|
Post by rimbecano on Aug 8, 2019 23:53:33 GMT -6
I believe that the CEP at daytime combat ranges is large enough compared to the size of a warship that any point on a ship is roughly equally likely to be hit, but I'm not certain of that.
|
|
pcasey
Junior Member
Posts: 58
|
Post by pcasey on Aug 9, 2019 8:58:22 GMT -6
I believe that the CEP at daytime combat ranges is large enough compared to the size of a warship that any point on a ship is roughly equally likely to be hit, but I'm not certain of that. I don't have the full CEP numbers, but I do know people who spotted naval artillery onto land targets (I used to work with a guy who spotted for the New Jersey in Lebanon) and his version was that you basically couldn't hit a target even as small as a building at range ... you sort of had to just range the shells onto the overall complex of buildings and then fire for effect until the specific building you wanted to hit happened to take a shell at random. I'd have to imagine that a moving naval target would be even more difficult to hit at normal combat ranges. The data I do have for long range shooting was from the Iowa in the late 1980s (so relatively modern fire control) at 34,000 yards (max range). She put 14 out of 15 shells into a 250m circle (which is pretty good) under ideal conditions e.g. stationary target, in the med, on a bright sunny day. Assuming the shell spread was linear with range (which is a little untrue because you tend to get more projectile yawing at long range and hence the distribution at close range is a little tighter, but this is close enough for government work: 17,000 yards (typical WW II combat range) ... 125m circle 9,000 yards (WW I combat range) ... 60m circle All in all I just don't think the data supports the theory that even a crack gun crew could hit something as small as a turret at typical combat ranges. Even in WWII it took some luck to even hit the target ship, much less a subsystem within it.
|
|
|
Post by director on Aug 9, 2019 9:31:06 GMT -6
pcasey - the Iowa class retained their original analog fire control equipment through their active service life. Analysts concluded that modern computers wouldn't increase accuracy. Basic gunnery hit percentages in WW1 were around 3% (0% to perhaps 5%). WW2 gunnery was not a lot better, though when actions were fought at close range the number of hits could be quite high (USS Washington versus Kirishima, for example). Fletcher Pratt's naval wargame, played with scale models on a ballroom floor, used a stick with a disk on the end and a pin off-center on the disk. On making a hit the player twirled the stick and stabbed a paper cutout of the target ship. Whatever the pin happened to strike got hit. A lot of active naval officers played; at one point the game had quite a mass following. www.si.com/vault/1963/09/23/616528/the-worlds-most-complicated-game In general, hits will tend to relate to surface area (including deck at some angles). So the probability of a turret hit is roughly the percentage of exposed area taken up by turrets, except that varies with target angle and range.
|
|
pcasey
Junior Member
Posts: 58
|
Post by pcasey on Aug 9, 2019 9:49:38 GMT -6
pcasey - the Iowa class retained their original analog fire control equipment through their active service life. Analysts concluded that modern computers wouldn't increase accuracy. Basic gunnery hit percentages in WW1 were around 3% (0% to perhaps 5%). WW2 gunnery was not a lot better, though when actions were fought at close range the number of hits could be quite high (USS Washington versus Kirishima, for example). I think what this is telling us though that you couldn't actually use battleship guns to target a specific subsystem (like a turret) on a target ship ... the accuracy just wasn't there at combat ranges. Would lead us back to the original premise that turret impacts should be proportional to the percentage of the ships cross sectional area covered by the turrets and/or barbettes.
|
|
|
Post by director on Aug 9, 2019 9:53:57 GMT -6
Sure - my edit crossed your response LOL.
I list the FP Naval War Game because they tried all sorts of complicated systems for gunnery before falling back on a simple, more general method.
|
|
|
Post by dorn on Aug 9, 2019 10:46:52 GMT -6
pcasey - the Iowa class retained their original analog fire control equipment through their active service life. Analysts concluded that modern computers wouldn't increase accuracy. Basic gunnery hit percentages in WW1 were around 3% (0% to perhaps 5%). WW2 gunnery was not a lot better, though when actions were fought at close range the number of hits could be quite high (USS Washington versus Kirishima, for example). Fletcher Pratt's naval wargame, played with scale models on a ballroom floor, used a stick with a disk on the end and a pin off-center on the disk. On making a hit the player twirled the stick and stabbed a paper cutout of the target ship. Whatever the pin happened to strike got hit. A lot of active naval officers played; at one point the game had quite a mass following. www.si.com/vault/1963/09/23/616528/the-worlds-most-complicated-game In general, hits will tend to relate to surface area (including deck at some angles). So the probability of a turret hit is roughly the percentage of exposed area taken up by turrets, except that varies with target angle and range. So it means that 20 % is practical maximum in ideal circumstances and WW1 style battle of two fleets the probability of turret hit would be something about 5 % even less.
