Long list of issues (Please read)

[Whomst'd've]

Would it be possible to have a free mode when placing barbettes and superstructures that only the player can activate and not the AI, and enable us to place barbettes and superstructures within existing mounting points by holding down left ctrl? This does not mean being able to place them outside of existing mounting points, but being able to place them in the large gaps between each mounting point, allowing for a more fluid and satisfying design. This would result in the Devs being able to keep restricting the players on where to place barbettes and superstructures (not ideal) but also allow the player to create a more efficient design within the restrictions forced upon them (which is good for the builder). 

Edited July 24, 2020 by Whomst'd've

[Whomst'd've]

Also this is more of a wish but I'll throw it out there anyway, have the turret size/diameter be based on gun size and the number of guns in that turret. If your confused look at the King George V class battleships (ww2). 

Edited July 24, 2020 by Whomst'd've

[subnorminalnt]

Whenever I try to update the game the launcher crashes. I have to reinstall the game instead of updating it. Also, the remember me button doesn't work on my launcher.

[Hardlec]

Range and bearing of Target ships is still missing.

 

It is still impossible to zoom out far enough to get a full context of the battle.  

[draconins]

In my humble opinion, making it realistic and ship design issues should be lower priority than making gameplay work at current point. They are issues but it does not stop play. I am still unsatified with current ship designer but it does not stop me to play. Even nonrealistic ship design can still be fun with good enough gameplay.

Gameplay issues mainly about control, for example torpedo not firing without clear reason, multiple formation control, ship turn in circle, ship evasion, ship ramming mechanics, sudden falling back of line logic. This kind of issues can make a lot of frustration in already mostly working games. Some of the points Nick already addressed earlier.

 

[Reaper Jack]

1 hour ago, draconins said:

In my humble opinion, making it realistic and ship design issues should be lower priority than making gameplay work at current point. They are issues but it does not stop play. I am still unsatified with current ship designer but it does not stop me to play. Even nonrealistic ship design can still be fun with good enough gameplay.

Gameplay issues mainly about control, for example torpedo not firing without clear reason, multiple formation control, ship turn in circle, ship evasion, ship ramming mechanics, sudden falling back of line logic. This kind of issues can make a lot of frustration in already mostly working games. Some of the points Nick already addressed earlier.

 

As others have pointed out before me as well, realism and gameplay are linked. Real life ship designs work or do not work because they are subject to real life conditions; if those same conditions are not simulated in game, then the ships that are best to use will not be anything like real life designs that worked historically. Right now this is why the ships that work in the single use scenarios are almost always the biggest, heaviest, most armored or bust. 

Most of the gameplay issues you mention are AI problems, and Nick has confirmed they are working on making the AI much smarter. 

[Hangar18]

Nick, i appreciate your response, Lots of good stuff there, and priorities seem well placed.

On 7/23/2020 at 12:08 PM, Nick Thomadis said:

because -unarguably- it is different to aim at targets that are slow or stationary than targets which are moving at high speed.

I think you are mathematically incorrect here. It is literally the exact same process, and speed being higher, or lower does not make the formula change.

If you know a target's speed, course, and range, there is nothing different from a jet boat going 50 knots, or a tanker going 6 knots. it changes Vt in the formula, that's it. It's just a variable, the only thing that changes is how long the shell will take to arrive, which could be sooner or later depending on course. If the ship changed course, you would have to restart this again, but if they havnt changed speed, then that really is not a huge deal either. It's just changing cos(a). If they change speed, and course, then you have to start the entire process from scratch. And trust me,  it takes freaking forever to do this on a ti 84, not including the time to recheck bearing, range and speed. That being said, on a ship it's semi automatic, you just have to plug in Vt, d, and bearing, and it keeps track for you.

(v_c² - v_t²) . t² + (2 . d . v_t . cos(α)) . t - d² = 0

t = (- b ± √(b² - 4 . a . c)) / (2 . a)
where:
a = v_c² - v_t²
b = 2 . d . v_t . cos(α)
c = - d²

@madham82

If you have the length of the ship (or anything close) speed is not hard to judge.

