German 8.8 Flak - Not Good Enough?

[Runyan99]

I am fed up with the inaccuracy of the German 8.8 flak. I know the topic has been discussed at length a few times before, but I am going to beat the dead horse.

I have used the 8.8 Flak in several different CM scenarios, and it seems to be an inaccurate weapon in the game. Some of my experiences are as follows:

1) In a CM operation, my opponent's 88 opens fire on a Sherman of mine which is halted on a road. The range is well over 1000m. The 88 and the Sherman excange fire for probably 20 seconds (the 88 misses the unmoving Sherman about 5 times as I recall). Eventually, the Sherman knocks out the gun with an HE shell (!)

2) In a QB with a human opponent, I have purchased an 88 and I have it trained on my right flank. My opponent sends a recon element consisting of some US armored cars moving along that flank. My 88 engages the armored cars at a range of approxamately 200-400m. The 88 fires 3-5 shots, misses all of them, and is later knocked out by US artillery.

3) In an operation I am currently playing, I have an 88 set back into woods. My opponent moves a Sherman into the gun's field of fire and begins area firing at a building. The range is about 350m. The Sherman is in open ground, and is not moving. My 88 crew is Veteran, is in command, and is unmolested by any enemy action. The 88 misses twice, and KOs the Sherman with the 3rd shot.

Three shots to hit a Sherman under these circumstances? Why? At anything under 1000m, I imagine the shell trajectory must be flat, which means hitting the target is a simple matter of point and shoot, with no calculation required.

I think CM under-estimates the accuracy of this weapon, both at short and long ranges. If the crew is inexperienced, out of command, or under fire, they may well hurry and miss a great deal. Under better circumstances however, given the sophistication of the gun and its targeting and range finding systems (which ALL of the evidence states is deadly accurate), a good Flak crew should be able to get a high percentage of first and second shot kills at almost any range. Particularly when the target is as fat as an immobile Sherman.

[Fishu]

Already Tiger crew should be able to hit targets from 1000 meters with first hit, or at least within 3 shots maximum.

(otherwise they WONT see Tiger as their tank)

88 Flak is much more accurate than Tiger and should be childsplay to kill tank 1000 meters away.

so if its like you said, then something is terribly wrong.

..and I've seen Tigers miss tanks from 200 meters.

[Scipio]

Gun accuracy has been often discussed here. The 88 was known to be one of the accuratest guns in WWII. IRRC even untrained troops were able to hit with it on distances of 1000m with the first or second shot.

Well, we can discuss this until the hell freezes, it doesn't helps. The only way would be that BTS publish the calculations and unit datas they use for gun accuracy, so we could compare it with the 'real world'. Of course there are always some things that influence the gun accuracy, but at last it comes all down to math.

[Andreas]

It does not really all come down to math. There is a difference between the 8,8cm gun in a tank and as an AA gun. The difference is the training of the crew, and the spotting, which AFAIK would be done through stereoscopic range-finders on the AA gun, while the tanks did not have that particular equipment.

Regarding whether the tank should hit first shot at 1,000m - IIRC there was an excerpt from German gunnery school posted here that called for 1 hit out of three shots @ 1,200m, which makes this rather unlikely to me.

Regarding hitting when close - it is my understanding that there is such a thing as too close, when your optics (which often were set a a certain distance, I have heard 800m as something of a 'standard') are becoming a problem. I do not however know when 'close' becomes 'too close'.

So really it is quite important to be clear about what is being discussed here. For the tanks, there have been lengthy discussions, and to my recollection there was no consensus at the end. For the AA gun, this was not really discussed at all so far.

Still holding out for DF airbursts from the 8,8cm. A man can hope

[Slapdragon]

Or, has been mentioned before, a person, if so inclined, work on the subject with a similar dedication to how Rexford worked with armor.

The 88 was not a supergun with all seeing all killing power. Nor was it that bad. It has been hugely over hyped, but it was an effective weapon.

So you could explore quantitatively and historically the issue.

First, look at the 88 from a quantitative realm. This is very tough, but base the data not on shooting tests but on ballistic reports and known modern ballistic models and you have a good start. Create a model which predicts accuracy of the weapon in a perfect shoot.

