Tank gun test, extreme ranges. (Warning: large pics inside)

[Andreas]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Jeff Duquette:

Environmental conditions and there effects on gunnery and range estimation are in fact discussed in the TigerFibel.<HR></BLOCKQUOTE>

Jeff, I still see a problem with some people here quoting what are probably one-offs to have the whole range of the gun shifted higher.

Since we both agree that CMBO does a good job on modeling the sheep, I think we can also both agree that the issue of the 'wolves', as you put it maybe something that needs looking at.

The reason I dredged up that quote last night was to give an antidote to the relentless optimism of the Tigerfibel ('Observe the rule and you won't miss at 1,200m' - not 'Observe the rule, be in the right territory, make sure noone shoots at you, be calm, grab enough sleep, and don't let your driver hit you with 'em nagtive waves and you won't miss at 1,200m'). I am sure it discusses environmental influence. Unfortunately the way it was presented here looked to me as if it should be evidence that the gun should have an automatic hit out to 1,200.

I also think that maybe better treatment of defensive preparation may help matters (ranging etc.)

[ElGuapo]

This is the grog'i-est thread I have ever seen.

[ 10-01-2001: Message edited by: ElGuapo ]

[Holien]

It might be groggy but it is quite interesting to us with limited knowledge.

I love spying on the giants of CM...

H

P.s. Keep at it lads you are doing something worthwhile!!!

[Jeff Heidman]

This is a great thread! I won't pretend to have enough knowledge to offer an opinion as far as whether the results are realistic, but just reading makes me wat to go play some more CM!

Anyway, I wanted to make a point about what the future might hold. It seems that a lot of the issues about CM tnak gunnery at range comes from the simplified model used to represent gunnery. I use the term "simplified" only in relation to real life, since it certianly seems to be highly sophisticated compared to any other game.

In CM, the engine takes a bunch of variables into account, and through some formula comes up with a percentage chance for this round to hit. Then, if it decides a hit is achieved, you see the gun fire and hit its target.

If it decides that there is a miss, another formula is used to randomly determine where the round lands, excluding the actual target. Then the engine checks if anything happens to be at that spot when the round arrives.

The problem with this method is that the adjustments made for subsequent rounds are abstracted into some percent chance increase in the first formula, but has no real effect on the rounds miss point in the second. Ideally, with a good crew, you should see each subsequent miss getting smaller and smaller until some point is reached where you have exhausted the ability of the crew to effect the CEP of the round.

So, would the "Holy Grail" of gun modelling be a system where rather than ever coming up with a percent chance for a hit, the engine just determines, based on all those variables, where the round is going to land, and then sees if there is something there when it does? In this case the engine would determine, for each shot, two items:

1. What point is the gun lined up with? I.E., if the CEP for this shot was zero, where will the shot land? This is going to be based greatly on the crews ability to estimate range, the gunners ability to lay the gun properly, and such. It would have very little to do with the actual characeristics of the ammunition or gun itself.

2. What is the CEP for this shot? This will largely be determiend by the characteristics of the round, gun, and environment. For example, no matter how good the gunner is, a 20mm shell is going to ahve a huge CEP at 2000m due to its low weight and velocity at that range.

Then, when the round is fired, the engine figures out #1, adds the variablity in #2, fires the weapon, and then sees what (if anything) it hits when the round arrives at the target.

For each subsequent shot, the crew should get better at #1, presuming they are decent gunners, and the shot pattern should see this decreasing range from impact point to target. Presumably the formula use could use the "short-long" bracketing system employed in WW2.

Since the crew really only has control over #1, we would see a consistently decreasing (with some random error thrown in) range from the aimed point to the actual target for subsequent shots, with some maximum being reached where the gunner cannot really aim any better, it just comes down to whether the CEP is small enough to get a hit at the range.

Did that make sense to anyone other than myself?

Jeff Heidman

[Slapdragon]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Jeff Heidman:

This is a great thread! I won't pretend to have enough knowledge to offer an opinion as far as whether the results are realistic, but just reading makes me wat to go play some more CM!

