German armour penetration overstated?

[Jeff Duquette]

My goodness. The Indirect Artillery model is wrong. Tank main gun accuracy is messed up. Now penetration ratings are screwy as well. Did CM get anything right? The casual reader might get the impression from this forum that CM is just another “souped-up” NINTENDO game.

[kipanderson]

Tom w, hi,

Yes, it could be that Charles was assuming that German armour was of higher quality than Allied armour. However CM has already got that angle covered. If you select a Panther in CM and then hit “enter”, thereby calling up the Panther’s unit details, you will see an armour quality number of 85% above the armour details. For the Tiger 1 CM has no armour quality rating thereby assuming that the Tiger 1 had 100% armour quality by WW2 standards. The Panther was known for its somewhat poor armour quality and the Tiger 1 for its high quality of armour, see the various Jentz books. So if you just take the Panther and the Tiger 1 as examples then CM have done a good job of covering the armour quality angle.

Andrew Hedges, hi,

I agree that the “pointer” rounds were optimised for penetration of vertical plate. In all of the above I am assuming the use of APCBC rounds that were optimised for use against angled plate and, as you say, were almost flat under their thin ballistic caps. I am assuming the use of APCBC rounds because these were by far the most common during the second half of the war.

BruceR, hi,

Yes it may be that some of Charles figures for the German guns are a little generous. However, on the whole, I feel that Charles has got the German figures about right. The figure Charles uses for the 75L48 gun is 108mm at 100m against 30 degree plate. The two figures you normally see the Germans use for this gun are 99mm and 106 mm. I would have gone for a figure of 103mm, i.e. half way between the two. It must be remembered that Charles is also using an armour penetration formula and no two formulas will come up with the exact same figures. Given that, I do not have a problem with Charles penetration figures for the German guns. I feel they are correct and are consistent with one another.

The problem is that “given” the figures Charles uses for the German guns his figures for a number of Allied guns are a bit off target, in my view.

All the best,

Kip.

[aka_tom_w]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Simon Fox:

I say its the superior armour, not the inferior quality of the round that could be

at issue here<HR></BLOCKQUOTE>Sheesh tom you've completely got the 'wrong end of the stick'. Armour quality is modelled seperately, which you should know. Penetration figures given in the game as a rough guide are against standard quality armour for all projectiles. It has to be that way because armour quality varies from vehicle to vehicle. The effects of armour quality can therefore only come into play when the target of the projectile is known. The issue is with the gun/projectile combination, armour quality has nothing to do with what kip is on about since the data is generated based on the same standard armour.

Hi Simon

I fully realize the that every armoured surface and the quality and thickness (to the mm) of every armoured AFV surface in the game is modeled in wonderful historical detail in CM.

No doubt about that.

I was only suggesting that the discrepancy that kipanderson commented on when he posted his essay about "German Armour Pentration Overstated?" could be interpreted another way, I admit I might be wrong, obviously of few of you here think I'm WAY wrong, but that's ok.

The question becomes, what can we look at to account for these conclusions:

"Both the US76mm gun and the German 75L48 gun deliver near identical energy, modified for penetration, to the target per square millimetre yet Charles has given the German gun 11% greater penetration. The British 17pdr and the German 75L70 guns also deliver identical energy yet Charles gives the German gun 18% greater penetration."

Well kip suggested we look at the quality of the round, and I suggested we look at the quality of the armour the round was penetrating as another possible source of the descprenacy.

The question is, is there a penetration modeled here that is not accruactely modeled?

If we are talking about that lets not also talk about exactly what is being penetrated here and not just assume (make and ASS of U and ME) that the CM modeling of all Allied and Axis AFV armour is 100% correct and perfect.

We should look at all the possible places where the descrepancy may have crept in, if there is one.

-tom w

[This message has been edited by aka_tom_w (edited 10-24-2000).]

