Matrix Games Forums

First a snippet from another thread....

We're just going to do a single shot here, though rate of fire is also modeled, as is turret speed. We'll just assume they're both facing each other head on. We will also assume they have sighted each other, which is not a given as there can be vast differences in your sighting range and effectiveness based on what your tank is doing and what the other tank is doing. For example, a PzKfwIVF2 rushing forward would almost certainly be shot at by that T-70 before it realized the T-70 was there.

Ok, now let's say the PzKfw IVF2 shoots its first shot at the T-70. First, the 75mm/L43 has a To Hit number of 5 at 500 meters. It's almost a sure hit inside of about 425 meters, but the difficulty starts to climb after that. So the base roll to hit is a 5 or more on a d10.

Since this is the first shot at a target over 300m away, that's modified up to a 6. The T-70 also has a size modifier of 1 (it's a small tank) which makes this a 7. There are no modifiers for previous shooting, for either tank moving or for shell type (HEAT and HE are less accurate), nor are there modifiers for cover in this case. We'll say the Panzer already turned its turret the previous phase, so no penalty for that either. So, basically, we have to roll a 7 or more on a d10 to hit with our first shot, which gives us a 40% chance. After the first shot, this goes up to a 50% chance. Once we actually score a hit (any hit) on the target, this goes up by another 10% to 60% as it's considered to be an "acquired target". If we were shooting at the side rather than the front there would also be a bonus. Note that the "to hit" roll is open-ended, so a result of 10 has a 50% chance to end up being a 11-15, which allows "lucky shots" to happen (i.e. driving forward, turret turning, snap shot at 500m, ping! - highly unlikely, but possible . Anyway, let's say we hit.



To summarize: the initial shot gets a 40% chance at 500m, the second shot is 50%, the third shot (if a hit is obtained) is 60%. Otherwise , its stuck at 50%.

Its my opinion that this is woefully undermodelled. Its also such an important part of the game that I HOPE the designers reconsider this modelling.

Matrix has stated that the game can be modded as far as these hit numbers, but I do not think that the released game should be as far off as I consider this to be. In other words, we can use 'Rexfords-Hit-Mod-Pack' or something like that.

For a weapon like the German 75mmL43, hittinmg a target at 500m is no great feat. in fact, the Germans preferred shooting it out at much longer ranges. At this time in WWII, the StuGIII with the 75mmL43 and the PanzerIV with the same gun were probably the best AT systems around.

The elements of gun accuracy can be broken down into:
1. Estimating the distance
2. Estimating the speed (stationary in the above example)
3. Laying the weapon (adjusting sights)
4. Firing and observing the shot
5. Corrections based on 4


At 500m, even a 20% error in distance estimation will only put you off +/-100m. The gun systems dispersion will then be the major factor in realizing a hit or a miss.

There are many other factors but these are the major ones. Since these are Vets in this game, I would not appreciably factor in 'human-factors' that much.

My general impression is that the modeling is off by at least a factor of 2. that is, what they are modeling at 500m might actually take place at 1000m. The followup shots are way off, especially if a BOT (burst on Target) is obtained.

Here is some data from the Rexford-Files. Note: This is not dispersion data but data based on the physics of projectile motion.

MAXIMUM TRAJECTORY HEIGHT
APCR VS APC/APCBC AMMO

RANGE..50L60.....75L48....88L56...88L71
500m...0.3/0.5..0.4/0.6..0.4/0.5..0.3/0.3
800m...1.0/1.5..1.1/1.5..1.0/1.4..0.7/0.9
1000m..............1.8/2.5..1.7/2.3..1.1/1.4
1500m..............4.5/6.3..4.0/5.5..2.7/3.2

Notes:
Max trajectory heights in meters
Slash separates APCR/AP trajectory data

This is dispersion data. (Note: Its for 1000m)

Here is a comparison of the 90% zones at 1000m for a few capped AP type rounds:

50L60 APC: 0.81m vertical and 0.73m lateral
88L56 APCBC: 0.98m vert and 0.56m lat
17 pdr APCBC: 1.19m vert and 1.01m lat
75L48 APCBC: 1.37m vert and 1.27m lat

