Sherman in CM:AK - peeve

[roqf77]

no. bit the point is its just a prediction. doesnt prove that its true tell we actualy find one. i.e mathmatican predicts that. gun a penetrates gun b at range x. then irl doesnt happen... who's right??

i see your point and agree. im just saying there is no substitute for a real test.

[MikeyD]

Getting hung-up on the math has always been a sticking point in the game. Real-life tank armor differed from vehicle-to-vehicle depending on the age of the castings, manufacturer, etc., etc. I recall hearing of one mid-production M4A1 that had the area next to the ammo racks thickened by beating the sand casts with a wooden plank - you can even see the wood grain preserved in the armor! Cast armor (aka Sherman turret fronts) quality appears to have varied pretty widely. It wasn't until late war that they felt they had got a firm grasp on quality control. The final production Shermans were the best of the bunch armor-quality-wise. So no matter how complex any game's penetration formula gets you're still not going to quite reach real-world equivalents. There will always be some compromise.

[ August 02, 2005, 02:04 PM: Message edited by: MikeyD ]

[flamingknives]

The variation is where the 'fudge-factor' comes in. Give a random chance that the tank has good, bad or indifferent armour.

Originally posted by roqf77:

because you dont actualy have a gun the ammo and the tank. therefore it will only ever be a prediction.

it will never be as good as actualy doing it.

Right. So barring digging up a Sherman and pelting it with AP shot, we shouldn't bother doing anything?

The properties of the materials involved in armour and projectile are fairly well known. Why should a mathematical model be inaccurate?

[roqf77]

no your over reacting thats not what i said okay. chill. there is no reason for a mathamatical model to be particulaly inaccurate. but thats only if you know all the factors beforehand, to an accuracy of 100%.

[Andreas]

So what you are saying is that mathematical models of WW2 armour penetration are particularly inaccurate?

[roqf77]

no. not all. but most i have experienced are. so it depends. it was more of a hypothetical/academic point. i didnt expect it to get so many people offended.

[JonS]

Originally posted by roqf77:

most [mathematical models of WWII armour penetration] i have experienced are [inaccurate].

How do you know this?

[roqf77]

BECAUSE THEY CONTRADICT REAL FIREING TESTS/ EXPERIENCE OF REAL LIFE TANK CREWS.(opps dont mind caps) or are you saying i should swallow everything thats put in front of me?

[roqf77]

sorry to cause the offence i have. im merely speaking from persoanl experience. the only god model i have experienced is combat mission.

[Michael Emrys]

Originally posted by roqf77:

sorry to cause the offence i have. im merely speaking from persoanl experience. the only god model i have experienced is combat mission.

I get your point. Most games are sloppily crafted to one degree or another. (Notice the "most" in that sentence. I am aware that some few miniatures game designers have taken great pains to "get it right". But for every one of them, there are three or four who will do little research and just slap something together and call it "realistic".)

However, and this is a big difference, most of the information used by the services themselves is reasonably well tested and reliable. Even there, there can be disagreement between different sources due to variations in testing procedures, materials, etc. but usually they are as accurate as anything you are going to get. And this is the kind of information that we are trying to track down and discuss. Much better than simply pulling numbers out of our butts, don't you think?

Michael

[roqf77]

yes. my point exactly.

[Bigduke6]

The point is that WW2 is undoubtedly the best-documented conflict in history, and there is plenty of evidence - first-hand accounts and the participating armies' internal communications - to look at, when you are trying to answer the question "What penetrates what, and when?"

So if you want to use the Einstein analogy, you have to ask if Einstein would have been using properly thorough methodology, had he thought about Black holes only with his brilliant brain, while having access to some empirical evidence of Black Holes' existence. (A little green man whispered a hint in his ear, some speculations by Copernicus, etc.)

I don't mean that using math and firing range data is an invalid way to predict armor/AP shell performance. Given the scope of WW2 and the nature of computer programming, it's clearly the only way.

However, that approach carries a risk: sometimes, maybe not often, and maybe even rarely, your mathematical modeling is going to produce results (predictions) that aren't the same as what happened in the war.

After all, even if you're using technical specs and firing table data to figure out AP performance, that basic data was produced by fallible humans.

As I have expounded ad naseum in other threads, a good example of this (in my opinion) is an early to middle-war StuG frontally invulnerable to most AT weapons of its day. Possible in CM, but as far as I know it never happened in the war.