[Guest Big Time Software]

Kip,

Thanks for the thoughtful analysis. I'll try to answer each of your points.

The Milne-de-Marre formula you're using is good for general approximations but is rather out of date. If I remember correctly, it's referenced in the British research paper from which I got most of CM's algorithms, but is considered somewhat obsolete, and the research was meant to improve upon it.

The unfortunate truth for the Americans is that their APCBC rounds by 1944 were still not as good as the German rounds. They weren't heat-treated to the same standards and therefore tended to shatter upon impact against thick, hard plate - which is what we find on the Tigers and Panthers.

The best example is the U.S. 76mm APCBC round. If it had been heat-treated properly, then the U.S. 76mm would have been able to take care of Panthers and Tigers pretty easily, even from the front, at least out to medium range. On paper it looks like this ought to be the case, given the 76mm's projectile mass and muzzle velocity and kinetic energy. And in fact this is one reason that before D-Day a lot of U.S. ordnance people figured that the Sherman 76 would have no trouble knocking out Tigers and Panthers. On paper, it seemed like it could do it. But in real life it didn't work out that way. The 76mm APCBC was structurally flawed, and tended to shatter when striking the thick German plate. That's why CM shows the 76 having noticeably lower penetration than the German 75mm L/48, which is otherwise a comparable weapon.

This is not to say that a US 76mm cannot knock out a Tiger or a Panther, because it can. But it has to get a good angle, generally at fairly short range, and it's still not a sure bet. It's right on the borderline. The reason is the APCBC ammunition which wasn't as good as it could have been.

A quick note: The reason you don't see this effect with the U.S. 75mm gun is that its projectile velocity was not high enough to stress the projectile to the 'shatter' point upon impact. But the 76mm passed that threshold.

Now, there is one thing you mentioned which still has me thinking... it's about the 17-pounder. My feeling is that the British APCBC ammuntion of the day was similar in quality to that of the Americans, i.e. somewhat prone to shatter. But I don't have direct evidence for or against that theory. So I had to make an educated guess. If anyone has evidence to support that the 17-pounder used better quality APCBC ammunition (compared to the American stuff) please let me know. If that turns out to be the case, I will remove the shatter penalty from the 17-pounder and you'll see its penetration stats rise noticeably.

Charles

[Paul Lakowski]

Theres a new chap who may be able to help here a Dr Elder , he just sent me a paper on decaping APC shells for BB. But he does cover WW-II test results...It seem that this is difficult to predict. Its the first time I've seen the 1950 Ordnance board report included as a source

OH yea one of his findings is that the spaced armor on the Pz-III front hull & turret forced the premature detonation of the 75mm APHE shot leading to no penetration of the main armor ....which may explain why the Brits removed the blasting charges from the rounds....come to think of it this is some thing Valera [RMZ] was saying.

Any way heres another quote from Livingston,

"Which leads to Paul's question about the welded head on the PzGr 39. This was an alloy conservation move and was noted as early as '42 in captured PzGr 39 for the 7.5. The head of the shot all the way to the bourrelet (where the ogive meets the sides) was of the richer alloy. This did not seem to affect the penetration significantly, except if the shot failed. There are some Geman combat stories which describe shots bouncing off when the official tables show a high percentage of penetration, which can be explained by premature shot failure, although the exact type of shot is not specified (these are combat anecdotes). Usually the German APCBC did hang together and do the job, far better than US AP and APCBC. "

It seems that the Allies were not the only ones with this problem. I also remember Valera on RMZ noted that Russian hardness of ballistic tips was too low and reduced penetration potential.

[This message has been edited by Paul Lakowski (edited 11-09-2000).]

[Dschugaschwili]

It has been stated by BTS in another thread that the listed armor penetration values are against enemy armor plate (probably 100% quality enemy plate).

And I think (I can't find this thread right now, so I'm not totally sure) that german armor plate is supposed to be better (harder?) than allied one at the same quality level. So that might explain a part of the observed differences too.