Then research accuracy of shooting by all soldiers no matter what nation and what method. Build a model for accuracy by an average soldier, and then work out from there.

Now combine the two models. You have a potential model simulating 88 accuracy at X range. Now you can check you model against personnel accounts, oral histories, unit logs, and the like. This is the point that simulations creators may add a small amount of fudge, or chuck the model all together if it has big problems.

Finally, publish your model with citation on the board for peer review. Or, if it is good enough, do what Rexford did and publish in print. You can also do a survey of how tank guns work in CM and publish a statistical analysis of 88 accuracy at different ranges to compare your model with CM. The greatest thing you could do would be to then suggest a new model for CM to adopt based on you pointing out the errors in the old model.

BTS wil then start to look at their model of accuracy, and look for changes. Now also remember that there are side effects to all this. Most people who ask for an unkillable king tiger or an all powerful 88 don't realize that if they got what they want, other models in the game will make the unit so expensive to purchase that they will disapeer from QBs. However, if the model is accurate, we have to accept the change in price.

Good luck in your inquiry.

[ 09-26-2001: Message edited by: Slapdragon ]

[Silvio Manuel]

For contrast, my only 88FlaK experience has been (both were in C&C)

1st (Vet) gun: gets 2 1st shot kills at 800m on Sherman M4A375w's

2nd (reg) gun: gets 1 richochet, then 2nd shot kill on M4A375w+

I couldn't complain. Those suckers re-load (predictably) slow though. Bad deal if you miss on a tank at close range.

[Måkjager]

I am playing an pbem Operation at the moment and its the first time i have gotten a 88mm Flak 18 and it is paying for its keep

First kill was at 920+m ...3 shots...3rd ko'ed the M4? Sherman ( the ? is because i do not have full ID of the type).

Second vehicle was also a M4? Sherman at 860m....semi hull down ..3 shots...koed on the 3rd.

The next ko was a M18 Hellcat at 800m+ ...1st shot hit the gun...2nd ko'ed the vehicle.

The lil' scapper is still hanging in there

Regards

Måkjager

[Panzer Leader]

Well, I love reading the stories of the omnipotent 88, but translating historical data, legend, hearsay, and personal accounts into a workabel system is no easy task.

Sure I wish they were more effective and I wish they were more accurate. I want my single (expensive!) gun to be able to tear up the enemy, but in CM that doesn't happen. Is it undermodelled? We will probably never know.

There is only so much data on this legendary gun and I can safely assume BTS used it to model the weapon. However like any gun, it is a fragile system that can be taken out easily. It is also a system with slow rotation and low ROF. If anything, the best argument would be for a reduction in price, not a revamping of the model.

As for the legend of the gun. It was definitely a useful and terrifyingly accurate weapon, but it was used in large quantities at long range for the most part. Put 10 guns 2,000 meters away from a sherman brigade and you will see. Its legend stems from the fact that it couold take out enemy armour well before they could even respond.

In CM terms I like the gun and smile when I am given one in a scenario but I probably wouldn't pick one in a QB -- there are too many variables working against its effective employment, the main one being range of engagement.

Still, though, a part of me would like to blast everything in sight to smithereens... Damn BTS and their oath not to use nationality modifiers!

[dragonwagon]

Well then , try using something else......

[Redwolf]

I could come up with quite a few first-hand accounts that attribute the 88, as a towed gun or in the Nashorn, with a much higher accuracy than other guns.

The problem, however ist to come up with any hard data. These quotes usually say "I've seen it shooting and it was incredibly accurate". I have not seen any information of the kind "x shots hit y times, as compared to gun foobar, which scored only z in a similar situation".

The Nashorn was equipped with more advanced optics and manned by people with excellent training in ballistics. But hard data? My new book "Panzerjaeger: Tank Hunter" by Averbeck/Folkestad has such a quote. The guy was manning AT guns all the war, at that time a Pak40 and he was deeply impressed by the Nashorn's accuracy. I would rate this as a reliable source and it would cause me to change my own game to make the Nashorn noticable more precise than the Pak40. But this isn't my game, and for a mass-market game it is probably better to base the accuracy on hard data, physical attributes, which minimizes complains.

