Suggestion for Infantry Firepower Modeling, CMX2

[billcarey]

I've been turning this over in my head for a while and wanted to run it by people. I think that the current modeling of infantry firepower may be the tiniest bit too abstracted. In situations where a squad has a "snap shot" at another moving squad sometimes it will unload on it with full firepower, and sometimes it will not shoot at all.

This means you have essentially a binary result: either the squad is firing or it is not. I suspect that this is different from the historical use of small arms where some portion of the squad would be ready to fire. Here's what I would suggest: Instead of two states, firing or not firing for the squad, CMX2 could track weapons at the individual level.

You would need three pieces of information for each weapon:

1. Firepower - already in the game.

2. Time to Reload - emperically determinable. (mostly)

3. Chance of Stoppage - emperically determinable (?)

I am assuming that currently the TacAI requests a squad to fire, and it then fires or not based on some internal timer. If this is not the case, then it will make my suggestion more difficult to implement.

Under the new system, when the TacAI requests a squad to fire, the squad would fire it's available weapons. What weapons are available would vary depending on the time spent reloading and stopped. Also, weapons that are patently out of range (SMGs) could be considered stopped for purposes of this little algorithm.

The advantages of a system like this are:

1. Squads will behave in what I believe to be a more realistic manner. They will fire much more frequently, but with a much reduced firepower. The product to firepower by time would remain unchanged.

2. Squads could then be partially suppressed (some weapons suppressed, some not) which would force more realistic infantry tactics and better use of suppressing fire.

3. I think it would be cool. This might model the advantages of the Garand as compared with bold action rifles and the advantages of bolt action rifles as compared with the gGarand.

The disadvantages of a system like this are:

1. Ammunition and Weapons States must be tracked at the individual weapon level instead of the squad level. I can imagine this taking a large amount of ram/processor cycles.

2. Players will begin to desire (and request) the ability to direct the fire of each weapon individually, which I believe they shoult not have at this scale and scope.

3. It could be difficult to program. I don't have a lot of experience with this sort of code, being more a html/php/perl guy myself. (?)

-b.

[Elmar Bijlsma]

I will leave details to better minds then mine but I like the basic you set out. Add to that the better ammo tracking that would be required for such a model and this could make infantry combat much more satisfying.

Perhaps then the Vickers MG will be better. I never thought it's ability for coninuous firing was accuratly portrayed.

[Hensworth]

Originally posted by billcarey:

This means you have essentially a binary result....

I am assuming that currently the TacAI requests a squad to fire, and it then fires or not based on some internal timer.

Sounds just as binary to me.

Already the squads are only firing the most effective weapons for the range at which they are engaging, i.e. you won't hear SMGs when a squad is shooting at something 500m away.

So the only difference your system would introduce is how many weapons are firing at the same time. For instance, only 5 rifles instead of 7, because the other 2 are reloading. This would influence the amount of firepower the squad puts out, but it wouldn't have much effect with regard to your initial thesis : the enemy being either fired upon or not within the window of opportunity.

I don't know anything about coding either, but any attempt so far to break into the sub-squad level has met with failure, due to the immediate and drastic multiplication of calculations that would entail.

Still, the Greeks thought that atoms could not be subdivided and it only took 2500 years to prove them wrong .

[flamingknives]

Tracking ammo individually (or at least per weapon type) would be the most important, IMHO.

In the current model, a volley of rifle and LMG fire takes ammo away from shorter range weapons such as SMGs, pistols and ARs. which use entirely different ammo.

[billcarey]

I think it would address the binary nature of things now. Maybe I didn't explain it well.

Right now lets say the squad fires once every ten seconds with a firepower of 100.

In my system it would fire four times every ten seconds. The first two time would only be with a fp of 15 because only three rifles are firing. The third time would be with a fp of 40 because the lmg engages. The fourth time would be with a fp of 30 because the lmg and some rifles are firing.

Each shot would only have a fraction of the squad firepower, but there would be more shots. None of this would be controllable by the player, though.

- b.

[tar]

I think that there should also be some thought given to the relative penetration ability of small arms -- something that is currently not handled at all.

I think that would help somewhat the current dominance of the SMG and the IMHO excessive strength of the US .50 MG.

Currently, the small arms have a single firepower rating which is then attenuated by the cover of the target. In reality, though, .50 rounds had much better penetration than .30 and both had better penetration than pistol rounds such as used in SMGs. If this could be reflected, it would, I think, help put the weapons in better perspective, since there would be another dimension along which the small arms would have advantages and disadvantages. If SMGs were attenuated by cover much more heavily than rifle fire, it would provide more of an incentive to choose rifle armed troops.

