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

[:USERNAME:]

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

I am merely showing that, with similar sized targets, the 88 has a better chance to hit than a 75 at long range. The chance isn't a whole lot better, but it is there. If you want to test something else be my guest. Show us your data and we can all examine it.

BTW, my previous post had a Firefly's chance to hit a Mark IV - compare the Firefly's chance to hit a Mark IV with a Tiger's chance to hit a Sherman if you want to find an optics 'bonus'. I suspect that the Firefly's chance to hit is similar to the Tiger's. <HR></BLOCKQUOTE>

Thanks. If you look up a few posts you will find that I compared already.

So the 88mmL56 is more accurate than a sherman 75mmL40 (or whatever it is). I agree with you and your findings. That is, the advantage is small.

Lewis

PS Perhaps the tenseness was just my subjective take. I dont find your post defensive in the least either.

[PzKpfw 1]

Length, width, & height comparison:

M4 Sherman:

Length - 5.89m

Width - 2.62m

Height - 2.74m

M4A3 Sherman 76mm

Length - 7.54m

Width - 3.00m

Height - 2.97m

Sherman VC:

Length - 7.77m

Width - 2.67m

Height - 2.74m

PzKpfw IV H - J:

Length - 7.02m

Width - 2.88m

Height - 2.68

Panther Ausf G:

Length - 8.86m

Width - 3.42m

Height - 2.98m

Tiger E:

Length - 8.45m

Width - 3.70m

Height - 3.00m

Tiger B:

Length - 10.29

Width - 3.76m

Height - 3.09m

Regards, John Waters

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

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

[Ron]

In the same vein as ASL's previous test I checked the following also. BTW flat open is the same as pavement when checking the %hit chance.

Firer Target 500m 1000m 1500m 2000m

M4A376 MkIV - 51% - 27% - 13% - 5%

Firefly MkIV - 51% - 30% - 17% - 9%

Tiger M4A1 - 51% - 27% - 14% - 7%

Not a lot to choose from.

Ron

[:USERNAME:]

http://www.onwar.com/tanks

heres a good site for tank data

[:USERNAME:]

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

In the same vein as ASL's previous test I checked the following also. BTW flat open is the same as pavement when checking the %hit chance.

Firer Target 500m 1000m 1500m 2000m

M4A376 MkIV - 51% - 27% - 13% - 5%

Firefly MkIV - 51% - 30% - 17% - 9%

Tiger M4A1 - 51% - 27% - 14% - 7%

Not a lot to choose from.

Ron<HR></BLOCKQUOTE>

This then contradicts the arguments that there is an advantage. The firefly has an advantage be it ever so slight.

Also see my post where I compared firefly to MkIV.

Lewis

[ASL Veteran]

Any advantage that is given must be primarily based upon the differences in the muzzle velocities in the various guns. This is consistent with what BTS has already described as being in the game. Now all we need is some real world data for some vehicles at firing ranges so we can compare them.

[:USERNAME:]

http://www.onwar.com/tanks/germany/profiles/ppz4f2.htm ]http://www.onwar.com/tanks/germany/profiles/ppz4f2.htm

[img=http://www.onwar.com/tanks/usa/profiles/pm4a3.htm]http://www.onwar.com/tanks/usa/profiles/pm4a3.htm

Heres some eye candy. Compare these two vehicles visually and then look up the heights and widths. Then please go back and carefully compare these two vehicles looking at the areas over the tracks and the width of the turrets, etc.

Any comments?

Lewis

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

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

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

[:USERNAME:]

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

Any advantage that is given must be primarily based upon the differences in the muzzle velocities in the various guns. This is consistent with what BTS has already described as being in the game. Now all we need is some real world data for some vehicles at firing ranges so we can compare them.<HR></BLOCKQUOTE>

I agree with you. Primarily is the key word.

But I disagree with silhouttes. Its key and we DO have alot of data on dimensions, etc. What good is all the firing data in the world if the silhouttes and the formula using them are incorrect?

Please see my post and comment on the PzIV and Sherman.

Lewis

[Guest Big Time Software]

OK, I have just read through the last 2 pages so I hope I address all the points raised...

