Jump to content

Elvis vs. JonS DAR Discussion Thread


Recommended Posts

Thanks for the DAR. If the game isn't slated to come out in the next two months or so, it would be fun to see another one done so we can see what's new/changed. Certainly would be a way to stay excited about the game getting a chance to see new and improved features.

Link to comment
Share on other sites

  • Replies 291
  • Created
  • Last Reply

Top Posters In This Topic

Useful DAR and AAR, guys, thanks. In a previous thread I wrote in part -

"I will nevertheless hazard a prediction.

Typical games of CMBNx2 will show dramatically higher casualty rates among the infantry, concentrated into much shorter periods of the overall engagement, than were typically seen in CMBB or CMAK. And they were not too low to begin with, but if anything too high to begin with."

"I make this prediction based simply on the likely impact of the "engineering" and 1:1 rather than "design for effect" and full squad approach being used. I consider it extremely unlikely that the real world aspects that reduce battle losses in real firefights will be simulated with anything approaching the accuracy and attention, that I expect will be lavished on basic interactions like spotting, shooting, flight of projectiles and hits or interception by cover.

"It is the perennial experience of "engineered" game design that the attempt to achieve realism by literalism of the simulation, achieves simulation of the most readily simulated bits of real combat, and doesn't even come close on others. And that the former will always include the most intentional and planned aspects of shooting range firepower, while the neglected elements will include a host of harder to simulate and disparate factors, that all tend to make for lower combat effectiveness per unit and moment engaged. With the result being a prediction of vastly higher lethality per engaged unit and per unit time, than seen in real combat, outside of the most extreme outliers of bloody massacres."

I was told in response -

"The whole point of writing it was that we're seeing the opposite than your prediction"

No, we are not. Having seen one instant of combat that impression might have been created, sure, granted at the time. But the overall battle was bloodier than ever, for exactly the reasons I predicted. No, getting rid of Borg spotting did not result in more realistic overall loss rates. Yes that is an improvement in realism, bully. No it does not replace design for effect considerations.

I should say in passing that other improvements in realism in the game are evident to all and I applaud them - immersion, graphics, recrewing vehicles, lots of stuff.

But the overall bloodiness remains definitely off in the "high" direction - exactly as has happened in every other tactical simulation design attempt that pursued engineering literalism over design for effect, since oh at least the 1970s. Not anybody at BTS's fault, it is just the predictable and predicted consequence of the modeling approach used. Always has been, always will be.

I was also told that high losses should be viewed as normal, because -

"the "scope" of CM is generally battles where both sides want a pretty decisive engagement and covers 1-2 hours of the heavy fighting"

That was before we had the full AAR to the end, to gauge the loss rate we are supposed to regard as typical. Now we have that result.

We just saw a typical result of that simulation. 60% casualties among those engaged in an hour or so. The attacking battalion took 255 casualties out of 435 engaged, and it continued its mission to a successful capture of its territorial objective.

255 casaulties in an attacking battalion was about the average loss rate experienced by the green British infantry on the first day of the Somme.

(17 divisions each of 12 infantry and 1 pioneer battalion attacked, and took 57500 casualties. That works out to 260 men per battalion - and the battalions were much larger, 700-800 men, not 435. Yes the worst hit lost more, some lost less, that is still an average first day of Somme loss rate). Which was the bloodiest one day shambles in the history of warfare, and remains unsurpassed to this day.

In Normandy, the US 30th Infantry *Division* lost 267 men on July 9th, in heavy fighting early in the push to St. Lo. That reflected a local counterattack by the Pioneer battalion of 2SS in addition to its usual attacks. Which had Panzer IV and SP gun support, and fire support from 4 battalions of artillery, 3 105 and 1 150mm, and the German artillery caused a high portion of the losses. This was a *division*, not a battalion, in a full day, not an hour.

The German counterattack "dissolved under heavy artillery fire", after initial success, in the word's of the German command diary. US div arty fired 9000 rounds that day and corps arty added to the total. As a point about the scale of effort involved, in case anybody thinks it wasn't heavy fighting or something.