It can be real nit-picky details. To me there seems to be evidence CM uses Soviet penetration tables, without taking into account the Soviet definition of penetration was a bit more rigorous than the German; and also that Soviet penetration data sometimes wasn't gathered empirically on a firing range, but mathematically extrapolated:

All I'm saying is the mathematical modeling ought to have a reality check, and when the results are out of whack with the reality (as we perceive it), the model needs some fiddling.

Clearly this is possible, as we have seen the designers thought hard about the front of a Sherman turret and made it "weaker", for empirical reasons, than you would expect if you were just looking at a sloped 89mm plate fronted by a curved 51mm plate.

[Michael Emrys]

I'm trying to figure out where the "89mm sloped frontal plate" came from. If it's fronted by a "curved 51mm plate" doesn't that mean we're talking about the 75mm gun armed Shermans? If so, all my sources give the turret front armor as 76mm sloped at 30°. So what's going on here? Where did that extra 13mm of armor come from?

Michael

[rune]

Big Duke,

It seems to me...

You would be wrong. Valera was on the Beta team from CMBB, and we knew this years ago.

"Perhaps I will find the whole text of those report (I don't have it now). However, you must understand there was a big difference between SOviet and german menthods for penetration tests. Very approximately, the difference was 25%: the armor considered to be penetrated if al least 50% of a shell happened to be found behind the armor. The Soviets used 75% value."

Also from valera:

"Also, it is important to understand that realistic penetration values in 1941-1943 was reduced significantly due to low quality ammo."

Rune

[flamingknives]

All I'm saying is the mathematical modeling ought to have a reality check, and when the results are out of whack with the reality (as we perceive it), the model needs some fiddling.

This is known as verification. Any mathematical modeller worth more than his computer will use verification as a matter of course.

As for knowing 100% of the data first, that isn't necessary to get a suitably accurate model. I've predicted system performance to +/- 10% without anything but the very simplest data on generic components. In that instance +/-10% was fine and achieved with the nastiest of assumptions.

Not so long ago I did some ballistic calculations for a mortar thread in this forum. I was 5% or less out with some fairly crude assumptions and no actual data about the specific system.

[Michael Emrys]

Originally posted by flamingknives:

Not so long ago I did some ballistic calculations for a mortar thread in this forum. I was 5% or less out with some fairly crude assumptions and no actual data about the specific system.

I bet you're hell on wheels at the gaming tables. They probably close the doors in Monaco when they hear you're in town.

Michael

[flamingknives]

A bit of hyperbole - I did know shell weight and muzzle velocity for a similar round. Coefficient of drag was the big guess.

The casinos won't let me bring my PC in with me

[rleete]

You can't do it in your head? Pfffft. Amateur.

[John D Salt]

Originally posted by flamingknives:

</font><blockquote>quote:</font><hr />All I'm saying is the mathematical modeling ought to have a reality check, and when the results are out of whack with the reality (as we perceive it), the model needs some fiddling.

This is known as verification. Any mathematical modeller worth more than his computer will use verification as a matter of course.

</font>

[flamingknives]

I should have been expecting that...

[John D Salt]

Originally posted by JonS:

</font><blockquote>quote:</font><hr />Originally posted by roqf77:

most [mathematical models of WWII armour penetration] i have experienced are [inaccurate].

How do you know this? </font>

[John D Salt]

Originally posted by flamingknives:

I should have been expecting that...

Indeed, if you were a logical positivist, you should have been able to make a verifiable prediction of it.

All the best,

John.

[flamingknives]

Unfortunately I don't think we're likely ever to get a really good theory of rigid-body penetration, since at penetrator velocities over about 1000 m/sec it all becomes plastic flows, and I bet this is the sort of stuff Mr. Incendiary Cutlery's colleagues spend their time pottering with.

Most, but not all. There's plenty of sub km/s stuff going on to be worried about.

[roqf77]

well largely i know its true, from playing games and getting panther glacias penetrations and single hit knockouts with stuarts 37mm. but not all. that all it was. not a insult against mathamticains or models in general.

"For people who like to contemplate the practically limitless expanse of human ignorance, consider that armour penetration is the thing we probably know the most about of all the things that go on on the direct-fire battlefield. Surveillance and target acquisition, suppression and situation awareness are all much harder topics and much worse understood." 100% true.

although if i were an ituitinist i would of just known that already,

[rune]

I now picture roqf77 dancing around the house singing:

"if i were an ituitinist Ya ha deedle deedle, bubba bubba deedle deedle dum"

Rune