Lewis, I did not comment on your size/% hit chance question because I lack the math skills necessary to answer it. I think one would also need to know the degree of missed shots outside the 1x1 or 2x2 area. On the face of it, assuming that both guns would miss proportional to the degree they can hit, I would take the 1x1 at 15%. A near miss would still likely be a hit since a tank is much larger than 1x1.

Also for Lewis, I do not have an answer for you about the HxWxL question. I know why you are asking about it and am unsure how it is calculated internally (and Charles is not around this weekend to ask). If the silhoutte is abstracted to be some sort of product of the three, this will yield a different value than if the silhoutte were actually "drawn" out as a 3D "box". I'll get back to you on this one.

An important note that has yet to be made...

When Allied tankers, commanders, historians compare the long range advantage of the average German tank versus the average Allied one they are MOST likely *NOT* comparing the 76mm gun and certainly not the 17lbr. As the war went on into 1945, perhaps, but definately not during the summer and fall battles. The would, instead, be comparing the short 75 found on the average US/Commonwealth tank for the first part of the war in NW Europe. Yet nobody is using this vehicle for comparison and instead using the less common ones. So I did...

At 1500m I simply took Regular 5 M4A1 (75) and 5 PzIVh and let them face off. I had each target a unique tank. Results were...

First Volley:

3 Shermans hit (2 brewed up)

0 PzIVs hit

Second Volley (now with more guns on the Shermans):

1 Sherman hit (brewed up)

0 PzIVs hit

Third Volley (5:1 this time)

1 Sherman hit (brewed up)

0 PzIVs hit

This is a more or less realistic test of two platoons meeting each other in combat. Within 40 seconds all Shermans were dead, the PzIVs not even phased.

So... how can one argue that the Germans don't have an advantage here?

Steve

[ASL Veteran]

I have located some accuracy data from Jentz’s book “Tank Combat in North Africa: the opening rounds”. In it he says the following on pages 57 and 59: “The estimate of accuracy is given by the probability 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 target. The first number shows the accuracy in percentage that was obtained during controlled test firing of the gun to determine the pattern of dispersion. The second number in parentheses was calculated by doubling the dispersion obtained from controlled test firing. Both the British and Germans considered that ‘doubled dispersion’ was a close approximation of the accuracy obtained by the troops in practice and, if they remained calm, in combat. All of these accuracy values were obtained from firing tables published by the respective armies during the war. The British calculated their dispersion based on a 90% zone and the Germans and Italians calculated the dispersion based on a 50% zone. The 90% zone from the British firing tables was used as the basis for calculating the percent accuracy against a 2.5m by 2 meter target so that it could be directly compared with the other nations guns.”

Okay, here is the table that Jentz created – it is located on page 58. I am only using the number located in parentheses – the one including the dispersion.

Weap mvel ammo 100m 500m 1000 1500 2000 2500 3000

2pdr 792m xxAP 100% x67% x26% x12% xxxx xxxx xxxx

25pr 472m xxAP 100% x66% x46% x28% xxxx xxxx xxxx

20mm 780m Pzgr 100% x87% x37% xxxx xxxx xxxx xxxx

37mm 758m Pzgr 100% x95% x47% x15% xxxx xxxx xxxx

47mm 775m Pzgr 100% 100% x89% x59% xxxx xxxx xxxx

50mm 685m Pzgr 100% 100% x96% x71% xxxx xxxx xxxx

50mm 835m Pzgr 100% 100% x95% x68% xxxx xxxx xxxx

75mm 385m Kgrp 100% 100% x73% x38% xxxx xxxx xxxx

88mm 810m Pzgr 100% x98% x64% x38% x23% x15% x10%

105m 395m Pzgr 100% x98% x63% x32% xxxx xxxx xxxx

20mm 840m Ital 100% x87% x37% xxxx xxxx xxxx xxxx

47mm 630m Md35 100% x95% x46% x17% xxxx xxxx xxxx

47mm 630m Md39 100% x95% x52% xxxx xxxx xxxx xxxx

The first two weapons are British and the last three are Italian. I had to keep all columns four places long so the table would work on the board. The only thing I would comment on is that the accuracy of the 88 Flak doesn’t seem all that fabulous here. The other thing is that muzzle velocity doesn’t seem to make all that big of a difference to accuracy.

