AE Naval and OOB Issues [OUTDATED]

[witpqs]

While I understand there should be some relationship, why should they follow a clear formula? Vessels are made with different designs of rudders and other relevant features that affect design. I would not expect a hard relationship. This is a general observation, I have no expertise in the subject and have no specific opinion on what you are pointing out.

[cardas]

I'd agree that in theory you could have a very complicated relationship between the game values and the real design that takes a multitude of factors into account. There are two things that says against that though; first the majority of ships do follow, at least with an allowance for some margin of error, a clear formula for their class. Secondly I find it difficult to believe that you'd be able to find enough data on all ships to make use of a very complicated formula, you could only do it for a subset of the vessels. So it's simply a matter of what the database in general seems to show and practicality.

That said, the issue I'm pointing out doesn't rely on there being a simple formula, it relies on the improbability of the Dutch ships being such outliers from their peers and that they all happen to have 30 as their maneuver value despite being so different.

[witpqs]

Yes, it comes down to if they had some information on those vessels you are seeing as outliers, and what that information is.

[cardas]

Indeed, it is of course impossible for me to know exactly what the reason behind the values are. I think my case is pretty compelling already, but the fact that the endurance/fuel values are also strange for the same ships (mainly the small AMc:s) lends further credence to my proposition that it's a database error.

I will use the Ardjoeno class as an example, it has an endurance of 600 with 120 fuel "units" at a cruise speed of 6 knots. So it has a very low cruise speed, only ARDs have lower cruise speed and no other ship cruises at such a low speed. Even at 7 knots you only find a very small amount of ships, mainly landing craft a single cargo class and notably also the Admiralty HDML for some reason (I think that's a bit strange as well) and the Dutch A class. The landing craft has a max speed of 8 or 9 knots and the cargo ship class (Std-E Cargo) also cruises at 9 knots. The Dutch boats and the Admiralty HDML are the only ships/boats with such a low cruise speed that has a max speed that is 10 knots or more. This is a major issue when you consider that 6 or lower means you do only 1 hex.

So moving on, the Ardjoeno also has an enormous amount of fuel and fuel consumption for such a small boat (tonnage: 75). In other words, not only does it cruise extremely slowly, it also pays excessively for every nautical mile it does. So you have a 75 ton diesel engined minesweeper doing 6 knots (1 hex) carrying more fuel "units" per ton of ship and consuming more fuel per nautical mile than, say, the Shell Type-A tanker that cruises at 13 knots and weights in at 9700 ton.

(I use fuel "units" so I don't anger Alfred [:)])

Edit:
I should add that I'm not entirely sure that the length/max speed values I've used are correct. If I continue with the Ardjoeno, I've seen several places that has it as 10 knots max speed rather than 15 knots, or with the length of the Ardjoeno class more or less switched with that of the Merbaboe class etc. Navypedia has it at 15 knots/1000 HP diesel while most other places puts it at ~10 knots/~135 HP diesel. It just underlines the difficulties of getting good data, especially when only relying on the internet.

While it does affect exactly what values they should have it doesn't take away from my belief that the original maneuver/fuel/endurance values are very wrong.

Most sources has Merbaboe as being heavier so it being longer would be logical, but breadth and draught is also a factor. I'd put it at 10 knots myself as apart from navypedia that seems to be the go to value, but as for the lengths who knows? Here's an example using some alternative length/max speed values that I've encountered. This also showcases that my formula isn't that sensitive within the 70+ maneuver span, a direct consequence of wanting to differentiate between the fast small boats and the slow small boats.

Code: Select all

 Length (m) - max speed - cruise speed - fuel - endurance - maneuver - ship name
 26,5       - 10        - 8            - 15   - 625       - 71       - Merbaboe
 22,7       - 10        - 8            - 13   - 545       - 71       - Ardjoeno
 

[US87891]

ORIGINAL: witpqs
Yes, it comes down to if they had some information on those vessels you are seeing as outliers, and what that information is.

