Turret hits and the poor ole PZ-IV

[Mr. Tittles]

This Panzer IV shows T34 tracks across the vulnerable areas. I find it impossible to believe the panzermen would not know about this being a weak area.

[Panzer76]

Originally posted by redwolf:

</font><blockquote>quote:</font><hr />Originally posted by Panzer76:

</font><blockquote>quote:</font><hr />Originally posted by Mr. Tittles:

Does anyone really know what 'rounded' armor means in game terms? How does the math work?

It varies from AFV to AFV. </font>

[Panzer76]

Originally posted by MikeyD:

I don't see where strengthening the PzIV turret front would be called for in the game. Sure its annoying to keep losing your tanks to 'turret front penetration' but it was an inefficient 1930's design and it gives about as much protection as an inefficient 1930s design would.

I disagree, the mantlet covers atleast 40 % of the turret. I agree that any skirts would not add any significant protection, and added tracks cant really be included as it varied from vehicle to vehicle.

[Treeburst155]

A test:

One Regular PanzerIVG, hull down (behind wall) vs one Elite, BORESIGHTED, American 37mm gun at 702 meters. Test run 400 times.

I checked the tank for a KO condition after one minute of combat. I defined a KO as any condition which would prevent the tank from firing its main gun again.

After 400 runs, 49.5% of the PanzerIVGs were KO'd.

I then ran the same test again with only one difference. I removed the wall so the tank was not hull down.

After 400 runs, 39.75% of the PzIVGs were KO'd.

Conclusion: In this particular situation, the PzIVs were better off NOT being hull down.

Note: During the hull-down test, the TacAI backed away from its hull down position. Had the tank stayed put, the KO % probably would have been higher. Still, inspite of the hull down status lasting only 10-30 seconds, the tanks behind walls fared significantly worse.

EDIT: This test leads me to believe that any situation where a turret can be penetrated, but not the hull, warrants avoidance of hull down status. This is especially true if a first round hit is likely whether hull down or not.

[ February 03, 2004, 01:45 AM: Message edited by: Treeburst155 ]

[Treeburst155]

Looking at the test scenario I used for the above test in the editor, the targetting tool revealed some interesting things.

I dug the Mark IVs in, then targetted them (in the editor). Chance to hit was 76% with kill chances "OK". When I "undug" the Mark IVs, chance to hit went up as one would expect to 93%; BUT, kill chances were reduced to "Low"! Even the editor tells us Mark IVs are better off avoiding hull down situations!

An increase in Chance to Hit means nothing when it occurs due to the exposure of impenetrable armor. The extra hits are harmless, and the chances of any one round hitting the vulnerable turret are reduced. So, the editor backs up my exhaustive testing.

Should your Mark IVs be hull down? It depends on whether the gun in question can penetrate the hull at the current range. If not, or if it's marginal, give 'em hull to shoot at!

[ February 03, 2004, 03:06 AM: Message edited by: Treeburst155 ]

[Michael Emrys]

Here's a question for discussion. Given two situations, in the first a tank of any given type is being fired on by a gun of any given type with AT capabilities and is hull-down and in the second situation conditions are exactly the same except that the tank is hull-up, should the gun on average score more hits on the turret of the hull-down tank?

My argument is no, or at least the difference should be miniscule. "Ah," you might counter-argue, "but in the case of the hull-down tank, the gun is aiming at the turret whereas in the case of the hull-up tank it is aiming at the center of mass, thus in the case of the former, more hits should concentrate on the turret." But look again, this may be a specious argument. For one thing, the turret does not present a larger target to the gun; it does not expand when the rest of the tank is invisible. For another, rounds aimed at the center of mass of the hull-up tank that are high may well find the turret. Rounds aimed at the turret that go high will simply not hit anything. Rounds that go low won't either, obviously.

I think my argument is stronger at longer ranges, say outside 500 meters than closer in. It also would depend somewhat on the accuracy of the gun and its sights. But I do feel that a hull-down tank should be harder to hit than is represented by the numbers that Treeburst discovered using his LOS tool.

Michael

[MAsta_KFC]

Your arguments are quite persuasive. Is there a chance BFC might be able to fix this in the next (and most likely, final) 1.02 patch? Hmm, dunno how they'd do it tho, would involve a bit more work then say, adjustments to To&E.

[Dschugaschwili]

Originally posted by Michael Emrys:

Here's a question for discussion. Given two situations, in the first a tank of any given type is being fired on by a gun of any given type with AT capabilities and is hull-down and in the second situation conditions are exactly the same except that the tank is hull-up, should the gun on average score more hits on the turret of the hull-down tank?

