accuracy/efficiency of machine gun fire

[pnzrldr]

Well, that is worth a chuckle. 7.62mm MGs are ridiculously easy to shoot fairly well. Our BN trains USAF and USN guys to shoot them. Previously untrained doctors, lawyers and log guys can usually manage to knock over targets out to 600m without undue difficulty using a few bursts to get oriented. A trained machinegunner would have great difficulty missing a man sized target out to 1000m with 3 or 4 bursts to get dialed in.

However, the game assumes you are being shot at and that conditions are less than a nice bowling alley range. I would presume that some aiming issues are introduced to enhance gameplay. Well trained machinegunners are lethal enough that one or two well emplaced ones who cannot easily be suppressed might make a scenario unwinnable without armor.

In juxtaposition, I've also noticed in CMFI that 60mm mortars are rather startlingly accurate. While as the attacking Ami's I appreciate the ability to diabolically land 12 consecutive bombs in the immediate vicinity of the appallingly non-camouflaged Pak, I cannot imagine anyone getting quite that performance out of that weapon. I have seen some guys who are pretty good, but....

I suspect that MG performance, like Pak camoflage, are gameplay tradeoffs. If your butt is truly on the line, you simply won't go there, but we want to play a scenario in an hour or two, so the designers introduce a few balancing workarounds that are close enough to real world to be absorbed into the full montage of combat they are creating. I'm a latecomer to the series, and bought CMFI first. Downloading CMBN 2.0 as I type.

[Sublime]

I think you'll be more impressed with 2.0 but thats just my opinion. I'd grab the CW module too. Interesting to hear comments from someone who's handled these weapons in real life. The mortar accuracy has been debated endlessly, many agreeing many disagreeing. They still are very accurate. And we're trying to push for them to introduce a camo option for units. Part of your problem may be playing the AI which can never place units as well and logically as a human. ATGs are much more fearsome when a human is controlling them.

[Sgt Joch]

The problem with modeling MGs is that there is surprisingly little useful data on their effectiveness in combat.

Shooting range data is useful, but everyone shoots better on a range, the targets are stationary, the distance is known and more importantly, no one is shooting at you. In combat, in addition to the stress, you also have the problem that enemy troops will generally move from cover to cover and hide as soon as they come under fire.

Combat also covers many situations, from a HMG in a fixed position/bunker where the crew is familiar with their sector and has measured the distances to a HMG which sets up in an unfamiliar spot and then has to immediately fire over unfamiliar terrain, although you could use TRPs to differentiate the two.

Plus, when you do find useful data, it is often open to interpretation. For example, this diagram is from the U.S. Army's FM 7-7 which came out in 1985:

The M60 is interesting since it has the same basic design as the MG34/42.

note that:

-the MG has a 50/50 chance of hitting a stationary man sized target at 600 meters;

-however, this drops down to a 50/50 chance of hitting a moving target at 200 meters.

While this is, of course, very useful info, it leaves many questions unanswered. i.e.:

-what does 50/50 chance mean? 1 round? 50%?

-what is a moving target? presumably man-sized, but moving which way? presumably towards the HMG, but what happens if it is moving at an angle or perpendicular to the LOF?

-where does this data come from? shooting range? combat reports? a bit of both? Is it the Army's best guess of the weapon's effectivenes in combat or a conservative estimate?

[Michael Emrys]

-what does 50/50 chance mean? 1 round? 50%?

On the page it says a 6-9 round burst.

Is it the Army's best guess of the weapon's effectivenes in combat or a conservative estimate?

Probably an optimistic one to give the troops confidence.

Michael

[Sgt Joch]

On the page it says a 6-9 round burst.

If 1 burst has a 50% chance of hit, then two bursts should have 100% chance. Two 6-9 round bursts works out to an average 15 rounds.

so worst case, "chance of target hit" means at least 1 out of 15 rounds or a 6.67%+ chance of any one round hitting a moving target at 200 meters.

[stikkypixie]

If 1 burst has a 50% chance of hit, then two bursts should have 100% chance.

How'd you work that out?

[womble]

If 1 burst has a 50% chance of hit, then two bursts should have 100% chance.

Nope. 75% chance that one or both will hit. Still a 25% chance of no rounds hitting home. OR you could say that it's always 50%, because anyone that's sucked up a 6 round burst of 7.62mm isn't going to be a target for the second burst, as they'll have gone down. Even given that there's no such thing as overkill, the second burst at a target that's been hit will be at a "stationary, prone" rather than an upright target, whether stationary or moving. So, like all probabilities, it's not simply additive.

[Sgt Joch]

How'd you work that out?

if each burst has a 50% chance of hit, if you fire two, at least 50% of the two or one of the two should hit, so if you fire two, you should have 100% chance of at least one hit...seems logical anyway.

Of course with the laws of probability, it won't work out that cleanly, it's more to get a ballpark figure.