Thanks for the info, it was as interesting as amusing.
|
|
|
Post by dorn on Aug 9, 2019 11:10:29 GMT -6
Quite interesting readings.
|
|
|
Post by axe99 on Aug 9, 2019 17:57:27 GMT -6
Assuming the shell spread was linear with range (which is a little untrue because you tend to get more projectile yawing at long range and hence the distribution at close range is a little tighter, but this is close enough for government work: 17,000 yards (typical WW II combat range) ... 125m circle 9,000 yards (WW I combat range) ... 60m circle All in all I just don't think the data supports the theory that even a crack gun crew could hit something as small as a turret at typical combat ranges. Even in WWII it took some luck to even hit the target ship, much less a subsystem within it. Thanks for the info - it's good stuff . Not suggesting for a second anyone could reliably hit something of that size at that range, but a 125m circle, for example, is around half the length of a capital ship of the era (less than half for a SoDak, more than half for an Iowa) - so assuming that their ranged in and actually landing where they want them to land, would it be fair to say that the actual part of the ship being targeted is less than the whole ship (for example, say the circle was centred on the middle of the ship, or the forward battery, or the rear battery), thus the probability of a turret hit is greater than the percentage surface area of the whole ship that the turret comprises? Ie, no fire control team is likely to centre their fire on the bow or stern, but rather the more more important parts of the ship, putting the CEP in an area where there's relatively more turret? All questions, I don't know the answers. I just think a "it's just the percentage of the ship covered by the turrets" feels overly simplistic (but I could be wrong here - some things really are quite simple, although most things aren't, and my general impression of long-range naval gunfire is that it wasn't).
|
|
pcasey
Junior Member
Posts: 58
|
Post by pcasey on Aug 9, 2019 19:01:02 GMT -6
Assuming the shell spread was linear with range (which is a little untrue because you tend to get more projectile yawing at long range and hence the distribution at close range is a little tighter, but this is close enough for government work: 17,000 yards (typical WW II combat range) ... 125m circle 9,000 yards (WW I combat range) ... 60m circle All in all I just don't think the data supports the theory that even a crack gun crew could hit something as small as a turret at typical combat ranges. Even in WWII it took some luck to even hit the target ship, much less a subsystem within it. Thanks for the info - it's good stuff . Not suggesting for a second anyone could reliably hit something of that size at that range, but a 125m circle, for example, is around half the length of a capital ship of the era (less than half for a SoDak, more than half for an Iowa) - so assuming that their ranged in and actually landing where they want them to land, would it be fair to say that the actual part of the ship being targeted is less than the whole ship (for example, say the circle was centred on the middle of the ship, or the forward battery, or the rear battery), thus the probability of a turret hit is greater than the percentage surface area of the whole ship that the turret comprises? Ie, no fire control team is likely to centre their fire on the bow or stern, but rather the more more important parts of the ship, putting the CEP in an area where there's relatively more turret? All questions, I don't know the answers. I just think a "it's just the percentage of the ship covered by the turrets" feels overly simplistic (but I could be wrong here - some things really are quite simple, although most things aren't, and my general impression of long-range naval gunfire is that it wasn't). The Iowa's 250M circle I was quoting at was in ideal conditions e.g. stationary target, nice sunny wind free day, flat calm, etc. I think they were shooting at a target somewhere off the coast of Crete if I remember. Was really more of a test of the efficiency of the gun barrels and ammunition than it was representative of what combat accuracy would look like. So accuracy will never be BETTER than that but it will almost always be worse . Other folks here are quoting an actual hit rate in combat conditions and typical ranges of like 3% to 5% of the shells fired hit the target at all. Some subset of the hits, of course, would be turret hits, but its not like any combatant in WWI or WWII was successfully hitting 100% of their shells into a 150M circle at 17,000 yards. If you look at something like the battle of the falkland islands, HMS Invincible fired off something like 600 main battery shells at Gneisenau and Scharnhorst and is estimated to have landed something around 10 long range hits (once Gneisenau was disabled Invincible did land a number of full broadsides at point blank range late in the battle). Admittedly that's an extreme case ... early WWI fire control, lots of funnel smoke to make the British gunners life harder, and an actively maneuvering target, but its still representative of the point. Big guns are just not that accurate under combat conditions so its a numbers game ... you throw a lot of shells at the target and statistically some of them are going to hit. If I remember the hit rates at Dogger Bank were not much better but I don't have that data handy (knowing this forum though somebody will dig it up .
