88mm KwK 36 L/56 accuracy test and some ideas

[aka_tom_w]

It has been suggested at least twice now that spotting has nothing to do with accuracy and should not be a part of this discussion.

I know we are only dealing with an abstraction here (at least I think we are dealing with an abstraction here) when we speak of the way CM models the increased chance to hit on the second third and forth shot after a target is acquired, so I had thought this would mean at ranges above 1000 meters the chance to hit should increase be the result of where the missed shot landed could be seen by the gunner.

On the real life test gunnery range when discussing the accuracy of the 88 the gunner is reported to have been able to spot the shot missing and adjust the range accordingly.

So I would conclude that if the gunner could spot the result of the shot miss at 1500 or 2000 meters then I would suggest that the gunner of a weapon that can penetrate 157 mm of armour at 2000 should be able to target and spot that far with his targeting optics.

It has been suggested here that we are only dealing with accuracy and we should leave spotting to another thread, but I feel that spotting and accuracy are the same issue because if a gunner can attempt to target an enemy tank at 2000 meters then it seems reasonable given the Real Life test range results quoted from Jentz that such a gunner can adjust for the shot drop and can spot where the shot fell and so if you can see were the shot fell so he can account for that and retarget, then I would presume the gunner could also spot a tank or vehicle, especially in the open on pavement at 2000 meters. My test range indicated there was no long range spotting ability beyond 1500m even when Stuarts and Shermans were firing there main weapons at 2000 meters trying to target the German KT's Tiger I' Nahorns and Jagpathers.

I am of the opinion that spotting and accuracy should go hand in hand because if the gunner cannot determine the distance with which he missed the target by seeing where the shot missed, there can be no logic for the increased chance to hit that CM does model on every susbquent shot fired once the target has been acquired.

If you are interested in modeling the increased chance to hit just dig in Sherm Jumbos at 1000 m 1500m and 2000 m make sure that they have no ammo of any kind then start firing away, they are hard to hit and you will see something like a 25% chance to hit after one minute of acquiring the target.

You can get as high as 65% chance to hit after one minute of aquiring the target if the Sherm jumbo at 1000 m is stationary (immobile) and not hull down.

This is a great discusion and it is very refreshing to read that is has been completely free of personal attacks.

Thanks So Much to Steve for reading and responding, your gunnery range test of the Pz Mk IV's and Sherms was very informative. I was very happy to see that at 1500 meters that the Mk IV's prevailed as I believe they likely would have in real life.

Lets keep testing this accuracy issue.

-tom w

[This message has been edited by aka_tom_w (edited 10-08-2000).]

[PzKpfw 1]

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

It has been suggested at least twice now that spotting has nothing to do with accuracy and should not be a part of this discussion.

<HR></BLOCKQUOTE>

Tom brings up a valid point, in WW2 tank engagements it was SOP for the TC to locate the target then guide the gunner to it, then the TC would order the corrections to hit while observing the shotfall.

I remeber reading somewhere a US Army report on US tank crew fire procedures, and that it estimated 25% of combat gunnery corrections by US TC's in combat, were faulty & resulted in misses, or such, then if the gunner had been allowed to fire on his own, I'll see if I can dig it out tommorow.

Regards, John Waters

------------------

Notice: Spelling mistakes left in for people who need to correct others to make their life fulfilled.

[PzKpfw 1]

Steve has a very valid point concerning the test accuracy.

The test results were not expected to represent the gun's actual performance in combat, they were to establish the overall accuracy of the system, and to show the results that could be obtained under ideal conditions, Ie, range to target was established before fireing.

Now to diferentiate the test result difrences from the expected crew practice results, the Germans, for example provided 2 tables, 1 was the 'official' test results that ASL posted & the other was the expected crew practice performance vs a 2m high x 2.5m wide target moveing at 20kph across the frontal arc. Ie, a German Tiger E crew training requirement was to score a hit by the 4th round on a moveing target @ 1200 - 2000ms.

The below examples highlight the difrences between the 'official' test accuracy & the expected actual crew practice results, on a German tank gunnery range, with the KwK.36, KwK.42, & KwK. 43 guns.

Test results vs expected Practice hit % results in ( )'s

8.8 cm KwK.36 L/56 Fireing Pzgr.39/42:

500m - 100% (100%)

1000m - 100% (93%)

1500m - 98% (74%)

2000m - 87% (50%)

2500m - 71% (31%)

3000m - 53% (19%)

7.5 cm KwK.42 L/70 Fireing Pzgr.39/42:

500m - 100% (100%)

1000m - 100% (97%)

1500m - 100% (72%)

2000m - 73% (29%)

2500m - 71% (31%)

3000m - 55% (18%)

8.8 cm KwK.43 L/71 Fireing Pzgr.39/43:

500m - 100% (100%)

1000m - 100% (85%)

1500m - 95% (61%)

2000m - 85% (43%)

2500m - 74% (30%)

3000m - 61% (23%)

3500m - 51% (17%)

4000m - 42% (13%)

Regards, John Waters

------------------

Notice: Spelling mistakes left in for people who need to correct others to make their life fulfilled.

[This message has been edited by PzKpfw 1 (edited 10-09-2000).]

[Guest Big Time Software]

<BLOCKQUOTE>quote:</font><HR>Tom brings up a valid point, in WW2 tank engagements it was SOP for the TC to locate the target then guide the gunner to it, then the TC would order the corrections to hit while observing the shotfall.<HR></BLOCKQUOTE>

This is correct. The gunners optics were, so far as I know, inadequate for spotting purposes. In other words, the TC had to get the turret to line up with the target and then guide the gunner to the orientate the gun until he could then get it into his scope's field of vision.

However... accuracy and spotting HAVE NOTHING TO DO WITH EACH OTHER. Even in the real world it has nothing to do with each other. If one unit can spot another better that does NOT improve accuracy. It improves the chance of seeing something to shoot at. Nothing more. To make an extreme point, I spot something at 2000m. I fire my gun without looking. Now... does spotting the target make my shot more or less accurate? No, in fact there is no effect.

In game terms there especially isn't any direct relationship between spotting and accuracy. If something can't spot something else, that is the result of the spotting code NOT the weapons code. The two are as unrelated as they can be, for vehicles as they are for infantry.

Steve

[Guest Big Time Software]

<BLOCKQUOTE>quote:</font><HR>Tom wrote:

I am of the opinion that spotting and accuracy should go hand in hand because if the gunner cannot determine the distance with which he missed the target by seeing where the shot missed, there can be no logic for the increased chance to hit that CM does model on every susbquent shot fired once the target has been acquired.<HR></BLOCKQUOTE>

The place for this is in the accuracy code, not the spotting code. However, I am not really sure how it is we are supposed to simulate this. The tank commander, buttoned or unbuttoned, is assumed to be using the tanks other spotting options (i.e. binoculars, rabbit ears, etc.) to follow the flight of the shell and communicate this to the gunner.