Anyway, I wanted to make a point about what the future might hold. It seems that a lot of the issues about CM tnak gunnery at range comes from the simplified model used to represent gunnery. I use the term "simplified" only in relation to real life, since it certianly seems to be highly sophisticated compared to any other game.

In CM, the engine takes a bunch of variables into account, and through some formula comes up with a percentage chance for this round to hit. Then, if it decides a hit is achieved, you see the gun fire and hit its target.

If it decides that there is a miss, another formula is used to randomly determine where the round lands, excluding the actual target. Then the engine checks if anything happens to be at that spot when the round arrives.

The problem with this method is that the adjustments made for subsequent rounds are abstracted into some percent chance increase in the first formula, but has no real effect on the rounds miss point in the second. Ideally, with a good crew, you should see each subsequent miss getting smaller and smaller until some point is reached where you have exhausted the ability of the crew to effect the CEP of the round.

So, would the "Holy Grail" of gun modelling be a system where rather than ever coming up with a percent chance for a hit, the engine just determines, based on all those variables, where the round is going to land, and then sees if there is something there when it does? In this case the engine would determine, for each shot, two items:

1. What point is the gun lined up with? I.E., if the CEP for this shot was zero, where will the shot land? This is going to be based greatly on the crews ability to estimate range, the gunners ability to lay the gun properly, and such. It would have very little to do with the actual characeristics of the ammunition or gun itself.

2. What is the CEP for this shot? This will largely be determiend by the characteristics of the round, gun, and environment. For example, no matter how good the gunner is, a 20mm shell is going to ahve a huge CEP at 2000m due to its low weight and velocity at that range.

Then, when the round is fired, the engine figures out #1, adds the variablity in #2, fires the weapon, and then sees what (if anything) it hits when the round arrives at the target.

For each subsequent shot, the crew should get better at #1, presuming they are decent gunners, and the shot pattern should see this decreasing range from impact point to target. Presumably the formula use could use the "short-long" bracketing system employed in WW2.

Since the crew really only has control over #1, we would see a consistently decreasing (with some random error thrown in) range from the aimed point to the actual target for subsequent shots, with some maximum being reached where the gunner cannot really aim any better, it just comes down to whether the CEP is small enough to get a hit at the range.

Did that make sense to anyone other than myself?

Jeff Heidman<HR></BLOCKQUOTE>

This makes perfect sense, and I see it as a second or third generation engine based on CEP modified by know variables. Of course it will always be a silumation, but it will progressively be a better simulation.

[c3k]

Gents,

Chiming in here. All I have read in Jentz (Panzertruppen, Panther, Tiger), Glantz, Ziemke, et al., supports the expectation of long range (1,000 meter or greater) gun accuracy. Much more so than is represented in this game.

We've seen mention of giving crews an accuracy bonus. Well, I think certain weapons need it as well. What kind of sights were used? For tanks, the chassis makes a critical difference in combat firing (platform steadiness, return to battery, smoothness of ride to track target, etc.).

No, JasonC, I don't have empirical evidence for this "gut feel". Not meaning to be insulting, but I can take the opposite tact that empirical evidence for LACK of accuracy is missing for a lot of these weapons.

Ken

[Jarmo]

Obviously a true shell flight path calculation would be better than the current system. But it would then be necessary to also change the way tanks and other units are done.

Currently a unit exists on one single dimensional point only. Although I believe tanks actually have two points to enable hull down. A tank cannot be one thread on the road and one on open ground, for example.

Due to this single-dimensional nature of units, it'd be almost impossible for a shell flight path to actually hit anything. (Except ground). (BTW, this is partly the reason why those AA-vehicles are so damn hard to kill.) It would be possible to calculate in some kind of "close-enough?" system, but that'd be unsatisfactory and still a compromise.

So it'd be necessary to also model tanks and vehicles in true-3D. At this point it wouldn't be enough to roll where the shell did hit. It'd be necessary to model the angles and thicknesses of all the armor plates in the tank, including weak points and thick gun mantlets of Tiger I. Guess it'd then also be reasonable to model what the shell did after it penetrated, did it hit the engine or the fuel tank?