Dschugaschwili

[kipanderson]

Charles,

Thanks for taking time away from your TCP/IP coding to deal with my post. I am one of those greatly looking forward to “live” CM play against friends.

When it comes to the Milne-de-Marre formula I agree 100% with what you wrote. However I was very careful not to over tax the formula. I was not using it to calculate penetration in the “raw”. I was taking your penetration figures as my constant and given one of your figures calculating the penetration figure for a “similar” diameter projectile. The figures I got for the 75L70 gun and the 88L71 gun are very similar to yours and well within the margin of error one would expect from two different formulas. For the limited purpose of testing the “consistent basis” penetration figures of similar diameter projectiles Milne-de- Marre does the job very well, in my view.

As with everything else in CM you have set new standards in the accuracy of your penetration figures. This is clearly illustrated by the thought you have given to the figures for the US 76 gun. I willingly bow to your greater knowledge of the problems experienced with that particular projectile and am happy to accept that your figure of 97mm at 100m against plate at 30 degrees as correct.

However as you imply in your post, and as the figures in CM illustrate, you have tarred British projectiles with the same brush. The US 76 gun had a muzzle velocity of 793meteres per second and you have downgraded its penetration by about 10% from that which it would have been using a German quality projectile. The British Mark V 6pdr had a muzzle velocity of 823 meters per second and you have also downgraded its penetration by about 10% from that which it would have been using a German quality projectile. The British 17pdr had a muzzle velocity of 884 meters per second and you have downgraded its penetration by 18% from that which it would have been with a German quality projectile.

I feel that all Allied projectiles should be “innocent until proven guilty”. By this I mean that unless there is consistent evidence that a given projectile, say the 17pdr, was of lower quality than German ammunition the assumption should be that it was of equal quality. The burden of proof should be with those who believe “German was better”. The decrease in armour penetration, due to increased propensity to shatter as velocity increased, of about 18% at a velocity of around 900 meters per second seems very harsh.

As to what the quality of British projectiles was, relative to US projectiles, this is a difficult question. One way to shine some light on the issue, even if only a little, is to look at how the various authorities in the US and Britain rated the “relative” penetration of different Allied projectiles. For the official British view I have turned to a table of data in the “official” British history of WW2. History of The Second World War, Victory in The West volume 1 by LF Ellis. This is one of a series of books that are the equivalent of the excellent US “green books” official histories of WW2; the British books are also of an equally high standard.

In the notes it is made clear that the penetration data comes from the Ordnance Board centre at Bovington, no surprises there. In the data US APC and British APCBC projectiles are listed separately. (The US APC round was of course a form of APCBC round.) The following penetration figures are against homogenous plate at 30 degrees at 100 yards.

75mmAPC/APCBC—--74mm

6pdrAPCBC-------93mm

90mmAPC---------123mm

17pdrAPCBC------149mm

Against homogenous plate at 30 degrees at 500 yards they give the following penetration figures.

75mmAPC/APCBC----68mm

57mmAPC----------81mm

6pdrAPCBC--------87mm

In the data it is specifically stated that the penetration figures for the US APC and British APCBC 75mm projectiles are the same.

For the official US view of the penetrating power of their own guns I have turned to the classic Catalog of Standard Ordnance Items December 1944. This is a document produced by the US military for their own, internal use. The penetration figures are against homogenous plate at 20 degrees at 500 yards.

57mmAPC----------84mm

75mmAPC----------74mm

90mmAPC----------130mm

Finally here is the penetration data from R Hunnicutt’s book on the Sherman tank, a highly respected source. It is against homogenous plate at 30 degrees at 500 yards.