Some more dispersion data 50% zones

75L48 APCBC
100m....0.1m/0.0m
300m....0.2m/0.2m
500m....0.3m/0.2m
800m....0.4m/0.4m
1000m..0.6m/0.5m
1300m..0.8m/0.7m
1500m..1.0m/0.9m
2000m..1.6m/1.3m
2500m..2.4m/1.8m
3000m..3.3m/2.3m

If you analyze the data closely, it becomes apparent that a weapon like the 75mmL48 can fire at targets within a reasonable distance, and due to the minimal trajectory height, combined with the small dispersion, hit virtually most stationary afv targets within certain ranges.

The Germans knew this and it was SOP to have a AP round loaded (in armor defeating weapons) and have the sights set at a fixed range. This was usually done when moving to contact so that a reasonable chance of quickly engaging a target could be achieved. The Germans said 'Shoot first, if you can't shoot first, then hit first'. Depending on the weapon and battle conditions, the sights were set for 500-800m range. Tank sized targets (especially large ones like the US sherman) could then be quickly fired upon without the usual drill of dialing in a range.

Some reports state that the Germans would use zones for firing at targets under 1000m or so. They would have a setting for close, near, medium ranges. Each of these settings would give the best chance of hitting a target in those zones. for shooting at targets beyond 1000m or so, they would use traditional gunnery concepts. This zone concept comes about due to the high velocity flat trajectory.

Targets at ranges of 500m are considered close. The US FM for tank gunnery defines this as being close to be specific. The Germans certainly did not want to engage the Soviets at such ranges. Virtually any trained gunner is going to hit you with a tank or antitank gun (within the first couple of shots). This is actually close enough that specific areas of a vehicle can be targetted (for a good gunner with a high velocity weapon after a few shots). Turrets or tracks may be cherry picked instead of shooting at center of mass.

I hope the designers can cite the sources they are using for the games gun accuracy system. I would be very interested in seeing it.

WO 291/1271 The ranging and dispersion of HE M67 (Anti-Tank) Shell in the
105mm Tank Howitzer
This gun ranges in close agreement with the firing tables for the 105mm howitzer in the M7 howitzer
motor carriage.
Percentage hits expected, once the MPI is adjusted on to the target:
Range (yds) Panther turret
5 ft × 2 ft
Panther head-on
12 ft × 10 ft
Panther side-on
23 ft × 7 ft
500 89 100 100
1000 43 100 99
1500 21 94 89
2000 12 76 73
2500 7 56 55
3000 4 35 36
3500 2 19 20
It is considered that a hit probability of 50% is a good line for "satisfactory" engagement range, so it is
recommended that engagements be conducted at 900 yards on hull-down targets and 2500 yards for
hull-up ones.

I am posting this data to make some quick points. The first is, just how precise the weapon system is. This data represents firing at a known range and moving the mean point of impact onto the target. In other words, its akin to zeroing data. Note that this is for a hollow charge shell. It would be perfectly plausible to zero this 105mm howitzer at 1000 yds in my opinion.

A very interesting point is that hitting the broadside of a panther does not make the numbers jump up. This is because these weapon systems are so accurate as far as breadth variations (rounds falling left and right) that it makes no difference. Height is the major factor in accuracy in direct fire HV weapons The panther turret target proves this out.

In real life, shooting at the side of a tank might really come into play when that tank is a moving target. The forgiveness in 'lead' would make it vulnerable is my point.

The data certainly shows just how hard hitting a hull down target can be as range increases. Mind you, this firer knows the range. In real life, guaging range to a hull down target is very difficult also.

At 500 yds, even with a 20% range estimate innacuracy (US tankers were tested to within 10%), most high velocity weapon systems have a very good chance of hitting a target like a tank. The second round followup shot even more so.