Sometimes a round would not be observed, sometimes it would. Generally the gunner knows what he is doing, so the chance would increase with each subsequent shot. On top of that, if the gunner fired at something and missed he could at least GUESS and make a correction. This alone should qualify the next shot for a slightly higher hit percentage.

If I am not mistaken... braketing was the deliberate system of under shooting to make sure the gunner/TC could see where the first round landed. The gunner would then try for an over shot by a set number of meters, and then do some quick math to figure out where the target was inbetween. This is really only necessary at long ranges or when the gun does not have a lot of velocity behind the shell.

Glad you liked my gunnery tests. Sherman 75s don't look so hot in those conditions, do they?

Steve

[PzKpfw 1]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Big Time Software:

This is correct. The gunners optics were, so far as I know, inadequate for spotting purposes. In other words, the TC had to get the turret to line up with the target and then guide the gunner to the orientate the gun until he could then get it into his scope's field of vision.

However... accuracy and spotting HAVE NOTHING TO DO WITH EACH OTHER. Even in the real world it has nothing to do with each other. If one unit can spot another better that does NOT improve accuracy. It improves the chance of seeing something to shoot at. Nothing more. To make an extreme point, I spot something at 2000m. I fire my gun without looking. Now... does spotting the target make my shot more or less accurate? No, in fact there is no effect.

Steve<HR></BLOCKQUOTE>

Agreed Steve spotting is a seprate issue, I was pointing out the relationship of the TC with the gunner concerning spotting as in guiding the gunner to the target & the TC's effect on accuracy after the target was aquired as I thought the report was interesting in that it stated the TC actualy hampered the gunner.

The the French tests in 1947 with the Panther showed the lack of a periscope for the gunner to use independantly, of the gunsight actualy hampered the Panther gunner's target aquisition in tests by upto 30 secs.

Thx for the replys in my other post I understand the process better now & want to add not to be argumenitive BTW, that their are at least 2 simulations that do model ballistics & armor on par with CM if not more detailed, they are HPS's Tiger's on the Prowl & Panthers in the Shadows.

Regards, John Waters

------------------

Notice: Spelling mistakes left in for people who need to correct others to make their life fulfilled.

[This message has been edited by PzKpfw 1 (edited 10-09-2000).]

[Guest Big Time Software]

Hi John,

Right you are about the TC/Gunner thing. Unfortunately, this is something we felt we couldn't even begin to quantify (or qualify for that matter) so we left it out. It is really amazing how LITTLE data there is on tanks out there. Sure, we know the firing sequence for the carbs of a Tiger, but do we know, in any sort of measurable way, how easy it was to lay that turret on a target? Nope

<BLOCKQUOTE>quote:</font><HR>Thx for the replys in my other post I understand the process better now & want to add not to be argumenitive BTW, that their are at least 2 simulations that do model ballistics & armor on par with CM if not more detailed, they are HPS's Tiger's on the Prowl & Panthers in the Shadows. <HR></BLOCKQUOTE>

Not to be argumentitive, but it is impossible for a 2D system to be on a par with a 3D system all things being equal. There have been a few discussions about the differences between the two models and IIRC HPS' games have a fair degree of abstraction in regards to positioning and, IIRC, they use penetration charts for the basis of determining penetration. As you know, we have a beef with penetration charts

Steve

[-Jochen-]

<BLOCKQUOTE>quote:</font><HR>CM absolutely, and without a question of a doubt, increases the chance to hit with each and every shot that is fired from a gun. It might not be reflected in the % hit figure when the Target order is used.<HR></BLOCKQUOTE>

Hmmm, I think I should conduct more test to get bullet proof data... But I think that once gunner scores first hit to target he knows the range (and other gun laying parameters) for sure and scoring further hits would be much easier than with first round. What my tests showed is that Tiger IE in CM did not gain measureble benefit from hits ie. successive shots tend to miss. This is apparent at ranges above 1500 m. I don't think gun's dispersion is that great, if gun points to same direction as with shot that hit target, dispersion is about only variable affecting fot fired round.

In short, I think once the correct range has been established by gunner it would be much easier to hit in succesive shots.

<BLOCKQUOTE>quote:</font><HR>We also did not "dumb down" any particular weapon arbitrarily. Instead, we looked at a whole bunch of battle descriptions, which actually made note of misses, and came up with our own "battlefield" probability to hit. It might not be exactly correct, but it is at least consistant internally.<HR></BLOCKQUOTE>

I did no suggest that some particular guns have been given a smaller hit percentage than others. As I understanded the tweak affected all guns.

<BLOCKQUOTE>quote:</font><HR>In general, remember that test range numbers are done under rather ideal conditions and not battlefield conditions.<HR></BLOCKQUOTE>

Thats why I also conducted my test in ideal conditions in CM.

<BLOCKQUOTE>quote:</font><HR>In CM the gunner is trying to fire pretty much as fast as he can. On a range, unless the test parameters state otherwise, great care is taken with each shot, which would increase accuracy quite a bit. Not to mention that the test gunner was not being fired upon or worried about unknown threats suddenly popping up from nowhere.<HR></BLOCKQUOTE>

I understand this reasoning. However German data that I have (and which is already presented elsewhere in this thread) also gives estimate of battlefield accuracy which is considerably higher than in CM. Data is taken from original KwK 36 manual which was produced pretty much trough the mid-late war period. If they had found out that their estimate of battlefield accuracy was far too optimistic I would think that they would have changed those numbers to reflect reality. I also think that they would not represent unaccurate information to troops using the guns.

<BLOCKQUOTE>quote:</font><HR>There is also the problem of the TacAI taking over and either moving or popping smoke during the test. If this should happen it will totally throw off the numbers since test range data is not against moving or obscured targets. Weather and partially broken terrain also factor in here, but they can be rulled out by using Clear weather and putting the targets on pavement (I would think).<HR></BLOCKQUOTE>

I did not see any other actions by TacAI than target buttoning up when it was hit. It did not move a inch (I watched the target from 50 m away and not from shooter perspective) during the test and I'm positively sure that it did not affect the hit propability. Target did not pop smoke because it didn't have any.

I used M4 Jumbo as a target which seemed to think that Tiger cannot penetrate it and did not bother to move anywhere, even at range as close as 100 m.

To conlude: target did not move and it did not pop smoke.

I think I will concentrate on one range and to gather large set of data and then analyze first round hit propability and hit propability after hit.

------------------

jochen

Kids today! Why can't they fetishize Fascist military hardware like normal people?

Ladysmith wants you forthwith to come to her relief

Burn your briefs you leave for France tonight

Carefully cut the straps of the booby-traps and set the captives free

But don't shoot 'til you see her big blue eyes

[-Jochen-]

Few things:

- Nobody should trust hit propabilites reported by CM and use them as basis for discussion! As BTS stated they are only approximations. I did find out in my tests that real hit propability is about half that is reported by CM. This is apparent at ranges above 1000 meters. So when you compare real life hit probability data and hit propability reported by CM, you really can't draw conclusions. For my limited testing CM reported hit probability 3 greater than it really was at range of 2000 meters ie. Tiger IE was reported to have hit probability of 0.24 but my test showed actual hit propability of 0.067. I admit that there is margin for error in my test but 3 fold error is quite large.

