol).

Each process ought to be designed independently because the PURPOSE of each IS independent. You want to know what the process is to decide if something hits what it shoots at. You want to know where it hits. You want to know what happens when it does. It is FAR better to have these split so you can see at a glance what any change of any parameter can and will affect.

The most IMPORTANT point is having each as a process means I can readily choose to address anything within the area that's relevant. If I want to make things more ACCURATE, I go to the GUNNERY process. If I am happy with hit rate but not AP performance, or relative HE performance, I can look at the HIT RESOLUTION or DAMAGE but NOT the GUNNERY. And so on.

The second relates to all the discussion about the effectiveness of secondary guns, particularly when it came to doing what people THOUGHT they were there to do, namely to damage/drive off DDs and/or sink things like transports.

It ought to be clear the root causes were potentially the DD "target ship size" (-70% or something in the "Rise of Heavy Cruiser"), "target high speed" (a further -50%) and the general accuracy of the guns. We can also see there were a variety of ways in which they could make changes designed to "make secondaries more effective".

Consider what they DID do. As far as I can tell they altered things in ALL 3 AT ONCE. The hit rates (gunnery), penetration (hit resolution) and damage (floods etc). When you look at it like that it's easy to see why there was such a clear and dramatic difference between v3 and v3 hotfix.

Altered the accuracy over range curve for guns less than 9" calibre (I think). They may ALSO have altered the way things like tower bonuses apply to secondary guns, too. They didn't really explain the how.

From what I've pointed out here, however, we can immediately see how and why doing that has a lot of OTHER effects beyond the principle 'complaint'. Specifically, secondaries now hit EVERYTHING more easily and reliably. In fact the secondary guns on a BC I had used for the "Armed Convoy" prior to the hotfix is now more accurate against TR, CL and CAs out to about 6-7km with the 6" secondary guns THAN THE MAIN 11" GUNS. Bear in mind those guns fire significantly more quickly than the main guns, AND there are more of them.

Thus we've got to the point where the secondaries have a HIGHER hit chance AND fire a lot more shells per minute as a battery than the main guns. Does anyone really think that makes sense? I don't. Seems the better answer might have been to tweak the table from which the "target ship size" modifier is generated so a DD sized target doesn't get -70%.

Note, too, it also means those same guns on DDs are boosted just the same way as they are on BBs or CAs. Is that a good thing? Does it make much sense? Is it what we were asking?

Plus the pen numbers seem to have gone crazy. But the armour DIDN'T CHANGE, as far as I can tell. This has done a few things.

Firstly, if HE pen is a function of AP pen rather than shell calibre (and I don't know), then the HE pen has dramatically gone UP. If the idea was to make AP more effective (which I believe is a general goal) then this is an important consideration. I would hope the two aren't tied, but HE pen is a black box right now.

Secondly, armour is now even MORE dependent on "angling" (sound familiar?). Why? Because my BC that was designed to deal with the sorts of gunnery I was likely to encounter is now, with its 18.4" effective belt from 90 degrees, totally INADEQUATE against guns with 30" pen that previously were much lower. Twice now I have had a main gun turret destroyed; that NEVER happened using the same ship before this hotfix.

Thirdly, the secondaries ALSO got a HUGE boost. Add that to the accuracy change and they rip CLs and CAs to pieces at earlier tech levels. I could just about turn OFF the main battery now and let the 5" or 6" guns kill them.

Sorry this has been so huge (I had time to kill before catching a plane, lol) but I wanted to lay out a bunch of stuff that I hope might illustrate the dangers of unintended consequences.

My own view is the hotfix is generally a step in several unwelcome directions. Which isn't to say it's a bad idea to try it to see the consequences, but I for one have seen enough.

REALITY was secondary guns were not at all good at killing ships. For GAME reasons we don't necessarily want it to be quite as bad as that.

Pre-hotfix I felt the secondaries on my ships of earlier tech levels (up to mark 4 or 5 at most) were fine at showering BBs with HE at range and ripping up Transports. Later scenarios they were hopeless at anti-DD, although frankly that also had a lot to do with the damage model (something I deliberately left out). Suffice to say "maximum bulkheads" turned things into the Bismarck, even transports.

The cascading effects of accuracy, penetration and altered damage/damage control, which is to say changes in ALL the 3 gunnery processes, have gone WAAAAY beyond what I believe most people thought was the problem. Why make changes to ALL of them? How about one, just the hit chance for example?

I hope all my rambling might give some ideas as to why, and also give us pause for thought when it comes to pressing hard for things we don't think "feel right".

I'm away for a week in 2 hours, so you'll all be spared any more essays.

Hope everyone gets a decent break.

Cheers

Edited December 22, 2019 by Steeltrap

[disc]

I do not mind the hotfix. I think it might have gone a little too far, but that's ok. I personally have not had trouble with overwhelming secondaries, but maybe I just haven't played the right missions again.

There are some other things to consider before balancing accuracy across the board. Though I think the separation into focus groups or processes is valid and a great approach, we have some gestalt stuff too.