Let's use a ship which we are guessing is roughly 124m long (this is actually a type 44 freighter). 

Mark the bow of the target in a bearing, start a stop watch, when the aft of the ship hits the bearing, stop the clock. in this instance, Let's say our ship took 40 seconds to clear the bearing. We're done here if the target has not changed speed, because we have Distance/time*1.94. So the ship in this scenario is going 6 knots. Because we have the speed, and we can time the ship, we know exactly how far the ship will go. At this point we are checking our ballistics charts.

Alternatively you can mark bearings over time, cross reference with range, and if you chart it out you will have distance and time, which is speed.

Here is what i used to really get the hang of this concept. You don't even need a visual on the target you can do this by sound.

http://ricojansen.nl/downloads/Sonar_Tracking.pdf

https://gameplay.tips/guides/4198-wolfpack.html

On 7/23/2020 at 12:08 PM, Nick Thomadis said:

USN battleships based on the Iowa hull can surely host 5'' secondary guns, and even larger. Can you share an image with the problem you anticipate?

My fault, I'm talking about the Large cruiser, and Battle cruiser hulls. It's the Iowa super structure, but you can't mount 5" guns on them. 5" guns aren't that large, but even on cruisers you can't stick em on. Not sure if this is intended or not.

On 7/23/2020 at 12:08 PM, Nick Thomadis said:

Very minor, since UI is still a work in progress. But we thank you for the report. 

I know it's bottom of the barrel priority, but it gives me headaches when comparing guns, it looks like the spacing is like .6 or .7 rather than .5

Pretty good WIP, I can't see any majot annoyances. The tool tips are really nice. There could certainly be a few QOL changes, something to compare mods/guns/ect.Actual values shown alongside percentiles would be helpful, but could cause clutter. An easier way to change small amounts of displacement on large ships. Maybe a better way to view technical stats? but i don't know how you'd do it. 

On 7/23/2020 at 4:34 PM, akd said:

nations tending to simply use the best propellants they had access to

That's the whole point of separating them, aside from just being pedantic about which is propellent or filler. Using the best is fine, but each nation might end up with something different. More options is also just cool in general, let's be honest here, customization is this games strength (aside from maybe the guy doing the graphics)

On 7/23/2020 at 12:14 PM, Nick Thomadis said:

About Armor, in general, we want to make it more complex. As we develop the game, we will present gradual improvements on this very important aspect of the game. Since the technology logic that will work for campaign is currently in process, and armor evolution is a big part of the technology, we cannot give specific details on how it will work as a further enhancement (Citadel, more armor zones etc.).

Armor penetration tables, will be more readable if we give the ability to switch armor values from "wrought iron" to more advanced armor types. It is in our list of future improvements.

Good luck, it's a huge and important task!

[Skeksis]

1 hour ago, Hangar18 said:

I think you are mathematically incorrect here. It is literally the exact same process, and speed being higher, or lower does not make the formula change.

Your equation doesn't take in account of 'trajectory conditions' and 'gunnery conditions', therefore with such conditions increased, for example: a stationary to a moving target, it is harder to hit, in RL.

That's what Nick is probably referring too. 

Edited July 25, 2020 by Skeksis

[madham82]

31 minutes ago, Skeksis said:

Your equation doesn't take in account of 'trajectory conditions' and 'gunnery conditions', therefore with such conditions increased, for example: a stationary to a moving target, it is harder to hit, in RL.

That's what Nick is probably referring too. 

Those conditions impact any firing solution. Speed does not create more deviation. Thats the whole issue with the penalty as is. It is as simple as moving the aim point as I stated earlier. Remember the firing ship is moving too.

[Cptbarney]

4 hours ago, madham82 said:

Those conditions impact any firing solution. Speed does not create more deviation. Thats the whole issue with the penalty as is. It is as simple as moving the aim point as I stated earlier. Remember the firing ship is moving too.

Basically world of warships gives the best example of this, if a ship is moving at 20 knots and your in an iowa you just place the cursor about 5 steps ahead and if 30knots 10 steps ahead. With the ship suddenly changing direction and speed you then have to try and predict where she will turn how far and if she might turn again as well which will mean you need to fire at angle (instead of a straightline) and also change the amount of steps to shoot at due acceleration and current speed.