- I think spotting is related to long range gunnery. If you limit spotting ranges to below realistic, you limit efectiveness of guns that are accurate and more likely to penetrate target on longer ranges. In my test regular unbuttoned Tiger IE was incapable to spot M4 Sherman jumbo at 2000 meters at level grass field. Would you be able to miss profile of M4 Sherman Jumbo at that range? Maybe even using a binoculars? Tiger spotted M4 after it had moved about 5 meters towards the Tiger.

- Question about target tracking is still open. This one requires quite a lot of testing to be completely answered and I'm aways from my computer for two weeks

- Gun differencies should be modeled beyond round weight and muzzle velocity. Same round fired from rifled and nonrifled barrels would have very different hit propabilities, even if they have same weight and muzzle velocity. I readily understand that quantification of such variables is difficult.

- I still question feasibility of tweaking down hit propability of all guns when compared to real life data. If both German AND Brits felt that battlefield accuracy (even approximated one) they gave was accurate, is it really wise to discard such data and trust your own approximation? After all, German and Brits had truly firsthand information that none of us cannot claim to posses today. Reasoning that gunners ability to hit targets were worsened by missing recently dead loved one or such can be reversed to a reasoning that gunners could perform better, desire of avenging death of recently fallen comrade etc etc.

We have real life data about certain gun's accuracy on battlefield, yet we cannot reproduce that accuracy on CM. As for my personal testing actual (not reported) CM hit probability can be as much off as 8 fold or more when compared to real life data gathered on similar situation at range of 2000 meters. This is far more important matter than arguing about few millimeters difference on guns penetration capability!

------------------

jochen

Kids today! Why can't they fetishize Fascist military hardware like normal people?

Ladysmith wants you forthwith to come to her relief

Burn your briefs you leave for France tonight

Carefully cut the straps of the booby-traps and set the captives free

But don't shoot 'til you see her big blue eyes

[ASL Veteran]

"Totally incorrect. The individual characteristics are in fact included. All weapons systems are simulated based on what they can do in mathematical theory, which is perhaps even more exact than the real world test firing range results. Then, and only then, are all weapons subjected to randomized factors."

Well Steve, then we have a bit of an issue here. It has been stated that each weapon displays its inherent characteristics in CM because an accurate mathematical formula is used and this formula correctly demonstrates the inherent differences in the various guns used by the combatants. This then is what is on the table. Problem is, we can't see this formula so we can only take the correctness of this formula on faith. I also wouldn't even want to look at this formula since it would probably overwhelm my puny brain. So, once again here is the data from CM and the data from Jentz:

Combat Mission

Weap 100m 500m 1000 1500 2000

50mm x87% x50% x26% x13% xx5%

88mm x87% x50% x27% x14% xx7%

Jentz

Weap 100m 500m 1000 1500 2000

50mm 100% 100% x95% x68% xxxx

88mm 100% x98% x64% x38% x23%

Now then, if the accurate ballistic data for each gun is included in the game, then the difference between the accuracy of the 50mm PAK and the 88mm Flak should be reflected in CM. Since CM is using a mathematical formula and using raw ballistics data, then perhaps the unadjusted Jentz accuracy figures should be used as a comparison

Jentz

Weap 100m 500m 1000 1500 2000

50mm 100% 100% 100% x99% xxxx

88mm 100% 100% x95% x77% x58%

Now, I am NOT claiming that these accuracy figures are the ones that should be in CM. I am only using these accuracy figures to compare one gun to another and to examine the differences in the guns themselves. I would also like to see if these differences are in fact reflected in CM. The problem arises when one looks at the hit probability that appears on the targeting line. This hit probability shows little or no difference between the 50mm PAK and the 88 Flak. In fact, it actually shows the 88 Flak marginally more accurate than the 50mm PAK. The real (dispersion) data would suggest that, at a range of 1000 meters, the 50mm PAK should be 30% more accurate than the 88mm Flak. The real (non dispersed) data shows that the 50mm PAK should be 22% more accurate than the 88 Flak at 1500 meters. The real data, both dispersed and non dispersed, is in direct conflict with the CM target line hit probabilities. This leaves us with three possibilities:

A. The hit probabilities that are shown on the targeting line are so 'rough' that they are nearly useless as an indicator of actual hit probabilities in CM. I say this not only because the 50mm PAK and the 88mm Flak show as equal hit probabilities on the Target Line, but because almost all guns in CM show a similar probability of hitting. On the face of it, this would indicate that there are no inherent differences in any gun at all - and this is why I made my original statement that there are no inherent differences in the guns themselves. I state that because there is no evidence (using the target line hit probabilities) that any differences exist. Can it be demonstrated in CM that the 50mm PAK is more accurate than the 88mm Flak at either 1000m or 1500m?

B. The data Jentz is presenting is wrong. The data that Jentz is using was from live firing these weapons at a range with the specific intent of isolating the inherent accuracy of each gun. However, this data could be disputed and debated - but to ignore this data entirely just because it doesn't agree with CM I think is the wrong approach. I would never suggest that these accuracy figures be the accuracy figures that CM uses. I think the focus should be on the difference in accuracy between these two weapons - is the DIFFERENCE modeled in CM?

C. The mathematical model in CM is either wrong, or missing some important variable. Heaven forbid if this is the case, but it is a possiblity nonetheless. I also wouldn't hazard to question the veracity of the CM formula until more accuracy comparison's can be tested.

Perhaps someone like Slapdragon would be kind enough to test CM to see if the 50mm PAK is more accurate than the 88mm Flak at all or if it is the other way around? I would be interested to see the results of such a test. I only pick Slapdragon because I believe he could create a fair and 'scientifically valid' test.

[aka_tom_w]

<BLOCKQUOTE>quote:</font><HR>Originally posted by -Jochen-:

Few things:

- I still question feasibility of tweaking down hit propability of all guns when compared to real life data. If both German AND Brits felt that battlefield accuracy (even approximated one) they gave was accurate, is it really wise to discard such data and trust your own approximation? After all, German and Brits had truly firsthand information that none of us cannot claim to posses today. Reasoning that gunners ability to hit targets were worsened by missing recently dead loved one or such can be reversed to a reasoning that gunners could perform better, desire of avenging death of recently fallen comrade etc etc.

<HR></BLOCKQUOTE>

That is a good point.

I believe the rational and logic behind tweaking down hit propability of all guns when

compared to real life gunnery range data, is IMO questionable. ( I do hope that is not too strong a word).

When we discussed modeling Long range German optics we were asked first to provide, historical data in a quanitifiable form to support our case. Then it seems we were asked to make a qualititative case for the need to model the quality of how and why the German optics bonus should be modeled.