1. Armor is wacky. It is not hard to build an invincible armored brick, so that guns that can't penetrate are worthless. Some battleships indeed had armor all over, but this tended to come at a cost, and such armor tended to be thin. A modern all-or-nothing battleship would have large unarmored areas. We cannot choose values for transverse bulkheads, turret sides, upper casemate belts, underwater protective decks, etc. So naturally the AP vs. HE dilemma is hard to solve. A good step may be some display of armor in port or in a viewer.
2. We have no crew. Although a small gun may not be able to damage the ship itself very much, it could kill crewmen. It may also cause morale effect through its great rain of fire. We do not have crew training, intraship communications, decrewing of exposed guns, or crew shock from impacts. Crew is not diverted from shooting, operating machinery, etc. to go put out fires or stop flooding. Spaces are not rendered uninhabitable by smoke or water.
3. Small guns have no bonus to accuracy against fast targets. A big gun with a forty second reload may find it difficult to find a continuing firing solution on a radically maneuvering target. A handy, rapid-firing gun can fire quick corrected shots and cope with quick changes. Though this advantage lessened with advanced fire-control, handy guns became very important against aircraft. I guess aircraft are kinda the elephant in the room with 1930s+ designs....
4. Small ships have no bonus for mounting small guns. A very lively ship like a destroyer may have problems keeping guns on target in a seaway. A light or short gun can make this easier on crew or machinery. Without automated loading, a heavy shell can be hard to handle and load in rough seas.
5. There is no apparent substructure for guns. A bigger gun needs heavier supports and heavier facilities for ammunition stowage and transport. Though this can more or less be subsumed in "gun weight" in the designer, I have never seen it discussed and thought it prudent to mention here.

It may also be said that a smaller gun can be less vulnerable to a "system" failure. A big-gun turret has a lot of parts. If one part is deranged, the whole turret might be disabled. A small gun can be more self-contained: it might need only an aimer, a loader, and a box of ready ammunition. On the other hand, larger guns might be built with greater system redundancy, so that any one issue can be bypassed or fixed.

The discussions in this thread have been thought-provoking.

Edited December 22, 2019 by disc

[RedParadize]

2 hours ago, Steeltrap said:

From a design point of view, accuracy/damage/armour really aren't intertwined other than in the sense of requirements and sequencing. I'm not saying this to nit-pick, lol, more as a general discussion about the challenges the devs face and how it affects us.

Its funny you said that. I actually PMed something along those lines to someone.I was not referring to the technical part. To clarify myself, they are intertwined in term of balance. This is far from being the last accuracy patch we get.

As for the rest of you text, I did not read it yet, but I will.

Edited December 22, 2019 by RedParadize

[arkhangelsk]

OK, I just did a Battle Off Texel Simulation. It didn't quite work out because of a bug, but the Full set of photos here:

https://photos.app.goo.gl/pQQZG9x7gmTpnS7T8

The test conditions. We are about 1910s, the German boats that were sunk were an 1899 design, so. I will design the destroyer and accept whatever light cruiser they give me.

 

Here is the destroyer. 3 single 4 inch guns, 29 knots. Close enough?

Here's the light cruiser they flipped me. Guns are a bit big, but I didn't choose it and I think 7" gun only made one hit the whole fight. Well, you can't have everything.

 

And here is the German torpedo boat. Lucky us, it's even 400 tons and has Few bulkheads, which probably closely resembles the real article.

 

Less than ten minutes in. Remember the real Battle off Texal's decisive phase lasted over 73 minutes. The whole battle from contact was over 3 hours.

 

Yes, that was the third boat. Unfortunately, they placed the fourth boat out of my visibility, and the bug kicked in when I tried to speed up and look for it. Despite my most damaged destroyer still capable of 27.4 knots, the entire formation stayed at 22 knots, and stayed at 22 knots, and stayed at 22 knots. By the time I noticed it was too late. If it wasn't for the bug, I can't see that last torpedo boat lasting more than 5 minutes under fire...

In short, the 4 inch guns are something like five times more effective than they should be, at least according to the Battle off Texal.

Edited January 13, 2020 by arkhangelsk
Deletion of old media to conserve quota

[Cptbarney]

I think people are going a bit mental with realism, if the game is not fun there is no reason to play it at all, i would rather have a more fun than realism but still having realistic mechanics and abilities that don't detract from the fun.

I think getting the game to a more stable situation and fixing whatever content needs fixing is more important then moving onto new stuff once that has been done.

It's a sandbox game and people seem to forget this for some reason, i get peeps don't want easy mode (neither do i) but balance is kinda a non-issue half the time in a singleplayer game (that doesn't mean it shouldn't be attempted).

Like i said for campaign they should have historical and non-historical with mechanics clearly reflecting the two modes in general in how they play and perform. Obviously AI needs to be improved and accuracy on secondaries could be tonned down a little (except a certain ranges i do expect a 100% hitrate unless the sea is that bad and the boat that nimble).

[arkhangelsk]

I'll actually disagree. First, realism is objective and universal. I don't think there are players that really hate realism - no one has ever been ashamed or felt bad about the fact the game they played has a high alignment with reality. Further, just the fact or even perception that something hard is realistic is a motivator for people to try to master it anyway.