WW2 hit probabilities
WO 291/180, Accuracy of anti-tank gunnery
Ranges in yards, target assumed to be Pz VI size.
Probability (%) of hitting static hull-up target with first round:
Gun OK for... 500 1000 1500 2000 2500
6 pdr line 100 100 96 87
range 87 33 13 3
both (hit) 87 33 12 3
17 pdr line 100 100 100 98 93
range 98 46 20 10 5
both (hit) 98 46 20 10 5
Probability (%) of hitting static hull-up target after first round:
Gun 500 1000 1500 2000 2500
6 pdr 100 86 59 41
17 pdr 100 94 71 50 36
Probability (%) of hitting moving target (direct-crossing at 15 mph) after first round:
Gun 500 1000 1500 2000 2500
17 pdr 86 48 26(e) 16
Probability (%) of hitting hull-down target with first round:
Gun 500 1000
17 pdr 59 18
Probability (%) of hitting static hull-down target after first round:
Gun 500 1000 1500 2000 2500
6 pdr 85 43 22 14
17 pdr 88 51 29 18 12
Comments and corrections
These values are those plotted on the graphs in the report; the value marked (e) is interpolated from
other data points. The overall shape of the fitted curves in each case is sinusoidal.
This report recommends that the maximum range of engagement for 6-pdr and 17-pdr ATk guns be
considered 800 and 1000 yards respectively. The criteria stated for maximum range of engagement for a
statically-sited ATk gun are:
50% chance of first-round hit on a static hull-up target;
90% chance of subsequent rounds hitting a static hull-up target;
50% hits on a hull-up direct-crossing target moving at 15 mph after MPI roughly corrected;
50% hits on a static hull-down target after MPI roughly corrected.
The first table clearly shows that errors in range have a much more important effect on accuracy than
errors in line.
Penetration ranges against Pz VI for each gun (ammunition not specified) are stated as being 800 yards
for 6-pdr on the front, 1600 yards on the side, and 2000–2500 yards for 17-pdr. Hit probability is
therefore regarded as a more important limitation on maximum engagement range than penetration.

The above data is more reality driven than systems driven. It exludes real battlefield dangers but gives a truer picture of what a HV gun can do.

Basically, the first round hit numbers for both guns at 500 meters (taking in range estimation errors) is very high. Second round followup even higher. Certainly getting a hit would mean a very high chance of followup shots hitting the target.

From Matrix games....

you disagree with any of our numbers, be it an armor rating, the hit location chart, or gun accuracy at various ranges, as long as you can edit a text file you can change it and the game will use your edited data. I'm sure that we will see not only texture mods and perhaps new models for the game, but also a lot of tweaks to the data once grogs sink their teeth into it. While we may not have the best "grass mod", there will likely be various different "stat mods" after release. Here's an example of the 75mm/L43 as a text (XML) file:

<gun caliber="75" type="75mm/L43">
<ammunition destruction="6" stun="10" type="APHE">
<tohits>
<tohit number="1" range="275"></tohit>
<tohit number="2" range="350"></tohit>
<tohit number="3" range="425"></tohit>
<tohit number="4" range="450"></tohit>
<tohit number="5" range="500"></tohit>
<tohit number="6" range="550"></tohit>
<tohit number="7" range="600"></tohit>
<tohit number="8" range="675"></tohit>
<tohit number="9" range="750"></tohit>
<tohit number="10" range="900"></tohit>
<tohit number="11" range="1200"></tohit>
</tohits>

If I follow the game's dynamics so far, at 1200m a PIVL43 has almost no chance of a first round hit on a T34. Second round hit is also very low as are follow up rounds till one of those low probabilities turns into a hit. Even then, the 10% hit percentage is exceedingly low.

In Panzertruppen2, onm pg.41., a battle report details use of the PanzerIVL43 during this period Feb-Mar 1943. It states:17 KV1, 26 T34, 1 T26, 1 MarkII, 3 MarkIII, 1 General Lee destroyed

Pzgr39 (AP) were fired at ranges from 1200-1600 meters. Every hit caused a destructive effect. Two to Three AP were expended per tank killed.

