Defense in CMBO and CMBB

[Specterx]

I know it's been discussed and well-beaten before, but the problems with defending in CMBO are still there, and from the looks of it, it doesn't seem like they'll be fixed in CMBB.

In short, defending in combat mission is, IMHO, too difficult. As long as your opponent uses some basic tactics (recon, smoke, narrow front advance, etc.), it's damn near impossible to win on the defense. The primary reasons for this are relatively well known:

1) Simplistic spotting. A concealed ATG 500m away is exposed for all units to see the moment it fires. Soldiers wearing camo in woods are spotted at 200m. Etc.

2) Few of the advantages enjoyed by RL defenders are given to CM ones. For example, there is no advantage for occupying a single space for a length of time, when IRL defenders could create thoroughly hidden positions, sight weapons etc. Since the attacking and defending troops are equal if foxholes are taken out of they equation (and they don't do all THAT much...) the defenders will always loose unless that player is very lucky or the attacker is simply inept.

3) The use of defensive equipment such as mines, wire etc. is not really encouraged, and in some ways discouraged. CM in a way assumes that barbed wire and mines are used in place of actual men, since both are drawn from the same pool. Given the choice between a few minefields and, say, an ATG, I'd take the ATG, since the mines have an overall lower probability of being useful at all and would be likely be less useful than the ATG.

4) Computer-generated maps. No water/bocage, scattered objectives, lack of map depth vs width...

5) Unrealistic modeling of certain weapons. Yes, it's a sin to mention it, but I feel confident that CM could be made much more realistic by simply adjusting the way some weapons work. Mortars in particular seem to have an effect more like that of a popgun than an 81mm bomb going off in your face. I don't know what the kill range of an 81mm mortar is, but it seems that it would cause many casualties if one hit 4-5m away.

MG's also feel "wrong". I know that tests have been done to demonstrate how MG's are designed to supress and not cause major casualties, but it seems that a pair of HMG42's firing on a platoon in an open field (or even under cover!) would cause the entire platoon to immediately hit the dirt and start crawling/running to the nearest cover. Right now they're sort of an annoyance but rarely cause much trouble unless they fire on a single enemy squad for 2-3 minutes.

(1) has an obvious solution - make spotting more imprecise. Guns in particular should be able to get off 3-4 shots without being spotted, or even more depending on the distance from the gun to enemy troops and what sort of cover the gun is in. For infantry contacts, the "Infantry?" marker should be retained, but it might be nice if it didn't give so much info about the enemy force, e.g. show a grey infantry figure standing still that "jumps" occasionally as the spotted unit moves until the contact firms up.

I can also think of an easy solution for (3), drawing somewhat on the system devised for use in the CMMC. The defending side in an attack battle would be given a certain amount of constuction materials (CoM) depending on the number of points in the battle (say 1/10, e.g. 2000pt battle = 200 CoM). Obstacles and fortifications would cost CoM rather than normal points. This could simulate the preparations by defending troops.

(2) requires some more thought...

I hope that some of these concerns will be addressed in CMBB - after all, the Germans didn't just roll over the Soviet defenses at Kursk.

[Gyrene]

I agree with much of what you said, and at least the MG gripes are getting fixed for CMBB.

The spotting problem is not likely to be fixed until afte CMBB

Gyrene

[LC-]

I dont see how the spotting problem can be fixed. If they did tweek it somehow then they would have to answer to "well I had a squad spot a AT gun but a squad 2 meters away didnt see it. FIX THIS!!" Its impossible to simulate someones vision. LOS is the best IMO that you can get for spotting. The other defensive problems I agree with.

[Viceroy]

Well, I'm obviously not a "pro" like you guys, but I win more often on defence than attack

I understand the points you're making but I think for more average players defence still has many opportunities. Obviously, having the right terrain is vital. I mean, if you can place your troops in a nice reverse slope postion things are pretty easy for the defenders. Being able to place guns in good keyhole positions, is also dependent on the luck of the map. On the defence you usually have the advantage of TRP's.

Maybe it's my style of play, but when I look at a map I always see the defending possibilites first and then the problems facing the attacker. Perhaps it's because I normally play smaller 1000pts games.

To try to give the defender more of a chance you could limit the number of turns, making the attacker have to hurry and not have the time for a full recon. Also, you can limit the amount of art available to the attacker so they can't have such a massive smokescreen.

Spectrex, I should put my money where my mouth is and let you challenge me to a PBEM game .... I'll defend naturally and you can show me how easy it is for the attackers!

[Schrullenhaft]

Most of these request can't be adequately fulfilled until the engine rewrite.