Plus TC and GC should be added or subtracted on really since they are their own equations entirely along with damage to the hull and damage to gunnery systems (rangefinders, radar etc.)

[akd]

9 hours ago, madham82 said:

Those conditions impact any firing solution. Speed does not create more deviation. Thats the whole issue with the penalty as is. It is as simple as moving the aim point as I stated earlier. Remember the firing ship is moving too.

Not quite that simple.  Greater speed will increase the effect of errors.  If a ship is moving at 15kn a 10% error in estimation of enemy speed might not cause the enemy to be out of the pattern size of the salvo, whereas a 10% error in the estimation of the speed of a ship moving at 30kn will be greater and might be sufficient to allow the enemy ship’s predicted future position versus actual to be outside of the pattern size.

Edited July 25, 2020 by akd

[akd]

4 hours ago, Cptbarney said:

Basically world of warships gives the best example of this, if a ship is moving at 20 knots and your in an iowa you just place the cursor about 5 steps ahead and if 30knots 10 steps ahead. With the ship suddenly changing direction and speed you then have to try and predict where she will turn how far and if she might turn again as well which will mean you need to fire at angle (instead of a straightline) and also change the amount of steps to shoot at due acceleration and current speed.

Plus TC and GC should be added or subtracted on really since they are their own equations entirely along with damage to the hull and damage to gunnery systems (rangefinders, radar etc.)

Based on your description, that is a terrible analogy (but I have never played WoW).  If WoW shows your targeting solution in direct relation to the actual current position of the target, then it is a much simpler gunnery problem.   The actual position, course and speed of enemy ships was an unknown which had to be inputted to generate a predicted future position.  You were not shown your predicted solution in relation to the actual current position of the target.

[Cptbarney]

4 minutes ago, akd said:

Based on your description, that is a terrible analogy (but I have never played WoW).  If WoW shows your targeting solution in direct relation to the actual current position of the target, then it is a much simpler gunnery problem.   The actual position, course and speed of enemy ships was an unknown which had to be inputted to generate a predicted future position.  You were not shown your predicted solution in relation to the actual current position of the target.

You've basically failed to understand my entire comment, im talking about how different speeds affect gunnery in a small way (as in you only have to calculate for a few more paces forwards or back to compensate ship speed) and if the ship is moving at a constant speed you don't need to do much to change how you shoot said ship in the first place.

It's when the enemy ship starts changing course that you now need to suddenly try to predict where it will go, how much speed it bleeds how big its turning circle is, whether it is slowing down on purpose or not.

The only thing wows shows you is how long your shells take to travel to the target based on nothing, but distance. It would be nice if people would not mis-qoute me.

[akd]

If it shows a predicted position in relation to an actual target position, it is not a good analogy for the real world gunnery problem, where everything about the target (speed, course and actual location) must be estimated based on observations.  You are correct that because of shell flight time, changes to speed and course are much more important than speed itself, but what you suggest above is that absolute speed has little effect as long as it is constant.  That is true only if you have perfect knowledge of the target’s speed, or are shown a predictive solution in relation to a target moving on a constant course at a constant speed.   If WoW shows both the actual target position and where your shells will land, then it is not a good analogy for the gunnery problem, where you must estimate both where the enemy is and where it will be.

Where Dreadnoughts errors is not in applying significant accuracy errors related to speed, but in making the error itself a constant.  If speed and course remain constant, then the speed penalty to accuracy should eventually cancel out entirely with repeated observations of target and shell spots.  I’m sure it kind of works that way in WoW too, but knowing the actual position of your target is going to greatly minimize the initial effects of speed on estimates in target future position.

Edited July 25, 2020 by akd

[madham82]

4 hours ago, akd said:

Not quite that simple.  Greater speed will increase the effect of errors.  If a ship is moving at 15kn a 10% error in estimation of enemy speed might not cause the enemy to be out of the pattern size of the salvo, whereas a 10% error in the estimation of the speed of a ship moving at 30kn will be greater and might be sufficient to allow the enemy ship’s predicted future position versus actual to be outside of the pattern size.