75mmAPC----------66mm

6pdrAPCBC--------81mm

17pdrAPCBC-------140mm

90mmAPCearly-----120mm

90mmAPClate------129mm

I apologise for including so much data in this post, but there is method in the madness. The reason I have included so much data is that there is no “correct” conclusion to be reached, only an opinion as to what the data “might” mean for the relative penetrating power of the above-mentioned guns. So, far better that I give all of the relevant data and people can reach their own conclusions rather than having to rely on my conclusions alone. Anyway here are my conclusions, however little they may be worth.

Looking at the data is seems to me that there is a question as to weather or not Charles has indeed been too harsh on both the 6pdr and the US version, the 57mm gun. Lets deal with the US gun first and compare the penetration of the 57mm gun to that of the US 75mm gun.

In CM Charles has a penetration figure of 77 millimetres against 30 degree plate at 100 meters for the 57mm gun and 76 millimetres for the US 75mm gun. Yet in the above data the difference between the two guns is never less than 10 millimetres of penetration, the 57mm gun being the more powerful. It is also worth remembering that on a “consistent basis” the 57mm gun should have a penetration about 10 millimetres greater than the 75mm gun. (See the start of this thread.) If you take into account all of the above and give due weight to the quality of the above sources, my view is that the “likely reality at the time” was that the 57mm gun did have a greater penetration than the 75mm gun. In CM I would increase the penetration of the 57mm gun from 77 millimetres to something above 82 millimetres.

When it comes to the question of the quality of British projectiles and the penetration of the 6pdr and 17pdr I also feel that Charles should increase their respective penetration figures in CM. If you look at the relative penetration figures for the 17pdr and the US 90mm gun you will see that the sources that mention both guns give the 17pdr considerably greater penetration than the US 90mm gun, far more than Charles does. There is also some suggestion from the data that the 6pdr did have a slightly greater penetration than its US cousin the 57mm gun. (See the second set of data above, the British figures for both guns at 500 yards.) Taking into account all of the above I believe the British projectiles should be given the penetration they would have had if the figures in CM had been calculated on a strictly “consistent basis”. That is if the British projectiles had been assumed to be of the same quality as German projectiles. Remember, as Paul points out above, the Germans also had quality problems. This would mean a penetration of 139 millimetres for the 17pdr and 85 millimetres for the 6pdr against 30 degree plate at 100 meters. Based on the very limited evidence I have I believe this to have been the “most likely reality at the time”. Which is what we are all after.

All the best,

Kip.

PS. Do take a moment to look at the data and reach your own conclusions.

[This message has been edited by kipanderson (edited 11-28-2000).]

[Guest Big Time Software]

Kip,

The effect you're noticing in CM is that CM penalizes most Allied AP/APC/APCBC rounds for poor manufacturing quality relative to the German rounds. Clearly this is appropriate for the U.S. 76mm round.

Originally we applied the same penalty to the British 17pdr, but in the upcoming v1.1 update we've already changed this, improving the 17pdr's performance. You'll see this in action soon.

So now the question is whether the same should be done for the 6pdr (i.e. treat its shells as "high quality"). If the answer to that is Yes, then should the same be done for the American version of that gun, the 57mm?

My hunch is to say that the American 57mm shells would be no better (or worse) than the American 76mm shells. So I would not think that the American 57mm deserves to be treated as having "high quality" AP ammo. But if the British 6pdr does deserve a "high quality" rating, then does this introduce an inconsistency in the game? Perhaps it doesn't if the American 57mm used American-built shells, and the British 6pdr used British-built shells.

But I'm no expert on this. Can anyone else comment?

Charles

[Guest wwb_99]

I was under the (mis-?)impression that much of the British munitions used came from the U.S.; i.e. the same factories that made the defective 76mm ammo. That might not apply to the 17pdr given that it was not used by the U.S., but might well apply to the 6pdr.

WWB

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

Before battle, my digital soldiers turn to me and say,

Ave, Caesar! Morituri te salatimus.

[Simon Fox]

bump

To answer the last post, no AFAIK 6pdr and 17pdr ammo was of British manufacture though 75 QF ammo was US but modified for British use.