1) Movement is actually one of the more likely ways to give away a position than an insufficient amount of camoflage. The perception of movement is our of our more acute visual capabilities. This problem basically boils down to Absolute Spotting and Relative Spotting - Unfortunately will have to wait until the engine rewrite before Relative Spotting can be implemented and it will be quite complex to model and code up correctly. Some of the adjustments to the current engine (the revision of it in CMBB) that you have suggested may be harder to implement than you think since a number of the features depend more on a sort of 'balance' than some easily assigned, constant, arbitrary value. Adjusting one set of spotting circumstances may unrealistically affect another spotting situation. Relative spotting takes on more importance, and its absence is more recognizable, with a 3D game than most of the 2D wargames that have come before CM (since LOS is much more complex, but highly desirable part of game play). However, there are some improvements, but none that approach the level of a relative spotting model. There is an 'Extreme FOW' option now that doesn't give as much info about units in contact that you get with CMBO currently (you won't be able to tell the experience of troops you're up against and the number of casualties, etc.). You'll also get more misleading/unknown info where you won't be able to accurately/dependably tell the stance of a unit (retreating, moving, taking cover, etc.) as easily as CMBO.

2) This is an admitted weakness with the system which does have several of its roots in the current absolute spotting model. Since there will be more 'fortification' types in CMBB maybe this will addressed in a mild manner, but not enough to please the criticism of the current system in CMBO.

3) This will be addressed for QB's by reducing the prices of defensive equipment. It will still come from the universal pool of points for all units, but it should allow for higher concentrations of field fortifications.

4) QB maps are being adjusted so that they aren't 'all width and no depth'. They may not please everyone, but they are an improvement over the current computer generated maps for QBs. I believe the ability to import a map for a QB should also be available.

5) The MG issue has been addressed so that they distribute their lethality and effects over more units and area than they used to. MGs should be quite formidible now, especially in the defense. I'm not sure what to say of the explosive effects of mortar shells, etc. One of the biggest advantages of mortars was their low velocity which made their approach audible only a few seconds before impact, thus being able to catch their targets in a less prepared and more vulnerable stance and create some quick casualties on the intial barrage. I believe BTS has considered the size of the round, the amount of HE filler and the fragmentation effects of the most common shells used for particular calibers of mortar tubes. The blast radius and its effects may be a bit arbitrary in their values, but all HE shells are measured on the same scale (to my knowledge). Increasing the lethality/effects of a mortar shell may possibly lead to unrealistic effects in the lethality of larger caliber artillery shells.

In short there are a lot of improvements in CMBB. So many in fact that the game will play a bit differently from CMBO (for the better). However there aren't enough improvements to cover all criticisms of CMBO, so there will still exist unrealistic effects in CMBB. That's acknowledged, and even the engine rewrite most likely won't be able to accomodate everyone's desires for the game. CM is evolving, just not at the rate that some (unrealistically) desire.

[Vanir Ausf B]

Other people have addressed the original poster's points well, but I'll add a few thoughts.

One thing that could be improved in CM is the near impossibility of hiding tanks in LOS of enemy units. I realize this is in part a engine issue that cannot be fully addressed at this time, but I have suggested previously that tanks that begin the game in trees be made significantly more difficult to spot until they move, at which time the bonus disappears for good. As it is tanks are just not as cost effective as AT guns on the defense.

Making the QB maps deeper is a huge plus, but it will all be for naught if the flags are not staggered more front to back allowing for true defense-in-depth. This would be a logical change to make with the deeper maps, so I am hopefull.

[Caesar]

If mines etc are used well with the assistance of helpful terrain they can be deadly. I have won two of the last three games where I was defending by purchasing mines etc to guide my opponents forces into a fire sac. I would rather they were a bit cheaper though or covered a bigger area. Currently I am being forced to consider which is the most likely point of attack and defend that with most of my force and hope that I am not wrong. (In the third of the above mentioned games the attacker decided to attack elsewhere. Not pretty!)

It is the vunerability of the Pillboxes that I find most annoying while defending. I have lost countless numbers of them to useless little light tanks and armoured cars etc. When attacking I often buy a couple of stuarts or daimlers or the like just in case I come across them.

[Cpl Carrot]

The old AT gun will survive a lot longer in the smaller battles, mainly due to there not be 10000 eyes looking at it. (By small I mean 1000 pts or less).

My two cents.

Cpl Carrot

[Graaf Spee]

Something I've been thinking about is that the defender can't choose where he will defend. Some maps are great and some are just impossible to make a stand on.

