Aggression Setting

[Leonidas]

CM is a fine game, but there is plenty of room for improvement. It's easy to pick on little things, but I suggest that we, the players, focus on suggestions that offer the most improvement to the game for the least amount of work for the developer.

I think that the developer has taken on too large of a task in trying to write AI that can distinguish various situations and come up with appropriate behavior every time. This is most glaring with tank AI, but it comes up some with other units as well. The more control the players have, the less work the AI has to do and the happier the players are because they feel closer to the battle.

Balanced against that, however, is the need for an elegant interface that newbies can understand and use without spending an hour reading the manual. CM does a fine job in making the game accessible to newcomers, and I'm sure they plan to keep it that way.

My suggestion is to add an aggression setting for each unit on a five point scale. This setting tells the unit how it should be balancing its need to stay alive with its need to attack the enemy. Sometimes you want a unit to keep its head down and try to stay alive until help arrives. Other times you want units to charge forward fanatically, spraying ammo at anything that moves. The aggression setting is the player's oppotunity to tell the unit what his role is for the coming turn.

Like most variables in CM, aggression would be introduced in various subtle ways, and the precise effect would vary with each unit. Overall, aggressive units fire at longer ranges, make less use of concealment and cover, and are more willing to burn up ammo on shots that will do little damage. Low Aggression units fire only at close range, make more use of concealment and cover, and fire only when their use of precious ammo is justified.

For infantry, high aggression would mean that the men fire more often, and a little less accurately, and they gain less defense when in cover. Low aggression means that the men are conserving ammo and trying to stay out of sight. The Hide command works as something of a low aggression setting for infantry, but you can't order troops to Hide when they are being fired on.

For tanks, low aggression means that the tank is ignoring targets of opportunity and looking only for units that can directly damage it. High aggression means that the tank will fire at whatever units present themselves.

For artillery, the aggression setting would determine the radius at which the spotter will stop the battery because of a chance of hitting friendlies. I don't know how historical it would be, but it would be interesting to have aggressive spotters be less accurate, but they get their orders through faster. Cautious spotters could take longer to start shelling and might interrupt firing to refine the aim, but they could be capable of more precise fire and efficient use of ammo.

Aggressive machine guns, fixed guns, infantry, and antitank teams will fire at anything they see. Low aggression units will hold their fire until they get a good shot against a valuable target.

It's tempting to try to link aggression settings with movement speed, such as making the Run command be Move plus High Aggression. That would simplify the interface somewhat, but it would also take some options away from the player. It's a balancing issue for the developer, but I favor player control. I can think of many situations in which I would want my infantry to run into a building or woods through enemy fire, but then hide themselves and go passive once they get there.

To me, this would take a big step toward making CM a more enjoyable game, and would reduce frustration with AI errors. I'm curious what others think.

[Terence]

I think no.

Part of the fun (to me) is that the units don't always do what I want.

The fact that you can;t set your units to be aggressive and have them do what you want --that sometimes they will ignore your orders and hide in their holes is a good thing, not a bad thing.

That's the way it was for the company commanders on the field, and that's the way it should be in the game. In my opinion.

[Michael Dorosh]

Does this not already exist - ie Fanatacism?

[Leonidas]

I'm not suggesting that anything be changed about whether your troops follow orders. That's a separate issue. I'm just saying that your troops need to know whether they should be attacking vigorously, or hunkering down, hiding, and saving ammo. That's information real troops would have.

[Terence]

not to be dense, but can't you do that now with the the hide and target commands right now?

[Clubfoot]

Leonidas,

The AI does determine the things you just mentioned. If a group of men are ordered across an open field to assault an MG bunker, they set to the task. Until they're either mowed into frothy, pink bits, or become aware of the bunkers' preoccupation with that engineer squad 200m away, or what have you. At this point, your troops will typically know whether they should, as you say, attack aggressively, hunker down, hide or save ammo.

The game is designed to avoid the micro-management of troops. I believe that the occasional "oh ****, WTF are my guys doing?"

only serves to enhance the game's fun factor.