Finally, although the 17pdr APCBC seems to be "fixed" what is the story with 6pdr.

A possible explanation why US 76mm APC ammo was substandard where 57mm APC was only slightly may be that the technical demands of effective heat treatment of the latter were more easily met whereas the 76mm took lonfer to perfect.

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

Muddying the waters as usual.

[rexford]

Jentz report on British tests vs. Tiger show 6 pdr APCBC shattering on a 30° hit when it should have penetrated, and British are quoted as being disappointed. They expected caps to prevent shatter.

Tiger tank in those tests have 250-285 Brinell Hardness armor, so wasn't that much different from U.S. and British penetration test plate.

6 pdr APCBC is solid shot vs. U.S. 57mm with HE burster, so there is about an 8% to 10% penetration advantage to 6 pdr right there.

Where does that come from? 75mm solid shot AP penetrates about 113mm at 0m/0°, 75mm APCBC with HE burster cavity penetrates about 91mm at same range/angle. That's a 24% increase due to differences in nose, burster and caps. We compared data for 37 AP and APCBC and figured that caps cost about 10% of penetration, bursters from 8% to 10%. Exact analysis buried in "my" storage shed somewehere (this clarifies that "I" do not live in a commune where everything is shared, including thoughts).

Projectile shatter is related to striking velocity and penetration/resistance ratio. Faint Praise indicates that 76mm APCBC didn't penetrate Tiger front hull beyond 50 yards.

Rounds shatter because they are too soft and pushing material out of the way at high speed creates stresses that the round can't handle. U.S. tests on nose hardness showed this, harden nose to German level (61 Rockwell C) and shatter failure disappears on hits that should penetrate.

Lowering penetration in CM may be a compromise solution, indepth analysis would have to compare striking velocity to shatter failure ranges and compute ratio of penetration/resistance to see if it fell into the apparent range for shatter gap failure, 1.05 to 1.25.

Should 17 pounder APDS use 76 HVAP slope effects? "I" have never seen a curve for 17 pounder that answers this question, so it is not resolved. Bovington may have something that can answer this.

90mm HVAP appears to have lower slope effects than 76mm, but this doesn't resolve APDS issue.

[Slapdragon]

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

Jentz report on British tests vs. Tiger show 6 pdr APCBC shattering on a 30° hit when it should have penetrated, and British are quoted as being disappointed. They expected caps to prevent shatter.

Tiger tank in those tests have 250-285 Brinell Hardness armor, so wasn't that much different from U.S. and British penetration test plate.

6 pdr APCBC is solid shot vs. U.S. 57mm with HE burster, so there is about an 8% to 10% penetration advantage to 6 pdr right there.

Where does that come from? 75mm solid shot AP penetrates about 113mm at 0m/0°, 75mm APCBC with HE burster cavity penetrates about 91mm at same range/angle. That's a 24% increase due to differences in nose, burster and caps. We compared data for 37 AP and APCBC and figured that caps cost about 10% of penetration, bursters from 8% to 10%. Exact analysis buried in "my" storage shed somewehere (this clarifies that "I" do not live in a commune where everything is shared, including thoughts).

Projectile shatter is related to striking velocity and penetration/resistance ratio. Faint Praise indicates that 76mm APCBC didn't penetrate Tiger front hull beyond 50 yards.

Rounds shatter because they are too soft and pushing material out of the way at high speed creates stresses that the round can't handle. U.S. tests on nose hardness showed this, harden nose to German level (61 Rockwell C) and shatter failure disappears on hits that should penetrate.

Lowering penetration in CM may be a compromise solution, indepth analysis would have to compare striking velocity to shatter failure ranges and compute ratio of penetration/resistance to see if it fell into the apparent range for shatter gap failure, 1.05 to 1.25.