[Sgt Joch]

Nope. 75% chance that one or both will hit. Still a 25% chance of no rounds hitting home. OR you could say that it's always 50%, because anyone that's sucked up a 6 round burst of 7.62mm isn't going to be a target for the second burst, as they'll have gone down. Even given that there's no such thing as overkill, the second burst at a target that's been hit will be at a "stationary, prone" rather than an upright target, whether stationary or moving. So, like all probabilities, it's not simply additive.

now you lost me, so what would be the correct percentage of the chance of any one round hitting?

[c3k]

.5 miss + .5 miss means .25 miss total of both.

Two bursts, each with .5 chance of hit or miss. 4 outcomes: Miss/Miss, Hit/Miss, Miss/Hit, Hit/Hit.

So, two bursts means 75% chance of hit.

3 bursts would increase that to 87.5%, or 7/8ths. (The progression is more easily seen with fractions: 1/2, 3/4, 7/8, 15/16, etc. Double the denominator and make the numerator=denominator - 1. Only for this particular case.)

[Sgt Joch]

thanks, I always hated math in school.

so the probability of a hit is even worse than I originally thought.

[womble]

now you lost me, so what would be the correct percentage of the chance of any one round hitting?

I'd assume the percentages offered in the M60 document were for "one or more" bullets from any given burst. The chances for any one round would be a different data set entirely.

[Sublime]

thanks, I always hated math in school.

so the probability of a hit is even worse than I originally thought.

I still suck at math. And when I transferred to UMass last fall my math didnt transfer so now I'll have to take another placement exam. I don't retain any of it so I expect to place around the 5th grade level. Perhaps 7th or 8th if I'm having a particularly auspicious day.

[DavidFields]

Well, that is worth a chuckle. 7.62mm MGs are ridiculously easy to shoot fairly well. Our BN trains USAF and USN guys to shoot them. Previously untrained doctors, lawyers and log guys can usually manage to knock over targets out to 600m without undue difficulty using a few bursts to get oriented. A trained machinegunner would have great difficulty missing a man sized target out to 1000m with 3 or 4 bursts to get dialed in.

However, the game assumes you are being shot at and that conditions are less than a nice bowling alley range. I would presume that some aiming issues are introduced to enhance gameplay. Well trained machinegunners are lethal enough that one or two well emplaced ones who cannot easily be suppressed might make a scenario unwinnable without armor.

In juxtaposition, I've also noticed in CMFI that 60mm mortars are rather startlingly accurate. While as the attacking Ami's I appreciate the ability to diabolically land 12 consecutive bombs in the immediate vicinity of the appallingly non-camouflaged Pak, I cannot imagine anyone getting quite that performance out of that weapon. I have seen some guys who are pretty good, but....

I suspect that MG performance, like Pak camoflage, are gameplay tradeoffs. If your butt is truly on the line, you simply won't go there, but we want to play a scenario in an hour or two, so the designers introduce a few balancing workarounds that are close enough to real world to be absorbed into the full montage of combat they are creating. I'm a latecomer to the series, and bought CMFI first. Downloading CMBN 2.0 as I type.

I am glad someone in Modern Warfare notices the mortar issue--not just us WW2 types.

My suspect in ghe MG performance issue: the world is analog, but computer games are digital. In RL there are an infinite number of ray trajectories, but that is not true digitally, and that would be emphasized with distance.

Just a guess--someone can now pipe in about how ridiculous that idea is.

[DavidFields]

The problem with modeling MGs is that there is surprisingly little useful data on their effectiveness in combat.

Shooting range data is useful, but everyone shoots better on a range, the targets are stationary, the distance is known and more importantly, no one is shooting at you. In combat, in addition to the stress, you also have the problem that enemy troops will generally move from cover to cover and hide as soon as they come under fire.

Combat also covers many situations, from a HMG in a fixed position/bunker where the crew is familiar with their sector and has measured the distances to a HMG which sets up in an unfamiliar spot and then has to immediately fire over unfamiliar terrain, although you could use TRPs to differentiate the two.

Plus, when you do find useful data, it is often open to interpretation. For example, this diagram is from the U.S. Army's FM 7-7 which came out in 1985:

The M60 is interesting since it has the same basic design as the MG34/42.

note that:

-the MG has a 50/50 chance of hitting a stationary man sized target at 600 meters;

-however, this drops down to a 50/50 chance of hitting a moving target at 200 meters.

While this is, of course, very useful info, it leaves many questions unanswered. i.e.:

-what does 50/50 chance mean? 1 round? 50%?

-what is a moving target? presumably man-sized, but moving which way? presumably towards the HMG, but what happens if it is moving at an angle or perpendicular to the LOF?

-where does this data come from? shooting range? combat reports? a bit of both? Is it the Army's best guess of the weapon's effectivenes in combat or a conservative estimate?