We'll get there eventually, maybe it wont take all that long either. The way computers are speeding up, we might have the game of grognards wet dream by the time CM has done the full circle and come back to Beyond Overlord.

Just another 10 years or so...

[M Hofbauer]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Bad Monkey!:

One thing that I think that should be thrown in here (though I don't know whether it's ever been mentioned or not) is that tanks should have modifiers just like HQ's do. You could have a driver's bonus (faster to get moving, less likely to get bogged), a gunner's bonus (accuracy), a loader's bonus (rate of fire), and a commander's bonus (spotting, other intangibles)<HR></BLOCKQUOTE>

this gem almost got lost amongst the firefight.

I really really like that idea. It would make tanks less generic, yet still variation would be limited to a certain envelope, just like you can expect a certain minimum performance from a HQ unit even without modifiers.

I think you really hiot onto a very good idea here. And it seems to me it should be possible to include this with a fair, feasible amount of effort (it worked for the HQ units).

[Slapdragon]

<BLOCKQUOTE>quote:</font><HR>Originally posted by c3k:

Gents,

Chiming in here. All I have read in Jentz (Panzertruppen, Panther, Tiger), Glantz, Ziemke, et al., supports the expectation of long range (1,000 meter or greater) gun accuracy. Much more so than is represented in this game.

We've seen mention of giving crews an accuracy bonus. Well, I think certain weapons need it as well. What kind of sights were used? For tanks, the chassis makes a critical difference in combat firing (platform steadiness, return to battery, smoothness of ride to track target, etc.).

No, JasonC, I don't have empirical evidence for this "gut feel". Not meaning to be insulting, but I can take the opposite tact that empirical evidence for LACK of accuracy is missing for a lot of these weapons.

Ken<HR></BLOCKQUOTE>

Actually, since the firing system for the game is based upon a model, the model working according to reality in many (or most) situations, the JasonC would not have to prove the game is right. The game is correct, unless we can prove it wrong by a balanced and credible display of evidence against the system, and supporting evidence for a new model.

So the opposite tact is indeed not possible.

[M Hofbauer]

well, I've rummaged through the Tigerfibel a bit. for anyone questioning the validity of the Fibel sometimes frivolous, less serious wording, it should be noted that the Fibel was an official Dienstvorschrift, technical SOP document, it was just an attempt to word it more fluent and more readable.

aaanyways, the Fibel in all the scetions on aiming, shooting, engaging enemy tanks seems to see 2000m as regular engagement range.

and mind you, that was the old Tiger I, with the KwK 36 L/56. I am not even starting to talk about the KT's KwK 43 L/71 or similar...

[JasonC]

"empirical evidence for LACK of accuracy is missing"

I don't think so. I think I have presented some suggestive evidence on that score. I don't claim it is more than suggestive, because there are certainly unknowns involved, but the suggestive evidence is extreme enough that it is hard to reconcile claims of high accuracy (say, ~20% to hit numbers at 2.5-3 km) with known facts.

Ignoring that evidence is not an argument. I am certainly open to alternate explanations for the evidence I've offered, that might be consistent with higher long range accuracy. In case anybody forgot, that evidence is that average AP ammo issued per weapon has 3 digits, while average KOs per weapon has 1 digit.

To face this evidence, those claiming high accuracy for even the longest range shots (e.g. only 1/3 misses at more than 2km, and any other field manual boosterism) have to explain what happened to all the AP shells that we know didn't kill enemy tanks. We know they didn't all kill enemy tanks because there are over ten million of the former and only a few hundred thousand of the latter.

Some might have been blown up in air raids, at factories or trains or supply dumps. Some might have been overrun in supply dumps. Some might have been left at KOed guns or inside dead tanks. The rest were fired.

A few might have been fired at buildings or bunkers. Some undoubtedly hit tanks that were already dead. Some of the lighter ones undoubtedly hit tanks without killing them. The rest either missed or killed tanks.