Should 17 pounder APDS use 76 HVAP slope effects? "I" have never seen a curve for 17 pounder that answers this question, so it is not resolved. Bovington may have something that can answer this.

90mm HVAP appears to have lower slope effects than 76mm, but this doesn't resolve APDS issue.<HR></BLOCKQUOTE>

One of the problems I have always had with the WW2 data on weapons penetrations (all the way through the 60s when standards seem to change) is the use of inspected or hand picked ammo. You read the actual test results and you find out that lots of ammo were rejected, and that once a lot was accepted for the test it was used through out the tests. Then other testers would test the same gun with a different lot of ammo and get different numbers.

In the US ammo was mass produced by companies that used to make tin cans, steel washing machine cases, farm tools, and about anything else you can think of. The result was inferior ammo. Tests were done though often (or at least the reports I have read) using armory ammo from long standing shell foundaries. The results could account for Kip's citation with much higher penetration than was the case.

The British, with a much smaller consumer good sector, did not use consumer product companies to the extent. This could explain why their ammo is better, and why UK ammo would perform closer to tests.

This extends to German testing, and to post war testing of captured equipment - since allied tests of German equipment used whatever shells were on hand, sometimes good, sometimes bad. Sometimes shells were hand turned, given data that would be so bogus as to be unusuable -- since hand turned equipment will always outperform mass produced on the average.

The way cannon shells are tested now is the same way that companies test acurracy and velocity of consumer handgun and rifle ammunition. Every shell made on every production line for a set period of time has an equal chance to be chosen for firing tests. A random set of random numbers are generated, in the 60's using a lotto ball wheel, now using a computer, and this number is used to determine what shell(s) from what lots are to be drawn for tests. The same goes for the armor to be penetrated in the tests, and anything else that can mess things up. Without this, official tables for ammo penetration, especially US ammo with known manufacturing problems stemming from war time conversion of factories to arsenals, would all be off by some amount, possible WAY off.

This is one of the reasons why Charles's discarding of these tables and using a physics model is so neat.

[rexford]

Good reasoning, and a decent alternative to blindly using data that has often proved to be bogus.

Shoeburyness tests in England during May '44 are consistent with TM-9-1907 U.S. penetration curves, although 76 APCBC outperforms published data (penetrates 100mm at 30° and 500 yards).

Then 76mm gets to France and can't pierce Tiger front beyond 50 yards (Faint Praise). This suggests large variations possible in field, which your discussion describes.

Would you support penetration data for U.S. projectiles that exceeds published data?

[Basebal351]

Alright, there are a lot of LONG posts in this thread, too much for me to read in one sitting. But, from what I have garnered here and elsewhere lately, is that people are trying to dispell the German "uber" myth of whatever. First of all, while I believe there was somewhat of an overstatement of German and German weapon capabilities made over the years in certain wargames, ALLIED soldiers themselves held, DURING WW2, some of the same beliefs and notions about German weapon superiority. What values do people think need changing? Do the German heavies need weakening? How far are we willing to go to change history to help these games become "balanced" for tournament play?? While trying to dispell the German "uber" myth, are we going to build up the Allies into a force they really weren't?

For "tournament play's sake", are we going to let Sherman 75's go toe-to-toe with a Panther, or a Tiger?? There was a reason American tankers felt LET DOWN by the manufacturers, there was a REASON why they were afraid of most German tanks, espescially heavies. Their tanks were...gasp, INFERIOR!! That wasn't very balanced, now was it? What did they do when their shots started ricocheting off of the Tiger's frontal armor? Did they try to change the Tiger's armor value, or did they vow to "play" only as the Germans from then on?

No, they developed the SAME tactics that have been recommended time and time again by intelligent people on this board. Flanking movements designed to catch the German tanks off guard, in the sides and rear. They used one of the only superiority factors the Allies held, STRENGTH in NUMBERS!! Your German opponent buys an "uber" Tiger, well surround that sonofabitch with a platoon of fast-moving Sherman 75's. Don't try to fight fire with fire by purchasing a disproportionate amount of Hellcats, or Jumbos for God's sake. And don't come crying to the message boards saying the Germans are too powerful.