|
|
|
Post by axe99 on Aug 9, 2019 19:20:45 GMT -6
Admittedly that's an extreme case ... early WWI fire control, lots of funnel smoke to make the British gunners life harder, and an actively maneuvering target, but its still representative of the point. Big guns are just not that accurate under combat conditions so its a numbers game ... you throw a lot of shells at the target and statistically some of them are going to hit. Absolutely - I'm coming from a statistical background in all this (and this is just my hypothesis, based on a training in stats and how things that are 'random' tend to work out once there are enough instances, not based on gunnery data) that the probability of it landing somewhere on the ship isn't equal for all parts of the ship. This is affected by the distribution of the shells within the CEP (which as you've mentioned is itself affected by the range of the action), where the centre of the CEP is aimed at, how accurate that aim is relative to where the actual thing that the CEP is aimed at really is at the time the shells land (so, in effect, how good is the fire control solution in that particular situation), the size of the CEP and the proportion of the area within the CEP that is made up of turrets relative to the size of the ship. If we establish that the CEP is smaller than the entire ship (which, as you say, isn't necessarily the case, given the size of your CEPs weren't in battle conditions), then at the very least, as long as all the turrets are in the CEP (which they may not be), then turrets should be hit at a greater percentage than their overall share of surface area. The other point I raised, but hasn't been addressed by anyone, is that are the hits in-game at normal real-world battle ranges? I've seen both from personal experience and what I've seen on the forums, that it's quite tempting to close the range more than may have occurred in actual battles (the consequences for a catastrophic explosion in RtW2 being a good deal more gentle than IRL ). The closer the ships are, the more likely that the CEP is going to only cover part of the ship, and presumably (but we don't know this - IRL and the game may both work differently - I have no idea of the targeting priority of ships at 'close enough range to make it worthwhile pointing at a specific part of the ship' would be) will be targeting the more important areas so more likely to have turrets in the CEP, all else being equal). Now, I may just being clueless, so all of the above factors aren't actually different - maybe the probability of a shell landing at any point in the CEP is the same as any other point (I'd find this very unlikely, given the way most statisics of physical observations work, but I don't have any gunnery data here, so it's all guessing), maybe the CEP is always larger than the ship, maybe there were never any targeting priorities even at close range, etc, in which case I'm sorry for the digression. I just think we should be careful about questioning in-game results relative to real-world results without a good body of evidence that they are, indeed, different enough to warrant suggesting changes (and they may well be different enough - I just haven't seen enough evidence in this thread yet that would satisfy me if it was up to me to make the decision - I'm afraid I don't have access to the kind of data required to make it either, or would just look it up ).