There aren't that many killed tanks. Any reasonable accounting for the other categories is going to leave an awful lot in the last alternate category, "fired at a live tank but missed".

Which means the average shots fired per kill is going to be reasonably high. And since the shorter ranged shots undoubtedly had high hit probabilities, the long range ones are going to be left with low hit probabilities, and a lot of rounds fired.

If the rounds made per kill achieved were on the order of 10, one might plausibly argue that the hit probabilities were still high, and explain all the rest of the difference by the various wastage factors above. But a factor of 100 or more is much harder to wave away like that. An army would have to be pretty bone-headed to need to make more than 100 shells to actually get 1-3 of them fired at their proper target.

I have an explanation for the long range fire reported, and the ammo numbers reported. My explanation is that they -had- enough ammo to engage in long range fire. They also had enough tanks, and high enough rates of fire, to deliver the ammo in tactically useful time-windows (minutes), with tactically useful effects (several dead tanks if they don't skedaddle out of LOS), even with low long range hit probabilities.

The manual gives 2 km as the normal engagement range for a Tiger I. The trial reports an average of 6 shells fired per kill at that range against stationary Sherman targets, perfectly believable. What does that lethality mean for realistic formations?

With less than one ammo load, leaving half for HE and saving some AP for close defense, a single platoon of Tigers could kill 20 tanks in 5 minutes. A company of Tigers could destroy a tank battalion in the same period of time. In other words, Tigers in position could easily prevent the approach within 2 km of even superior bodies of tanks, with little danger to themselves.

What about longer ranged fire, up to 3.5 km? At that range, the tests report 40 shots for 1 kill. Some regard this as an excessive expenditure of ammo and think it means they couldn't fight at that range, so the hit probability must be higher. But a Tiger company could destroy a tank platoon spotted at such range with only that accuracy, in 2-3 minutes, using less than 1/3rd of its available AP (thus saving most for closer range work).

If the tanks firing at range are up a step in organization size, they can kill easily at extreme range with an acceptable ammo expenditure. If the tanks firing at 2 km are even down a step in organization size, they can kill easily with an acceptable ammo expenditure. Somewhere between the one and the other, they can KO a unit of equal size.

Those results fit well with a doctrine of 2km as the normal engagement range. That range lets a small group dominate a larger one, while still not letting them get close enough for underpowered guns to be effective against Tiger or Panther armor. It also fits regular use beyond that range, effectively. When the ammo is there, the numbers of tanks are there, or the need is high enough, whole enemy tank units can be defeated at ranges well beyond 2 km. Even with only the hit probabilities seen in the tests.

It also fits the ammo totals - sometimes 40 per kill, sometimes 6, sometimes in between at 15, sometimes close range and only 1-3 needed. Most of the killing probably happened around the 6 each distance, but a lot of the shells were probably fired at the 15 distance. The shots fired per kill might average 15. Overkill and all the forms of wastage might plausibly account for the rest of the gap up to the total produced.

When you fight in companies and battalions, and fire for minutes not seconds, it just isn't necessary for every shot to be a hit, or even for every 3rd or 5th shot to be a hit. Tank companies have a lot more friendly shells aboard than enemy tanks they will face at one time. Each tank does not need to KO 8-16 enemies per outing - even with some own side losses, 2-3 will do fine. And they didn't kill 8-16 on average, even over their whole operational life, because there aren't that many dead Allied tanks.

That is evidence. You can face it and wrestle with it, and if you see some other way to explain all aspects of it, go right ahead and please share it with us. But just leaving out the part that doesn't fit your own idea (the ammo and kill data), after it has already been pointed out to you, is not reasonable.

[Andreas]

<BLOCKQUOTE>quote:</font><HR>Originally posted by M Hofbauer:

aaanyways, the Fibel in all the scetions on aiming, shooting, engaging enemy tanks seems to see 2000m as regular engagement range.

<HR></BLOCKQUOTE>

Yes, it also suggests that you can not miss at 1,200m.