It is my impression that it is an adequate solution to the problem to make 88 towed guns and Nashorns crack or elite. That helps the accuracy and some rate of fire and these weapons wouldn't profit much from the other experience attributes, so it wouldn't inbalance much else (they don't cut through infantry MG blazing). That doesn't help in Quickbattles, but sounds right for scenarios.

I can come up with a number of quotes that the 17pdr had a problem with accuracy. Hard data? Nope.

A wholly different issue is that in my personal estimation, the open-top tank destroyers (especially M10 and Nashorn) should have an edge over tank with the same gun in accuracy and especially rate of fire. The three men manning the gun in the open turret of a M10 must be good for something. Also they had training more concentrated on this specific task compared to the allrounders in the tanks. But then again, hard data?

[M Hofbauer]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Panzer Leader:

Put 10 guns 2,000 meters away from a sherman brigade and you will see. <HR></BLOCKQUOTE>

...which is exactly what I did.

I put up a setup between 10 x 8.8cm FlaK and 50 x M4 (plain vanilla) Shermans, facing each other at a distance of 2,000 meters.

Within one turn (which took awfully long to compute) it was all over.

All ten 8.8 FlaK were killed, at a loss of only one M4 Sherman.

edit: in a second re-run of the test four shermans were killed before all of the 8.8 guns were destroyed.

[ 09-26-2001: Message edited by: M Hofbauer ]

[Lanzfeld]

Well, I have to also chime in and complain about the 88's accuracy in this game compared to other guns.

I just played "Last Defense" as the Germans. I had a veteran tiger which I keep back to fight the three hellcats when they arrive as and experiment. I keep the tiger back because the hellcat shells will bounce off at this range. The hellcats (which I think are regular crews) were hitting the tiger four or five times EACH for every 88 hit. This was not a one time thing either. Try it for yourselves. The 88 mounted on a tiger just HAD to be more accurate than this! For crying out loud! Just point and fire at this range!

[M Hofbauer]

redwolf,

I made a lengthy series of tests to see if my suspicion that the Nashorn's excellent long-range accuracy was mismodelled was true or not. Too often I had lost them to simple 37mm or 40mm - equipped Armored Cars.

The results so far show that indeed the little guns are just as accurate and better than the big ones.

So far, Greyhound vs 8.8, at a distance of 1,000 meters the 8.8 is about a third more likely to die. And the score would be even worse if not a couple of them small hits would only damage the Nashorn. In other words, the accuracy is just as good as that of the logically inherently more accurate bigger caliber.

[M Hofbauer]

uh, btw, everybody/everything was "regular" quality in those tests, vehicles/guns were padlocked to face each other.

Nashorn-Greyhound tests are done one on one, with seperated firing lanes, while the 10 vs 50 one-time fun shootout was a bunch vs a bunch.

[Redwolf]

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

So far, Greyhound vs 8.8, at a distance of 1,000 meters the 8.8 is about a third more likely to die. And the score would be even worse if not a couple of them small hits would only damage the Nashorn. In other words, the accuracy is just as good as that of the logically inherently more accurate bigger caliber.<HR></BLOCKQUOTE>

That is the result of the current model, which bases accuracy exclusivly on the speed of the shell. The 40mm and 37mm shells are very fast.

The Nashorn has a much bigger silhuette, the small vehicle has a much better rate of fire so the Nashorn looses. The latter two points are OK and somewhat realistic, a Nashorn shouldn't tangle with Greyhounds, it is too slow (traverse, reload) and especially costly for that.

I am not quite sure about the physical effects of the caliber. If you are talking about the effect of side wind, the issue is obviously more complex, because the length of the projectile plays a big role. The mass is higher, but the precision effect of the mass is modified, possibly multiplied, by the amount of rotation of the projectile in flight.

[offtaskagain]

Don't forget the great majority of those accuracy charts are based on a gun already centered on a 2x2.5m plate. They merely show the dispersion pattern. It has nothing to say about the ability of a crew to accurately train the gun on a target 2000m away. That complicates matters greatly. It's like giving Joe Schmoe a Barrett .50 rifle and telling him to shoot that jeep 1200 yards away. The gun may be accurate, but the hit is totally based on the skill of the shooter.