On the worst sectors of Omaha beach on D-Day, half the men made it to the shingle.

Half of Elvis's men didn't make it to the "Alamo". That is the kind of outlier of bloody shambles this loss rate represents.

At Omaha, losses of 33 to 50% in the initial wave were sufficient to reduce the survivors to a pinned down and disorganized shambles that managed to accomplish nothing for another hour or two. Units no longer existed, so many commanders were down and so many of the remaining men were either busy tending to wounded or morally shattered, that the combat effectives remaining from the first wave were isolated knots of handfuls of men. On a few sectors, companies had reached the shingle with practically no losses or in some cases, about 25%, and retained combat effectiveness. But were still

seriously delayed, disorganzed, and shaken by the entire experience going on around them. Yes some still accomplished miltiary tasks - over the next 2 to 4 *hours*. Rally took that long, not 2 to 4 minutes.

JonS wrote "As has been pointed out, in the real deal once Elvis' first attack on Hill 154 foundered, and especially when his casualties started to mount, it seems to me unlikely that Elvis' boss would have pressed him to keep going. Or that his own men would have kept going, regardless of what the boss wanted. Fundamentally troops are too willing

to continue to engage in combat, because their commanders - all of us, as players - don't really care whether those little pixeltruppen live or die. We continue to mash our virtual troops together long after real soldiers would have called it quits." He says he sees no alternative, and we can discuss that (I will below). But he's simply right about the realism point.

Elvis commented at one point that in real life, his force would have packed it in after the initial failed attack on the woods. He added he'd probably get court martialed over the butcher bill for that part of the affair. He is also just right, realism wise. (Leaving aside a tendency for brass to regard high losses as proof of trying that might have prevented literal court martial, it still would be considered a fiasco).

Next question - are games doomed to get this wrong because no real lives are on the line and players will always push their men excessively? Answer, yes there is a tendency in that direction, and *if* an engineered design does not simulate the real world forces of confusion and morale failure that actually prevent such outcomes, then yes games will continually err in the direction "way too bloody". But no, if design for effect systems are added to address the subject. Which is hardly new ground for simulations. We know why the issue arises and ways to address it.

Early grand tactical board wargames introduced morale considerations at the individual unit level. This was already a huge advance in realism, and CM has it - bully. But consider the late 1970s era board game system of Terrible Swift Sword as an example.

Its individual units were infantry regiments (the strength of battalions in modern terms), with 100 man step losses. Individual units were subject to rout, and when weakened their morale fell and they became more susceptible to rout again.

It was found after the system had been played for a while that players readily adapted to the brittleness of individual units, cycling the strong ones through hot action, and collectively pushing the whole army far harder than historical counterparts could ever dream of pushing them. With losses too high over long periods as a result. The designers did not simply accept this; they viewed it as a problem that called for a refinement of the game system.

So the system introduced a design for effect system called Brigade Combat Effectiveness. The next step up in scale above the individual units, and the focus of the game's simple command and control system, the brigades were now rated for a loss tolerance. As soon as the cumulative losses to all the units of the brigade combined exceeded its BCE level, all the remaining units of the brigade became subject to additional morale penalties. They became more likely to rout, unable to attack in melee, harder to rally, etc. As a result, a brigade pushed too hard was effectively "wrecked" and unable to withstand heavy action against a fresher opposite number in the enemy army.

And immediately, there was a strong incentive on players to manage their casualties much more carefully. The same losses spread over the entire army would put out of action fewer units than all the losses piled onto the men up front, which would BCE whole brigades and thus effectively multiply the men rendered ineffective by a factor of 2 or 3.

Design for effect plus a morale system above the engineer-depicted base unit scale, created a *command incentive* to treat losses in a manner closer to the way the historical commanders treated them. A commander who recklessly mashed any unrouted unit into the enemy any time he could, would see his army come apart on him, two to three times faster than a commander who used ranks and reliefs and whenever possible relied on fresh troops, and avoided putting 33 to 50% losses on individual subformations.