No offense to anyone, but I've noticed a lot of people on this board seem to have forgotten that glaring acts of individual stupidity happen in real life, too. So the next time your Tiger doesn't take that shot (you know the one, it makes you pull your hair while shouting "SHOOT, SHOOT you idiot" at the monitor) or your company HQ meanders about in a meadow while being chewed by an MG42 instead of taking cover in those dense woods a gnat's ass away, chalk it up to the "dumb-ass factor" present in ALL aspects of life.

Sorry for the rant.

Clubfoot.

[Leonidas]

Terence,

You can somewhat do what I want with Hide and Target, except: 1) You can't make men Hide (and stop wasting ammo) if someone is firing at them, even if the fire is ineffectual. 2) You can't Hide MGs, which makes it very dangerous to move them when they are concealed, as they will fire at anything in sight. 3) You can't override the 'smoke and reverse' syndrome so common to American tanks, even when you set up a perfect shot for them.

And isn't it kinda silly to have to use the Hide command when what you really mean is 'Don't waste your ammo and give away your position until you get a good shot'?

Clubfoot,

I'm new to this forum, so I don't know what others have said, but I don't think I'm getting my point across.

Aggression is not morale. Sometimes soldiers are ordered to be aggressive, and other times they are ordered to be passive. Whether they actually follow those orders is totally separate, and CM already models that very well. Sometimes you order troops to stay hidden, and they get up and fire at long range anyway. Sometimes you order men to charge, and they decide they would rather stay in their foxhole.

All I'm saying is that I would like the chance to give that order more precisely than Hide and Ambush, which are imperfect workarounds.

I also disagree about taking AI errors and pretending that the program is modeling acts of individual stupidity. If CM is supposed to have a random 'individual stupidity' factor that causes units to run amok, that's fine. But systematic AI weaknesses are not modeling that. It's just that AI is very hard stuff to write, and it invariably makes mistakes. Can we at least agree that it's a worthy goal to try to eliminate those mistakes, instead of papering them over by calling them realism?

[Clubfoot]

Leonidas,

First of all, sorry if I came off harsh. I don't mean to be grumpy.

I would have agreed it was worthy to tweak and add, then tweak and add some more to CM before the release of the v1.1 patch. But with every request for a code tweak or added function, you must realize that the price you would pay would be further delay of CM2.

The men who brought you CM were but two. Although their staff has doubled to a whopping four, they have stated that every bit of buggering about with CM puts off CM2, and we can't have that!

Now, "Don't waste your ammo and give away your position until you get a good shot" falls under the category of an ambush. Units set to ambush an ambush marker will not waste their ammo, or give away their position until they get a good shot. Even if it isn't always near the ambush marker. If your boys in ambush get a primo oppor-tunity to hit something good, even if it's not close to the ambush marker, they'll take it. Or they won't. I've been surprised at the differences in the same turn when played over and over. Given the same opportunities and environmental circumstances, units that are not in panic, routed, or shaken, will react differently to the same stimuli. This is a result of many interacting probabili-ties arriving at a percentage chance that a unit will do this, that, or the other thing.

This is a function of AI "decision making", not morale. Some decisions seem "smart" and others "brain-dead", and although morale plays into this and is often the pivotal factor, it often isn't as well.

What you consider ineffectual fire at a group of hidden soldiers, even if their attrition rate is less than say, 1 man per two turns, they and their little randomized computer brains do not consider ineffectual.

They are being fired upon and might soon be hit ( and 1 or two may have been). CM units do pay attention to squad attrition and react to it, regardless of whether their morale is affected. Besides, hiding while being fired upon means you're no longer hiding, but taking cover.

Tanks that smoke and reverse may have decided to based on a perceived threat, not changing morale. And the desire to override their decision to do so is a plea for micro-management.

MG's can hide. They can't sneak. This is because they are toting a big fat hog and in many cases, hundreds of pounds of ammo.

You are correct that AI is imperfect, but I think programming in the ability to alter or override decisions made by individual soldiers or tank commanders is probably not the answer.