Steve and Charles, to the best of ANYBODY'S knowledge, got these armor penetration numbers RIGHT. And that's the only way to look at it. Nobody knows, beyond a reasonable doubt, the correct armor penetration numbers. But, what we do know through anecdotes AND tried-and-true armor penetration values, was that German armor, in MOST instances, was superior. The Allies weren't "slighted" in Combat Mission. That couldn't be further from the truth. The Allies were given footing on equal ground, and the people making and playing this game have gone the distance in trying to make the armor differences as historically accurate as possible, without skewing history in the process. And I say, well done. But, perhaps it's time for all of the Allied players using improper tactics to stop complaining about German "superiority."

And think I'm a "German-only" player? That couldn't be further from the truth. I play the Americans 75% of the time, and actually find it easier than playing as the Germans.

And, darn it all, I've made ANOTHER long post. D'oh!

Thanks,

Jim

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

"...you're just jealous that the UK didn't get to join the war before the Americans took Berlin. But, I WILL give credit to where credit is due. If it wasn't for America's almighty industry, we might not have been able to win the war single-handedly. You UK-landers would have been the first people we would have called, though, if we needed some help... or some more tea. We promise."

[This message has been edited by Basebal351 (edited 01-15-2001).]

[Slapdragon]

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

Good reasoning, and a decent alternative to blindly using data that has often proved to be bogus.

Shoeburyness tests in England during May '44 are consistent with TM-9-1907 U.S. penetration curves, although 76 APCBC outperforms published data (penetrates 100mm at 30° and 500 yards).

Then 76mm gets to France and can't pierce Tiger front beyond 50 yards (Faint Praise). This suggests large variations possible in field, which your discussion describes.

Would you support penetration data for U.S. projectiles that exceeds published data?

<HR></BLOCKQUOTE>

From the literature, I would have to say US tanks would not exceed printed penetration listings, in fact I would say that most tanks overstate penetration in testing for reasons I have outlined above. It is possible though that they would understate penetration IF for example the steel plate chosen was much harder than expected, or if other testing variables were not properly controlled. Industrial testing in the 1940s was primitive. Statistics was not used farther than a simple statement of percentages, and lurking variables could be left in that would never make it today.

I always take printed tests from before the 60s with a grain of salt because they were not designed to be repeated, and they are not written in a way that allows someone to say: yes, this is the same test done in the same way as this other one. A few, such as the firings of the Panther after the war, were done on so few shells, and against such a wide variety of targets that the testers may as well of just thrown some dice and yelled "Yatzee". The problem being that these numbers statistcally could be off by dozens of millimeters, and methodologically may be off another couple of dozen milimeters depending on how it was conducted. So I would have to say yes, the numers generated could be low, but they also could be high and we have no way of knowing now.

There is one factor we are missing though, but which may not be worth messing with in game turns. The literature is pretty well established that US 76mm shells performed under expectations despite testing in the US. The reason like I aluded to is that many consumer goods factories were making shells. Likely these factories produced substandard work for their entire life in this industry.

The main prewar arms factories though were still in business and making shells. They made a lot of shells. Compared to the other factories, maybe 10% of the shells were from these more experienced foundaries, a number pulled from my ass since I doubt anyone has ever done a study of this (I am just assuming that since the 1940 shell production rate was 10% the 1944 rate that this represents the professionals).

So, 9 in 10 76mm shells would perform close to Charles chart, lower than the pen tables produced in the US claim. But every so often a tanker is going to luck into a batch of 76mm properly face hardened with a much greater penetration. A tanker can't tell the difference between the good stuff and crap, so they likely would not take having above average shells in their racks into account. Still, one 76 Sherman in 10 concievable would be much more powerful for having loaded a case lot of good ammo in at ressupply.

