Bismarck and her contemporaries

[Byron Angel]

With a bearing accuracy of </= 1/10th of a degree (</= 1.8 mils) are we saying that Seetakt had full blind-fire capability in early 1942? This level of accuracy is, at least at first glance, equivalent to US Mk8 FC radar performance.

[Dave Saxton]

Actually the Calias B was demonstrating a full blind fire capability as early as Sept 1940, and the Calias A probably had the capability but using a different lobing technique earlier than that. The Admiralty asked an RAF specialist to go down and see how how the Germans were directing such deadly accurate long range coastal artillery in the dark of night with no aid from artificial light. He reported that it was indeed radar operating on 80cm with a PRF of 500, which reverberated through the Admiralty like a thunderbolt, and many refused to believe it. Howse alludes to this when he describes when the decision to attempt to jam the Calias B Seetakts was first made in Feb 1941. A coastal convoy was just being cut to pieces by coastal artillery late one stormy night, but the situation was aleivated quite a bit when they turned on the jammer.

Aboard ship however, the fine bearing accuracy that could be attained by the radar itself could be lost in the traverse mechanicals of the range finder mounting. These were not directors, as the Germans used seperate optical instruments for range and for bearing. A solution to this problem was first applied in late summer 1941 on Tirpitz. There may have been a progressive improvement of the effectiveness of the remedy to the problem, with the net bearing accuracy being further improved over time. It was reported that tests conducted on Prinz Eugen during 1943 proved that no further improvements in Seetakt's bearing accuracy aboard ship were needed.

[Karl Heidenreich]

Dave:

Actually the Calias B was demonstrating a full blind fire capability as early as Sept 1940, and the Calias A probably had the capability but using a different lobing technique earlier than that. The Admiralty asked an RAF specialist to go down and see how how the Germans were directing such deadly accurate long range coastal artillery in the dark of night with no aid from artificial light. He reported that it was indeed radar operating on 80cm with a PRF of 500, which reverberated through the Admiralty like a thunderbolt, and many refused to believe it. Howse alludes to this when he describes when the decision to attempt to jam the Calias B Seetakts was first made in Feb 1941. A coastal convoy was just being cut to pieces by coastal artillery late one stormy night, but the situation was aleivated quite a bit when they turned on the jammer.

Aboard ship however, the fine bearing accuracy that could be attained by the radar itself could be lost in the traverse mechanicals of the range finder mounting. These were not directors, as the Germans used seperate optical instruments for range and for bearing. A solution to this problem was first applied in late summer 1941 on Tirpitz. There may have been a progressive improvement of the effectiveness of the remedy to the problem, with the bearing accuracy being further improved over time. It was reported that tests conducted on Prinz Eugen during 1943 proved that no further improvements in Seetakt's bearing accuracy aboard ship were needed.

This is really awesome!

[IronDuke]

in many ways accurate fire is actually more difficult to achieve in training than in battle.

I simply do not believe this: For example in battle one cannot pick the state of the sea or of visability. The target ship will most likely be moving at high speed on an unpredictable course, The Captain Officers and crew may or may not be very tired, they will certainly be more stressed than on a practice shoot and, perhaps most importantly, the enemy will be firing back and, possibly, hitting you...

[Bill Jurens]

You will note the key phrase "in many ways". Some things are easier to do in practice shooting. Others are actually easier to do in combat.

The point I was trying to make is that the differences between the accuracy of training shoots and combat shooting is typically fairly predictable, and often not as great as some might believe. In peacetime, and even in war, difficulties in scheduling and the need to control costs meant that big gun training shoots were often fired under far less than optimum conditions even in peacetime. Along similar lines, one is typically heavily restricted in the amount and type of shooting one can do in training. In combat, 'throwing away' two or three salvos to get an early spot would probably be acceptable; in training one does not even have have the luxury of of firing 'warming rounds' etc. Nor can one typically fire (and spot) big juicy eight or nine gun patterns -- it's just too expensive -- and one thus has to work with three-gun, cold-gun salvos, perhaps with someone deliberately 'spotting you off' so that you get extra practice in retaining (or trying to retain) a three-splash straddle. Further, training regimens -- at least in the USN -- very typically included the simulated loss of directors or the failures of important pieces of equipment in order to see both how the crew and the remaining equipment might react.