In the end, I imagine we'll have to agree to disagree. And that's ok!

Clubfoot.

[Leonidas]

Clubfoot

I'm guessing that you're a fan of the Close Combat series. I played the first Close Combat a little and couldn't stand it, because it gave so little control over my units. It didn't feel like a game. I gave a few orders, and as soon as the shooting started my men did whatever they wanted, which was mostly cowering. It may have been accurate, but it wasn't fun. It was more like watching a randomly generated WWII tactical movie than playing a game.

To me, games are about choices. It may be historically accurate to show that any particular commander in WWII had only a little control over only a few things. But that doesn't mean that giving the player only a little control over a few things will make for a good game. I'll take a good game over a good simulation any day.

CM is both a game and simulation. It seems to me that CM would be a better game without being any less of a simulation if the player could convey to the troops exactly the kind of information that real troops receive. I'm not suggesting this as a patch - I'm suggesting it as a feature for CM2. But if the response I've garnered on this forum is representative of the developers' thinking, it sounds like CM2 will be more of a WWII movie generator than a game.

[IntelWeenie]

Interesting idea, but I have no idea if it passes the cost/benefit test. There have been times when I thought a setting like this would come in handy, usually related to tanks firing at non-threatening (to the tank) infantry or units shooting small arms halfway across the map at some Infantry? target when I would rather them save ammo.

When I read the title of the thread my first thought was a commander setting like in the old Broderbund "Sun Tzu's Ancient Art of War" game. (Was it just me, or was "Crazy Ivan" always the most fun to play against? )

[Spook]

Leonidas:

You've provided an interesting subject of discussion. But now borrowing a sample from your first post:

Like most variables in CM, aggression would be introduced in various subtle ways, and the precise effect would vary with each unit. Overall, aggressive units fire at longer ranges, make less use of concealment and cover, and are more willing to burn up ammo on shots that will do little damage. Low Aggression units fire only at close range, make more use of concealment and cover, and fire only when their use of precious ammo is justified.

For infantry, high aggression would mean that the men fire more often, and a little less accurately, and they gain less defense when in cover. Low aggression means that the men are conserving ammo and trying to stay out of sight. The Hide command works as something of a low aggression setting for infantry, but you can't order troops to Hide when they are being fired on.

I think the problem here is what you are using to define "aggressive behavior." Right now, in terms of fire combat, this "aggressiveness" is already there in a different way. Whether green or veteran, troops will frequently engage each other even beyond moderate ranges if an enemy unit pops into view unexpectedly. But if you wish your troops to "control" their engagement range by using ambush markers, only the more experienced troops will abide by these regularly and hold their fire until the right moment.

In terms of infantry popping away at longer ranges, I don't personally view this to be "aggressive." I view it to be inexperience. If you want your troops to be more effective in modeling "aggressive" behavior, they would have to seek closer engagement ranges, even if that means the enemy's return fire will be more effective also.

You see, my view of "aggressiveness" is the willingness to engage the enemy AND stay engaged within ranges where relative firepower and "force" is decisive. For most infantry squads, firepower will not be very decisive per squad at ranges over 150 meters.

And here again, CM already allows for this. The more experienced squads are not only more effective in firepower (with all other things being equal), but are also more willing to return fire when fired upon in turn and taking losses.

Or if you want to model "aggressive" behavior with inexperienced troops, then you can adjust a side's relative fanatism setting when you create a scenario. By example, Green SS squads with the highest fanaticism setting would be a way to model the 12th SS Panzer in the Ardennes battle.

As to the idea of "agressive" artillery spotters willing to plot fire missions close to friendly troops, that's interesting to propose. However, the premise here is that the artillery FO is able to know where ALL friendly forces are at ALL times during the battle. Not a likely historical premise.

[Guest Andrew Hedges]

I dont' think that the TacAI needs aggressiveness tweaking: the fact that some troops can't stay hidden is a realistic one and is based on the units experience. To get troops that follow orders better, buy more experienced troops (or place more experienced troops in the most critical locations).