|
|
|
Post by director on Aug 9, 2019 23:21:55 GMT -6
The whole subject is murky. Navies didn't make a point of telling the public (or other navies) what their accuracy was. Some navies didn't practice firing live ammo much - too much wear and tear on the barrels and liners, not to mention the cost of propellant and ammo. I don't know what it costs to loft a small car out to the horizon, but I'll wager it isn't cheap. Gunnery practice was almost always slated to make things look good - stationary or slow targets, good daylight weather, short ranges and so forth. My belief is that combat was always poorer shooting than exercises (not a surprise to anyone here) and that long-range gunnery (that war-winning tactic so loved by the Japanese and US navies) was a fantasy. As axe99 says, in our wargames we find ourselves constantly closing the range to get hits. We don't do that because we are stupid or reckless (well, you guys aren't) but because we (almost always) must get hits to win. What do we see from actual combat in WW1 and WW2? Navies closing the range, or not getting any significant number of hits. In actual battle, nobody could shoot well enough to make long-range fire effective - you ran out of ammo before enemies. I also think it is revealing that naval gunnery reached a practical 'plateau' sometime in the 1930s and, with the obvious exception of radar range-finding, did not improve for a generation. US analog fire control on the Iowa-class was still substantially un-improvable in the 1970s and 80s... the limitations were not in the fire control but in the myriad factors not subject to control. Indeed, the first time we see accuracy increase again is with the introduction of active guidance - a theoretical possibility that I'm not sure was made practical for warship shells. An active-guided rocket-assisted shell would make touching the enemy at 40 miles a definite possibility, but that system can't compare in flexibility, lower cost and with the even longer ranges offered by missiles and aircraft. axe99 - Your point about us 'amateur admirals' closing the range in combat made me think of another anecdote from the Fletcher Pratt wargames. Evidently some of the most-feared opponents were the amateurs and not the actual experienced naval officers. The amateurs would manuever fearlessly and close the range without regard to casualties while the real navy men acted to preserve their treasured ships and men. One other: a Prussian army officer on holiday taught his two friends - university professors - to play the 'kriegspiel', only to be horrified as they slaughtered thousands of their 'men' in pursuit of victory. They had, he thought, not the proper spirit...
|
|
|
Post by dorn on Aug 10, 2019 0:50:38 GMT -6
It is quite similar what happened at Battle of River Plate. Tactically Graf Spee has all advantages on her side but at the end she was badly hurt be fearlessly attack much better armed ship.
Relating to capital ship hits etc. I think that Royal Navy was quite right over fighting ranges of battleships in WW2. In reality almost all battles were fought at that distances and the ones which were fought at large distances have scored not so many hits to be decisive.
And at that combat ranges any hit is just luck of statistics and any particular hit on certain part of ship even more. They can be same change of probability of hit by targetting certain part of ship but I would expect that it would be insignificant to be meaningful.
|
|
|
Post by axe99 on Aug 10, 2019 4:36:56 GMT -6
I don't know what it costs to loft a small car out to the horizon, but I'll wager it isn't cheap. Great post Director The deeper stuff is gold as ever, but for whatever reason I particularly enjoyed this bit On the ridiculously optimistic about long-range firing, apparently (according to Kaigun) in one 1939 gunnery test the IJN reported that with spotting aircraft and using indirect fire, at a distance of 32,000m, 12 per cent of rounds fired hit. Needless to say, while I'm a _long_ way from a gunnery expert, I was quite confident not to take this figure seriously (either as necessarily the results of the test, and absolutely definitely as something that could ever be achieved in a battle situation)! I'll stop rambling on now about turret hit percentages - I do think we probably don't know enough to be sure about the turret hit thing (and my position doesn't need repeating), but I don't think I'll have anything new to add.
|
|
|
Post by director on Aug 10, 2019 11:10:30 GMT -6
Something to bear in mind: you don't have to hit a turret to put it out of action.
A graze to the barbette or a nick on a barrel (admittedly, physical contact made if not a full-on impact) can cause a crew to cease fire while an inapection is made. Damage to engineering, in the way of electrical and/or hydraulic failures, will do the job. In low visibility, loss of radar, or damage to range-finders and gunnery control stations can do it. If turrets share a common magazine, power line, hydraulic piping, ventilation, etc and et al then ship damage can absolutely affect its firepower even if the turrets are not actually hit.
|
|
|
Post by dorn on Aug 10, 2019 11:31:57 GMT -6
Something to bear in mind: you don't have to hit a turret to put it out of action. A graze to the barbette or a nick on a barrel (admittedly, physical contact made if not a full-on impact) can cause a crew to cease fire while an inapection is made. Damage to engineering, in the way of electrical and/or hydraulic failures, will do the job. In low visibility, loss of radar, or damage to range-finders and gunnery control stations can do it. If turrets share a common magazine, power line, hydraulic piping, ventilation, etc and et al then ship damage can absolutely affect its firepower even if the turrets are not actually hit. I think you are right. And such hits even outside turret or barbete is reported in RTW as turret hit.... Probability of such hit is impossible to estimate.
|
|