I am not sure what makes you think that targets in training always moved at low speeds and on predictable courses. In an offset shoot, you are actually visually firing at a full-size (almost completely) freely maneuvering target. Certainly course and speed changes during firing, by both target and firing ship were common, though sometimes somewhat restricted due to safety considerations.

I have never argued that the results of target/training shooting might be expected to yield results identical to those to be expected in combat. In THEORY, the match should be close to congruent, but in practice -- as might be expected -- it appears that combat shooting is almost always somewhat less efficient. While one can discuss the detailed reasons for this discrepancy at length, the exact causes are in and of themselves somewhat academic, so long as one can obtain and employ the appropriate corrective multiplier. Large military organizations obtain, and refine, these proportional multipliers routinely.

While the extrapolation of training results into combat situations is admittedly open to error, in practical terms it's the best we have -- basically the only game in town -- and certainly much better than what we might achieve via other, usually undefined, intuitions. A model, however flawed, is at least open to analysis and discussion. Without such a model, coherent and meaningful discussion tends to be difficult or impossible. Basically, in order to discuss something, one has to have something to discuss...

Bill Jurens.

[IronDuke]

Any weapon, still less a weapon system, that performs about two thirds as well in action as it did under testing and training is generally doing pretty well. In actual warfare there is just too much friction (to use the Carl von Clausewitz phase) ranging from
the obvious enemy intentions and actions through crew morale, training and tiredness to technical malfunctions, misunderstood orders and climatic conditions, etc, etc.

Battleships were big, very impressive machines, they were largely made obsolete by aircraft, especially aircraft flying from Aircraft Carriers (more big impressive machines).

We would be well advised to remember that, in most circumstances, it is more the Officers and men who man the machines, not the machines themselves, who make the difference between victory and defeat. This has always been true, was true in WWII, and remains true to this day.

Standards of skill, training, experience, morale, operational doctrine and intelligent aggression, not to mention a little luck, all play at least as big a part in any battle as the machines the fighting man uses.

[phil gollin]

[1] ............... The point I was trying to make is that the differences between the accuracy of training shoots and combat shooting is typically fairly predictable, and often not as great as some might believe. ..........

[2] ........ I am not sure what makes you think that targets in training always moved at low speeds and on predictable courses. In an offset shoot, you are actually visually firing at a full-size (almost completely) freely maneuvering target. Certainly course and speed changes during firing, by both target and firing ship were common, though sometimes somewhat restricted due to safety considerations. .........

[3] ........ so long as one can obtain and employ the appropriate corrective multiplier. Large military organizations obtain, and refine, these proportional multipliers routinely. ...............

[4] .......While the extrapolation of training results into combat situations is admittedly open to error, in practical terms it's the best we have -- basically the only game in town ........

Re. 1 - There is no evidence for that, the data for real life combat firing is too small


Re. 2 - Each type of exercise "should" if being more or less representative of real life show less or more accuracy and these should be extraced out as separate data sets to be examined. Unless each restriction and the knowledge that the firing ship has of those restrictions is known (and any quirks of the exercises) are all known and built into the analysis then one is merely adding together a mass of different conditions.


Re. 3 - There it is, the big assumption - where you mix up "now" with "then". Your casual phrase which throws away Operational Research and decades of increasing sophistication regarding military testing and exercises. The only way such a figure could be reasonably arrived at is by examining in detail the differnces in results between exercises and many combat examples and analysisng operationally the reasons behand the differences. The Operational Research was not done at the time and the number of combats were so few that it gave insufficient information. One cannot have "the appropriate corrective multiplier" as each combat situation would be different


Re. 4 - It's only a game worth playing if it gives worthwhile results, for which there is no evidence.

.