I wonder if the StratAI (I think that's what it's called) couldn't be tweaked to add some variation in strategies used by the computer. Something like, say, 80% of the time the StratAI plays the way it currently does, and 20% of the time it does something human-like (i.e., unexpected). Maybe it would decide to keep all of its armor hidden until turn 15, or until the opponent crossed a specific place, or killed a squad; other times, it would decide on putting most of its forces in a narrow location and make a hard push there. Other times, the StratAI could make a diversionary attack with one platoon and then attack with two platoons somewhere else.

I have no idea how hard this would be to code, though, or whether, having been coded, it would work out as I imagine.

But I'm happy with the TacAI.

[Leonidas]

I agree that experience has a great deal to do with whether infantry is willing to hold its fire until the enemy is at close range. But there are a few situations in which long range fire is appropriate, such as providing suppression fire in an overwatch, or firing at retreating enemy squads at the end of a battle.

Will Crack or Elite troops automatically hold their fire, even if they aren't told to Hide or given an Ambush? I assume that they will fire at extreme range unless you stop them, but I hadn't experimented with that.

Maybe 'Aggression' isn't the best term. I was just looking for a single concept that could be applied in different ways across a range of units and situations, to give players a broader range of commands without cluttering the interface. I wanted to address a number of CM deficiencies in a single control:

Tanks and other vehicles should know whether to expect enemy armor, infantry AT defenses, or no AT defenses. A real WWII vehicle with a working radio would have that information.

Defending infantry that is under fire should be able to attempt holding fire and hiding, so that the attacker does not know whether the infantry has retreated.

Machine guns should be able to move without firing at any visible target.

Units generally should have the option of increasing the rate of fire and ammo consumption at the cost of accuracy, or firing more slowly, conserving ammo, and doing more damage per shot.

I think that giving the player more options - not necessarily more control - would make CM an even deeper game.

[Michael Dorosh]

<BLOCKQUOTE>quote:</font><HR>Originally posted by Andrew Hedges:

I dont' think that the TacAI needs aggressiveness tweaking: the fact that some troops can't stay hidden is a realistic one and is based on the units experience. To get troops that follow orders better, buy more experienced troops (or place more experienced troops in the most critical locations).<HR></BLOCKQUOTE>

Even better will be a campaign that lets your troops gain this experience on their own. But good advice.

I didn't like Close Combat either - I found the control of my troops adequate, and realistic, I just didn't like the small scale (much more flexibility in CM - where you can field a brigade or a company) nor playing on the same crummy maps over and over (again, CM is much more flexible with random terrain).

[Spook]

Will Crack or Elite troops automatically hold their fire, even if they aren't told to Hide or given an Ambush? I assume that they will fire at extreme range unless you stop them, but I hadn't experimented with that.

The TacAI behavior isn't absolute, but yes, troops will typically fire at moderate to long ranges as a default.

Maybe 'Aggression' isn't the best term. I was just looking for a single concept that could be applied in different ways across a range of units and situations, to give players a broader range of commands without cluttering the interface.

What it sounds like you want is what is in games like TacOps, Brigade Combat Team, or Steel Beasts---a "standard operational procedure" or SOP, for fire control. This has been discussed in the past on this forum, and you could try "SOP" as a search topic to read what earlier views were offered here. But as you can see, SOP isn't a feature per se in CM.

(snip)

Units generally should have the option of increasing the rate of fire and ammo consumption at the cost of accuracy, or firing more slowly, conserving ammo, and doing more damage per shot.

I think that giving the player more options - not necessarily more control - would make CM an even deeper game.

It can be accepted that providing more options in scenario design without expanding player control during gameplay can be useful. However, I think that setting "fire range" or "fire intensity" across the board for all troops on one side in a scenario is not a good idea. Sometimes troops in real life would pop away at extended (and ineffective) range, while others might choose to wait until closer range. It was typically situation-dependent for each unit or squad, and it should be left to the unpredictable control of the TacAI.