[Slapdragon]

Hof,

I know what is wrong with your test. I too did that test, only I set up 62 88mm 2200 meters from the starting point where 12 Sherman M4A3(75) sat. But instead of putting everybody in the open, which is unrealistic, I put the guns in some woods. The goal of the tanks was to advance 200 meters to a victory flag. The goal of the guns was to stop them.

3 turns later (3 minutes !!!) 12 Shermans were dead. They never even spotted the 88s. This hardly seems an imbalance, unless we assume that the 88 was only ever used from open ground. Since we know it is not, then a 2000 meter battle from concealment is devestating, the tanks did not fire shot 1.

[JasonC]

Being a bean counter by temperment, I offer the following check on claims of extreme first shot accuracy with AT weapons.

1st proposition - the average AT weapon, AFV mounted or towed, KOed no more than 2.5 enemy AFV over its entire life. Evidence - add up the number of AT weapons fielded by the Germans, and assume they KOed every allied AFV ever made. You can exclude all the ineffective early war "doorknockers" if you like, it won't change the result appreciably.

2nd proposition - most of the ammo produced for AT weapons wasn't produced on a lark. It was meant to be used, and a significant fraction of it, at least, was used. Some rounds "die" in dead tanks or gun positions, some are overrun in depots or blown up by aircraft, etc. But 1 out of 3, as a lower bound, actually get fired at enemies.

3rd proposition - most effective AT shells KOed target they hit. 75mm long and up against the mix of Allied tanks, allow for some overkill, about 2 hits would result in a kill. But be generous and up it to 3.

Deduction - then the average shot taken at tanks by AT weapons probably had a hit probability with 1 digit. Or, under 20% to be exceedingly generous about it.

You see, there are really only two ways you can arrive at proposition 1 above. Either AT weapons generally shoot with high first round hit probabilities, in which case the low number of kills each results from a "six gun shoot-out" in which the first to the draw wins. In this case, the low overall kill number for each reflects early destruction of the AT system itself. Or, the low number of overall kills also reflects low hit probabilities (or shots at such ranges that penetrations become rare events, or both).

The problem with the first of those two rival theories (call it the "quick draw" theory) is that it predicts extremely low ammo loads produced, compared to the number of tubes produced. If an AT weapon generally only shoots half a dozen times before dying, then it is to say the least rather strange that the combatants bothered to make ~250 anti-tank type rounds for each gun they fielded. They ought to have achieved a much greater combat effect by producing more tubes, and fewer shells. But they did not try to do this.

Of course there were passages of arms in which the quick-draw theory was how a local fight played out. If the enemy creeps deep enough into LOS of lots of AT weapons, which are not spotted in turn, and then all open up from ambush at close ranges, then no doubt high chances of hits with each shell will be achieved. And the guys on the wrong end of such an ambush will be toasted with little loss and low ammo expenditure too.

But those have to be exceptions. Some on both sides, but still exceptions. Otherwise there is no way to make sense of the enourmous average ammo loads produced per gun. And there is an obvious reason to think this might well be the case, besides the production figures.

Usually the sighting differential is not all that large. And if it isn't, and both sides can see each other, then it simply becomes physically impossible for large numbers of AFVs to live long enough to get down to short ranges where the hit probabilities are high. I mean, if the hit probability is 10% per round, then the tanks can still fire fast enough to KO an equal sized force in a few minutes. Ten minutes of that and surviving tanks within LOS of each other will have to be exceptions rather than the rule.

I think the way this played out in reality is a lot of creeping into and out of LOS and extreme range, or top-hat low-ski games with the same effect. If a sizeable number of side As tanks or PAK are in range and sight of point X, then no side B tanks remain alive close to point X. One side or the other then moves. A few vehicles move into sight, and they exchange their shots in turn.

These effects are somewhat masked in CM, because the maps are so small. Cover tends to be abundant too. And armor forces on each side tend to be very close to even and small in absolute numbers and dispersion in space. These factors were not present in the real war, to the same degree anyway.