Typical observed losses in games fell significantly. In part this was simply due to lower average effective morale, which meant units on the whole spent more time routed (because some formations still fought BCE'ed). And some of it reflected the losing side coming apart at a more realistic loss level and retiring from the field, intact or routed

depending on how well it was commanded, but "blown" all the same. A portion, however, came from commanders exercising great loss avoidance themselves, voluntarily, in response to realistic command *incentives*.

JonS said "'real' battles are kind of boring: a long period of nothing. A few troops move, one or two get shot. Some artillery is called in. A lot more nothing. Some more movement, a couple more deaths. More artillery. Some mortars. Some more movement, then side or the other calls it quits and either cancels the attack for the day or

pulls back to the next stream/ridge/hedgerow. That doesn't much sound like the kind of game I'd like to play".

He is entitled to his opinion, surely. It is a game I'd love to play, especially if the dilemmas incentives and realistic alternatives involved are right. I'd settle for half way between that description and our present norm of all out kamikaze action up at the grand tactical level.

Here is the US army green book's description of why and how 50% losses to companies on Omaha sufficed to put them out of effective action for hours rather than minutes.

"Command was generally one of the gravest problems faced by assault units, not only because officer casualties were high and mislanding of command groups had left many units leaderless, but also because of extreme difficulties of communication... Furthermore, in the confusion of the mixed units, which were under heavy fire in some places, their men huddled along the shingle embankment or sea wall and generally shaken by the shock of the first few minutes of severe action, it would have been impossible for any commander to exercise control over more than a small group

of men on a relatively narrow sector of the front."

Some of that is simply the old "panic" morale state in which a unit stops responding to orders. But some is a one to one simulation issue of units being abstractions and dissolving completely under sufficiently poor morale or high casualty states. Coupled with *limited command attention*.

Imagine you had to give orders to each individual man on Omaha beach using a CM interface. You could not do it. The organization structure of units responding meaningfully to commands and each doing their job without further input from higher echelons, is as vital a part of a unit's combat power as its weapons. And organization structures can and do *dissolve* under personnel losses and morale failures.

Some of this is also the old global morale, which may simply be set way too high for realism. Not that outright surrender should be common at low levels of losses, but a willingness to break off the action should be.

I proposed a version of this back in CMx1 days that players could implement themselves by voluntary agreement before the game. Just have a scenario set global morale level for each side that triggers a mandatory offer of ceasefire. If say your side's threshold in the scenario is 60% global morale, then as soon as your global morale reading drops below that figure, you have to depress the "offer ceasefire" button each turn it stays that low.

You can always offer it just because you want to. If both have to offer it because both are below their respective thresholds, then the action will break off. This creates a strong incentive to avoid losses or morale state dangers that could drive your force below its morale threshold, and to wait for units to rally before attempting another risky action, and the like.

Another possibility to deal with the grand tactical level is something like a company specific BCE system. Which might work through rally speed, and be modified by the command ratings of the company. As the company takes losses, all units within it suffer a slow down in their rally speed. It might be something like twice the actual physical casualties as a percentage, plus the units in red morale states. If both combined

are driven high enough, rally speed would slow to a crawl, and effectively units in that company that pin would stay pinned, that break would stay broken. This would reflect the way trama to other units within the command sop up all the available command attention, as well as scaring the remainder as individuals.

The issue is there and it is addressable. You can view these suggestions as design for effect, or as trying to engineer these command, confusion, and morale contagion issues, not just the physical spotting and shooting ones. No doubt there will remain room for improvement either way, but that isn't a reason to settle for first day Somme loss rates as supposedly normal.

Use or discard as you like, one man's opinions, for whatever it may be worth, etc...