Again, play the game some more. Experiment with units of varying experience (including having leaders of differing "combat" capabilities). I am of the impression that what you are asking for is already there. And while I wouldn't mind SOP's on a per-unit basis someday, we have to admit to such a feature gives more direct player control instead of less.

[Maximus]

"What we have here...is failure to communicate. Which is the way he wants it...so he gets it."

Leonidas,

Everything that you have explained, wanted, etc, is already in the game. They just come with a firm understanding on the "tools" of the game.

Just as several of our fellow board members have explained, all the tools are there for you to get the behavior you want. You just gotta know how to use them. In other words, learn the game's conventions. Me finks you are just too used to playing other RTS games where the AI isn't very complicated as to where it requires your "agression" settings. Trust me, if those "other" games had half as much Tac-AI as CM does, they wouldn't need aggression settings.

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

"Live by the sword, live a good LOOONG life!"--Minsc, BGII

"Boo points, I punch."--Minsc, BGII

"Buttkicking for goodness!"--Minsc, BGII

"Aim small, miss small."--Mel Gibson, The Patriot.

[This message has been edited by Maximus (edited 01-22-2001).]

[jasoncawley@ameritech.net]

I don't mind dumb sometimes, because it does indeed add to the game and realistically so. But I can mention a few pet peeves in the dumb department that are a little too regular.

#1, what bright idea do bazooka teams and the like have, firing off all of their available ammunition (without a targeting order), shooting at vehicles they can indeed penetrate that are indeed in range, but at ranges where the hit probability is 4%? Does the idea of "wait until he is a little closer" demand too much?

How about, as a rule of thumb "if my hit probability times my remaining rockets is less than 1/2, I think I better save this little puppy". So a zook team with 8 rounds would merrily fire at extreme range with only a 6% chance of hit, which is dumb but may seem live-preserving to a scared corporal. But one round left, and he would wait until either #1 he has a definite order to expend it (player designated target) or #2 his estimate of his hit chance is 50%+.

If this seems like it would make them hold their fire too long, then make it ammo times hit chance equals .25 or whatever. But some reluctance to fire off every round at 4% range would seem to be in order.

(Objection - "Hey, just move him closer". Answer - they'd burn him; he has to wait for them to come closer, and they probably will which is why he is there. Won't do any good if he is out of ammo by then, though).

#2, what is the idea of scared guys in the open routing away from their own rides? Now, I understand, frazzled is frazzled and seeing armor, even friendly armor, moving toward you might be nerve-testing. But you'd think they might grok that an armored ride out of a hot place (if they rally that much - and cover and supporting fire in the meantime) is not quite the same as "private pancake" on the menu. I've parked an AFV between troops pinned in the open and an enemy MG, only to see them get so scared of their own ride they routed off the map - LOL.

Perhaps if troops associated any vehicle they've already ridden on, with safety rather than danger, might help things a bit. I can understand running away from a vehicle even as "likely to attract fire", but I expect a slightly greater sense of cooperation between a halftrack and the squad its driver and gunner are actually members of, in real life - as an example.

Tweaks, yes, but I would appreciate them.

[Michael Dorosh]

As I've said elsewhere TANKS DRAW FIRE. If you are an American GI and you spot enemy armour coming you are NOT going to go running to jump on the back of a Sherman because you know in your heart of hearts the damn thing is going to in all likelihood brew up - and even if not, it is going to draw a lot of nasty attention to it!

So in all likelihood you are going to RUN the hell away from it!

[Leonidas]

Ok, guys, I give up. No more suggestions from me. After reading similar threads, I have yet to read a serious discussion on this forum about ways to dramatically improve CM. Even most suggestions for minor tweaks are shot down.

My conclusion is that any suggestion on this forum on how CM could be improved will be interpreted as an attack on CM, and the loyalists will launch a counterattack. Satisfying as this must be for the loyalists, it does a disservice to CM's long term future.

Ironically, I would bet that this same attitude was a major stumbling block to the development of CM: "What's that? A wargame without hexes?? Blasphemous! If you think eliminating hexes would improve wargames, then you obviously don't know much about wargames!"