You highlighted the suggestion I made in the earlier post. The game doesn't simulate estimated speed (or margin of error for it). We are given it exactly. So therefore if speed is known (not estimated), there is no margin of error. A 20kt ship and a 40kt one will not change the calculation in this case. 

So this reinforces how the current penalty is completely without basis. Because as you mentioned, the variation is in estimation of speed vs actual speed, not the numeric value itself. 

Edited July 25, 2020 by madham82
wrong quote

[akd]

We (the players) “know” the speed because enemy information is displayed without proper fog-of-war.  There is no reason to assume that this being known (viewable in UI) is factored into the game’s gunnery model in any way (and I don’t believe it is a factor in the accuracy table), just like the AI totally ignores and can not make use of the “known-to-player” torpedo launching status viewable in enemy ship panel.  That is a big problem with not applying proper fog-of-war to information: it creates unrealistic expectations or assumptions about AI behavior.

Edited July 25, 2020 by akd

[SonicB]

I agree with a couple suggestions here and would like to thank @Hangar18 for the mathematical proofs. It seems to me that the solution is as simple as applying speed as a multiplier to the 'target maneuver' malus rather than as a separate modifier. The only reasonable effect (within an achievably simple model) that speed can have on targeting is increasing the area that a ship could possibly be occupying after it completes a maneuver.

If I'm shooting at a 5kt target that starts a turn, I know it'll still be within a reasonably small cone on my gunnery plot even if I don't know its turning radius or how much rudder it has on. A 40kt target starting a hard turn, however, could generate anything up to a few hundred yards of lateral separation from its previous course in a matter of seconds.

Regarding estimating speed: it's always going to be a factor in gunnery, and I'm willing to be corrected by anyone with gunnery or Naval experience, but if I could put in a little bit of personal observation: I'm a yacht skipper in my off-time with a few thousand nautical miles' experience of playing Frogger in busy shipping lanes, and unlike estimating a turning vessel's position, the difficulty of estimating a vessel's speed doesn't exponentially increase with that speed. To elaborate, you can estimate pretty much any speed within a fairly comfortable range based on:

- Bearing and course observations over time, combined with your known course and speed.
- Direct visual cues (bow and stern waves, wake height, exhaust smoke, interaction with sea conditions, spacing and location of the standing waves versus a vessel's likely displacement)
- Knowledge of the vessel type, its capabilities and its likely operating conditions
- If you have it, radar or AIS (vessel transponder) signals

In fact, based on the first two datasets which are most applicable here, I would probably have a greater margin of error in estimating the speed of a vessel doing 10kts than one doing 30kts, because there is less variation between the regular bearing and course observations, as well as fewer and smaller visual cues. Of course, the consequences of getting it wrong tend to be on you much more quickly at 30kts, but that's a story for another time.

TL;DR: targeting difficulty due to maneuvring increases exponentially with target speed. Targeting difficulty due to target speed does not.

[madham82]

2 hours ago, akd said:

We (the players) “know” the speed because enemy information is displayed without proper fog-of-war.  There is no reason to assume that this being known (viewable in UI) is factored into the game’s gunnery model in any way (and I don’t believe it is a factor in the accuracy table), just like the AI totally ignores and can not make use of the “known-to-player” torpedo launching status viewable in enemy ship panel.  That is a big problem with not applying proper fog-of-war to information: it creates unrealistic expectations or assumptions about AI behavior.

1. It is being utilized, look at how the "fast target speed" penalty changes directly when the speed in the target display. You are assuming it isn't because that would make logical sense, but the impact it has to accuracy directly contradicts that.

2. Exactly how many salvos would it take to adjust for "bad" estimate on speed? A few perhaps (assuming the target maintains speed), but again actual game experience shows you can fire 100s of salvos with no improvement (or hits for that matter).

3. If an estimated speed was impacting accuracy, why does it not impact all speeds? Why does the fast speed penalty start around 28-30kts and peak at 37kts? 45+kt designs do not see any increase in the penalty, which would be the case if if it was being applied like you believe.