I do not see how PC can possibly model real world conditions. Especially since actual German training expectations for completion of the course could not possibly be met given the hit percetages listed?

Its my general opinion that the game could basically double all the range values and still be conservative given a Veteran crew.

I also think that follow up shot modifiers should be reflective of the crews abilities. Giving all units equal modifiers makes the tanks 'robotic' and does not reflect battlefield realities such as being unbuttoned, optics, 3 man turrets, etc. Not to mention experience and training.

I won't be able to provide a 100% detailed answer here, but there are a number of key factors around combat -- it is actually quite a detailed simulation.

Factors that go into "hitting" include the gun capabilities, type of ammunition, terrain between the targets, armour rating (including density and angle of entry) of the target (if applicable), if the target is moving, if the firing vehicle has moved recently, etc.

Once a target is "hit", you need to determine the type of damage done, primarily of the shell has penetrated its target -- depending on where it hit (i.e., gun, treads, front, side, cupola, etc.) Again, this is made by a number of factors, especially the ammunition type.

Finally, you should understand that the die is actaully not a simple 10d. It is actually a -4 to 15 roll. (If you get a 1 or 0 on the first die roll, you roll a second time to determine the actual value.) Because of this, the distribution of potential rolls is also not uniform -- i.e., you have a 10% chance of rolling a 2, 3, 4, 5, 6, 7, 8 and 9, but only a 5% chance of rolling a 1 or 10 , and 1% chance of a 0, -1, -2, -3, -4, 11, 12, 13, 14 or 15.

Deride

ORIGINAL: Deride

I won't be able to provide a 100% detailed answer here, but there are a number of key factors around combat -- it is actually quite a detailed simulation.

Factors that go into "hitting" include the gun capabilities, type of ammunition, terrain between the targets, armour rating (including density and angle of entry) of the target (if applicable), if the target is moving, if the firing vehicle has moved recently, etc.

Once a target is "hit", you need to determine the type of damage done, primarily of the shell has penetrated its target -- depending on where it hit (i.e., gun, treads, front, side, cupola, etc.) Again, this is made by a number of factors, especially the ammunition type.

Finally, you should understand that the die is actaully not a simple 10d. It is actually a -4 to 15 roll. (If you get a 1 or 0 on the first die roll, you roll a second time to determine the actual value.) Because of this, the distribution of potential rolls is also not uniform -- i.e., you have a 10% chance of rolling a 2, 3, 4, 5, 6, 7, 8 and 9, but only a 5% chance of rolling a 1 or 10 , and 1% chance of a 0, -1, -2, -3, -4, 11, 12, 13, 14 or 15.

Deride

That still means for a 75mmL43 firing at 1200m at minimal chance of scoring a first round hit.

<tohit number="11" range="1200"></tohit>

If I understand you correctly (I believe you mean to say '0 or a 10', no?), rolling an 11 is a 5% chance. Most of the modifiers you mention will actually decrease the to hit further?

what Yoosername said,

we took a lot of flak over the first round hits in Squad Assault, but it really should be that way

E

ORIGINAL: Yoozername
That still means for a 75mmL43 firing at 1200m at minimal chance of scoring a first round hit.

<tohit number="11" range="1200"></tohit>

If I understand you correctly (I believe you mean to say '0 or a 10', no?), rolling an 11 is a 1% chance. Most of the modifiers you mention will actually decrease the to hit further?

The tohit number provides the base chance to hit which is then further modified. For example, you can get a -1 if you are firing at the side of a stationary target or a +2 if the target has moved already during the turn. There are also +/- modifiers for size of your target. Again, there are a very large number of factors that I can't completely get into on a forum. But, there are a few negative modifiers.

Once you have a modified tohit chance, you determine the number of hits based on the weapon's rate of fire. You can score up to 8 (!) hits during a turn phase (although, for tanks, you are looking at 2 or 3 as an absolute maximum.)

To see if you get a hit, you rool a d10+/-5. If you are >= the modified tohit, you hit the target. So, if your modified chance to hit remained at 11, you would have a 4% chance to hit for each potential shot (which, again, may be 2 or 3.) For each hit, you then determine penetration. I.e., see if there is any real damage to the target.