If 50 tanks in a battalion draw within range of 4 tanks, each able to penetrate the other, the result is unlikely to be a quick-draw duel at ranges with 90% hit probability. Because 1 minute of fire by the 50 will clobber all 4 tanks even if the hit probability per shot is tiny; a few percent would do it.

Notice, this point about the hit probability of the average -shot- is distinct from the question of the hit probability of the average -kill-. Shorter range, higher hit probability shots, in "quickdraw" fashion, may account for an important chunk of the dead AFVs that occur. But they must then be balanced, in the per shot average, by lots of ranged fire under conditions in which the hit probability per round is quite poor.

We can probably say with reasonable safety, though, that most AT rounds fired in the course of the war had no more than modest hit probabilities. On the order of 10% or less. Otherwise, all the shell production was the purest waste, and the combatants would not have continued it.

Food for thought...

[Maastrictian]

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

A wholly different issue is that in my personal estimation, the open-top tank destroyers (especially M10 and Nashorn) should have an edge over tank with the same gun in accuracy and especially rate of fire. The three men manning the gun in the open turret of a M10 must be good for something. Also they had training more concentrated on this specific task compared to the allrounders in the tanks. But then again, hard data?<HR></BLOCKQUOTE>

I don't understand. Why are the three men in a M10 turret better trained or capable than the 3 men in a M4 turret?

--Chris

[Maastrictian]

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

I put up a setup between 10 x 8.8cm FlaK and 50 x M4 (plain vanilla) Shermans, facing each other at a distance of 2,000 meters.

<HR></BLOCKQUOTE>

Did the FlaK have fox holes? Were they in some sort of cover or just open ground? What kind of cover?

--Chris

[Redwolf]

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

I don't understand. Why are the three men in a M10 turret better trained or capable than the 3 men in a M4 turret?

--Chris<HR></BLOCKQUOTE>

I said three men manning the gun in the M10. Two loaders, one gunner. Plus commander and driver.

[ 09-27-2001: Message edited by: redwolf ]

[StellarRat]

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

In other words, the accuracy is just as good as that of the logically inherently more accurate bigger caliber.

<HR></BLOCKQUOTE>

Just because a gun is bigger does not mean it is more accurate. What is your reasoning for this assumption?

[Slapdragon]

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

I don't understand. Why are the three men in a M10 turret better trained or capable than the 3 men in a M4 turret?

--Chris<HR></BLOCKQUOTE>

Maast, they may not have been, but the M18 and M10 had 2 men loading the main gun and could fire 20 rounds a minute if they cared too. In fact, SOF in an ambush was to send round out the barrel 10+ rounds per second and overwhelm the enemy. This worked better with the M18 than the M10 because of turret design flaws in the M10, but ROF superiority made them better AT weapons than even the E8. Of course speed helped the M18 a lot, and we know the M10 did not really do all that well as an AT weapon.

[Maastrictian]

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

Maast, they may not have been, but the M18 and M10 had 2 men loading the main gun and could fire 20 rounds a minute if they cared too. In fact, SOF in an ambush was to send round out the barrel 10+ rounds per second and overwhelm the enemy. This worked better with the M18 than the M10 because of turret design flaws in the M10, but ROF superiority made them better AT weapons than even the E8. Of course speed helped the M18 a lot, and we know the M10 did not really do all that well as an AT weapon.<HR></BLOCKQUOTE>

Ahh, I assumed that the American TDs had a crew of 4. But infact they had a crew of 5, so (correct me if I'm wrong) there were *4* men in the turret, 2 loaders, 1 gunner, and one TC.

--Chris

[M Hofbauer]

Slapdragon and Maastrictian,

the issue to my understanding was not inherent advantages of camoflaged AT guns, but it was all about accuracy.

original quote from Runyan99, opening sentence of this thread:

"I am fed up with the inaccuracy of the German 8.8 flak."

Hence, hiding and foxholes were avoided and both sides put into the open, to ensure that both had the same chances to see each other, so that only pure accuracy would count.

Again, the issue raised here in this thread is accuracy as far as I am concerned, not hiding the AT guns etc.