The realty is this penalty is not based in science, it is arbitrary. It is attempting to convey an weakness in gunnery that is not factual. This is easy confirmed by building a 37+kt DD and sailing in a straight line. Any ship targeting this one will not see the penalty ever change, which clearly indicates the "estimated speed" has no bearing on it. It can be 1km away with 100% identification and the penalty remains. 

What we are asking for is a penalty that represents a ship actually changing it's range/bearing (by maneuvering or changing speed) in between salvos, not sailing in a straight line that can be easily predicted. We already have pieces of this, we just need to strip out this one which is broken and creates an easy exploit that fundamentally changes the nature of ship design. Speed in naval design was never used a defense against gunfire, it was used to prevent getting in a gun fight you couldn't win. That was the whole idea behind the BC and Panzerschiff. 

[akd]

1. It is being used as a direct input to accuracy, not because “it is known” and being inputted into the calculations the ship would be making to produce a gunnery solution.  Note how it changes instantly and precisely.  There would be no way to gather such precise information instantly in reality.  Lack of proper fog-of-war is leading you astray.

As noted above, if the speed of an enemy ship were known precisely and constantly updated perfectly, it would basically be a non-factor in gunnery unless target course or speed changed after the guns were fired.

2. Yes, it is a constant because it is not a variable being deduced from observation and used in gunnery calculations, but a constant malus that has nothing to do with what your ship does or doesn’t know.

3. Estimates are not, as per above.  It is just a fixed malus based on absolute speed. That is a problem with the gunnery model.

[Hangar18]

6 hours ago, akd said:

There would be no way to gather such precise information instantly in reality

Sure there is. I can do it in a sim. This is not a long process, it takes more time to properly ID ships than to create a solution. Once you have the ship ID, and therefore length (you only need to be in the ballpark), probably about 5 minutes to have a perfect solution. I mean 99% accurate. I'm not good at it, and it is generally done by more than one person anyway. I suspect you could form a full solution in sub 1 minute with people who are in practice. I've seen some people who can eyeball the angle off bow (usually takes me about 2-3 for a confident result).

How fast you are going doesn't matter to me. honestly if you are going faster, it makes things easier because your time to cross distances becomes shorter, which therefore decreases the time I need to make my solution. Instead of taking bearing every 20 seconds, I can take em at every 10. When a targets are slow, you have to wait an eternity for them to travel to the point you can take bearings with a high confidence.

The same is true for getting speed. I may have to wait awhile for a ship going 6 knots to travel it's length (around a minute), but when you are doing 30 knots it's only going to take a couple seconds.

6 hours ago, akd said:

if the speed of an enemy ship were known precisely and constantly updated perfectly, it would basically be a non-factor in gunnery unless target course or speed changed after the guns were fired.

Ships don't change speed quickly for the most part, you'd almost need a course change to bleed off a meaningful amount.

10 hours ago, akd said:

We (the players) “know” the speed because enemy information is displayed without proper fog-of-war.

Unless you have no visual, poor visual due to weather, or range, unknown ship type (with no nearby known references), and no radar, the process to get the targets speed should not take more than a minute. 

Technically you only need course and range to solve for speed. 

8 hours ago, SonicB said:

targeting difficulty due to maneuvring increases exponentially with target speed

Yeah but the speed penalty has zero maneuvering involved. That's the issue.

16 hours ago, akd said:

Not quite that simple.  Greater speed will increase the effect of errors.  If a ship is moving at 15kn a 10% error in estimation of enemy speed might not cause the enemy to be out of the pattern size of the salvo, whereas a 10% error in the estimation of the speed of a ship moving at 30kn will be greater and might be sufficient to allow the enemy ship’s predicted future position versus actual to be outside of the pattern size.

None of that is really related to speed. The errors were caused by external factors which caused an incorrect speed. And the amplification of the errors is a function of range. 

 

Hopefully I mathed right, I'm exhausted.