My argument is no, or at least the difference should be miniscule. "Ah," you might counter-argue, "but in the case of the hull-down tank, the gun is aiming at the turret whereas in the case of the hull-up tank it is aiming at the center of mass, thus in the case of the former, more hits should concentrate on the turret." But look again, this may be a specious argument. For one thing, the turret does not present a larger target to the gun; it does not expand when the rest of the tank is invisible. For another, rounds aimed at the center of mass of the hull-up tank that are high may well find the turret. Rounds aimed at the turret that go high will simply not hit anything. Rounds that go low won't either, obviously.

I think my argument is stronger at longer ranges, say outside 500 meters than closer in. It also would depend somewhat on the accuracy of the gun and its sights. But I do feel that a hull-down tank should be harder to hit than is represented by the numbers that Treeburst discovered using his LOS tool.

Michael

Your argument would only make sense if

a. the shot distribution of the gun was a uniform distribution and

b. the range of the shot distribution included the entire turret when aiming at the center of mass.

While we can usually expect ( to me true most of the time, the shot distribution is much closer to a normal distribution in reality.

This means that an area near the aim point will be hit more often than an area of equal size that is farther away from the aim point. So hitting the turret more often if one is aiming at it (read: hull down target) is realistic.

Dschugaschwili

[Redwolf]

Originally posted by Michael Emrys:

Here's a question for discussion. Given two situations, in the first a tank of any given type is being fired on by a gun of any given type with AT capabilities and is hull-down and in the second situation conditions are exactly the same except that the tank is hull-up, should the gun on average score more hits on the turret of the hull-down tank?

BFC mad a clear-cut decision very long ago that this is what happens in CM.

A CM AT shooter always shoots at the center of what he sees. No aiming for weak parts. No compensation for lower probability to hit the turret in. If it is hulldown then the turret gets more hits overall.

This is why it would be so nice to get my solution for the small turret model into the game. Then a tank like the Pz IV wouldn't suffer from going hulldown but you wouldn't have to fuss around with super-aimed shooting either (which BFC tried and dropped in a CMBO beta).

[Michael Emrys]

Originally posted by Dschugaschwili:

Your argument would only make sense if

a. the shot distribution of the gun was a uniform distribution and

b. the range of the shot distribution included the entire turret when aiming at the center of mass.

While we can usually expect ( to me true most of the time, the shot distribution is much closer to a normal distribution in reality.

This means that an area near the aim point will be hit more often than an area of equal size that is farther away from the aim point. So hitting the turret more often if one is aiming at it (read: hull down target) is realistic.

Your counter-argument only makes sense if the CEP of the firing gun and its sights at whatever range we are discussing is no larger than the frontal aspect of the target turret (especially its height). Do you know this to be the case? I certainly do not.

Michael

[Michael Emrys]

Originally posted by redwolf:

This is why it would be so nice to get my solution for the small turret model into the game. Then a tank like the Pz IV wouldn't suffer from going hulldown but you wouldn't have to fuss around with super-aimed shooting either (which BFC tried and dropped in a CMBO beta).

This may be a good quick fix for the Mk. IV and a few other tanks, but does not address the more general problem of too many observed turret hits on all tanks. I am thinking more in the long term for CMx2.

Michael

[Redwolf]

Originally posted by Michael Emrys:

This may be a good quick fix for the Mk. IV and a few other tanks, but does not address the more general problem of too many observed turret hits on all tanks. I am thinking more in the long term for CMx2.

Well, BFC tried to implement targetting the weakest point and it failed miserably, creating killer bees. I am sure they will not try again.

In my opinion, you could reward hulldown position overproportionally.

Example: lets say for a hull-up tank the probablity to hit the turret is 35% of all shots. So that we don't mix up percentage bases, let us talk in absolute numbers.

200 shots. 100 miss. The rest, lets say hull-up 30 hit lower hull 35 upper hull and 35 the turret. The overall hit probablity is 50%.

Now, if the thing goes hulldown, then BFC raises the probablity to hit the turret, because the aim point is now higher. Of these 200 shots now 140 hit the ground, 45 the turret and 15 the upper hull. The overall hit probability is less (30%) but the number of hits on the turret raised.

I would be perfectly happy with keeping the 35 on the turret and just dropping the upper hull from 35 to 5. That would lower the overall hit probability to 20% and keep turret hits exactly the same.

That would benefit hulldown more than it does now, which might be a good thing to reward good tactics and it also makes the irritation of more turret hits go away.

[Michael Emrys]

Originally posted by redwolf:

I would be perfectly happy with keeping the 35 on the turret and just dropping the upper hull from 35 to 5. That would lower the overall hit probability to 20% and keep turret hits exactly the same.