That is exactly it. John did work with the maneuver parameter in Babes, but mostly for the main ship classes. There will be lots of outliers that he did not, frankly, get around to defining and that the rest of us did not think significant enough to update.

He had a simple model based on speed/length that worked for the majority of vessels. The very small and/or very fast vessels, (like PTs, etc..) needed some abstraction help to keep the values within the limits set by Gary Grigsby’s calculation code. It would not do to have invulnerable vessels under those conditions.

John has posted his maneuver calculations before, so I’m sure they are open source. If you want, I will ask that he post it again, if he still has it. It may be useful for your modifications of the outliers.

Matt

[cardas]

I do have some formulas that seems to give reasonable results for the majority of ships, but having the actual ones used for DBB would definitely be a plus if it isn't a bother.

[Dili]

Cardas can you share your formulas?

With caveat i must say in the research i have been making i doubt a formula can explain many (too much) reality. For my Mediterranean mod my research led me to tactical diameter of British ships and there are extraordinary variation independent of ship type. I know tactical diamater is not all that is in maneuverability but has a big enough impact.

Some examples by Class.

BB Nelson 670yd
BB KGV 970yd
CV Furious 1320yd at 25kt
CV Ark Royal 1010yd
CV Illustrious 1050yd
CV Eagle 800yd at full speed
CLAA Dido 600yd at 14kt
CL Fiji 750yd at 14kt
CA Kent 1040yd at 14kt
CL Edinburgh 640yd at 12kt
CL Southampton 780yd at 14kt
DD Tribal 815yd at 33kt
A Class Destroyer 600yd at full speed
A Class destroyer Leader 980yd at full speed(longer hull)

CA Dupleix 1000 at 11kt, considered a very bad value
CL Dugauy Trouin 750m at 28kt - this being French data doesn't mean it is precisely the "tactical diameter" of Royal Navy. This value is considered very good and contrary to most French ships of the period.

[cardas]

Sure, but they are simple linear functions based on ship length and max speed.

WL = ship waterline length in metres
S = max speed in knots

Class
DD: 70*S/WL + 46
CL: 160*S/WL + 15
Rest: 200*S/WL

CLs should reasonably merge with the rest when the S/WL ratio gets to 0,375 but there isn't any such CL in the original database. That would also be at a maneuver value of 75, while I think that somewhere around 68-72 you have to start using another formula so that you don't get very high maneuver values for fast MTBs. The crude one I used in my suggestions for the Dutch ships is that any non-DD gets the following: 15*S/WL + 65 at maneuver values greater than 70.
You could make compelling arguments for a less abrupt change by not using linear function at that point. Still, you'd have to use a function that can handle S/WL ratios from around 0,35 to over 2 (Japanese MTB-T1: ~18 m and 38,5 knots).

I won't claim this value accurately shows how manoeuvrable a ship is but they do agree reasonably well with the database values of existing (war)ships. Essentially this would mean that the maneuver value doesn't actually tell the full story about how manoeuvrable a ship is, tactical diameter etc is completely ignored. You could say that the maneuver value in the database is simply a value on how much of a ships own length it can clear in a given amount of time. Still, many other database values aren't exactly 100% accurate to their name either, the most important part is to use values that fit in with the rest unless you plan to rework the entire database.

This is my functions compared to DBB database values of ~250 ships. X-axis is the S/WL ratio times 100.

As you can see it's not exactly the same but not horrendously bad either. Some of the outliers might either be due to bad speed/length data on my part or on the developers part or some other special considerations. Some notable differences are PC Argo (DBB: 43, calc: 57), AM Bird (DBB:59, calc: 48), DE Black Swan (DBB: 60, calc: 44) and PC Grimsby (DBB: 53, calc: 40). Landing craft are also quite different and so is obviously the Dutch vessels I've mentioned in this thread (they are the ones at 30 maneuver value).