Time of flight for AP Shell with MV = 2,500 fps (762 mps)
   10,000 yards (9,140 m): 13.2 seconds
   20,000 yards (18,290 m): 29.6 seconds
   30,000 yards (27,430 m): 50.3 seconds
   36,000 yards (32,920 m): 66.1 seconds
   40,000 yards (36,580 m): 80.0 seconds

If we are to take the center of a DD (Fletcher) as a point of aim we would have 57M to bow or aft before we will shift our point of aim away from the ship due to error. At 10ky

57/13.2*1.94 = 8.4 knots of error. At which point you are so far off you would be unlikely to not notice you have made an error. If you were in a pinch you have enough room for error, you could just assume 30 knots, and you're still likely on target. Even then the dispersion is so large, the ship is unlikely to be safe.

Because destroyers are small (and we already have size penalty) that's the best case scenario

If we are to take a BB (Iowa) and aim at the center of the ship, we have 131m of error. At *40ky* 

131/80*1.94= 3.2 knots 

At this range it is technically possible to not notice your error, but even if you are off by (what coincidentally happened to be 10%) 3.2 knots, you aren't off by much. If you overestimated speed, I would suggest you would actually notice, because a BB at 36 knots is odd. The dispersion pattern of the shells is 200m across, which means almost half the ship will still be inside an area you will expect to hit anyway.

This is a range the navy has zero data on in terms of accuracy. They never bothered with anything above 30ky.

Someone re math if I messed up, like I said, I'm tired, and I'm also, pretty garbo at math anyway.

Edited July 26, 2020 by Hangar18

[SonicB]

3 hours ago, Hangar18 said:

Sure there is. I can do it in a sim. This is not a long process, it takes more time to properly ID ships than to create a solution. Once you have the ship ID, and therefore length (you only need to be in the ballpark), probably about 5 minutes to have a perfect solution. I mean 99% accurate. I'm not good at it, and it is generally done by more than one person anyway. I suspect you could form a full solution in sub 1 minute with people who are in practice. I've seen some people who can eyeball the angle off bow (usually takes me about 2-3 for a confident result).

Out of interest, are you talking sim or IRL experience, because agreed, immediately judging the angle on bow of a ship in visual range is very possible. On a clear day, close enough to make out the ship's main features, with a good set of compass binoculars and a non-maneuvring target... I mean, completely innocent cargo vessel... within five degrees is pretty achievable and I don't do it for a living.

3 hours ago, Hangar18 said:

How fast you are going doesn't matter to me. honestly if you are going faster, it makes things easier because your time to cross distances becomes shorter, which therefore decreases the time I need to make my solution. Instead of taking bearing every 20 seconds, I can take em at every 10. When a targets are slow, you have to wait an eternity for them to travel to the point you can take bearings with a high confidence.

The same is true for getting speed. I may have to wait awhile for a ship going 6 knots to travel it's length (around a minute), but when you are doing 30 knots it's only going to take a couple seconds.

Completely agree, given that the most reliable way of calculating speed is also through bearing and course observations.

4 hours ago, Hangar18 said:

12 hours ago, SonicB said:

targeting difficulty due to maneuvring increases exponentially with target speed

Yeah but the speed penalty has zero maneuvering involved. That's the issue.

Exactly the point of the second half of my sentence

[SonicB]

Okay to return to the original question (sorry Nick but hope the discussion was useful!)

There's a bug when grouping ships of different classes which I don't think has been mentioned recently. For example, if I select a DD and attach it to a CL, sometimes the formation will end up with the DD as the leader. This makes little historical sense and also contradicts the game UI - logically, the way to get the DD as the leader is attach the CL to the DD.

If this is difficult to fix, could we please have a simple option to designate the selected ship as formation leader?

[akd]

8 hours ago, Hangar18 said:

Sure there is. I can do it in a sim. This is not a long process, it takes more time to properly ID ships than to create a solution. Once you have the ship ID, and therefore length (you only need to be in the ballpark), probably about 5 minutes to have a perfect solution. I mean 99% accurate. I'm not good at it, and it is generally done by more than one person anyway. I suspect you could form a full solution in sub 1 minute with people who are in practice. I've seen some people who can eyeball the angle off bow (usually takes me about 2-3 for a confident result).