Link to comment
Share on other sites

Jason C, the higher level morale idea you bring up is not without interest, But CMX2 already has the means to set any casualty limit you wish. You can tack on such a large point penalty for any loss percentage you care to specify that it will make victory unachievable for that side. You can set that for 5% losses and, to use the extant example, give the American player 4 hours just to take hill 154. It would probably be more realistic. But you are conflating scenario design with engine flaws, they are not the same thing.

And the purpose of an AAR like this one is not to produce a perfectly historical overall result, but to illustrate as many specific game mechanics as reasonably possible, and be entertaining to read.

Link to comment
Share on other sites

Jason C, the higher level morale idea you bring up is not without interest, But CMX2 already has the means to set any casualty limit you wish. You can tack on such a large point penalty for any loss percentage you care to specify that it will make victory unachievable for that side. You can set that for 5% losses and, to use the extant example, give the American player 4 hours just to take hill 154. It would probably be more realistic. But you are conflating scenario design with engine flaws, they are not the same thing.

Correct. The engine and victory point system allows for exactly what Jason is putting forth. If the scenario had been made to give me more points for conservation of units than for territory to be taken my plans would have been very different. As they are/were in the NATO German campaign battles I have mentioned. The game has the flexibility to cater to both tastes in war game play. It is not a function of the engine itself but of the parameters set forth by the scenario designer. Do you prefer chocolate or rocky road ice cream? Or both? You've come to the right ice cream shop because you have both and many other in between.

Link to comment
Share on other sites

It looks like battles, because of the shakier morale, progress much slower than in, say, CM:BO.

Would that be accurate? It was sometimes hard tell how much was happening in what period of time in the DAR because I didn't always look to see how many turns any particular report covered. Or, it could be that lots of stuff was happening, but neither participant decided to write it up.

Link to comment
Share on other sites

JasonC,

Sounds like you want something that CM never was, never should be, and never will be. What you're advocating is a higher scale of combat than CM portrays, so many of your observations are colored by that focus. Others have debated you about this in the other thread, so I have I. I don't see much point in doing so again since it doesn't appear you will change your perspective.

Steve

Link to comment
Share on other sites

Jason C, the higher level morale idea you bring up is not without interest, But CMX2 already has the means to set any casualty limit you wish.

Correct. Elvis also spoke about this. JonS could have used the Surrender feature too, or simply fled from the battlefield instead of trying to get in a few last licks. If Elvis didn't have some tanks that might have actually worked.

The point here is to remind everybody that CM is a sim at heart, but still a game. Games are meant to be fun, so there's a limit to how strict a simulation can possibly impose itself onto the player before fun goes out the window. The threshold for this varies from player to player. The vast majority of gamers out there wouldn't touch CM of any flavor ever because there's no power ups, spawn points, running over civilians with tanks, the latest graphics crazed features, etc. So in terms of producing a "fun game", Combat Mission is already behind the 8 ball. Constricting the customer base further is something we have to be very careful about unless that customer is a military agency looking for a dry sim to run map exercises with. In that case, the criteria changes dramatically :D

You can set that for 5% losses and, to use the extant example, give the American player 4 hours just to take hill 154. It would probably be more realistic. But you are conflating scenario design with engine flaws, they are not the same thing.

Correct. We've had people complain about CM:SF battles where they took 4 or 5 individual casualties and were told they lost the battle. "WTF?!? I only lost 2 guys and a Humvee! I killed 40 of the enemy and took the building I was tasked with taking! That's unfair!".

Well, according to the guy who made that particular scenario, that player should have been able to get the same objectives with fewer casualties. Whether it is really possible for a player to pull that off or not is yet another question. Therefore, the system must be viewed separately from any one individual scenario's desired (or accidental) encouraged outcome.

And the purpose of an AAR like this one is not to produce a perfectly historical overall result, but to illustrate as many specific game mechanics as reasonably possible, and be entertaining to read.