We have now found and presented reliable historical test fire data which speaks directly to the question long range accuracy and we find that there even figures for what an "average cool gunner" could expect to achieve in combat. When we compare these figures to what we find in our collective CM gunnery test range results we find that CM does not model that higher level of accuracy that was found in the Real Life test range. Even for that which was expected under combat conditions by an "average cool gunner".

When we ask why not? we are told:

"We also did not "dumb down" any particular weapon arbitrarily. Instead, we looked at a whole bunch of battle descriptions, which actually made note of misses, and came up with our own "battlefield" probability to hit. It might not be exactly correct, but it is at least consistant internally.

In general, remember that test range numbers are done under rather ideal conditions and not battlefield conditions. In CM the gunner is trying to fire pretty much as fast as he can. On a range, unless the test parameters state otherwise, great care is taken with each shot, which would increase accuracy quite a bit. Not to mention that the test gunner was not being fired upon or worried about unknown threats suddenly popping up from nowhere."

I would suggest the grounds upon which CM chose to "dumb down" the accuracy data for targeting all weapons in CM is perhaps even more questionable than the grounds for the inclusion of gyro stabelizer which then gives the Allied tanks equiped with this dubious piece of equipment a slight edge in targeting and accruacy I presume, while on the move, I think this seems historically accurate, but what we are concerned about is that all accuracy for all weapons systems has arbitartily been "dumbed down", while an accuracy, "while targeting on the move bonus has been confered upon those tanks in the Allied force that historically had Gyro stabilizers.

Again I would like to see some kind of long range advantage modeled for German gunnery, especially the 88 as per this quote:

"The Tiger I and Tiger II could open fire for effect at 1200 meters and 2000 meters respectively. At greater ranges, bracketing was to be employed with jumps of 200 to 400 meters, switching to fire when within 100 meters of an armoured target. On practice range, an average Tiger I was expected to hit the target by the fourth round at 1200 to 2000 meters. Exceptional individual Tiger I's had scored hits against stationary tanks at 2500 meters and concentrated platoon firing could be used against targets at 3000 meters. The Tiger II could fire at stationary targets at 4000 meters. The same applies to moving targets for both Tigers."

from: http://redrival.com/leibstandarte/fire.htm

source: Jentz?

In game terms I would suggest these results from this statement:

" On practice range, an average Tiger I was expected to hit the target by the fourth round at 1200 to 2000 meters. Exceptional individual Tiger I's had scored hits against stationary tanks at 2500 meters"

OK lets double that 1 in 4, so in combat conditions in CM a Reg Crew should hit a 2 x 2.5 target, Lets say a tall skinny stuart at 2000 m 1 in 8 or for odds 1 to 8 or begining with the first shot at 12% the second shot at 18%

Tiger I Reg Crew targets Stuart at 2000m

Proposal:

shot con green Reg Vet Crack Elite

first 3 6 12% 18 24 32

second 6 12 18% 24 32 43

third 12 18 24% 32 43 55

forth 18 24 32% 43 55 70

fifth 24 32 43% 55 70 85

sixth 32 43 55% 70 85 90

seventh 43 55 70% 85 90 90

eigth 55 70 85% 90 90 95

This may be EQUALLY slippery logic but I just took the quote "an average Tiger I was expected to hit the target by the fourth round at 1200 to 2000 meters" and doubled it for combat conditions for a Reg crew then because nothing is 100% in CM (which is IMHO as it should be) figured if you were targeting at 85% accuracy by the 8 th shot that should meet the combat expectation of one shot in eight hitting as per the doubling the quote of a hit by the forth shot.

Question? Should the first round fired by the elite gunner (32%) be 10 times more likly to hit than the conscript gunner (3%)? This may, now that I think about it, be WAY out of line?

I'm refering to IDEAL conditions above, bright day, no smoke, no arty, both vehicles stationary, PERFECT line of sight, both vehicles in the open.

I would say that above proposal would un-dumb down the long range chance to hit accuracy algorythyms.

Comments?.

Its STILL fun to play here in a Flame free zone (this thread), thanks to all for the congenial atmosphere of discussion.

ASL Verteran, I just read your last post and thought it was BRILLIANT, I agree completely, as you can see here.

Thanks especially to Steve for listening, I know you don't really agree with most of what some of us here are suggesting, but we are (ok, I am) thrilled you are listening and participating

Thanks

-tom w

[This message has been edited by aka_tom_w (edited 10-09-2000).]

[ASL Veteran]

I think it best to leave the Gyro issue alone,and I don't see any reason to question BTS's criteria for 'dumbing down' accuracy. We don't need to question reasons, let us only question data.

[The DesertFox]

Tom,

<BLOCKQUOTE>quote:</font><HR>"The Tiger I and Tiger II could open fire for effect at 1200 meters and 2000 meters respectively. At greater ranges, bracketing was to be employed with jumps of 200 to 400 meters, switching to fire when within 100 meters of an armoured target. On practice range, an average Tiger I was expected to hit the target by the fourth round at 1200 to 2000 meters. Exceptional individual Tiger I's had scored hits against stationary tanks at 2500 meters and concentrated platoon firing could be used against targets at 3000 meters. The Tiger II could fire at stationary targets at 4000 meters. The same applies to moving targets for both Tigers."

from: http://redrival.com/leibstandarte/fire.htm

source: Jentz?<HR></BLOCKQUOTE>

saw you asking...yes the above text is a Jentz quote. This and much more can be found in :

"Germany's Tiger Tanks : Tiger I & II : Combat Tactics" by Thomas L. Jentz (February 1997) Schiffer Publishing, Ltd.; ISBN: 0764302256

Chapter 7 pages 8 - 22 "Operational Characteristics"

Cheers

Helge

------------------

Sbelling chequed wyth MICROSOFT SPELLCHECKER - vorgs grate!

- The DesertFox -

Email: desertfox1891@hotmail.com

WWW: http://www.geocities.com/desertfox1891

[aka_tom_w]

<BLOCKQUOTE>quote:</font><HR>Originally posted by ASL Veteran:

I think it best to leave the Gyro issue alone,and I don't see any reason to question BTS's criteria for 'dumbing down' accuracy. We don't need to question reasons, let us only question data.<HR></BLOCKQUOTE>

That's fair, I stand corrected.

And yes it is best to focus on the data, you are correct, thanks for the reminder.

-tom w

[aka_tom_w]

Earlier this morning I proposed this table of hit accuracy..

Now before any one else trashes it and calls it absurd let me do it!