Did Bobby Woll not set his sights to 800m?

The manual on what the 8,8cm Flak 18/36 can actually achieve is frontal kills out to 1,500 (turret) and side armour out to 2,000m. It never goes beyond that, and 2,000 is not the regular but the maximum range that is indicated in the penetration drawings (or whatever you want to call them).

Having said that, for AT guns it makes more sense to wait than for tanks, since then you have a chance to destroy the enemy, instead of just scaring him and inviting artillery on your position. A tank can relocate much easier.

[Grisha]

I think as Jeff D pointed out much earlier in the thread, the only way to better model gunner methods in CM is by actually modelling the way real gunners fire, ie 'bracket'. I'm not sure how CM is currently modelled to fire cannon, but if shell dispersion is a random process that is modified by a sort of RL_pattern parameter, then CM isn't modelling gunner methods. Of course, by keeping the hit function in CM's code to a minimal size by merely factoring accuracy variables is probably done to keep the game from being a CPU bear. To actually code a hit function that would incorporate the gunner practice of bracketing, then overlay that with variable gunner skill levels, would probably bloat the function/s by fivefold easily. If Charles opted not to make a RL hit function in CM, then it was probably for frame rate and CPU processing reasons, especially when considering that he's trying to keep this game playable for a wide range of computer systems.

I'm certain it's not that Charles can't model RL gunnery methods of target acquisition. It's just that when combined with the level of graphics quality (3D accelerated environment), adding that level of AI might require a system more on par with current flight sims, ie PIII600+ with GEForce2 cards and 256Mb+ RAM.

[PJungnitsch]

<BLOCKQUOTE>quote:</font><HR>Originally posted by JasonC:

The Germans made 2 million AP rounds for 88 long. Not the FLAK, nor the gun in the Tiger I, nor HE. For only 5000 weapons (70% of them towed PAK, incidentally). There were 400 AP rounds made for each of those weapons. If, as some seem to think, the kill chance per round had to be 1/5 or something, and if at least half the rounds weren't KOed or overrun unused, then every one of those weapons would have KOed 40 enemy tanks.<HR></BLOCKQUOTE>

That's a LOT of building on a shaky foundation. Stockpiles of certain late war German weaponry were huge compared to the numbers that made it into the field, due to the collapsing transportation system. Other factors, like the tendency to hit enemy tanks not once, but again and again until they burn, or the massive late war withdrawals in bad conditions (in which the big awkward 88's and their ammo was probably the first to be left) all could put a monkey wrench into careful calculations like that.

Still an an interesting theory but there must be figures from the field on actual shell consumption that would depend on a less convoluted chain of reasoning.

[JasonC]

"building on a shaky foundation"

I admit it is suggestive, not dispositive. But where do you propose the shells went?

The late war collapse story won't work for the bulk of the shells, because too many of the shells were made and issued by 1943 or early enough in 1944. The overkill story will only work for about a factor of 2, or 3 at the outside, based on data from wrecks recovered on hits per kill.

A factor of 100 goes a long way with such ideas. If there were only one order of magnitude for such wastage, such explanations (between them) might be believable. But there are two of them.

Reports from the front are often anecdotal, and usually focus on reporting the results of shooting, not shooting with few results, let alone with none. Where total ammo usage stats are available, they are for things like an armor battalion over a month or more, typically not one engagement. In those cases, you see a reasonably high number of expended AT type rounds per claim - on the order of 7-15.

And enemy kill claims are often overstated, we know from other data, by around a factor of 2, compared to what is seen on the other side of the hill. A drawback to such battalion-month figures is that one doesn't have any hard data on the engagement ranges, but one can assume they will be in some distribution, over that long a period and size of unit.

But you certainly do not see overall ammo usage only 2x the number of kill claims, so the "can't miss at 2km" sort of claim is obviously bunk. (That much ought to be obvious from every unit report, anyway).