[Sigurd]

Flamingknives,

I just hought at the truth you said while re-playing Yelnya stare with 1.03. In the standard setup, most units are half-squads. while the russians are at long ranges, i requested only the LMG equipped hal-squads to fire, preserving the ammno of the other equipped with SMG for the final assaul of the russians. when they finally came at close range, my MP40 were on 100% amno to welcome the russkis

Had I merged the half-squads, I would have had no ammo at all to repell the final assault. So I didn't change the initial setup

so, to simplify the initial idea, perhaps just a distinction between long range amno (LMG, rifles) and short-range ammo (for SMG) would already help a lot the current system. Providing you the advantage of 2 half sqads and their separated ammo counts, without the disavantages of such an organisation (lower morale)

what do you think of this intermediate term of the initial idea ? It would only multiply by 2 the ammo tracking memory

[Moon]

It seems that you might not know that in CM as it is, extreme and long range fire costs less than a full ammo point, thus simulating the use of only a fraction of the available arms.

Martin

[Runyan99]

Originally posted by Moon:

It seems that you might not know that in CM as it is, extreme and long range fire costs less than a full ammo point, thus simulating the use of only a fraction of the available arms.

Martin

Martin, I have played a lot of CM and I have never, ever seen this happen in-game. In fact, I would be interested in hearing from anyone who has actually witnessed this feature in action.

Billcarey, I support your suggestion. Tracking small arms at the individual level is something I have wanted for some time, and something I certainly want to see in CMX2.

[Kingfish]

I think what Moon meant to say was that at extreme and long ranges a squad will go thru it's ammo load at a slower rate. This is due to the fact that at such ranges you will have the only the LMGs firing (maybe 1-3 rifles included) while at much closer ranges everybody joins in the fun.

[JasonC]

I think this is a Bad Idea .

It assumes all weapons currently loaded will fire whenever there is a target for them. This is a most dubious assumption in actual combat. The reality is that of all the weapons that theoretically can bear at a given time, only a fraction of them fire, and often none do.

Men happen to be heads down that instant. They happen to be looking the other way. They are engaged in some other task, besides reloading or firing. They don't see the enemy. They mis-ID them. They cower. Only a couple enemy are exposed at a time, too briefly. Someone just manages to sight in but hesitates a moment too long and the target has gone prone. This sort of thing happens constantly. It is perfectly realistic that infantry fire comes in lumps, in flurries and lulls.

Every weapon fielded has a theoretical rate of fire so high it could go through its entire practical battlefield ammo load in a few minutes. Yet battles lasted hours, without everyone running out. They were not pushing the ROF envelope, except perhaps in short, exceptional "mad minute" windows at the most exposed targets.

The observation was that the squad fire seems "lumpy", sometimes a briefly exposed enemy is not shot at at all and sometimes it gets a full blast. That is not a problem to be corrected, it is an accurate reflection of the uncertainty and intermittent nature of small arms firing in actual combat.

Multiplying small shots also tends to reduce all firing outcomes to an "ergodic" statistical average. The more random rolls you make, the tighter you squeeze the expected variance. You'd wind up with predictable "melting rates", as smooth and continuous as thermodynamic behavior. Moreover, this then faces rally rates, like a "headwind". Fire below a certain intensity would have no lasting effect, adding suppression only as fast as it rallies away.

There is already so much more of this in infantry combat than in armor combat that the former is noticably more predictable in outcome, relative losses, etc. There are already scads of shots in any significant infantry vs. infantry confrontation. There is no need to multiply them even higher.

[ June 16, 2003, 04:24 PM: Message edited by: JasonC ]

[billcarey]

Originally posted by JasonC:

I think this is a Bad Idea .

It assumes all weapons currently loaded will fire whenever there is a target for them. This is a most dubious assumption in actual combat. The reality is that of all the weapons that theoretically can bear at a given time, only a fraction of them fire, and often none do.

I would humbly disagree. I am most certainly not assuming that all weapons that are hot will fire. Much to the contrary. The game right now is assuming that all weapons that are hot will fire. All at once.

Right now either all weapons are firing or no weapons are firing.

My suggestion is that instead of the two binary states, firing and not firing, each individual weapon have a firing / not firing state. This would mean that the firepower exerted by a firing squad would, instead of being a discreet curve, be a more stepped smoother curve (obviously not a smooth function, but a much smoother that the current binary one).

I would think that this would simulate the historical reality you describe, Jason, better than the current system. I think the effects on the game would be less dire than you suggest as well. The most significant change that I percieve fromt the current system is that it would require better suppression to effectively move units, penalizing those who tear around the map willy nilly.