I've had some time to think about this and to suggest an elite German crew would have a 32% first shot chance to hit a Stuart at 2000 meters may make me look like a fool...

so Lets take another look at that table:

Tiger I Reg Crew targets Stuart at 2000m

Proposal:

.

shot....con... gr ...Reg ..Vet ..Crk .Elite

first......3... 6... 12%.. 18... 24... 32

second..6... 12... 18%.. 24... 32... 43

third....12... 18... 24%.. 32... 43... 55

forth....18... 24... 32%.. 43... 55... 70

fifth... 24... 32... 43%.. 55... 70... 85

sixth....32... 43... 55%.. 70... 85... 90

seventh.43... 55... 70%.. 85... 90... 90

eigth..... 55... 70... 85%.. 90... 90... 95

Question? Should the first round fired by the elite gunner (32%) be 10 times more likly to hit than the conscript gunner (3%)? This may, now that I think about it, be WAY out of line? Probably crazy but then look what Jentz has to say about accruacy:

Jentz :

"The following tables show estimated accuracy, of both the 8.8 cm Kw.K.36 L/56

and 8.8 cm Kw.K.43 L/71, in hitting a target of 2 meters high and 2.5 meters wide,

based on assumptions that the actual

range of the target has been correctly determined, and that the distribution of hits

is centered on the aiming point. The first column shows the accuracy obtained

during controlled test firing to determine

the pattern of dispersion. The second column (in brackets) includes the variation

expected during practice firing due to differences between guns, ammunitions and

gunners. All estimations are in

percentage and do not reflect the actual probability of hitting a target during under

actual combat condition. However, the average, cool gunner, after sensing the

tracer from the first round, could achieve the result presented in the second column: (50) in brakets below

8.8 cm Kw.K.36 L/56

Ammunition Type

Pzgr.39

Range @ 2000 m

87 (50)

So Jentz says "the average, cool gunner" can hit "a target of 2 meters high and 2.5 meters wide" at 50%

SO in CM what does that mean?

What experience level do we call the "the average, cool gunner" and after how many rounds in combat conditions would that crew level be firing rounds at a 2 meters high and 2.5 meters wide target that he would hit 50% of the time?

In the above proposal a Reg gunner is targeting at 55% accuracy by the sixth round fired and an Elite gunner is firing at 55% by the third round.

The question is to what degree will BTS attempt to model 50% chance to hit at a target "2 meters high and 2.5 meters wide" fired by an "average, cool gunner" at 2000 meters in the game?

The way the targeting at 2000 meters is modeled now it is NO where near that generous chance to hit percentage.

Everything you have just read about how the varying levels of crew experience are modeled and how the chance to hit percentages in the above table were modeled to be increasing after every round fired at an acqyired target, ALL came from my fertile imagination and is NOT based in ANY way on how I understand CM to model these variables because I have NO idea how CM does indeed model the increased chance to hit variable for an acquired target, but I do know from my tests the chance to hit, does appear to rise after a minute of shooting at a target that has not been hit in the first minute.

I would say that that increased chance to hit that acquired target is unrealistically low for the main weapons in the game that fired the very high velocity rounds that traveled along straight flat and more predictable trajectories.

Any new data from your gunnery test ranges to present?

-tom w

[This message has been edited by aka_tom_w (edited 10-09-2000).]

[This message has been edited by aka_tom_w (edited 10-09-2000).]

[This message has been edited by aka_tom_w (edited 10-09-2000).]

[Guest Big Time Software]

Oh boy, many different things to respond to...

First, Jochen's two posts...

<BLOCKQUOTE>quote:</font><HR>In short, I think once the correct range has been established by gunner it would be much easier to hit in successive shots.<HR></BLOCKQUOTE>

It is. However, there is still a margin for error ESPECIALLY at very long ranges. Hitting once is NOT a guarantee of hitting with the next shot. This is basic laws of probability mean that the greater margin of error the greater the chance for a miss, regardless of knowing the range (although that is the most significant part).

<BLOCKQUOTE>quote:</font><HR>Thats why I also conducted my test in ideal conditions in CM.<HR></BLOCKQUOTE>

Ideal conditions in CM are NOT the same as ideal conditions on a peaceful test range. Not even close. So although you can make the testing circumstances in CM more "ideal" do not be fooled into thinking you have created something similar to a test range. CM is a COMBAT simulation, not a test range simulation

<BLOCKQUOTE>quote:</font><HR>If they had found out that their estimate of battlefield accuracy was far too optimistic I would think that they would have changed those numbers to reflect reality. I also think that they would not represent unaccurate information to troops using the guns.<HR></BLOCKQUOTE>

True, but no nation had a statistician riding into battle. Those numbers quoted by the Germans you mentioned are an educated guess. What you see in CM is "actual". My guess is that the Germans, and Allies, had a rather simplistic and generalized system to downgrade the ideal numbers based on a scientific guess after talking with tank crews. The question is... what were the skill levels of the tankers they talked to? In any case, I very much doubt there were any sort of scientific tests done on the battlefield.

<BLOCKQUOTE>quote:</font><HR> I think spotting is related to long range gunnery.<HR></BLOCKQUOTE>

I will only say this one more time. It does not have anything to do with accuracy. But OBVIOUSLY if you can't see something you can't shoot it, so it does have something to do with the ability for the vehicle to shoot (accuracy not even a factor). This is something totally different that needs to be looked into. Since 2000m+ ranges were next to impossible to find in NW Europe we didn't weigh spotting abilities towards greater ranges.

<BLOCKQUOTE>quote:</font><HR>We have real life data about certain gun's accuracy on battlefield, yet we cannot reproduce that accuracy on CM.<HR></BLOCKQUOTE>

I disagree. We have "real life guesses that we don't know how they estimated them". CM's hit probability might be too low, but I do not think these guesses are irrefutable and unquestionable figures.

-------------

OK, now for ASL Veteran,

<BLOCKQUOTE>quote:</font><HR>The hit probabilities that are shown on the targeting line are so 'rough' that they are nearly useless as an indicator of actual hit probabilities in CM.<HR></BLOCKQUOTE>

Uhmm... I said as much at least 3 times already This negates your further two points:

<BLOCKQUOTE>quote:</font><HR>The data that Jentz is using was from live firing these weapons at a range with the specific intent of isolating the inherent accuracy of each gun. However, this data could be disputed and debated - but to ignore this data entirely just because it doesn't agree with CM I think is the wrong approach.<HR></BLOCKQUOTE>

We do not ignore data, but we also do not fudge our equations to use various bits of this or that chart made by this or that country at this or that time for this or that weapon. That is bad science.

<BLOCKQUOTE>quote:</font><HR> I think the focus should be on the difference in accuracy between these two weapons - is the DIFFERENCE modeled in CM?<HR></BLOCKQUOTE>

The only way to do this is to set up a target range in CM and do repeated tests as clinically as possible, remembering that the results will never be "perfect test rage quality".

<BLOCKQUOTE>quote:</font><HR>The mathematical model in CM is either wrong, or missing some important variable. Heaven forbid if this is the case, but it is a possiblity nonetheless. <HR></BLOCKQUOTE>

Yup, there could be a flaw in the equations' design, data, or implementation (i.e. a bug). If we can isolate such a problem we can fix it. However, the testing thus far is quite inconclusive.

----------

Now, on to Tom...