How fast you are going doesn't matter to me. honestly if you are going faster, it makes things easier because your time to cross distances becomes shorter, which therefore decreases the time I need to make my solution. Instead of taking bearing every 20 seconds, I can take em at every 10. When a targets are slow, you have to wait an eternity for them to travel to the point you can take bearings with a high confidence.

The same is true for getting speed. I may have to wait awhile for a ship going 6 knots to travel it's length (around a minute), but when you are doing 30 knots it's only going to take a couple seconds.

Ships don't change speed quickly for the most part, you'd almost need a course change to bleed off a meaningful amount.

Unless you have no visual, poor visual due to weather, or range, unknown ship type (with no nearby known references), and no radar, the process to get the targets speed should not take more than a minute. 

Technically you only need course and range to solve for speed. 

Yeah but the speed penalty has zero maneuvering involved. That's the issue.

None of that is really related to speed. The errors were caused by external factors which caused an incorrect speed. And the amplification of the errors is a function of range. 

 

Hopefully I mathed right, I'm exhausted.

Time of flight for AP Shell with MV = 2,500 fps (762 mps)
   10,000 yards (9,140 m): 13.2 seconds
   20,000 yards (18,290 m): 29.6 seconds
   30,000 yards (27,430 m): 50.3 seconds
   36,000 yards (32,920 m): 66.1 seconds
   40,000 yards (36,580 m): 80.0 seconds

If we are to take the center of a DD (Fletcher) as a point of aim we would have 57M to bow or aft before we will shift our point of aim away from the ship due to error. At 10ky

57/13.2*1.94 = 8.4 knots of error. At which point you are so far off you would be unlikely to not notice you have made an error. If you were in a pinch you have enough room for error, you could just assume 30 knots, and you're still likely on target. Even then the dispersion is so large, the ship is unlikely to be safe.

Because destroyers are small (and we already have size penalty) that's the best case scenario

If we are to take a BB (Iowa) and aim at the center of the ship, we have 131m of error. At *40ky* 

131/80*1.94= 3.2 knots 

At this range it is technically possible to not notice your error, but even if you are off by (what coincidentally happened to be 10%) 3.2 knots, you aren't off by much. If you overestimated speed, I would suggest you would actually notice, because a BB at 36 knots is odd. The dispersion pattern of the shells is 200m across, which means almost half the ship will still be inside an area you will expect to hit anyway.

This is a range the navy has zero data on in terms of accuracy. They never bothered with anything above 30ky.

Someone re math if I messed up, like I said, I'm tired, and I'm also, pretty garbo at math anyway.

I think you are generalizing from submarine sim experience in a way that does not fully reflect the long range gunnery problem, especially when we are talking about fleet engagements.  But the conclusions are the same. Whether precisely known or not, target speed itself is not the dominant factor in accuracy, but it is also not irrelevant.

Edited July 26, 2020 by akd

[SonicB]

14 minutes ago, akd said:

I think you are generalizing from submarine sim experience in a way that does not fully reflect the long range gunnery problem, especially when we are talking about fleet engagements.  But the conclusions are the same. Whether precisely known or not, target speed itself is not the dominant factor in accuracy, but it also not irrelevant.

Would just like to add that @Hangar18's conclusions, while they may be drawn from subsims, are still broadly accurate in the real world where you're trying to figure out the speed of large ships at fairly long ranges. In general, I don't think it's harder to estimate a higher speed. Under some conditions, it is easier to tell the speed of a faster vessel through visual observation, due to the increased difference of bearing and target angle over time, and if you have access to additional readings from a rangefinder that's wider than most of the boats I've sailed, so much the better. (Of course, even WWII gunnery officers didn't have 4kW, >50nm range modern marine doppler radar with computerised course plotting, so I guess that's a fair trade.) Under other conditions, it can be harder, especially in difficult weather and sea conditions when observation and visual cues are limited - but if we abstract that into game terms I genuinely don't think it should be a factor on its own.

Once again, disclaimer is that my experience is based on trying to avoid large ships, not trying to sink them. As satisfying as that would be on occasion.

[akd]

no one claimed it was harder to estimate the speed of a faster vessel.