[lwd]

... Re. 4 - It's only a game worth playing if it gives worthwhile results, for which there is no evidence.

That's a rather strong statement and seems to contradict the evidence supported so far.

[phil gollin]

That's a rather strong statement and seems to contradict the evidence supported so far.

What evidence ?

There is much statistical evidence for exercises (great) - but very little for actual combat, and no way of bridging that gap.

The statistical work on the exercises is great, it just shouldn't be expanded into "real life" combat. It is no substitute for proper Operational Research and decades of changes to, and sophistication of, naval exercises.

.

[Bgile]

What evidence ?

There is much statistical evidence for exercises (great) - but very little for actual combat, and no way of bridging that gap.

The statistical work on the exercises is great, it just shouldn't be expanded into "real life" combat. It is no substitute for proper Operational Research and decades of changes to, and sophistication of, naval exercises.

.

I don't understand why, if a system achieves certain results in practice that can't be translated into real life. A good part of it is just a matter of physics and training.

[lwd]

Indeed many of the divergences are explainable by lack of training. Others by combat or training effects. As with any statistics one should not expect a one to one correlation. Indeed such a correlation is very suspicious in and of itself. Phil's position seems to be that statistics and experiments are worthless.

[Karl Heidenreich]

phil gollin:

The statistical work on the exercises is great, it just shouldn't be expanded into "real life" combat. It is no substitute for proper Operational Research and decades of changes to, and sophistication of, naval exercises.

True! That has been my heresy for a lot of guys here, specially if you don´t abide the dictate of Okun´s followers of the navweaps.

[Bill Jurens]

While -- at least with regard to large-gun ship to ship naval combat -- there are admittedly relatively few actual engagements upon which the results of training exercises might be calibrated, this typically represents much less of a problem in the sister services, i.e. armies and air forces, where actual engagements are much more frequent. There are even some good examples in the naval area as well, e.g. with regard to the relative performance of torpedo bombers in training and combat roles.

In the absence of other data to the contrary, it would seem reasonable to assume that the corrective multipliers derived in other situations, as mentioned above, would be fairly similar to those which might be expected in gun-to-gun naval engagements as well, at least so far as general trends are concerned.

What else might one do?

Bill Jurens

[phil gollin]

What else might one do?

That's simple - nothing !

Why try to make a case for something for which there is insufficient data ? Why try to give what is essentialy a vague approximation (or even guess) some cover of respectibility by associating it with a large statistical survey ?

You also ignore the whole discipline of Operational Research, which even in its simple WW2 form showed how complex were the factors involved in trying to match exercise results with those of "real life" and find ways of increasing weapons' effectiveness.

Armies and Air Forces had no better Operational Research (except possibly for the RAF's Coastal Command who invented the discipline) both in terms of techniques or application. There are numerous examples in all three services of the lack of effective analysis due to lack of "real life" results, and even where there were large numbers of results (e.g. ASW attacks) the analysis was incnsistent due to the large numbers of factors which were intermingled in any one attack. To try to ignore what has possibly been the biggest factor affecting weapons procurement, testing, exercising and useage since WW2 is a quaint conceit.

A simple "corrective multiplier" has no place alongside a proper statistical analysis. Fine, add it as a jokey aside as a "piece of fun", but not as a serious adjunct t the proper analysis. And always accept that it doesn't cover all factors or circumstances.

[Karl Heidenreich]

phil gollin:

That's simple - nothing !

Why try to make a case for something for which there is insufficient data ? Why try to give what is essentialy a vague approximation (or even guess) some cover of respectibility by associating it with a large statistical survey ?

Mr. Gollin, you are on the way of become a real enemy for a group of USN Battleship partisans around here. That, or simply being ignored. Anyway, for the last weeks an amount of arguments and... evidence (shall we say?) has been acumulating against certain myths.

But your mistake is that they will learn. I can tell you what is going to happen: discussion will soon die and, in a couple of weeks, one of these guys will re start a thread or will post something exactly with one of the arguments that everybody thought was over and done. It´s a waste of time, to be sincere.

Regards,