Sorry that I'm being somewhat short in the description. Some of these factors are still being tweaked for the game, and it would require an extensive post to cover all the factors that will make the final version of the game. Beta testing should provide a good balance of how well these factors work, and, of course, all data is open to modification.

Deride

Once a target is "hit", you need to determine the type of damage done, primarily of the shell has penetrated its target -- depending on where it hit (i.e., gun, treads, front, side, cupola, etc.) Again, this is made by a number of factors, especially the ammunition type.

Finally, you should understand that the die is actaully not a simple 10d. It is actually a -4 to 15 roll. (If you get a 1 or 0 on the first die roll, you roll a second time to determine the actual value.) Because of this, the distribution of potential rolls is also not uniform -- i.e., you have a 10% chance of rolling a 2, 3, 4, 5, 6, 7, 8 and 9, but only a 5% chance of rolling a 1 or 10 , and 1% chance of a 0, -1, -2, -3, -4, 11, 12, 13, 14 or 15.

Deride

Thanks but please clarify that

(If you get a 1 or 0 on the first die roll, you roll a second time to determine the actual value)

actually means ..1 or a 10..?

So a final To Hit number of '9' is a strait 20% chance?

And a final To Hit number of '10' is a 10% chance

And a final to hit number of '11' is a a 5% chance? Because 11, 12, 13, 14 and 15 gives 5% And increments of 4%, 3%, 2% and 1% are possible? Is the second die roll a 10 sided die? If the second die roll is 6-10, it remains '10' and if its 1-5 its 10 plus 1,2,3,4 and 5 respectively? Thats how you get 11-15?

ORIGINAL: old man of the sea

what Yoosername said,

we took a lot of flak over the first round hits in Squad Assault, but it really should be that way

E

Can you link to a thread?

I am not just arguing about the first round hit percentage. the second round and subsequant rounds after a BOT (burst on Target) are too low.

As an example, at 1200m a 75mmL43 firing at a T34 (2.5 meters tall) would have a base number of 11. On the first shot, it gets a +1 to a 12. So a 4% chance of hitting on the first round. The next round is an 11 or a 5% chance. This is a 1% chance improvement. IF this round hits (20:1 shot), THEN it gets to be 10% chance?

This is fundamentaly wrong. ROF does not factor into this. Unless the game has some more negative type modifiers, the math just does not bear out considering actual battlefield reports or actual training completion requirements.

Seems like you aren't letting me get away from this, so I better respond in some better detail. [:D]

ORIGINAL: Yoozername
Thanks but please clarify that

(If you get a 1 or 0 on the first die roll, you roll a second time to determine the actual value)

actually means ..1 or a 10..?

A d10 roll results in 0, 1, 2, ... 9.. there is no 10. This is just a programming thing, I guess. A 0 actually represents a 10 and a 1 represents a 1. In the case of rolling a 1, you have the chance of being 'under' -- i.e., -4, -3, -2, -1, 0 or 1. In the case of rolling a 10, you have the chance of being 'over' -- i.e., 10, 11, 12, 13, 14 or 15.

(The way you determine this is by rolling the d10 again. If you rolled a 1 the first time, you will get the value of 1 for 6-0, but a roll of 5 gives you 0, 4 gives you -1, 3 gives you -2, 2 gives you -3, and 1 gives you -4. Thus, rolling a 0 represnts 5%, -1 1%, etc.)

ORIGINAL: Yoozername
So a final To Hit number of '9' is a strait 20% chance?
And a final To Hit number of '10' is a 10% chance
And a final to hit number of '11' is a a 5% chance? Because 11, 12, 13, 14 and 15 gives 5% And increments of 4%, 3%, 2% and 1% are possible? Is the second die roll a 10 sided die? If the second die roll is 6-10, it remains '10' and if its 1-5 its 10 plus 1,2,3,4 and 5 respectively? Thats how you get 11-15?