[:USERNAME:]

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

Lewis, I did not comment on your size/% hit chance question because I lack the math skills necessary to answer it. I think one would also need to know the degree of missed shots outside the 1x1 or 2x2 area. On the face of it, assuming that both guns would miss proportional to the degree they can hit, I would take the 1x1 at 15%. A near miss would still likely be a hit since a tank is much larger than 1x1.

Also for Lewis, I do not have an answer for you about the HxWxL question. I know why you are asking about it and am unsure how it is calculated internally (and Charles is not around this weekend to ask). If the silhoutte is abstracted to be some sort of product of the three, this will yield a different value than if the silhoutte were actually "drawn" out as a 3D "box". I'll get back to you on this one.

<HR></BLOCKQUOTE>

Syeve

Thanks for the feedback.

Lets hear from Charles then. Enjoy your weekend.

Lewis

[Guest Big Time Software]

Did a few more repeats of the above test. All results were after roughly 20-40 seconds of firing:

Test 2

4 Shermans KOd, one immobilized and hiding behind smoke. 1 PzIV KO'd, another hit but no damage.

Test 3

4 Shermans KOd, one hiding behind smoke. 1 hit on a PzIV caused it to button up, no damage.

Test 4

4 Shermans KOd, one hiding behind smoke. 2 hits on the same PzIV caused it to button up, no damage.

Test 5

3 Shermans KOd, 2 hiding behind smoke. 2 PzIVs KOd by turret penetrations, 1 immobilized, and one other hit that caused no damage.

Test 6

4 Shermans KOd, one hiding behind smoke. 2 PzIVs KOd by turret penetrations, one other hit that caused no damage.

Test 7

5 Shermans KOd. No PzIVs KOd, 3 hits that caused no damage.

Test 8

4 Shermans KOd, one hiding behind smoke. No PzIVs KOd, no hits of any sort.

Test 9

4 Shermans KOd, one hiding behind smoke. One PzIVs KOd, one hit no damage.

Test 10

4 Shermans KOd, one hiding behind smoke, one of the shots first caused an immobilization. One PzIVs KOd, no other hits.

Shermans - 36 out of 50 KOd (72%) and the others "fled"

PzIVs - 7 out of 50 KOd (14%) and the others "won"

Interesting is to note the hits registered. 94% of the Shermans were killed after only one hit, while almost half the hits on the PzIVs did no damage.

Conclusion...

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

When comparing the average tanks of the day, which is what the veteran's and broad historical statements are based on, the Shermans were dog meat at 1500m range. Also, if the PzIVs scored a hit it was almost certainly a kill even against frontal armor. While the Shermans had a much lower chance of hitting and only about a 50/50 chance of causing any damage.

So, explain to me again how the historical advantage of the Germans at long ranges isn't simulated in CM?

Steve

[ASL Veteran]

I also want to add that Jentz wrote "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 gunner could achieve the accuracy shown by the number in parentheses after adjusting his fire onto the center of the target - if he remained calm."

Unfortunately, this accuracy data is from 1941 and has the 2 pounder and the 25 pounder for British guns so we can't really do any direct comparisons to the game. I just thought it was interesting data to share.

[Guest Big Time Software]

Did a few more repeats of the above test. All results were after roughly 20-40 seconds of firing:

Test 2

4 Shermans KOd, one immobilized and hiding behind smoke. 1 PzIV KO'd, another hit but no damage.

Test 3

4 Shermans KOd, one hiding behind smoke. 1 hit on a PzIV caused it to button up, no damage.

Test 4

4 Shermans KOd, one hiding behind smoke. 2 hits on the same PzIV caused it to button up, no damage.

Test 5

3 Shermans KOd, 2 hiding behind smoke. 2 PzIVs KOd by turret penetrations, 1 immobilized, and one other hit that caused no damage.

