akd

Yes, but with early radar there would be significant range error and even more significant bearing error, and no ability to observe fall of shot. So "seeing" with radar is not the same as visual observation. Later radar designed for fire control would improve on that, but with the caveats above. For finding the enemy's rough location, the advantage is almost absolute (perhaps debatable with early radar in the most ideal visual conditions), but for shooting at the enemy it is much more conditional and interrelated with visual conditions.

Yes, but radar does not visually see and ID the target. I disagree, however, that it is necessarily higher than the highest visual spotting position. Probably often true, but not always. Regardless, that is not the point. Seeing with radar in this era was not the same as visually observing a target.

When we speak of LOS here, I think we are talking about literal line of visual sight.

Yes, but I think simple penalty is insufficient. That still leaves you with the possibility of a single destroyer conducting fire control for an entire fleet. I think the rule should be simpler: either the ship itself or divisional ships sailing within XXXm must be able to visually see the target to fire on it with the following limitations: Before the advent of complex fire control computing and dedicated fire control radio frequencies, ships firing on same target should suffer a significant "concentration fire" penalty. (And this should be per ship selectable technology, not unlocked and applied to all ships subsequently built.) Radar II (assumed to be second generation search + fire control radar, not first-gen search only) can allow ship (or division with above restrictions) to fire on an unseen target, but only within 18,000m (as noted above this was broadly the range for large caliber splashes to be spotted on radar, allowing for adjustments). Possibly blind fire beyond that range might be allowed with a significant penalty for not being able to spot and adjust. Radar should not give any information on targets that are out of visual sight except possibly a broad classification such as small, medium or large and course + speed.

Well, not really, because all radar currently does is extend LOS, and then gets restricted by the same factors that affect LOS. It is a poor representation of radar.

Lyddite is not a propellant. Or at least not for anything but martyrdom operations. Might be a source of some confusion in the various effects.

He was not controlling the BC in the screenshot. At the same time that the AI would not fire at the destroyer because of accuracy, his destroyer apparently scored 73 gun hits on the BC. That the BC fired 8 shots at beginning (presumably after first sighting the destroyer) then stopped firing after I think can be attributed to the following recurring issue that seems mostly repeatable (but hard to test since there is no way to do multiple runs with identical conditions): The first shot fired will almost always "lock" the target, then is followed by a salvo of all guns that can fire. This is true even if it is 1890 and both ships are closing on eachother at full speed and fire opens at 6km+ range and you are only firing with a single gun. Only after this first ranging shot + subsequent salvo does the relative motion of the two vessels seem to get taken into a account, which is then followed by loss of the target lock and several more salvos to again acquire a lock even if the ships are then sailing on roughly the same course and at roughly the same speed. Under some conditions, the target lock will never be reacquired. This can best be observed if you acquire the enemy first and you set your guns to aggressive (that way you can often see 1 shot target lock even when the initial accuracy before firing is like 0.6%). Locking the target (having a firing solution for long-range gunnery) on first shot in these conditions should be impossible. The target has just been sighted, so there is no time to determine a course and speed (which anyways would require range-finding not available in 1890), and the conditions are the worst possible: the ships are approaching each other at full speed, imposing the highest possible range rates, while often also sailing on courses that are not reciprocal adding a bearing change problem).

Well, as Nick has indicated, the AI is not going to fire with 0.1% or 0% accuracy.

7.3% is the baseline for the gun, not relative to current target and conditions. The BC likely fired one ranging salvo and one full salvo because of a problem / bug in the gunnery model I will make a separate post on. As Nick noted, you can see the current accuracy of 0.1% in your screenshot, he just thought you were controlling the BC.

Target size and speed factors dominate the accuracy calculation (but this should not be the case).

I don’t think AI ever uses aggressive.

Would be fun if it were a negotiation mini-game, rather than just having a historical or random treaty imposed (although that might be your lot as head of the Navy).

You really think the whole point of that mission is to teach you to click one time? It didn't used to be... Anyways, it should establish a baseline for the entire era. It is essentially 1890 tech at 1890 engagement distance (with opportunities to modify one or two specific techs beyond the baseline).

Caveat: once central firing of main battery is developed, centerline and side turrets with same guns should be treated as a single battery, as should turrets with the same gun but in different numbers.

This is not simply how it worked in reality and was highly situational and dependent on ship-specific technologies, most of which would only be available at the very end of the era, and even then arguable in fully achieving what you describe.

Play the first academy mission "Target Practice" with light armament then come back and let us know how things "feel." (It is objectively absurd, but also “feels” wrong, or at least pointless.)

Abstraction of human factors is necessary, especially command and control. Abstraction (or extreme fudging based on consensus of “feel” or whoever shouts loudest) of factors like armor penetration, armor protection, accuracy and spotting is not. It is actually a huge waste of time as it just leads to constant back and forth based on subjective feel (see post below). Better to make objective what can be objectively established, then shape abstraction and “feel” factor around elements that must be abstracted.

You are confusing realistic tactics and technology with the reductio ad absurdum of historic verisimilitude. The former is possible with limitations. It cannot be denied as fun simply because it objectively relates to reality. In fact, it could be argued that the sandbox aspect of the game is only really fun if you face the same relationship between tactics and technology that existed historically and then get to make your own decisions in that context.

Encountered same, with a consistent alteration between all turrets firing, then only one turret firing through several salvos. Reported in game.

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).

They are closely interrelated and you can’t really have one without the other. Look at the quote in the bottom of my post. I’m not sure this is true. The big problem was more specific: ships were incapable of hitting destroyers or TBs with guns, even within ridiculously close range, and of course showering them with close misses has no physical or morale effect. Now the problem broadly remains: big ships are still mostly incapable of hitting small ships even at ridiculously close ranges, but now those small ships can achieve 100% hit rates with guns vs. big ships even at what is for them medium range, and across the board accuracy is generally boosted out of line with historical limits, especially for the bookends of our broad era (although with some limited testing I’ve found hit rates for middle-caliber guns circa 1910 to be in line with trials, but with “target locking” occurring much too fast). This suggests that largely the issue was not baseline accuracy, but more specific to the interaction of the various malus / bonus factors and to a gunnery model that treats the long-range gunnery problem the same as the short-range gunnery problem (they are in fact vastly different.) Wanting “incentive” to add big secondary batteries to pre-dreadnoughts highlights a pretty fundamental difference in how you might approach this. What if their presence in reality was actually a fallacy based on bad presumptions? You want the game to justify this fallacy post-facto? Wouldn’t it be better to have the actual relationship between the technology and tactics play out in game as in reality, then you can try the historic designs, but also be rewarded for taking different approaches.

Some more from D.K. Brown’s The Grand Fleet on secondary batteries and their effects during Jutland:

Number of side main turrets is a campaign tech unlock, so you would need to play with the year to get it working.

You could probably generalize safely to say that "one good hit" would not do unless that hit: 1. Detonated a main magazine or initiated a chain of events that directly led to main magazine detonation. 2. Caused sudden, large buoyancy loss on one side of the ship causing it to capsize (typically only the result of torpedo or mine strike).

There needs to be some allowance for minor flooding that is not repairable, such that cumulative gunfire damage can lead to sinking. I am seeing single mid-caliber shells (10-12”) cause flooding on par with a torpedo hit to early BBs, but if the ship does not sink, I have also seen this sort of flooding completely repaired and all buoyancy subsequently restored and the ship is back to square one.