[StellarRat]

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

My only concern came after a test of 40 hot seat games over two months with 4 people randomly fighting each other on completely random boards fighting random sides, and finding that Germans did statistically better when everything else was random. This is not a fine enough study to say WHY they did better - but with all other variables controlled or randomized their is a small advantage German in quick battles more than is explained by chance.<HR></BLOCKQUOTE>

Perhaps there are psychological reasons why the Germans are doing better? Some people feel the Germans were the underdogs in WW II (because they lost.) It stands to reason that some players would try harder when playing the Germans to see if they could beat the historical odds. Also, could it be possible that players normally favor the Germans therefore the have more experience playing the Germans hence more wins?? Just some ideas.

[This message has been edited by StellarRat (edited 01-15-2001).]

[Paul Lakowski]

Another problem with the comparison is that German test plate was reportedly 'pound for pound'supposed to be 10% better than allied plate, so on the face of it 100mm penetration against german test = 110mm Ve allied test.

German test methods refer to 2/3 of test batch having to meet criteria and test methods speak of 'average results'. It may not be know to some but theres a ± 7% variation in all penetration figures. So if you fired 10 shots you'll get 10 penetration figures all different but with in a certain range of results.

The faint praise test don't mention how many shots were fired but it does appear that about half penetrated at 200-300 yards Vs Panther glacis.

I found that allied projectiles suffered more in slanted impact compared to Russian & german results due to sharper nose designs ...which are penalized in such cases.

[Slapdragon]

One of the most important advances of this game is the physics model as opposed to tests.

I have messed with pistol and rifle ballistics in reloading, and then went to a range and did fairly good tests of penetration using wall board (measuring number of centemeters penetrated) and in wound effects ( for simplicity using modlers clay and measuring wound size by pouring water into the excatvated cavity). In all cases handling variables was a major pain. You need to know the exact mass of the bullet going down range, what the velocity of the bullet is, you have to hit the target as close to center as possible, you have to control for wind, humidity, and the difference between grass under the bullet or black dirt, all of which turned out to cause a noticeable variance in the testing.

If I went to the store and bought a bunch of 20kg blocks of modelling clay and checked the size of the blocks, it would sometimes be bigger than 6 months before when I bought the same clay. Bigger in terms of volume by mass means the clay was less dense, and when I used the clay to fire a series of 9mm bullets into it with a I got greater excavated volume than the last time from the same case of ammunition.

This all got way too complicated after I found how many things could really mess with my answers. The best I could usually do was assign a general style of ammunition a range of penetration or wound trench figures based on two hundred test fires under a 5 different situations and then use some simple statistics to say -- this is what I have, your mileage may vary.

Now, better guys than me have done these tests in labs with lots of money using a special gelatin that is always consistant, and they have created a huge set of so called wound trench effects tables included over penetration estimates, and bullet failure estimates. As lame as my little hobby shooting was, the numbers correlated to some extent to these formulas. In other words, they predicted what I found, even if what I found was not useful in a standard sense since I did not control variables well enough.

For most of these data sets from the 40s and 50s, what is missing is more precise discussion of what conditions the tests were conducted under, for example the density of my clay bought cut-rate at a modelling warehouse because I was a cheap bastard, or the fact that my round-nose ammunition was blunter than I had used before. Then, as Paul pointed out, they should list a range of penetrations values rather than just giving what they consider a median. We could then take the formula Charles used and say -- hmmm, does it predict within barking distance or do we have a problem such as US shattering effect. But many penetration studies just whipped out the old cannons, grabbed the plate they had, fired a lot of shells, sometimes reusing the plate, and come up with one number that proves to be bogus in actual use: ie. US 76 versus official figures.

This is one of the reasons when we have a topic that starts out quoting a table from Jentz or a military manual for penetration that I am automatically a sceptic.