Test 6

4 Shermans KOd, one hiding behind smoke. 2 PzIVs KOd by turret penetrations, one other hit that caused no damage.

Test 7

5 Shermans KOd. No PzIVs KOd, 3 hits that caused no damage.

Test 8

4 Shermans KOd, one hiding behind smoke. No PzIVs KOd, no hits of any sort.

Test 9

4 Shermans KOd, one hiding behind smoke. One PzIVs KOd, one hit no damage.

Test 10

4 Shermans KOd, one hiding behind smoke, one of the shots first caused an immobilization. One PzIVs KOd, no other hits.

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

Results...

Shermans - 36 out of 50 KOd (72%) and the others "fled"

PzIVs - 7 out of 50 KOd (14%) and the others "won"

Interesting is to note the hits registered. 94% of the Shermans were killed after only one hit, while almost half the hits on the PzIVs did no damage.

Although I did not count how many 1st shot hits were registered, it was a huge number. But putting that aside, the most any PzIV fired at the Shermans was 4 rounds. So all the destruction noted above was done in 1-4 shots.

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

Conclusion...

When comparing the average tanks of the day, which is what the veteran's and broad historical statements are based on, the Shermans were dog meat at 1500m range. Also, if the PzIVs scored a hit it was almost certainly a kill even against frontal armor. While the Shermans had a much lower chance of hitting and only about a 50/50 chance of causing any damage.

The number of shots per hit for the Germans seems to be very realistic. A large number of 1 shot one kill, but the rest resulting in kills after only 2-3 more shots.

So, explain to me again how the historical advantage of the Germans at long ranges isn't simulated in CM?

Steve

[PzKpfw 1]

Speaking of 76mm Sherman's & a general comment as I haven't started testing this yet.

I was playin a QB today on flat open terrain where max engagement range was 1003m. I had 5 PzKpfw IVJ Vet, vs 5 M4A1 76mm Sherman Vet.

The initial exchange ran from 760m to 961m 4 PzKpfw IV J's were brewed up in 1st round hits, while takeing out 2 M4A1's in return & the surviving PzKpfw IV took out 2.

The last PzKpfw IV & M4A1 duked it out at 961m head on faceing each other, with the IV fireing 10 shots in 2 turns and in those 10 shots the % to hit never went up, it stayed at 29%, the M4A1 fired 8 - 10 as well. The PxzKpfw IV got a TF penetration on the 11th shot.

I would have expected with each shot the % to hit would have risen with the gunners makeing corrections, the exchange went on far to long for vet crews IMHO.

As to the 76mm & 17lb they certinly helped level the feild, but as I posted before neither was in quantity during the initial stages of the fighting, Steve is correct in that the anecdotal experience comments are generaly concerning the 75mm armed Shermans.

But you can find comments from 76mm armed Sherman & TD crew in the report so often quoted here concerning it it was an improvement, but comments concerning its usefulness, against the Panther & Tiger it gets as derided as the 75mm. But again the comments have nothing to do with accuracy they concern the inability of the 75 & 76mm to defeat German armor.

The 17lb was a very potent gun on par with the German KwK.42 etc in terms of MV & trajectory, has anyone found any reports with the 17lb etc, similar to the US ones discussed here?, I have seen British tankers deride the 75mm vs German armor but very little concerning the 17lb & it's performance.

Regards, John Waters

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

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

[This message has been edited by PzKpfw 1 (edited 10-08-2000).]

[Guest Big Time Software]

<BLOCKQUOTE>quote:</font><HR>I would have expected with each shot the % to hit would have risen with the gunners makeing corrections<HR></BLOCKQUOTE>

It most certainly does, but as I have stated a couple of times already, I am very sure the displayed hit % does NOT factor in everything. That value is there ONLY as a guide to know ROUGHLY how likely it is to hit the target. It is not designed to be exact as that would be too much info.

<BLOCKQUOTE>quote:</font><HR>the exchange went on far to long for vet crews IMHO.<HR></BLOCKQUOTE>

Hey, stuff happens In my example Regular PzIVs were hitting Shermans at 1500m within 4 shots. So either something on the battlefield is causing it to be a more difficult shot than your example implies, or you just got unlucky. Even Veterans can have a bad day

The only solution to (realistic) outliers like this is to have some sort of magical cut off, drawn arbitrarily, that would say "after x number of shots the to-hit chance is 100%". Anything short of that will allow examples like what you descibed to happen. Obviously, we are not in favor of this since there is no rational basis for it.