<BLOCKQUOTE>quote:</font><HR>I believe the rational and logic behind tweaking down hit propability of all guns when

compared to real life gunnery range data, is IMO questionable. <HR></BLOCKQUOTE>

This is rather ridiculous. Test range data is completely and utterly divorced from reality of the battlefield. No serious student of science or someone with real life military experience would buy this argument for a penny (not even a Canadian penny ).

<BLOCKQUOTE>quote:</font><HR>I would suggest the grounds upon which CM chose to "dumb down" the accuracy data for targeting all weapons in CM is perhaps even more questionable<HR></BLOCKQUOTE>

Forgive the strong words here but... total and utter BS. Your premis that test range data is only marginally less valid than battlefield data is utterly, and completely, based on not one shread of evidence. On the contrary, test range data is *always* questioned by everybody I have ever read and disccused such issues with. Having extensive experience with small arms I can prove to you, without a question of a doubt, that test range results are utterly overblown compared to real life use of the weapon.

The ohter thing you have missed, yet again, is that ALL weapons have been dumbed down. So whatever precious and in your head benefit there should be for German guns has been downgraded in proportion to that of the Allied guns. So the margin of superiority REMAINS EXACTLY THE SAME.

<BLOCKQUOTE>quote:</font><HR> than the grounds for the inclusion of gyro stabelizer which then gives the Allied tanks equiped with this dubious piece of equipment a slight edge in targeting and accruacy I presume, while on the move, I think this seems historically accurate, but what we are concerned about is that all accuracy for all weapons systems has arbitartily been "dumbed down", while an accuracy, "while targeting on the move bonus has been confered upon those tanks in the Allied force that historically had Gyro stabilizers.<HR></BLOCKQUOTE>

In general would you PLEASE stop comparing completely unrelated things and drawing totally unsupportable conclusions? Specifically I point to you dragging the Gyro thing into this discussion. First of all... the gyro is only a factor when firing on the move. Since we are testing against STATIONARY targets, just as the clinical test firings, the gyro has absolutely NO part in this discussion at all. And on top of that, we CAN prove that it was a valid benefit for firing on the move (something you keep questioning without providing even one half assed attempt to discredit, I might add). Finally, the bonus for firing on the move if a gyro is present is tiny.

Tom, I know you mean well, but tossing in a bunch of unsupported and rather questionable logic like this does not contribute to this discussion in a positive way. Instead, I have to spend my time refuting things that should not even be in this discussion to begin with. Please keep that in mind for future posts, here and in other threads.

<BLOCKQUOTE>quote:</font><HR>The question is to what degree will BTS attempt to model 50% chance to hit at a target "2 meters high and 2.5 meters wide" fired by an "average, cool gunner" at 2000 meters in the game?<HR></BLOCKQUOTE>

Remember that a 50/50 chance can still allow for 20 misses in a row. The probability of this is low, but some people have the misconception that 50/50 means if you miss the first time you hit the second time. Sorry, but that is mathematically incorrect.

I would classify the average "cool" gunner to be a Veteran or higher. Regular gunners would tend to be a bit less "cool", Conscripts no where near "cool". So an Elite gunner hitting 10 times more frequently than a Conscript doesn't seem at all incorrect.

Steve

[PzKpfw 1]

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

So Jentz says "the average, cool gunner" can hit "a target of 2 meters high and 2.5 meters wide" at 50%

-tom w

<HR></BLOCKQUOTE>

I Think this needs some clarification, somehow, the 2nd table data has been infered as an combat hit % Ie, 50%; this is not the case, the 2nd table was, derived to show, the expected crew gunnery results, on a fireing range after the test result data, had been scaled down to remove the controlled effects used in the tests . The table 2 results are not derived from any actual combat results.

At best the table 2data indicates the accuracy a regular crew could be expected to achieve against a stationary 2m high 2.5m target, or a target moveing 20kph across the frontal arc of a stationary tank on a gunnery course.

Regards, John Waters

------------------

Notice: Spelling mistakes left in for people who need to correct others to make their life fulfilled.

[This message has been edited by PzKpfw 1 (edited 10-25-2000).]

[PzKpfw 1]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Big Time Software:

I will only say this one more time. It does not have anything to do with accuracy. But OBVIOUSLY if you can't see something you can't shoot it, so it does have something to do with the ability for the vehicle to shoot (accuracy not even a factor).

This is something totally different that needs to be looked into. Since 2000m+ ranges were next to impossible to find in NW Europe we didn't weigh spotting abilities towards greater ranges.

Steve<HR></BLOCKQUOTE>

Steve, 2000m + engaements were far from the norm, but they did occur, and some have been cited here, Ie, 3600yrd engagement near Gereonsweiler, 2000+ yrd engagement outside Ubach etc.

I hope you get to look into spotting as a seprate issue from this as IMHO spotting distances should not be constricted by an percieved engagement range limitation, it should reflect the max range in LOS and any benifits devices like gunsights & bino's give as well.

<BLOCKQUOTE>quote:</font><HR>

So whatever precious and in your head benefit there should be for German guns has been downgraded in proportion to that of the Allied guns. So the margin of superiority REMAINS EXACTLY THE SAME.

<HR></BLOCKQUOTE>

Agreed, the question then remains on what % would any gun be more accurate then another, under those conditions. As well as the problem in actualy compiling crew gunnery range data for each nation then modifing it to represent combat %.

Regards, John Waters

-----------

"We've got the finest tanks in the world. We just love to see the German Royal Tiger come up on the feild".

Lt.Gen. George S. Patton, Jr. Febuary 1945.

[This message has been edited by PzKpfw 1 (edited 10-09-2000).]

[Guest Big Time Software]

<BLOCKQUOTE>quote:</font><HR>John wrote:

At best the table 2data indicates the accuracy a regular crew could be expected to achieve on a gunnery course. <HR></BLOCKQUOTE>

DOH! I totally missed that John, thanks for pointing that out. Oh boy, that is a critical point and I think its implications are rather obvious. At least I hope they are

As for the 2000m+ engagments being rare, they were. But as you say, they did happen. I think we get into one of those "it was written about because it was unusual" aspects of battle documentation. This is a rather understandable, and obvious, tendency to write about the things that were not normal or average while not writing about the opposite. So I think we are both in agreement. Engagements did happen at these ranges, but quite rarely.

Heck, they happened far less frequently on the Eastern Front as a whole than most people would like to believe (although much MORE than Western Front). However, the quantity of these engagements (as opposed to proportional) is rather large and therefore we will be looking at long range engagements quite a bit as we move to the Eastern Front.

Steve

[aka_tom_w]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Big Time Software:

Remember that a 50/50 chance can still allow for 20 misses in a row. The probability of this is low, but some people have the misconception that 50/50 means if you miss the first time you hit the second time. Sorry, but that is mathematically incorrect.

Steve<HR></BLOCKQUOTE>

Hi Steve

Thanks for your very informative and detailed reponse, I will need some more time to read over it further and reply at length.