- b.

[Moon]

Originally posted by Kingfish:

I think what Moon meant to say was that at extreme and long ranges a squad will go thru it's ammo load at a slower rate. This is due to the fact that at such ranges you will have the only the LMGs firing (maybe 1-3 rifles included) while at much closer ranges everybody joins in the fun.

Yep. The ammo loadouts don't show fractions of ammo "points", but when you observe closely you will sometimes see more than one burst per ammo "point" spent, especially when units are firing close to extreme ranges (i.e. mainly when not all weapons are eligible to fire, as Kingfish mentioned above).

Martin

[Moon]

Billcarey, I agree with JasonC here, though for a slightly different reason. I don't think an abstract system necessarily gets better by taking away "some" but not all of the abstraction.

Once you decide to go with an abstraction (and the alternative would be to show every individual, check individual LOS, morale and so on - something outside of what hardware was able to handle when the CM engine was made), the main goal is to make the abstraction work within its own framework to produce believable results. The current system does that, and if you imagine each "burst" in CM to represent a number of individual shots spread out over a few seconds, it's not even much different from "reality" and/or the system you're suggesting - but it's much easier to handle for the soft- and hardware.

Ask yourself this - if the current CM engine would show you individual soldiers firing their weapons one by one, but the underlying combat resolution would be exactly as it is now (in other words, we'd fake it and you wouldn't know) - would it bother you?

To be honest, personally I think the real choice is - abstract system or individual tracking, and there is nothing in between ("less abstract") which would warrant the effort to code it. Who knows, maybe the new engine will be powerful enough to go the individual route?

Martin

[Sigurd]

Martin

I didn't know at long range you could consume fewer than 1 amno point...

Yet the 2 ammo stockpiles idea would go further in diminishing the extreme/long range ammo consumption.

I thought it could be welcome for desert combat, where LOS can be very long, cover very thin, rendering long range squad fire more attractive...

[Michael Dorosh]

I agree with JasonC and Moon. Thinking there are only two states for a weapon to be in - firing and reloading - is a grossly inaccurate assumption, upon which the rest of the suggested changes are heaped.

The basic premise is faulty, as is all that came after it.

Firepower at the squad level was based on the LMG. Perhaps a better way of tracking ammo would be how well the riflemen were passing belts/magazines of ammo to the LMG number 2, and how well the number 2 was at changing barrels and reloading the MG. This is actually more important to tracking or modelling infantry firepower than how prone a Mauser was to jamming. (Immediate Actions on a Mauser are pretty simple. Open bolt, clear obstruction, close bolt!!) A clumsy No. 2 could ruin a section's whole day. So could a warped barrel.

Riflemen in action were probably just as concerned about passing off their MG ammo to the machinegunner (in the CW, every rifleman actually carried Bren mags for his section's Bren, in the German squad, several men carried boxes or loose belts of MG ammo, or in a pinch recharged the 50 round non-disintegrating links from their Mauser pouches), or in throwing grenades, seeking cover, spotting targets for the MG, etc. as in reloading or firing their own rifles.

But then again, as Moon points out, do we really want to see that modelled in the game? Or just take for granted what is happening by using three soldier icons on the screen blazing away?

I'd also direct the original poster to the MP44 thread or somesuch, for a discussion of infantry tactics in WW II, and how little "advantage" an M-1 Garand really had over a Mauser, for all the (little) difference it made when 9 of "theirs" lined up to blaze away at 12 of "ours". The MG made the difference, and few riflemen on either side directly killed one another with rifle fire.

[ June 16, 2003, 09:44 PM: Message edited by: Michael Dorosh ]

[Michael Dorosh]

Originally posted by JasonC:

It is perfectly realistic that infantry fire comes in lumps, in flurries and lulls....it is an accurate reflection of the uncertainty and intermittent nature of small arms firing in actual combat.

My favourite quote, which I have posted here several times, does well to illustrate some of what JasonC refers to:

Farley Mowat, a platoon commander in the Hastings and Prince Edward Regiment, wrote about the first time he used a Bren Gun in action in Sicily in his book And No Birds Sang. His company had been ordered to withdraw:

Alex (the company commander) left us Nine Platoon's three Brens to free their crews from the weight so they could help with the wounded. Six light machine guns gave me a lot of fire power...or would have done except that we only had one or two magazines remaining for each gun.