Steve

[This message has been edited by Big Time Software (edited 10-08-2000).]

[PzKpfw 1]

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

Hey, stuff happens In my example Regular PzIVs were hitting Shermans at 1500m within 4 shots. So either something on the battlefield is causing it to be a more difficult shot than your example implies, or you just got unlucky. Even Veterans can have a bad day

<HR></BLOCKQUOTE>

Thx Steve was just curious why the pk never went up assuming the gunner was makeing adjustments. The only onstruction was a bush . Ahh could you explain what occurs to us so we understand the process better on what occurs in these lil duals with the tac ai & gunnery.

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

The only solution to (realistic) outliers like this is to have some sort of magical cut off, drawn arbitrarily, that would say "after x number of shots the to-hit chance is 100%". Anything short of that will allow examples like what you descibed to happen. Obviously, we are not in favor of this since there is no rational basis for it.

Steve

<HR></BLOCKQUOTE>

Ack don't want that either, just trying to get a handle on whats occuring how it all works.

Regards, John Waters

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

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

[Guest Big Time Software]

John wrote:

<BLOCKQUOTE>quote:</font><HR>Ahh could you explain what occurs to us so we understand the process better on what occurs in these lil duals with the tac ai & gunnery. <HR></BLOCKQUOTE>

I don't know what the code says, but it is pretty much everything you would expect. Range, weather conditions, hull down or not, silhouette, crew experience, gun, speed of target, speed of own vehicle, terrain that each is on, etc. etc.

In all of this there is a degree of variability. So there is no "perfect" set of circumstances that will allow a given situation to always produce the same results. There is any number of real world reasons why the Vets in your scenario might have continued to miss. One simple one is that they were tyring too hard

I mean, have you ever been in the middle of something, and even though you think you "almost have it" an hour later you still don't? Then you come back from supper or something and spot the problem and make the correction perfectly within seconds. While we do not pretend to simulate stuff like this SPECIFICALLY, the oddball results that come up could be explained by such reasoning.

The key here is... overall, do the battle results appear to match the overall battle results noted by veterans and historical accounts? And as for outliers, who knows. Could be divine intervention A real life example...

A friend of mine, who has hunted all his life, was a game warden for 20 years, and makes a living off of "guiding" hunters told me a story about the biggest buck white tailed deer he had ever seen (happened years ago on our property in fact ). The buck was standing in a bog only 400ft away. It was overcast but there was one ray of light right on the poor creature. He took his time, aimed, and fired. Aimed and fired. Did this two more times and yet he neither hit nor did the buck move. After the 4th shot my friend, still absolutely shocked that he had missed, put down his gun and cracked open his thermus for a cup of coffee. He sat there for 20 minutes until the buck had walked back into the woods. As my friend said, "If I could fire four shots, in those conditions, and miss each one, it clearly was not that buck's day to die"

As for rating, I would have rated my friend an "Elite" or "Crack" hunter

Steve

[ASL Veteran]

Okay, while it is not possible to test the accuracy of Allied tanks vs German tanks from 1944 until some data arrives, it is possible to test the German guns against themselves. I set up a firing range with the 50mm PAK 38, the 88 Flak, and the Lynx at the end of the paved lanes. I placed Shermans at 100m, 500m, 1000m, 1500m, and 2000m and read off the accuracy that the targeting lines displayed. I then compared these readings to the real life accuracy data for these guns as given by Jentz in “Tank Combat in North Africa”. Here is a raw comparison:

Combat Mission

Weap 100m 500m 1000 1500 2000

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

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

20mm x99% x87% x56% x27% xx9%

Jentz

Weap 100m 500m 1000 1500 2000

50mm 100% 100% x95% x68% xxxx

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

20mm 100% x87% x37% xxxx xxxx

I am going to make some comments on this data and see what you guys think. First, the target used by Jentz’s data is 2.5m x 2m. This is approximately the size of a Sherman, but not exact – the Sherman is a bit bigger. The first thing that I notice is that the drop off CM gives these weapons at 500m seems to be overstated. Out beyond the range of 500 meters the 88 seems to be about right in CM if everything was moved over one column, but the 50mm PAK is really underrated out to 1500 meters while maybe it is overrated beyond that range. The 20mm on the Lynx in Jentz almost matches CM exactly, except that CM doesn’t model the drop off at 1000 meters and is probably overstated beyond 1000 meters. If we apply the 100m accuracy in CM as a base, then we could subtract 13% from each gun at each range to arrive at a modified ‘CM’ accuracy. Perhaps the 20mm should just use the Jentz figures since they seem to match CM so closely.

Modified CM accuracy

Weap 100m 500m 1000 1500 2000

50mm x87% x87% x82% x55% xx5%

88mm x87% x85% x51% x25% x10%

20mm x99% x87% x37% xx5% xx1%

As you may note, the accuracy of the 88 Flak using this method is almost the same as it is in CM now. You just have to move the CM accuracy one column over to the right out to 1500m. Since the Sherman’s dimensions are actually a little bigger than 2.5m x 2m one could argue that the modified CM accuracy should be even higher, but these numbers seem reasonable as second or third shot probabilities. Of course, these are just preliminary findings and much more data would need to be sifted before coming up with any type of accuracy ratings for each gun.

[:USERNAME:]

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

As for rating, I would have rated my friend an "Elite" or "Crack" hunter

Steve<HR></BLOCKQUOTE>

I would rate the deer lucky or very stealthy. If your friend was using a scoped rifle and could miss at that range against a still target and there was nothing wrong with the sights well, he just could have been tricked by an illusion. I suspect the sights also. They might have been maladjusted somehow.

I have fired high powered hunting rifles and you do not percieve the bullets strike. That is, you have to check the target to see where the bullet went because the immediate kick is tremendous and you are kicked back. Its not exactly analogous to firing a tracered tank round. The mass of the vehicle and the recoil mechanism takes the kick. I bet everyone on earth would at least blink. But a tanker can observe the error. A rifleman cant without a buddy with binos.

Also, the light thing is tricky. I remember being in the service in Hawaii and walking through a dense jungle and coming across an opening into a field. The field was in direct sunlight and everything was crystal clear (no haze, no wind, no motion in the grass). I saw some movement and being far ahead of my company, thought it was the enemy. It was and I thought they were only a few dozen yards away!! But its an illusion because I percieved them with such clarity that I thought they were closer. They were 150 yards away!!! I was ducking and there was no chance in hell they could see me if I jumped up and down! The dark US cammies make you almost invisable in shadows.

Lewis

PS check your weasal for bullet holes. That guy might also need glasses.

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

[PzKpfw 1]

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

John wrote:

I don't know what the code says, but it is pretty much everything you would expect. Range, weather conditions, hull down or not, silhouette, crew experience, gun, speed of target, speed of own vehicle, terrain that each is on, etc. etc.

In all of this there is a degree of variability. So there is no "perfect" set of circumstances that will allow a given situation to always produce the same results. There is any number of real world reasons why the Vets in your scenario might have continued to miss. One simple one is that they were tyring too hard

The key here is... overall, do the battle results appear to match the overall battle results noted by veterans and historical accounts? And as for outliers, who knows. Could be divine intervention A real life example...

A friend of mine, who has hunted all his life, was a game warden for 20 years, and makes a living off of "guiding" hunters told me a story about the biggest buck white tailed deer he had ever seen (happened years ago on our property in fact ). The buck was standing in a bog only 400ft away. It was overcast but there was one ray of light right on the poor creature. He took his time, aimed, and fired. Aimed and fired. Did this two more times and yet he neither hit nor did the buck move. After the 4th shot my friend, still absolutely shocked that he had missed, put down his gun and cracked open his thermus for a cup of coffee. He sat there for 20 minutes until the buck had walked back into the woods. As my friend said, "If I could fire four shots, in those conditions, and miss each one, it clearly was not that buck's day to die"

As for rating, I would have rated my friend an "Elite" or "Crack" hunter

Steve<HR></BLOCKQUOTE>

Thx Steve, question is odball reasoning and the other RL examples you used such as trying to hard modeled in CM? I agree in RL things can go totaly off the wall, but I didn't realise you had coded all these variable factors into the men in CM.