[offtaskagain]

Just to throw in my little bit about ammo expenditure, almost every picture I've seen of a KO'd tank was one that was almost see through with holes. Jason says a little bit of that surplus ammo was fired at dead tanks, but almost every picture I see shows numerous insurance holes. Yes I know this isn't concrete proof, but it does seem to show a pattern. Here are some of the pictures I'm referring to, from The Russian Battlefield.

Dead Cats

[Jarmo]

I'd take the "you can't possibly miss at 1200m" to actually mean "you can't miss unless your aim is off".

Just like saying "at 300 meters, you can not miss with a rifle". Well, you obviously can, but not because the rifle wouldn't shoot straight.

Meaning, if you hit an unmoving target once, don't move yourself and dont change the aim, the subsequent shots will also hit.

[c3k]

JasonC,

As always, your postings are very appreciated for their depth. I do not agree that comparing total ammo production numbers to total AFV's destroyed and ATG's in action yeilds a valid prediction of tactical effectiveness. Another view of the large ammo production is that the ATG's were SO effective that the Germans pulled out all the stops to ensure that any ATG with tanks to fire at (since some % of ATG were in quiet sectors of the front) had a plentiful supply of ammo. I don't know.

While not as well founded as yours, I've done a bit of research on the 8.8 cm Flak in the anti-tank role. (Rustle, shuffle) Ahh, here it is: in my Panzerblitz game the 2 guns in a Flak 88 piece have a "20-A-20" attack/range factor. The T-34/85 company of 10 tanks has a defense of 10. Therefore, it's obvious to us grognards how effective the Flak 88 was. (Tongue firmly in cheek.)

I cannot as readily dismiss "anecdotal" evidence as some others. Another word for "anecdotal" is "after action report" or "eyewitness" or "confirmed post-battle intelligence". These German long guns (75's, 88's of various caliber-lengths) had an awesome reputation for accuracy both sides. There must be something to it.

Respectfully,

Ken

[JonS]

JasonC,

two categories for the 'lost' shells could be "used in training", and "lost in the supply chain". The second should be taken to mean both destroyed by enemy action, as well as being held at various levels of depot.

Eg. a divisional PaK abeitelung(sp?) has, say, 20 guns. Each gun tractor holds say, 50 rounds (=1000 rounds so far), and the abeitelung(sp?) holds say 500 as an immediate reserve. The division has a fump of another 500 shells, and the korps has a dump of, oh, say 1500 (for 3 divs). Now there are 6000 shells being held by the korps in total. Three korps to an armee gives 18-20000 shells, which is an order of magnitude larger than what is at the coal face, so to speak, and we're only up to armee level

Now, my numbers are plucked, but the point is that very large numbers of product (any product) can be tied up in the supply chain. Remember this was in the days before JIT and all the other fancy business management acronyms. Sure, the reason for having a supply chain is to keep the guns fed, but at any given point in time many more shells will be in the chain than at the end of it, and making things worse is that the German supply chain supposedly wasn't the most efficient of beasts.

Regards

JonS

[ 10-01-2001: Message edited by: JonS ]

[rexford]

Past posts have clearly illustrated how excitement and fear can freeze guns crews, or lead to wild eyed mistakes and way out misses. But the cool clear headed accuracy of professionals has also been displayed, such as the M18 commander who thought out the ballistics issues "by the book" and knocked out German tanks, one after the other, at 2000 yards.

What can a careful, well trained pro do with an 88L56 tank gun?

We have a computer program that bases first shot accuracy on errors such as range estimation and ability to zero in on the center of the observed target, and does NOT crank in "fear", "excitement" and "just plain dumb" factors. Follow-up misses use bracketing technique to zero in, and the assumption is crews with better range estimation talents are also better at bracketing.