Wriggling forward to the edge of the knoll, I passed the word to shoot at anything that moved - but to make every bullet count. Behind me I could hear stones rattling as Eight and Nine platoons broke cover and began their rush up the steep slopes. Instantly the metallic hail from an MG-42 swept over our heads in vicious pursuit of our retreating comrades.

I had my binoculars to my eyes at that moment and by the sheerest fluke glimpsed a flicker of flame and a filmy wisp of smoke coming from a pile of brush on the far side of the road. Mitchuk was lying next to me behind his section's Bren, and I grabbed his arm and tried to make his see what I had seen but he could not locate the target. After a moment he rolled over and pushed the butt of the gun toward me.

"You take 'em, Junior!" he said...and grinned.

The feel of the Bren filled me with the same high excitement that had been mine when, as a boy during October days in Saskatchewan, I had raised my shotgun from the concealment of a bulrush blind and steadied it on an incoming flight of greenhead mallards.

There was a steady throbbing against my shoulder as the Bren hammered out a burst. A stitching of dust spurts appeared in front of the patch of brush and walked on into it. I fired burst after burst until the gun went silent with a heavy clunk as the bolt drove home on an empty chamber. Quickly Mitchuk slapped off the empty magazine and rammed a fresh one into place.

"Give 'em another!" he yelled exultantly. "You're onto the ****ers good!"

Maybe I was. It is at least indisputable that after I had emptied the second magazine there was no further firing nor any sign of life from the brush pile. On the other hand, I never actually saw a human target, so I cannot be haunted by the memory of men lying dead or dying behind their gun. And for that I am grateful.

[billcarey]

I'm being misunderstood, and it's my fault for not communicating what I mean clearly, so I'm going to try again.

1. I recognize that there are many more states a weapon can be in that firing and reloading.

2. I recognize that in the vast majority of circumstances a relatively small percentage of a unit's weapons were firing simultaneously.

3. Having read these boards actively for four (?) years now, I've tried to pick up as much history and information as I can from the posters who obviously know their stuff.

Here's my suggestion:

As I understand the current state of the game squads have only two states: firing and not firing.

If the squad is firing, then it is simulated as firing all weapons, and if it is not firing, then it is simulated as firing no weapons. I would suggest (and I think that the historical anecdotes presented in this thread bear this out) that very often it was only some fraction of a squad's firepower that was being employed at any given time.

That being the case, I would suggest that squads project firepower more often (if not more or less continuously) but at only a fraction of currently rated firepower. The product of firepower and time would not change from what it is now.

Maybe that will clear up what I'm suggesting.

- b.

[tar]

That being the case, I would suggest that squads project firepower more often (if not more or less continuously) but at only a fraction of currently rated firepower. The product of firepower and time would not change from what it is now.

I'm not really sure, in the grand scheme of things, what difference this will make. At an individual squad level some things may change, but given the large number of such interactions, I would expect that the aggregate result would not be much different.

If my conjecture is true, then it is probably not worth the effort to change.

[aka_tom_w]

Originally posted by billcarey:

Here's my suggestion:

As I understand the current state of the game squads have only two states: firing and not firing.

If the squad is firing, then it is simulated as firing all weapons, and if it is not firing, then it is simulated as firing no weapons. I would suggest (and I think that the historical anecdotes presented in this thread bear this out) that very often it was only some fraction of a squad's firepower that was being employed at any given time.

That being the case, I would suggest that squads project firepower more often (if not more or less continuously) but at only a fraction of currently rated firepower. The product of firepower and time would not change from what it is now.

Maybe that will clear up what I'm suggesting.

- b.

I am not sure that is an entirely accurate observation of what the game is doing.

Moon tells us (and I assume he KNOWS) that only the long rang weapons (rifles) fire at long range targets, and at extreme range when they fire a burst it only counts as a fraction of a point of ammo expenditure. (not a full point he tells us)

ALSO in CMBB there is a firing state that is HOT and MAX ROF when in close the AI will "turn it up a notch, to 11 if you are into Spinal Tap, he he)" and produce more firepower out of MG's in the "expend all rounds mode" (as I understand it this is an improvement over MG performance in CMBO)

so I would say there seems to be AT least three states of ammo expenditure.

SO I would conclude there is not JUST two states, firing and none firing.

i)there is long range firing that expends VERY little ammo

ii)I assume there is average range average firing that expends average ammo

and

iii)we know there short range firing at extreme ROF that expends max ammo.

I think the game handles this aspect of small arms firing and firepower pretty well myself.

What I just described sort of sounds like a slightly varied approach to what you were actually suggesting, which I think the game accomondates fairly well

IMHO

-tom w

[ June 17, 2003, 04:13 PM: Message edited by: aka_tom_w ]