Heh, when I was young I worked at an Nature Center in NY, I saw an ancient 18pt buck one day while eating my lunch on a game trail he was around 50ft away from me with 5 doe, he never spooked, even aproached closer, my boss told me the buck had been seen since he took over running the center, 10yrs earlier, that was the largest buck I ever saw & I hunted deer 15 yrs.

Regards, John Waters

[Guest Big Time Software]

ASL Vet wrote:

<BLOCKQUOTE>quote:</font><HR>Perhaps the 20mm should just use the Jentz figures since they seem to match CM so closely.<HR></BLOCKQUOTE>

Couldn't do it even if we wanted to since CM doesn't use hardcoded values for any weapon at any range. All done through equations.

Keep in mind that the figures Jentz uses are... test range figures For example, there are figures in there that are 100%. Sorry, but nothing on the battlefield has those sorts of odds. So as has been stated before, by myself and others, test range data can not be used without a heavy dose of salt.

One thing to keep in mind when playing around with test range data is that the margin for error becomes greater the further away the target is. The controlled test tries to neutralize these variables as much as possible to find the "true" or "inherent" ability of the weapon. However, the crews in the front lines have no such luxury. So the margin for error (i.e. a miss) increases in a non linear way the further out the target is from the shooter.

Think of it like a pistol. If you are at a target range and take great care you can probably hit a target at 50m quite regullarly and in a tight group (assuming you are a decent shot). Now... try firing at Lethal Weapon (or insert favorite cop movie here ) rate of fire or in a non ideal environment. You might be lucky enough to just hit the paper. But take the same target, and the same tests, and put it 10m in front of you... the results will be much more favorable. Just more food for thought.

Lewis, it is possible that his sights were off. But then again, one would expect that he checked this sort of thing before going off into the woods. For all we know he might have had too much coffee BEFORE shooting which threw off his aim by just a hair. But by the time that round go out to the buck, the shot went wild by dozens of CMs.

Your point about spotting the impact of the round is of course valid. However, if you overshoot, and the round impacts behind an obstuction, you might very well not see the impact. I got to watch a M3 Stuart and a 57mm AT gun fire AP rounds at about 200-300m. Even though there was a backstop made out of dry and dusty dirt, the visual clues from the hit were not all that great. At 1000, 2000, or 3000m I could see this being a huge problem. What I am saying here is that observing a missed round is not a sure thing even with a second set of eyeballs looking for it.

John wrote:

<BLOCKQUOTE>quote:</font><HR> I agree in RL things can go totaly off the wall, but I didn't realise you had coded all these variable factors into the men in CM.<HR></BLOCKQUOTE>

They are not coded into CM, but rather we dumbed down the test range, pefect condition clinical results in order to abstractly account for them. There are hundreds of little things that could affect accuracy in a given firing, so there is *no* way we can simulate even a fraction of them. Things like a mathematical miscalculation, not having enough sleep, thinking about a loved one who was just killed, a momentary glint of sunlight, misjudging target speed or direction, etc., etc.

As was discussed back in the optics thread, abstractions are necessary because it is impossible to code every single thing that goes on inside a human mind, machine, teamwork, physics, etc. The difference in realism between different abstractions of the same element in war come down how well the abstraction is combined with finite rules of reality.

CM has, for example, far more detailed physics and ballistics simulation than other wargames do, and probably more detailed than the vast and overwhelming majority of tank sims. However, it does not simulate everything. Since no other game simulates everything either, and does a less accurate job with physics and ballistics, CM is (in theory) a better abstraction on the whole. One of the reasons is that all of the hundreds of things that we can't simulated are not rulled out by using the precise nature of the physics and ballistics models unchecked by outside factors. In a board game you would roll dice to get these outside factors. In CM, we tweaked down probability in some cases and give more weight to parameter variables (like Experience, Weather, etc.) in others. Overall this makes CM more realistic.