Here is what a professional Tiger tank crew could do against a 2m x 2m target (rolling terrain blocks bottom of vehicle from view):

10% AVERAGE RANGE ESTIMATION ERROR

(This is elite type unit)

500m: 100% probability on all shots

1000m: 65% on first shot, 85% on second, 95% on third

2000m: 14% on first shot, 38% on second, 47% on third, 60% on fourth, 65% on fifth

3000m: 3% on first shot, 14% on second, 23% on third, 24% on fourth, 28% on fifth

25% AVERAGE RANGE ESTIMATION ERROR

(this is average ability for WW II crew)

500m: 85% on first shot, 95% on second, 100% on third

1000m: 30% on first shot, 55% on second, 70% on third, 75% on fourth, 80% on fifth

2000m: 4% on first shot, 17% on second, 23% on third, 33% on fourth and fifth

3000m: 3% on first shot, 6% on second, 9% on third, 13% on fourth and fifth

5% AVERAGE RANGE ESTIMATION ERROR

(range finder use, ranges marked in field or on maps by defensive units, very best crews)

500m: 100% all shots

1000m: 85% on first shot, 100% afterwards

2000m: 28% on first shot, 57% on second, 65% on third, 70% on fourth, 75% on fifth

3000m: 7% on first shot, 21% on second, 27% on third, 32% on fourth, 36% on third

Note that tanks with 5% and 10% average range estimation errors, which are the Michael Wittmann types with nerves of steel, can, with a few shots at a target, attain close to 100% accuracy at 500m and 65% to 75% accuracy at 2000m.

Just as importantly, common everyday crews (25% average ranging error) who go about their business carefully can score 85% accuracy at 500m on the first shot against a partially obscured target.

Jeff's comment about a few tanks doing most of the killing seems right on the mark. Reports on Tiger performance usually center about a few tanks in a unit doing most of the kills, with little or nothing from the others. It is probable that the best Tigers not only shot more accurately but fired more often.

3000m kills within a few shots by Tiger tanks are possible based on the computer analysis, and would be even more likely by the higher velocity Panther, Nashorn and Tiger II weapons (plus Panther 75mm has less dispersion than Tiger 88L56).

[TigerChow]

Just to put my two cents in...

In reference to the anecdotal evidence to long range kills, is there any mention of whether the target and/or gunner was in motion? I can see how it might only take a few bracketing shots to hit a stationary target (say you just crowned a ridge and found a platoon of T-34's having breakfast 2km away). However, I have a harder time believing that it is that easy to hit a moving target at that distance. If it takes 4-5s (I'm quoting from memory of earlier post - may be wrong) for the shell to travel that distance, it seems to me that any "juking" that the target does could easily produce a miss.

Just a thought for further conversation...

[rexford]

What can a Tiger crew see at 2000m and 3000m?

With 2.5x magnification, a 2m x 2m target would appear to be:

0.03" high and wide at 2000m

0.02" high and wide at 3000m

With 5x magnification, Tiger I observed target size is twice as large. The computer model assumes 2.5x magnification, so Tigers, Panthers, Tiger II's and JagdPanther accuracy would be boosted at the longer ranges by virtue of a larger target on the sight.

The model assumes that the target stands out to some degree against the background and has the characteristics of German gun sight optics, which would lead to accuracy decreases for most American gun sights in reduced light conditions.

The computer model includes penalties for inability to precisely aim at the center of target as range increases. Note also how important care of the gun becomes at long range, everything must be well aligned and maintained to hit at 2000m and 3000m range given a fairly small observed target.

How many crews could maintain their weapons, as well as their nerves, for high accuracy at 2000m and 3000m? Not everybody, but certainly quite a few.

[Jeff Duquette]

<BLOCKQUOTE>quote:</font><HR>Marcus Said: well, I've rummaged through the Tigerfibel a bit. for anyone questioning the validity of the Fibel sometimes frivolous, less serious wording, it should be noted that the Fibel was an official Dienstvorschrift, technical SOP document, it was just an attempt to word it more fluent and more readable.

aaanyways, the Fibel in all the scetions on aiming, shooting, engaging enemy tanks seems to see 2000m as regular engagement range.

and mind you, that was the old Tiger I, with the KwK 36 L/56. I am not even starting to talk about the KT's KwK 43 L/71 or similar...<HR></BLOCKQUOTE>

Agreed.