I would however like to post this about odds, I figure it is somthing we all know because we've been rolling dice since we were old enough to underwrap a wargame:

from :http://www.math.byu.edu/~jarvis/gambling/gambling/gambling.html#35

"One of the easiest mistakes to make with gambling is thinking that past gambles influence future ones. This common mistake is sometimes called the "gambler's fallacy," and it often leads people to bet more money and to bet more often than they otherwise would. For example, many people know how to figure that there is only a one in sixteen chance that a fair coin will come up heads four times in a row. But if the coin has already come up heads three times in a row, then the chances that it will do so a fourth time are the same as they would be if it had never been tossed before--one in two. However, it is easy to make the mistake of thinking that this coin has only a one in sixteen chance of coming up heads. It seems that the coin should make the average of

past tosses "come out right." But in reality, the coin does not remember past tosses and feels no obligation to even out the number of heads and tails that have come up before. As we make more and more coin tosses, the ratio of heads to the total number of tosses will approach one half, but this does not mean that there will be exactly (or even close to) the same number of heads as tails, nor does this mean that in the course of a few tosses things will come up anywhere near even. Misunderstanding this fact leads gamblers to believe they have more information than they really do, and can cause them to be

more willing to gamble than they otherwise would.

Simply multiply the odds of getting heads with each toss . 1/2 x 1/2 x1/2 x 1/2 = 1/16

Steve says:"a 50/50 chance can still allow for 20 misses in a row. The probability of this is low, but......"

odds of heads coming up (or missing a 50/50 chance to hit) 20 times in a row on a 50/50 coin toss is 1/1,048,576

So the odds of hitting something 20 times in a row on 50/50 odds are 1/1,048,576 and the odds of MISSING somthing 20 times in a row on 50/50 odds are 1/1,048,576.

Those are extremely low, bordering on riduculous odds.

I agree completely that LUCK is and should be a HUGE factor on the battle field and despite all the bitching and moaning and complaining , AND positive and constructive suggestions we all make the game models tank gunnery better than any simulation I have ever played.

So we are all still just trying to make the game more realistic.

More later,

-tom w

[This message has been edited by aka_tom_w (edited 10-10-2000).]

[Guest Big Time Software]

Tom,

Good post about probability. This is something that people very often forget. It is also interesting to note that with each successive shot the odds of hitting gets better, but when the odds are rather poor the chance of missing the next shot is still rather high.

<BLOCKQUOTE>quote:</font><HR>So the odds of hitting something 20 times in a row on 50/50 odds are 1/1,048,576 and the odds of MISSING somthing 20 times in a row on 50/50 odds are 1/1,048,576.<HR></BLOCKQUOTE>

I'll take your math at face value here But this appears correct. I picked a particullarly high number of misses for a 50/50 shot to illustrate a point that it is possible. At long ranges, with hit chances in the low double digits, the odds of missing 10 times in a row are not all that bad.

Charles just mentioned something to me I did not see in any of the discussion above. The test range % chance of hitting was (often or always??) based on the gunner already knowing the range to target. Which means, the percentage to hit as stated does not include the 2-4 shots necessary to establish the range (long shots at least). Which is yet another reason why those numbers should not be taken at face value.

Steve

[aka_tom_w]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Big Time Software:

Tom,

Good post about probability. This is something that people very often forget. It is also interesting to note that with each successive shot the odds of hitting gets better, but when the odds are rather poor the chance of missing the next shot is still rather high.

I'll take your math at face value here But this appears correct. I picked a particullarly high number of misses for a 50/50 shot to illustrate a point that it is possible. At long ranges, with hit chances in the low double digits, the odds of missing 10 times in a row are not all that bad.

Charles just mentioned something to me I did not see in any of the discussion above. The test range % chance of hitting was (often or always??) based on the gunner already knowing the range to target. Which means, the percentage to hit as stated does not include the 2-4 shots necessary to establish the range (long shots at least). Which is yet another reason why those numbers should not be taken at face value.

Steve

<HR></BLOCKQUOTE>

OK on the first point, at long range if the first shot is 10% chance to hit and the chance to hit, only goes up a little then sure each shot following still has a large chance to miss so missing 4-5-6 or 7 in a row when all the chance to hit percentages all along are under say 30% is very feasible no problem there.

The math was just 2x2x2x2x...... 20 times, easy. about a million to one odds, but it COULD happen

But, on the second point "Which means, the percentage to hit as stated does not include the 2-4 shots necessary to establish the range (long shots at least). Which is yet another reason why those numbers should not be taken at face value."

I think what we are really trying hard to present is that the German's had optical aids in their sights that helped them determine range very quickly and accurately. This is Especially true for the highly effective Donkey eared range finder (Ok I "should" find out its official german name and some real accuracy stats on it), but it is known to be standard equipement with an 88 mm flak team and it DID predict range with a very high degree of accuracy allowing that flak 88 to put the first round right on or very near the target at any range without the need for 2-4 range finding rounds. My point is this technical advantage is not modeled in CM.

What is at issue here, (and sadly the best way to support this arguement is with gunnery optics as modeled in Panzer Elite) is that German tanks had optical sights and equipment that gave them the ability to establish range quickly and accuratly in battlefield conditions.

Just to recap this issue came up way back last May when Heinz 25th PzReg posted this..

"Heinz 25th PzReg

Member

posted 05-15-2000 09:28 AM

Greetings all

I have been playing the CM golde demo for some time now and it is great. But I am a litte interested in how the game simulates the tank gunners actions. I have seen tanks miss their target completely at ranges under 300m. Ok, that probably happened sometimes during the war, but the when a second shot also misses, something is wrong. Are the gunners blind?? Do they need glasses?? The German Zeiss optics was actually quite good and they where very accurate. The American optics was not that good in comparison. The German gunners should have a distinct advantage just because of their optics. Is this more clear at engagements over long ranges in CM??

And does the hit probability increase after the first shot?? I have seen my StugIII gunners miss the first shot against Shermans hoping they would nail them with the next shot. But sometimes the second shot can be even further of target. Is there a logic to all this or is it just chance??

Qualified answers would be greatly appreciated.

Many thx

Former tank gunner

Heinz 25th PzReg"

http://www.battlefront.com/discuss/Forum1/HTML/004572.html

What he and I and others are suggesting (ok what I know I am suggesting) is that for the 88 flak the range finding ability of the donkey eared range finder should give it a better first shot chance to hit than an 88 in a tank, (tests by Ron indicate no such range finding first round chance to hit advantage is modeled)

"

Vet.Tiger vs Vet.M4A3(76)(1500m/50x)

Tiger

Hitchance - 16%

255shots/73hits

avg. shots 1st hit - 4.15

avg. shots per kill - 5.21

1st shot hits - 14%

worst case for 1st hit - 11shots

5 or more shots for 1st hit - 38%

The tests are of Veteran AT guns in woods versus Regular stationary tanks at 1500m. I ran

them 50 times, the tanks never returned fire as they didn't spot the AT guns, with the

following results:

88Flak vs M4A3(75)

Hitchance - 15%

278shots/87hits

avg. shots 1st hit - 4.1

avg. shots per kill - 5.56

1st shot hits - 16%

worst case for 1st hit - 11shots

5 or more shots for 1st hit - 38%

AND I would suggest German tank gunners that fired the 76 and 88 mm weapons that were of Veteran, Crack, and Elite crew status SHOULD have their chance to hit percentage raised after their first shot miss at stepped increase per subsquent round fired that is higher than what is presently modeled at long ranges over 1000m as an advantage for acquiring the target and adjusting for the shot drop after a first shot miss, because they had the techical aids and 4 years (some of them) battle experience to help them do that more accurately. I would also agree that there is good data that suggests that the 17 pounder and the allied 76 mm firing high velocity rounds by Veteran, Crack, and Elite crews should also be modeled with the same higher chance to hit step increase on subsquent shots after a first round miss.