[Spook]

Leonidas, it's really unnecessary for you to feel discouraged already in offering your views on revisions. Heck, look at the topics started up by rexford. That guy has been throwing out all kinds of posts that challenge the "physics" and data presently applied to the CM direct-fire routines. And BTS has accepted some of rexford's commentary as a basis for future changes.

I don't fault the concept of "aggression" that you are looking for. I just happen to feel that CM already applies it, in different ways, to give your desired end effects.

In fact, some form of "fire control" on a PER-UNIT basis could be argued for with some validity, especially for defensive fire weapons like AT guns. Providing fire control as a unit menu option, though, increases direct player control, which I think we both know. And I don't think that BTS seeks added micromanagement to that level.

The bottom line here is that if you desire a future change to CM and seek discussion on it, invoke BTS directly for their views if necessary. I've done that a few times, and some BTS member is usually receptive.

[Tiger]

Leonidas, please grow up. This is called a discussion forum for a reason. If you'd stop and think about what people are saying instead of getting your ego bruised, you may accomplish more in the long run.

Tiger

[Guest Andrew Hedges]

Originally posted by Leonidas:

After reading similar threads, I have yet to read a serious discussion on this forum about ways to dramatically improve CM. Even most suggestions for minor tweaks are shot down.

My conclusion is that any suggestion on this forum on how CM could be improved will be interpreted as an attack on CM, and the loyalists will launch a counterattack. Satisfying as this must be for the loyalists, it does a disservice to CM's long term future.

Ironically, I would bet that this same attitude was a major stumbling block to the development of CM: "What's that? A wargame without hexes?? Blasphemous! If you think eliminating hexes would improve wargames, then you obviously don't know much about wargames!"

There are countless - really - examples of CM being improved by player input. Look for the Rexford posts, or a search on seating capacity of the M3A1, or do a search on the German optics thread, or hull/turret rotation, or Allied use of tungsten, or off-road speed of wheeled vehicles, or spotting rules for fast-moving vehicles, or crew spotting, or crew armament, or crew morale, or...well, they really are countless. And those only include the changes I remember off the top of my head since I got the game in October.

So it could not be farther from the truth that BTS doesn't value, or incorporate, user input; they do so more so than any other game I've ever heard of (which may explain why people sometimes seem so protective).

But to incorporate changes, you do need to make a better case than it would be neat, if...

Especially dealing with something that has been discussed as exhaustively as the TacAI.

(And I really do encourage you to do a search on the topics I've listed; it will provide a more accurate description of how BTS responds to input than the posts of the last week or so, which are not unencumbered by a certain amount of history).

But of course people will present compelling arguments for the status quo; people have different opinions, after all, and just because something is a new idea doesn't mean that it is better than how things already work.

[Vanir]

After reading a comment from Leonidas in another thread, I think what he really wanted, but did not explain quite right, is SOPs like in TacOps. Anyone who has played TacOps knows how nice they are and I've frequently thought they would be a welcome addition to CM/

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

You've never heard music until you've heard the bleating of a gut-shot cesspooler. -Mark IV

[Clubfoot]

Leonidas,

It would indeed be a shame if you ceased to post your opinions for game improvement.

Player input has been a (maybe THE) major factor in all the games changes, thus far.

Your posts clearly represent your opinion and you've managed to avoid acrimony.

Just because some of us don't agree with you, does not mean our opinions are any more valid than yours. We're simply (you included) engaging in debate. A good thing!

And most of us do so love CM (you know, like a lad loves a kissing cousin)that we do on occasion defend it's current merits quite vigorously.

Please don't interpret posts of diverging opinion as counter-attacks. They're (usually) not. The fact that a number of your respondents began with messages like "you've provided a good topic for discussion" and the like should be a sign of encouragement. Your topic has produced a number of thoughtful, lengthy missives in response. And that's a good thing as well.

Don't throw in the towel!

Clubfoot.

[This message has been edited by Clubfoot (edited 01-23-2001).]