Correct, and correct! I know I enjoyed it even though I have the latest full version on my desktop :D

Steve

Link to comment
Share on other sites

Its impossible to follow everything that happened plus the comments but I'd hope that any snipers would play a relatively minor role in the outcome and that the bombardment effect on morale would be less than predictable. If troops that were bombarded are always reduced in effectiveness for the rest of the game then that would suck, as well as being completely unrealistic. I can't recall the influence of leadership here, if any, but realistically at the platoon and squad level it it can be crucial, particularly immediately after a bombardment and when attacking where the unit is taking casualties.

The other thing was the surrender bit. Attackers don't usually surrender, they may go to ground or run away but they usually don't surrender. Defenders may surrender, particularly when attackers get close where they usually start to gain an advantage. So basically attackers may go to ground or withdraw while defenders may withdraw or surrender.

While the infantry aspect seems a lot better than in CMx1 I'm yet to be convinced that its good enough to spend my time on.

Link to comment
Share on other sites

The other thing was the surrender bit. Attackers don't usually surrender, they may go to ground or run away but they usually don't surrender. Defenders may surrender, particularly when attackers get close where they usually start to gain an advantage. So basically attackers may go to ground or withdraw while defenders may withdraw or surrender.

I'm curious as to what this is based on - real life, or expected behaviour in a game?

Link to comment
Share on other sites

In CMSF it does seem as if the allied forces are akin to supermen - always obeying orders, fighting to the death. But, that may be a political issue since the game environment is too close for comfort to today's situation and in deference to the high % of mil profs. in this market.

It has nothing to do with military profits :rolleyes: and everything to do with morale, C2, and skill level.

Link to comment
Share on other sites

Use or discard as you like, one man's opinions, for whatever it may be worth, etc...

Thank you, Jason. I think you put your argument very clearly and cogently. As dan and Elvis stated in their replies, it may well be possible to address your concerns within the existing system. At least I am willing to give it a try before requesting additional modifications. But the fact that you stated your case as precisely as you did makes it easier to give precise replies. I hope this will become a trend in posting on this board.

Michael

Link to comment
Share on other sites

Its impossible to follow everything that happened plus the comments but I'd hope that any snipers would play a relatively minor role in the outcome and that the bombardment effect on morale would be less than predictable. If troops that were bombarded are always reduced in effectiveness for the rest of the game then that would suck, as well as being completely unrealistic.

It is situationally dependent. But the primary reason Elvis' forces were brittle and under performing is he lost a LOT of guys. So it wasn't a simple "shots hit here, morale totally crushed". Not even close to that. There is Global Morale behind the scenes, but not exactly like in CMx1 in that you don't get a running display of it. You get a sense of it as did Elvis, when your shattered remains start thinking they'd rather sit this one out. That is completely realistic because in real life Humans aren't robots. The game is already too generous with the player's ability to run them to their death, as JasonC pointed out (and we've been discussing here since 1998).

The other thing was the surrender bit. Attackers don't usually surrender, they may go to ground or run away but they usually don't surrender. Defenders may surrender, particularly when attackers get close where they usually start to gain an advantage. So basically attackers may go to ground or withdraw while defenders may withdraw or surrender.

Uh... thats not true. When soldiers get into a situation where they either surrender or die, they often choose to surrender. It doesn't matter what's going on around them in the Big Picture at that moment. Since the attacker usually doesn't put too many soldiers into that position, the attacker usually doesn't produce many POWs for the defender. But absolutely, without ANY doubt, attacking forces did lose people as POWs to the other side.

One very famous incident in Afghanistan had a US Navy Seal surrendering to the defender. It happens.

While the infantry aspect seems a lot better than in CMx1 I'm yet to be convinced that its good enough to spend my time on.

That's what demos are for :D

Steve

Link to comment
Share on other sites

Useful DAR and AAR, guys, thanks. In a previous thread I wrote in part -

"I will nevertheless hazard a prediction.

Typical games of CMBNx2 will show dramatically higher casualty rates among the infantry, concentrated into much shorter periods of the overall engagement, than were typically seen in CMBB or CMAK. And they were not too low to begin with, but if anything too high to begin with."