Good German gunners firing 88's should be much more accurate on the second, third and forth shot following misses than the hit percentage they are currently modeled at, at ranges over 1000 meters.

This opinion is not new.

-tom w

[-Jochen-]

<BLOCKQUOTE>quote:</font><HR>It is. However, there is still a margin for error ESPECIALLY at very long ranges. Hitting once is NOT a guarantee of hitting with the next shot. This is basic laws of probability mean that the greater margin of error the greater the chance for a miss, regardless of knowing the range (although that is the most significant part).<HR></BLOCKQUOTE>

I understand that you don't neccessarily hit target again right after first hit.

I try to sneak to my computer to do conclusive testing during this week. I will concentrate on eiter 1500 m or 2000m range and gather larget set of data that is usefull enought to determine to what extent hit propability increases after range is known (after first hit).

<BLOCKQUOTE>quote:</font><HR>True, but no nation had a statistician riding into battle. Those numbers quoted by the Germans you mentioned are an educated guess. What you see in CM is "actual".<HR></BLOCKQUOTE>

Are you implying here that Germans and brits had less accurate educated guesses during WWII than you have made 55 years later?

What I do know is that BTS has made hit propability smaller by some artificial number that does not rely on scientific data. This is also aducated guess at best and I fail to see how it is any better than what was made 55 years ago by persons that might have had firsthand knowledge about battlefield conditions.

To support my view, I again present some test data. At 2000 meters I found out that CM 88 mm KwK 36 had hit change of 0.067 against M4 Sherman. Real life data of 88 mm KwK 36 has hit propability of 0.5 (0.87 under ideal conditions) against similar sized target. The hit propability difference between real life data and CM data is almost ten fold. Which one is more likely to be off?

<BLOCKQUOTE>quote:</font><HR>My guess is that the Germans, and Allies, had a rather simplistic and generalized system to downgrade the ideal numbers based on a scientific guess after talking with tank crews.<HR></BLOCKQUOTE>

But is that almost exactly what BTS has also done? Some real life test results (possibly even the same we have used in our discussion?) have been looked upon and then they are downgraded by simplistic and generalized system? Why modify the numbers yet again? I fail to see this as an way to improve overall accuracy of simualtion.

<BLOCKQUOTE>quote:</font><HR>The question is... what were the skill levels of the tankers they talked to? In any case, I very much doubt there were any sort of scientific tests done on the battlefield.<HR></BLOCKQUOTE>

I hope you don't take this as an any sort of flame but I cannot refrain to use US Gyro's as an point.

BTS stated that gyro effectiveness was only assessed in tests on firing range in scientific conditions, not on battlefield conditions, yet their effect on accuracy has been succesfully added to CM.

What prevents using real life shooting range test data in CM? It has been gathered on conditions much similar to gyro testing as far as I know. Even the battlefield accuracy which could be reached by those tested guns are estimated by contemporary specialists have been given, in form that should be directly usable in CM without any further modification.

<BLOCKQUOTE>quote:</font><HR>I will only say this one more time. It does not have anything to do with accuracy. But OBVIOUSLY if you can't see something you can't shoot it, so it does have something to do with the ability for the vehicle to shoot (accuracy not even a factor). This is something totally different that needs to be looked into. Since 2000m+ ranges were next to impossible to find in NW Europe we didn't weigh spotting abilities towards greater ranges.<HR></BLOCKQUOTE>

I agree. I think I haven't said that accuracy and spotting are the same thing but that they are related: you cannot possibly exploit your accuracy advantage if you cannot spot enemy. I also agree that spotting is issue that should be looked upon in future, I don't think that regular Tiger IE crew can miss sight of M4 Sherman at 2000 meters away in level grass field. Atleast it is highly unlikely.

<BLOCKQUOTE>quote:</font><HR>I disagree. We have "real life guesses that we don't know how they estimated them". CM's hit probability might be too low, but I do not think these guesses are irrefutable and unquestionable figures.<HR></BLOCKQUOTE>

I agree that the real life figures we have for basis of our discussion are not the whole truth. But taht is all we have. If BTS (or anyone) cannot present more information, those real life figures are what CM should produce because no-one really knows better.

I think none of us has the knowledge and experience to say what the exact hit propabilities should be. So why don't we just use the numbers that were obtained during WWII?

------------------

jochen

Kids today! Why can't they fetishize Fascist military hardware like normal people?

Ladysmith wants you forthwith to come to her relief

Burn your briefs you leave for France tonight

Carefully cut the straps of the booby-traps and set the captives free

But don't shoot 'til you see her big blue eyes

[PzKpfw 1]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Big Time Software:

Charles just mentioned something to me I did not see in any of the discussion above. The test range % chance of hitting was (often or always??) based on the gunner already knowing the range to target. Which means, the percentage to hit as stated does not include the 2-4 shots necessary to establish the range (long shots at least). Which is yet another reason why those numbers should not be taken at face value.

Steve

<HR></BLOCKQUOTE>

Actualy I brought this up back when I explained the diference between the test results & the expected results on the gunnery range, thats one of the diferences in the practice results % from the test results, Ie, the gunners had to establish range and hit a moveing target etc, in the tests range was predetermined & the target was static.

Regards, John Waters

------------------

"We've got the finest tanks in the world. We just love to see the

German Royal Tiger come up on the feild".

Lt.Gen. George S. Patton, Jr. Febuary 1945.

[This message has been edited by PzKpfw 1 (edited 10-10-2000).]

[aka_tom_w]

<BLOCKQUOTE>quote:</font><HR>Originally posted by PzKpfw 1:

Actualy I brought this up back when I explained the diference between the test results & the expected on the gunnery range, thats one of te diferences in the practice results % the gunners had to establish range etc.

Regards, John Waters

<HR></BLOCKQUOTE>

Hi John,

I always enjoy your posts and I'm sorry to ask, but can you explain your point or rephrase your statement, I'm sorry but I'm not sure what point you are making, and I am interested in understanding it. Perhaps just elaborate a little more about what your are saying about gunners on the gunnery test range having to determine the range to the target, and the % results.

-tom w

[This message has been edited by aka_tom_w (edited 10-10-2000).]