That would benefit hulldown more than it does now, which might be a good thing to reward good tactics and it also makes the irritation of more turret hits go away.

That sounds just like what I'm asking for.

Michael

[Mr. Tittles]

The game downplays the reality of hulldown in many ways. Some fixable at this point, others not.

Hulldown tanks are harder to spot. But with the shared spotting, this defeats this battlefield fact.

Hulldown tanks should not experience 'concentrated' hits on the turret as others have pointed out. This is just wrong.

Hulldown tanks with turret inferior armor or mixed protection on the turret are modeled as rectangular slabs of one armor thickness/angle (in the case of the Panzer IV, the weakest part).

I would like to see something like redwolfs reprogramming or just an attempt like curved armor/averaged armor/anything-else to address the problem. For the H/J to be so mismodeled kills the experience for me.

I also think size for tanks should be reduced when hull down by factors such as: Gun depression, experience, leadership (when in radio contact). So a hull down Panzer IV is not the same as a hull down T34.

Height is the biggest factor when trying to hit a tank at range. A good test would judge the difference between a hulldown tank an non-HD recieving hits.

[Treeburst155]

At some fairly close range, the exposed turret of a hull down tank would be big enough in the sight reticle that Chance To Hit would not be much less than if the same target were not hull down. In such a case, if we assume the gunner is aiming for center of visible mass at all times, the weaker turreted tank is better off exposing the hull, thus lowering the enemy gunner's aim from the weak turret COM to the turret ring.

The question then becomes one of how much hit percentage should fall off for a hull down target at various ranges compared to the same target at the same ranges with the target hull up. What should the curve look like?

In my test situation above, Chance To Hit dropped from 93% to 76%. This 17% drop (18.3% of original chance) does not seem like enough at a range of 700 meters considering about half the target is now hidden; but I'm only going on intuition here. Also, the fact that my 37mm was boresighted and elite may account for the fairly small drop in Chance to Hit. More ordinary circumstances are easily checked in the editor. I will do this today.

As for the front armor thickness of the PzIV, I'll let you grogs debate that issue. I'll concentrate on the actual effects hull down status has on enemy gunners in the game.

[ February 03, 2004, 12:09 PM: Message edited by: Treeburst155 ]

[Treeburst155]

If you remove the TRPs from my test scenario, and change the elite 37mm gunners to regular, the Chance to Hit the dug in Mark IV is 19%. C to H goes up to 34% when the tank is hull up. This represents a 44.1% decrease in Chance To Hit when the tank goes hull down. This seems about right to me for a regular gun crew at 700 meters. I'm happy with CM's treatment of hull down as far as Chance To Hit changes are concerned. It's the weak turret armor on the PzIV compared to the hull that makes hull down of questionable value. If hull down tactics are to be rewarded, the best fix would be to beef up the PzIV turret in some way IMO.

[CombinedArms]

I will bet dollars to donuts that we aren't going to see a fix in this area until the new engine. I see good suggestions here--some of which might be practical even with the current engine-- but BFC has left this situation unchanged through two new game releases (CMBB and CMAK), so I can't imagine they will try to address it in a mere bug-fix. No doubt CMx2 will have a more sophisticated model for armor penetrations. For now, I'm guessing that we are stuck with this unfortunately "gamey" anomaly.

This makes me want to return to the issue in tactical terms. I'm now quite convinced (when I started my own tests I doubted it) that there are some situations where the PzIV is better off hullup than hulldown. Treeburst's test seems even more persuasive than my own, and I trust that Redwolf's results, which I haven't seen, were also compelling.

So my question is: Are there any other common tanks or AFV's where this also applies? It would have to be a tank with relatively weak turret armor and/or upper hull armor and relatively strong lower hull armor, esp. with reference to the guns it most commonly confronts (Of course, the reference gun for the PzIV is the Sherman 75--if this were a Firefly no doubt you'd want to revert to hulldown).

Off hand, I can't think of any other tanks to which this hullup preference might apply (except maybe the Grant, so it can use its main gun), but I'd be glad to hear suggestions.

[Redwolf]

Originally posted by CombinedArms:

I will bet dollars to donuts that we aren't going to see a fix in this area until the new engine. I see good suggestions here--some of which might be practical even with the current engine-- but BFC has left this situation unchanged through two new game releases (CMBB and CMAK), so I can't imagine they will try to address it in a mere bug-fix. No doubt CMx2 will have a more sophisticated model for armor penetrations. For now, I'm guessing that we are stuck with this unfortunately "gamey" anomaly.

It's do sad. CMAK is about as close to the perfect game as anything I have seen, except DOOM2 maybe.