[Dili]

I think your values line up very nice.
About you oddities maybe they have 2 rudders, or draught was accounted for.



The problem is that a formula doesn't deal with reality, and i mean lot like i posted above. An Iowa class could reportedly run inside a Fletcher DD turn.

From a comment in a defense website . The Iowa-class battleships had a tactical diameter of 814 yards @ 30 knots, whereas the much smaller Fletcher-class destroyers had a tactical diameter of 950 yards @ 30 knots.

Note that speed is apparently not much important to tactical diameter value. See paper link below.


So how to work the values in game?

To take to extreme how to handle the values of 670yd of Nelsons and the 980yd of A Class leader destroyer?

Tactical Diameter is an important value of maneuverability in escaping a torpedo attack. Another issue is also we don't how the game handles "maneuverability" . If for example includes the ship physical size.


Here is a paper to the formulae
http://www.academia.edu/225461/On_Turning_Ships

His formula:
DD x5.5 the length
Cruisers and BB's x3.5 the length
Carriers x4.5 length

He also says that tactical diameter increases or decreases only 10% to 30kt or -10% to 7kt from a base value of 15kt so says don't bother at least for the game kind is talking about.

[US87891]

ORIGINAL: cardas
I do have some formulas that seems to give reasonable results for the majority of ships, but having the actual ones used for DBB would definitely be a plus if it isn't a bother.

No bother. In an abstraction, anything that works, works. Your formulas seem to work. John's are not all that different.

SQRT[1000(WL/Sp)] = Raw Num
WL = waterline (or overall) length in meters. Doesn't really matter that much which one is used.
Sp = "practical" top speed in knots. This is not the biggest number one can find for trials.

Then adjust Raw Num into various "type" boxes. Box multipliers were developed from HP/Wt, waterplane, prismatic, and block coefficients, and specific acceleration. Take a ship class' Raw Num and multiply by its type "box" value.

Num x 2 for old 20kt BBs
Num x 2.3 for other BBs
Num x 2.5 for BCs (including Alaskas)
Num x 3 for CAs
Num x 3.3 for "big" CLs
Num x 3.5 for ordinary CLs
Num x 4 for DDs

SQRT function is just to un-scatter and smooth the data so they fit within the endpoints defined by GG's code, and return sensical result values from the different combat calculations that use Mvr as an input variable.

Matt

[cardas]

I see, interesting that there were that many categories. I tried it out and . I actually think I will stick to my own simple functions for now because, as you said, if it works then it works. I appreciate knowing what was actually used for DBB though, so thank you!

(I assume that SQRT[1000(WL/Sp)] = Raw Num is a typo and it should be SQRT[1000(Sp/WL)] = Raw Num)

ORIGINAL: Dili
So how to work the values in game?

To take to extreme how to handle the values of 670yd of Nelsons and the 980yd of A Class leader destroyer?

Tactical Diameter is an important value of maneuverability in escaping a torpedo attack. Another issue is also we don't how the game handles "maneuverability" . If for example includes the ship physical size.


Here is a paper to the formulae
http://www.academia.edu/225461/On_Turning_Ships

His formula:
DD x5.5 the length
Cruisers and BB's x3.5 the length
Carriers x4.5 length

He also says that tactical diameter increases or decreases only 10% to 30kt or -10% to 7kt from a base value of 15kt so says don't bother at least for the game kind is talking about.

Neat, but as written in the paper there is no clear correlations from the simple base physical parameters and tactical diameter. Basing maneuver on ship length and speed is appealing as it takes both physical size into account to a degree and also an important part evasive part (speed). This information is also mostly readily available (speed is admittedly a murky area if you want to use "real" practical speed). It does ignore other important aspects such as how quickly the speed can be changed and the tactical diameter but such information can be much harder to come by, especially if you are only a casual modder without access to a lot of sources of information. The box values that US87891 posted is kind of a generic catch-all for this and can be likened to the generic formula for tactical diameter in the paper.