Hitting means that your roll is greater than or equal to the final to hit number. For example, if the final to hit is 6, then a roll of 6, 7, 8, 9, 10, 11, 12, 13, 14 or 15 will score a hit. So, you need to add all of those percentages up -- in this case 50%.


Yoozername, I think your issue is really around the 75mm/L43 and its ability to hit at range. At 1200 (GSUs = game scale units; 1 GSU = 1 m), we have modelled a base chance to hit of 4% when using APHE ammunition.

Now, you might argue that this is not a realistic characterization of this weapon during that time period, but we are using data that we did not make up. This core of this data has been used previously in other simulations. I also think it is important to let the game go through some beta testing to understand if this is a reasonable capability for this weapon or not. And, certainly, I am no weapon expert myself, so I have no ground to argue one way or the other.

Thanks for your interest... And hopefully this provides a bit more insight.
Deride

Russ,

I posted a detailed example in another thread and we're looking at this from the game balance side. I don't think there's any real confusion as to how the game works, just a question of whether Yoozername's referenced data is valid for combat situations or not. I'm doing some additional research on that.

Regards,

- Erik

Thanks for that Erik...

Deride

http://gva.freeweb.hu/weapons/german_accuracy5.html

Some data comparing the German 75L24 (stump) and the Panther 75mmL70.

Note that dispersion data can be read to mean a known range AND the mean point of Impact (MPI) is adjusted onto a target. It is system precision.

Noteworthy is that the 75mmL24 is just barely 'zero-capable' at 1000m. This is for a 2.5mx2m target. a 75mmL48 and certainly a 75mmL70 are perfectly capable of zeroing the weapon at this range. In fact, the 75mmL70 was certainly capable of destroying armor in the 1500-2000m range within 4-5 shots.

From the website...


Gun Accuracy Data
The Gun Accuracy tables show the results of two types of tests: dispersion and firing.

Dispersion tests show the percentage of projectiles that will hit a 2.5m × 2m target during controlled test firing. The pattern of dispersion is assumed to be centred exactly on the target. These results give a good theoretical comparison of guns and ammunition types, without considering the complicating effects of human error.

Firing tests show the expected percentage of projectiles that will hit a 2.5m × 2m target by a gunner during practice firing on a gun range. It is obtained by doubling the dispersion pattern obtained from the dispersion test data. The British, Germans and Italians all considered this to be a close approximation of the accuracy obtained by troops in practice firings and, if they remained calm, in combat when the range to the target is known. Due to errors in estimating the range and many other factors, the probability of a first round hit was much lower than shown in these tables. However, the average, calm gunner, after watching tracer from the first round, could achieve the accuracy shown on subsequent shots.

So what does this all mean? Lets just discuss stationary targets for now.

Basically it can be summarized that first round accuracy is dominated by human error. The greatest human error is range estimation. As I have said before, not all range estimation is the same. In fact, in some cases, like an antitank gun defense that has time to set up and actually survey and measure range, its minimal. First round accuracy in these situations might be better than the commonly used 'double-disperion'.

In cases such as when an AFV advances into unknown terrain, First round accuracy cen be speculative. Second round accuracy depends largely on observation of the first round but might approach or exceed double-dispersion. This is an iterative process that reduces the error till we approach system precision (dispersion). We will only get as good as that.

The ideal game, in my opinion, would use dispersion data as a base to hit number. It would then have modifiers to these numbers based on a range estimate and other factors to come up with a true first round hit number. Subsequant to hit numbers could factor in being unbuttoned, opticsm, etc. to iterate each shot towards MPI and system precision.

An interesting thing is that a hit could be an outlier. A shot on the edge of the distribution that dispersion implys. So a game that allows a big bonus for this is not always correct.

While anectodal, this oft quoted account (Its a KV2 I believe) shows the accuracy of these weapons.