So although we can't account for the minute physical and mental characteristics of the gunner, we can allow a gun crew to either perform above, at, or below average on a situation by situation basis. Luck plays a HUGE part in real warfare, and so it should obviously play a significant role within CM. The trick is to make sure the average results are in line with the average historical results and that the outliers are few in either direction (beneficial or harmfull). And when there are outliers to make sure they are in some way plausible. A Veteran gunner missing a stationary target at 100m, while remaining still, 40 times in a row points to a problem with the system, but 3 times is not totally out of the question.

Steve

[:USERNAME:]

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

CM has, for example, far more detailed physics and ballistics simulation than other wargames do, and probably more detailed than the vast and overwhelming majority of tank sims. However, it does not simulate everything. Since no other game simulates everything either, and does a less accurate job with physics and ballistics, CM is (in theory) a better abstraction on the whole. One of the reasons is that all of the hundreds of things that we can't simulated are not rulled out by using the precise nature of the physics and ballistics models unchecked by outside factors. In a board game you would roll dice to get these outside factors. In CM, we tweaked down probability in some cases and give more weight to parameter variables (like Experience, Weather, etc.) in others. Overall this makes CM more realistic.

Steve<HR></BLOCKQUOTE>

Steve

I hope you see my point that vehicle size is certainly documented and that CM should model this as closely as possible as it relates to spotting (another thread perhaps), target size and hit distribution, etc.

Its apparent to me that no other game has the input from customers that CM has. I think your models should work around what is best known first and the abstract around that. You model armor penetration with very little abstraction and I hope the spotting, size gets the same consideration. Accuracy, repeatability, etc is then the abstracted variable in my thinking.

Lewis

[ASL Veteran]

Hmmm, reply with quote doesn't seem to be working properly. Anyway, while there are a lot of factors that alter accuracy on the battlefield, I don't think that the inherent accuracy of the weapon itself can be completely ignored. Since all those battlefield factors are going to be affecting all the gunners equally, these various effects can be abstracted and applied to each weapon equally. In that case, the only difference between the guns is the inherent differences between the guns themselves. If the individual characteristics of each gun is not included, then you have a situation where each gun has an equal chance to hit at a given range. This is essentially the case in Combat Mission.

However, the firing range data shows that certain guns are inherently more accurate than others. The test range indicates that the 50mm PAK is much more accurate than the 88 Flak at 1000 meters and one could presume that - all other factors being equal - the 50mm PAK is a more accurate weapon and should have a greater chance of hitting than an 88 at that range.

When you start comparing the 2 Pounder to the 50mm this starts to become a serious problem. The 2 Pounder really drops off after 500 meters while the 50mm (both of them) remains significantly more accurate. If the inherent accuracy of the 50mm is not included in the equation somewhere, then the British would get an artificial advantage in CM 3 Boy, will we see the German Optics crowd scream then!

[Guest Big Time Software]

Lewis, couldn't agree with you more. To the extent that the CPU can handle the known stuff, it should be incorporated as is. Things like measurements of a vehicle are one of the few things that is documented in a consistant way If Charles is abstracting this I am sure it is for CPU reasons, which means if it can be improved in the near future.

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

ASL Vet wrote:

In that case, the only difference between the guns is the inherent differences between the guns themselves.<HR></BLOCKQUOTE>

You are correct here. CM does simulate the inherent qualities of the guns, which is then reduced to simulate realistic reduction in accuracy on the battlefield.

<BLOCKQUOTE>quote:</font><HR>If the individual characteristics of each gun is not included, then you have a situation where each gun has an equal chance to hit at a given range. This is essentially the case in Combat Mission.<HR></BLOCKQUOTE>

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

Since the abstracted battlefield limitations are evenly applied (in regards to accuracy), and modifying each weapon's "best case" inherent capabilities, you STILL get weapons that are inherently different from one another. However, all are consistantly, and proportionally, less accurate than they are in theory.

<BLOCKQUOTE>quote:</font><HR>If the inherent accuracy of the 50mm is not included in the equation somewhere<HR></BLOCKQUOTE>

It is. See above and previous 2 posts.

Steve

[This message has been edited by Big Time Software (edited 10-08-2000).]