If you look at all the highly complex things coded in, from all the 3D models with independently rotating turrets, the TacAI, exploding houses, passenger, explosions, all kinds of detailed angle computation for AA shot - and now it is not even possible to fix a knockout point issue, correct artillery pricing, not have a barrage screwed up by temporary obfuscation of LOS before FFE or insert a minimal modification for hit probability distribution.

This makes me want to return to the issue in tactical terms. I'm now quite convinced (when I started my own tests I doubted it) that there are some situations where the PzIV is better off hullup than hulldown. Treeburst's test seems even more persuasive than my own, and I trust that Redwolf's results, which I haven't seen, were also compelling.

I didn't keep results. I was just running a few quick tests to make sure there is no hidden mechanism in CM to correct the situation - that a turret hit become more probable when you go hulldown. From that on the given numbers lead to Treeburst's results with normal calculation. So it is in fact that a tank with an extra small unangled turret surface is punished heavily.

So my question is: Are there any other common tanks or AFV's where this also applies? It would have to be a tank with relatively weak turret armor and/or upper hull armor and relatively strong lower hull armor, esp. with reference to the guns it most commonly confronts (Of course, the reference gun for the PzIV is the Sherman 75--if this were a Firefly no doubt you'd want to revert to hulldown).

The early T-34 obviously.

This becomes complicated and gets out of hand quickly if you want to become more sophisticated. There are many tanks which have about the same basic strength on turret and upper hull but they gain strength from thickness on the turret and strength from angle on the upper hull. If you now start taking ammunition into account that reacts in non-standard ways to angles (most tungsteen shots) then things get messy quickly.

The same applies when you start taking elevation into account. If the shooter is higher or lower than the target (or the target is on a slope) then you quickly lose most of the angle on the hull but the turret becomes actualy stronger.

If you want to look into this the first thing we would need is numbers about the "curved" surfaces in CM. These numbers, the distribution of angles, are not given out by BFC but shooting results suggest that some of them might be off - in particular for the Jagdpanther and Hetzer.

[Michael Emrys]

Originally posted by Treeburst155:

In my test situation above, Chance To Hit dropped from 93% to 76%. This 17% drop (18.3% of original chance) does not seem like enough at a range of 700 meters considering about half the target is now hidden...

Actually closer to two-thirds, depending on which tank we're talking about.

I'll concentrate on the actual effects hull down status has on enemy gunners in the game.

Thanks, Tree. You don't know how much I appreciate your efforts to clear up these kind of gameplay issues.

Michael

[Treeburst155]

You're welcome, Michael. I'm just trying to improve my game here.

I figured out a good way to test hit location percentages!! It's tough to explain, but fairly easy to do. I'll go into the details if somebody REALLY wants me to.

The results of Test One (hull down Mark IV):

Out of 512 hits on a dug-in Mark IV, 73.6% were turret hits. This is counting "Gun Hits" as turret hits, which seems logical to me. So, Panzer76 was correct in his assertion that about 25% of hits on hull down vehicles is to the upper hull.

It will be interesting now to see what percentage of hits are on the turret when the tank is NOT hull down, all else being the same.

[Michael Emrys]

That may be interesting, but what we need to know is does the absolute number of hits on the turret change between hull-up and hull down, and if so by how much? In other words, if 1,000 shots are fired in each test, and 370 strike the turret when the tank is hull-up but 560 hit it when it's hull-down, then something is rotten in Denmark.

Michael

[Treeburst155]

My next test will answer these questions, Michael. :cool:

[Redwolf]

Originally posted by Michael Emrys:

That may be interesting, but what we need to know is does the absolute number of hits on the turret change between hull-up and hull down, and if so by how much? In other words, if 1,000 shots are fired in each test, and 370 strike the turret when the tank is hull-up but 560 hit it when it's hull-down, then something is rotten in Denmark.

Michael

I never gathered this data but this is what BFC said it is.

Don't ask me for a thread reference, but it was discussed a lot of times in CMBO times. Steve said that the aim point is always the center of what is visible and hence the aim point gets up for hulldown and hence you have a higher absulte number of turret hits.

[Treeburst155]

I remember that CMBO discussion too. CM gunners ALWAYS aim at center of VISIBLE mass. In the case of hull down they will hit the turret with 75% of all hits. How much is this turret hit percentage reduced when the target is NOT hull down? This is the real question I think.

[Treeburst155]

With center of visible mass aiming, you would only have a higher absolute number of turret hits on a hull down vehicle if the Chance To Hit did not go down enough to counteract the center of visible mass aiming. The percentage of hits that impact the turret SHOULD be higher because you can't hit the lower hull at all.