CHAPTER 1
INTRODUCTION
Lithuania, 24 June 1941
About half-way to Rossienie . . . the only supply route to the bridgehead was blocked by a heavy tank of the KV type . . . . an antitank battery with 50-mm guns was ordered to work its way forward and destroy the tank . . . . The first round, from about 600 yards, was a direct hit. A second and third round followed. By the time the eighth hit was scored, the Russian tank crew had discovered the position of the firing battery. Taking careful aim, they silenced the entire battery with a few 76-mm shells. . . .
Since the 50-mm antitank guns had failed to pierce the 3-inch armor, it was decided that only the 88-mm flak gun with its armor-piercing shells would be effective. . . . Well camouflaged with branches and concealed by the burned-out German tanks lining the road, the gun safely reached the edge of the forest and stopped 900 yards from the tank.
Just as the German crew was maneuvering the gun into position, the tank swung its turret and fired, blasting the flak gun into a ditch. Every round scored a direct hit, and the gun crew suffered heavy casualties.
[The next morning, following an unsuccessful attack by combat engineers], a feint frontal attack was to be executed by a tank formation. . . while another 88-mm gun was to be brought up. . . .
The German armor deployed and attacked. . . while the 88-mm gun took up a position to the rear of the [Russian] tank. The very first round was a direct hit and, as the crew tried to turn the gun to the rear, a second and third shell struck home. Mortally wounded, the tank remained motionless, but did not burn. Four more 88-mm armor-piercing shells hit their mark.
The Germans closest to the tank . . . found that but two of the 88-mm shells had pierced the tank armor, the five others having made only deep dents. Eight blue marks, made by direct hits of the 50-mm antitank guns, were found. . . . No trace of the fire from the German tanks [conducting the feint] could be found. . . . Suddenly, the gun barrel started to move again, and most of the Germans scattered. Quickly, two engineers dropped hand grenades through the hole made by the hit on the lower part of the turret. A dull explosion followed, and the turret cover blew off. Inside were the mutilated bodies of the crew.1
Account from elements of the 6th Panzer Division, Army Group North, in the opening days of Operation Barbarossa.

http://img339.imageshack.us/my.php?image=hollow2oz.png

This data may be of interest. Note the 'beaten-zone for a 2m high target.

This data also shows, for larger sized hollow charge weapons, that they were not as innacurate as many assume. Smaller weapons like 75mm IG or L24 were (after 600m or so) but the US 105mm and this weapon were not.

The beaten zone concept shows the forgiveness in range estimation in my opinion.

While the 75mmL70 is not in the game, it is interesting to look at its accuracy

http://www.geocities.com/Area51/Capsule ... stics.html

--------------------------------------------------------------------------------
Name Caliber/Length Projectile Maingun Accuracy (%)
Type Name Weight(kg) Muzzle Velocity (m/s) Test Type Range (m)
500 1000 1500 2000 2500 3000
7,5cm KwK 42 75mm / L70 APCBC PzGr 39/42 6,8 935
Tests 100 100 100 92 73 55
Practice 100 97 72 49 29 18

APCR PzGr 40/42 4,75 1120
Tests 100 94 76 56
Practice 100 56 32 20
HE SprGr 42 5,74 700
Tests 100 100 90 66 42 28
Practice 100 97 44 24 13 8

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

The 7,5 cm KwK42 L/70 was a very accurate gun capable of first round hits at ranges exceeding 1000 meters. The estimated accuracy is given as the probability (in percentage) of hitting a target 2 meters high and 2.5 meters wide, representing the target presented by the front of an opposing tank. These accuracy tables are based on the assumptions that the actual range to the target has been correctly determined and that the distribution of hits is centered on the aiming point. The Test data show the accuracy obtained during controlled test firing of the gun to determine the pattern of dispersion. The Practice data include the variation expected during practice firing due to differences between guns, ammunition, and gunners. Both data were reported in the accuracy tables from an original report on the 7,5 cm KwK42 L/70 as shown in the Accuracy table. These accuracy tables do not reflect the actual probability of hitting a target under battlefield conditions. Due to errors in estimating the range and many other factors, the probability of a first round hit was much lower than shown in these tables. However, the average, calm gunner, after sensing the tracer from the first round, could achieve the accuracy shown as the Practice data.