...etc etc,

Use or discard as you like, one man's opinions, for whatever it may be worth, etc...

I very rarely post on the outer boards but I really can't resist.

Jason, again as usual you have all of the facts exactly correct and the context dead wrong. CM is a game. Plain and simple. It is not a combat simutator (it could be adapted to one but that is another discussion).

The issue here is not in the machine. CM is very accurate, scary accurate in some places in providing an atmosphere and environ in which to play the game. To try and advance this artificially into a simulator will not, nor ever will work.

Why? Because despite your dusty books and facts, combat and warfare is still in large part a human experience. The player, not the machine is the center of this issue. Two players could easily simulate real world casualties in this game, so the machine is not broken. The players are.

To try and bring a player into the mindset of a battlefield commander and have him react in a similar manner delivering realistic results would reqr him to live and train with a human being for every digital soldier he has. Then he would go home and every real world avatar would stand in a field with someone behind them holding a gun. Injured in-game get limb shots and dead get..well dead.

Imagine what you average player would do in that situation? 99% wouldn't even play. The rest would be exceptionally cautious to the point of inaction. It takes us years to get people to the point where they will send men to die knowingly, and trust me it never gets easy even then.

Now to try and "force" a player to play towards some historical norm, is a recipe for disaster. It only leads to extreme frustration, hell you can already hear some of that from newbies coming over from the Company of Heros world.

There is always room for better simulation of battlefield friction and adding stress to the player to push him into playing more realistic to win. But if I want to do Day 1 of the Somme, well I can. If players want to play bloody, they can go ahead. If they want an historical enactment, well go right ahead.

CM is the sandbox and in my opinion should never try to "replace" the player in order to achieve historical accuracy.

Link to comment
Share on other sites

Will there be a option for the scenario designer to have the AI to give up or route at a specific casuality level or if a specific object is taken ?

So if the player inflicts 100 casualitys to the AI or if he captures a dominating hill the AI will route (what would mean surrender). It would be up to the scenario designer how much beating the AI will take.

Link to comment
Share on other sites

Uh... thats not true. When soldiers get into a situation where they either surrender or die, they often choose to surrender.

Very true, but as far as I could see from the DAR those weren't the only 2 options to the troops who threw their hands up in the DAR, or from what I could tell (I'm mainly thinking of the troops in the woods who just needed to retreat a few action spots to be out of the fire)

Although of course as I wasn't actually playing the game I could be totally wrong, perhaps elvis could enlighten me?

Link to comment
Share on other sites

Do troops still rout as well, or has surrender essentially replaced that functionality? As structured it really seems like a last ditch warning to the player that a given unit has been used up. You can get them somewhere safe or they will decide to take the POW camp instead.

Its my understanding that once a unit is in the surrender state, there is nothing you can do with it. You can not move it, you must move friendlies close enough to rescue them. See this post by JonS

http://www.battlefront.com/community/showpost.php?p=1228741&postcount=49

Link to comment
Share on other sites

Very true, but as far as I could see from the DAR those weren't the only 2 options to the troops who threw their hands up in the DAR, or from what I could tell (I'm mainly thinking of the troops in the woods who just needed to retreat a few action spots to be out of the fire)

Although of course as I wasn't actually playing the game I could be totally wrong, perhaps elvis could enlighten me?

Some did. I each situation of the DAR that I had men throw their arms up there were others in their unit who retreated rather than surrender. The tank crew had 4 men, of which one surrendered. The squad that ran onto Hill 144 had 9 men and 4 threw their arms up and all ended up being rescued. The squad on the cart path had 6 men and 2 surrendered. Surrender is not a unit-wide thing that happens.

Link to comment
Share on other sites

But if thet were under merciless mortar fire, but not in close proximity to any enemy soldiers, that had brought them to the same state in terms of morale failure what would they have done?

Died.

Mord.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.

×
×
  • Create New...