Second, fun is subjective and nonuniversal. I believe you said you want secondaries buffed because you like "pew-pews". Well, not everyone does. But we'll have one more ground to say Why Not then you do. Besides, if you play a realistic game and you get say your cruiser to kill a battlecruiser, you can be rightly proud of your accomplishment. If you play a game where the little guns have been buffed beyond all recognition, you can't rightly claim the same credit, can you?
I was so proud when I finally beat the odds and got past the DD vs TB mission in Alpha 2. You know why? Because it is realistic, and hard. I watched TortugaPower, a veteran naval game Youtuber, fail time and time again until he gave up playing it (he did "beat" it once, with the cheat on and him having Tech 5 and them having ... Tech 2?). And here I am. I did it. Doesn't that kind of pride motivate you?

I've been told that mission got easier now, and given what happened above, I believe them. But can you feel the same pride?

And if balance is a "non-issue", maybe we should think about making it real, should we. Plus, real does come with a balance built in, and one that's hard to argue against. For example, after beating DD vs TB, I have come to very realistic conclusions when I play the campaign. First, I must remember to build escorts. Lots of them so I will never have to fight 2.5 to 1 odds. Two, I must remember to stay ahead in Tech even if it means a few fewer ships (in that scenario, they had Tech 2 and you Tech 1, and that's one reason why it's so hard). You see, I will automatically be motivated to do the correct thing, without artificial buffs to "make secondaries and destroyers useful".

I'll say make the historical mode first. As we've just seen, it is actually extremely easy and quick to create a "non-historical" mode that satisfy those palates once you've gotten it right anyway - just insert absurd values at suitable points. It took them something like a year, or more, to get to Alpha-2, another couple to Alpha-3, and only one week to make this "hotfix". They can probably make a different version once a week and sell them as DLC. Getting it right is the real challenge and one I think we should all participate in.

Let me conclude by putting it this way. How much of the recent results would the developers be willing to put on their webpage? Ultimate Admiral Dreadnoughts, the realistic naval combat simulator where you can destroy battlecruisers with a destroyer's gunfire! Where everything is so over-lethal one maneuvering mistake will be irrevocable! Where battles are resolved five times faster than the historical equivalent!

Yeah, these things are rather unprintable, are they? I thought so.

P.S. Ok, to be fair to you, it has been universally acknowledged that very close-range hit rates are a problem. I won't mind if they just have across the board 100% hit rates for 1000m for Mark 3 and up, and even for Mark 2 and 1 I won't mind giving every caliber whatever was the best hit chance in Alpha-2 or pre hotfix Alpha 3 for that Tech level was (so say all 90 or all 80 or whatever). For 2500m, I won't necessarily begrudge them reducing the difference between the calibers, but at the very least they should have avoided inverting the hit rates so the small guns have a higher number. For one thing, given the greater facilities for the big guns, it isn't plausible, and remember that even at the same hit rate, the smaller gun will already hit a lot more simply because of how much more often it fires. By 5000 we should definitely be returning to the "regular program".

Edited December 22, 2019 by arkhangelsk
Add a point about hit rates at close ranges.

[Tousansons]

I ll try to develop a bit on realism, even if I agree with most of arkhangelsk said.

You can't emulate reality in a video game. Some things will always be abstracted or removed because they are not interesting, too difficult to implement in the engine. For this reason, I don't think you can call it objective or universal in this particular environnement. You can use realism as a selling point and make sure your video game use accurate historical values, but in my opinion there will always be some concession needed. A video game is first and foremost a game, the "fun" factor is an important part of it and reality is not necessarly a fun game. Though like you said, fun is subjective and what the person X will like is not necessarly the taste of Y.

In UA:D the game part is the way the player can design and pit massive lumps of steel with guns against each other in a somewhat believable 1900+ era sandbox. This is already not a grognard wargame with detailed OOB and a hundred pages manual. The realism part in the final product should be the use more historical values in battles and ships designer, but the realism part will stop the moment you name you're 40k tons japanese battleship BAEruna and refit her with 16 inch guns in 1935.

Now to push it a bit further and still ignoring the whole alpha patching we have so far in UA:D, I ll be taking another exemple. The simple fact that you give orders to ships by a simple clic on the screen is already an abstraction of the whole (and often complicated) command chain used to move a fleet. In real life they used signals, later they used radio, what about the time needed to pass the orders among the ships officers and decipher them if they are coded? For movement orders alone you can have different level of realism (Rule the waves use a simple abstraction on the whole signals layer).

That's also why we have things like the game Armored Brigade. Here you have an interesting take on this chain of order. There is a timer on each command given to a unit on the field, depending on different factors. The more orders there are pending, the more time units will have to wait before receiving the orders to move/disembark at a location. There is other abstractions in it (like off map artillery) but I hope your getting my point here.

I honestly like both approaches as long as it bring something new to try and ultimately master in a video game.

Now with what I just read in the topic I want to try and sink a BC with a single destroyer while it last.