Grisha:

Very erudite response. Thanks. Presumably all that trouble-shooting you do for the U of W network has given you some unique insight into this question.

===========================

Just to add a bit more on the subject of long range tank gunnery (be warned as a stream of consciousness is about to follow):

<UL TYPE=SQUARE><LI>The reticle for the gunners sight of the Tiger I…Tzf9b (Turmzeilfernrohr 9b) and the latter upgraded Tzf9c instruments were graduated in 100 meter intervals out to 4000 meters.

<LI>The reticle for the gunners sight of the Tiger II…Tzf9d instrument was graduated in 100 meter intervals out to 6000 meters.

<LI>Penetration statistics for both the 88L56 and 88L71 obviously do not drop to zero at 2000meters. The voluminous BIOS Report on German Tank Armor generated by the British Army from captured German documents following the war details penetration testing conducted by the Germans for various tank main guns. Testing was conducted for impact velocities equivelant to the weapons muzzle velocity, and proceeds downward from there to impact velocities of less than 500 m/s. This lower magnitude of impact velocity(i.e 500 m/s) for both the 88L56 and 88L71 corresponds to velocity decay associated with ranges in excess of approximately 4500m.

<LI>German wartime firing tables for both the 88L56 and 88L71 are laid out to ranges of 4000m. Firing tables for the Flak88 firing pzgr (armored piercing ammunition) is detailed up to ranges of 4000m. Firing tables for the FLAK88 firing spgr (high explosive) is laid to ranges in excess of 14,000m.

Battlesight Gunnery Someone brought up Bobby Wohl and his habit of setting his gunsights to 800m. This isn’t really a trick Wohl invented to allow him to run up high kill counts. Rather this is SOP based upon Wohl’s training and no doubt reinforced by his practical experience. Battlesight gunnery takes advantage of the relatively flat trajectory of armored defeating ammunition to ensure respectable first round hit probability against surprise targets. An armored piercing round or high explosive round is preloaded (An AP round if hard targets were anticipated, HE if infantry or anti-tank guns were expected) and the gunner pre-indexes his range setting.

It implies the gunner wants to be able to fire quickly at surprise targets and still have a reasonable chance of hitting what he is aiming at without a more involved range assessment procedure (read here seconds are critical in these situations). Battlesight gunnery would be employed more often in the Attack, although situations when Battlesight is employed are obviously not limited to the attack…ala a T34 just appeared in the woods 500m to your right flank. A Battlesight fire command sequence by the Tank Commander would be something like ”GUNNER – BATTLESIGHT – TANK – FIRE”. Fire commands for PRECISION gunnery typically include both an ammunition type command as well as a range to target command.

I had come across excerpts from German WWII gunnery training manuals indicating recommended pre-indexed ranges for Battlesight gunnery. I can’t at the moment find this information, but will dig about a bit more tomorrow. However, US Army gunnery training for the old M48A5 stressed a 1200m Battlesight pre-index range for SABOT and 900m for HEAT. Ultimately Battlesight gunnery is tangent to what is being discussed here. The focus of the thread is obviously long range tank gunnery, which implies precision gunnery techniques.

[ 10-02-2001: Message edited by: Jeff Duquette ]

[Jeff Duquette]

Rex:

Excellent insights. Thanks for posting.

[Andreas]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Jeff Duquette:

<LI>Penetration statistics for both the 88L56 and 88L71 obviously do not drop to zero at 2000meters. <HR></BLOCKQUOTE>

Of course not. The training drawings for the gunners that I have seen do though. You probably know the ones that have black for the vulnerable areas and indicate the distance to which the armour can be defeated for various types of shell. In Piekalkiewicz he has a print of these for T-34, KW-2, Churchill, Sherman etc. and they never go beyond 2,000m. I post this here tomorrow.

I am perfectly aware that the Tigerfibel was an official document. I am however also perfectly aware of some of the comments I have read about training in the German army during the war and its relationship to reality. Just because it is in the Tigerfibel does not mean it is the Ultimate Truth.