Very nice info.

But another complicating issue (there are a gazillion) is that I believe the doctrinaire uses of an HMG is not just to kill, but to break units, or deny territory. Running toward an unsupressed HMG would be, I think, in general, thought to be ludicrous.

Thus, to ask what the kill ability of an HMG is, on a technical level, is like asking what would the kill level be of a squad running across a minefield. In reality, maybe some troops would get across. But no commander in WW2 in the Western Theater, other than Russian with the NKVD behind them, is going to order an open ground assault against an unsuppressed HMG. [someone is now going to post the 1:million exception]

There are even some cultural issues. A "hit" in the Western Theater generally, I think, meant out of effective action--this is not like fighting drugged up Phillipinos during the insurrection against the US earlier in the 20th century (this is not derogatory--I am just alluding to the issue of the need in that case for the development of 45 cal pistols)

Whatever tweaks need to be made to make that a CM2 reality also, "need" to be made. Maybe they already have been made. I certainly trust Battlefront will make them.

[it would seem there are only 2 threads in CM2. 1. HMG and mortar threads. 2. All the others]

[Sgt Joch]

I hope you guys don't get the impression nothing was changed in 2.01 regarding HMGs. I just wanted to point out it's not a black and white issue. I think most players will be very happy with 2.01.

[c3k]

Shhhhhh!!!!

[acrashb]

In RL there are an infinite number of ray trajectories, but that is not true digitally, and that would be emphasized with distance.

Technically correct, but perhaps not relevant in real terms.

If the game uses 32-bit floating point (the current default float) to represent a 3-dimensional coordinate over, say, a 10km square battlefield, the granularity is going to be about 1mm (asusing about 7 decimal places of precision) - as this is less than the diameter of any projectile, it should be good enough. With a 32-bit integer (allowing for a total of about 4,295,000,000 values on a given axis), the granularity is even better.

Even if my math is off by an order of magnitude, the point is sound: a modern computer can get enough granularity to represent which earlobe is shot off, let alone the general fall of shot of an MG burst. CPU considerations aside, of course, which is why we have action points to speed up some calculations.

[DavidFields]

Technically correct, but perhaps not relevant in real terms.

If the game uses 32-bit floating point (the current default float) to represent a 3-dimensional coordinate over, say, a 10km square battlefield, the granularity is going to be about 1mm (asusing about 7 decimal places of precision) - as this is less than the diameter of any projectile, it should be good enough. With a 32-bit integer (allowing for a total of about 4,295,000,000 values on a given axis), the granularity is even better.

Even if my math is off by an order of magnitude, the point is sound: a modern computer can get enough granularity to represent which earlobe is shot off, let alone the general fall of shot of an MG burst. CPU considerations aside, of course, which is why we have action points to speed up some calculations.

Nice info. Thank you.

[PzKfW]

2.01- much improved. Thank you.

[Sublime]

Yes very much so, thank you as well. +1

[herr_oberst]

.5 miss + .5 miss means .25 miss total of both.

Two bursts, each with .5 chance of hit or miss. 4 outcomes: Miss/Miss, Hit/Miss, Miss/Hit, Hit/Hit.

So, two bursts means 75% chance of hit.

3 bursts would increase that to 87.5%, or 7/8ths. (The progression is more easily seen with fractions: 1/2, 3/4, 7/8, 15/16, etc. Double the denominator and make the numerator=denominator - 1. Only for this particular case.)

Er, not to quibble BUT .5 miss * .5 miss = .25 both miss. Straight probability.

And yes, while adding bursts would change the fractions incrementally, for the ultra-grogs it STILL isn't straight math, as the bursts aren't (hopefully) "random independent trials". Any reasonable gunner would start "walking" his rounds, reducing his change for 'error' (i.e., a miss).

If we're not careful, we might actually distract Charles from the coding he's currently working on (whatever that is) by bringing up statements like 'the bayesian probability of an experienced gunner not under fire missing on his fourth burst is (hopefully) lower than his chance of missing on his first burst. Each successive burst would be aimed based the accumulated knowledge gleaned from prior bursts at that target. An inexperienced gunner would likely see a reduction in his error as well, but not to the degree that an experienced gunner would. After all, that experience should be worth something."

Gods curse you all, I've tried mightily to forget my tenure working in sadistics, er, statistics, and now you've brought it all back.

[noxnoctum]

So was this accuracy issue resolved with the 2.01 patch? (or 1.01 for FI)

[Killkess]

The fire seems to be more accurate in 2.01 and the ROF has been tuned up as well. I am not sure which of both makes the MGs stronger, but they are definitly much better than before. I cant say that i´ve seen successive correction from the gunner, but the higher rof with a better overall accuracy enables a HMG to keep 2 squads at bay quit reliably.

[Albe Pavo]

As far as you know, these changes will be added also to non 2.0 CMBN or to CMFI?