So how can that change? One solution is of course to use homemade maps in a QB. But what I've been thinking about is a little more random. Lets say the map should be 1.5km wide and 1km deep. The defender should then get a map that is 1.5km wide and maybe 4km deep. Then he gets to choose where on that big map the actual playing area should be placed. It's still random and you can get stuck with very bad terrain, but I think the defender would have a better chance. I also think that would be a little realistic since usually (but not always) he who defends would choose to make his stand where he would have he greatest possible chance of holding.

Comments?

/Kristian

[SlowMotion]

I think this would be a very interesting new feature. But it shouldn't be just for defense QBs. In a MeetingEngagement players could make a deal that one player can decide map generation parameters (hillines, trees and so on) and the other player can choose the place from this bigger map.

[Chupacabra]

IIRC you'll also be able to properly entrench on the defense, rather than just starting with foxholes.

[Redwolf]

<blockquote>quote:</font><hr>Originally posted by Specterx:

In short, defending in combat mission is, IMHO, too difficult. As long as your opponent uses some basic tactics (recon, smoke, narrow front advance, etc.), it's damn near impossible to win on the defense. The primary reasons for this are relatively well known:

<hr></blockquote>

Well, contrary to what others said, I think that low-point defenses are very diffucult, at least when you don't play Fionn-75 or so.

It is just impossible to buy stuff that can withstand both a Churchill-VIII company attack and a human wave of small-squad British Glider troops with no armor (except 5 Wasps).

<blockquote>quote:</font><hr>

1) Simplistic spotting. A concealed ATG 500m away is exposed for all units to see the moment it fires. Soldiers wearing camo in woods are spotted at 200m. Etc.

<hr></blockquote>

This is very hard to fix. Even the relative spotting in the engine rewrite as describben by Steve in some post will only be a partial solution. Once somebody spotted the gun, the player will still be able to drop area fire on it or direct a platoon there which in real life would have no idea that easy prey is waiting.

<blockquote>quote:</font><hr>

2) Few of the advantages enjoyed by RL defenders are given to CM ones. For example, there is no advantage for occupying a single space for a length of time, when IRL defenders could create thoroughly hidden positions, sight weapons etc. Since the attacking and defending troops are equal if foxholes are taken out of they equation (and they don't do all THAT much...) the defenders will always loose unless that player is very lucky or the attacker is simply inept.

<hr></blockquote>

For me, the defender's lack of stockpiled ammo is most annoying.

Also, foxholes in CMBO do not offer very much cover (see recent thread by Wreck on the T&T forum) and if they would offer much stronger cover, they would do so in every direction, where in real life a good foxhole in good only in one direction and for the other directions it will ether offer less cover or handicap the occupant's fire.

What I think would be a very cool feature is that the defender can buy "terrain change points". You buy 10 such points, and you can change ten things in the terrain, as in the editor. Make that spot a little higher, that a little lower, place a wall there (stack sandbacks), reduce that woods patch to scattered trees, blow up that house, plant some flowers.

<blockquote>quote:</font><hr>

3) The use of defensive equipment such as mines, wire etc. is not really encouraged, and in some ways discouraged.

<hr></blockquote>

The stuff is just too expensive. Also I want cover from the roadblock, like a wall gives. Since it doesn't move, that *should* be implementable.

<blockquote>quote:</font><hr>

5) Unrealistic modeling of certain weapons. Yes, it's a sin to mention it, but I feel confident that CM could be made much more realistic by simply adjusting the way some weapons work. Mortars in particular seem to have an effect more like that of a popgun than an 81mm bomb going off in your face. I don't know what the kill range of an 81mm mortar is, but it seems that it would cause many casualties if one hit 4-5m away.

<hr></blockquote>

I think CMBO mortars only suffer from too few ammo, especially for defenders. And maybe they uffer from overly determined attackers, mortars shells will confuse a real attack without communications on the run much more than the "yes, you go there, and I can promise you the other buddies will be there" CMBO attackers. Green troops advertised here.

MGs have been discussed.

<blockquote>quote:</font><hr>

(1) has an obvious solution - make spotting more imprecise. Guns in particular should be able to get off 3-4 shots without being spotted.

<hr></blockquote>

That is no solution, making spotting just harder. Because if you approach with a single unit into a gun LOS, and it shoots at you, it will be almost impossible to detect it. That will lead to a very unrealistic "spotting advance by mass" behaviour. That is no good, a single recon unit shot at by a gun should have a fair chance to spot it.

The only practical solution here is to make all units not directly shot at or directly in the shot path "deaf". Only units the gun barrel aims at have a chance to spot the gun.

That is still a very artificial thing to do, as there is no reason why the guy standing 20 meter besides the gun position shouldn't hear it and open fire on his own, but I guess that this distortion is less annoying than the immedeate total exposure of guns we have right now.

Still, even with that system in effect, once the gun is actually spotted (by the units in the shot path), it is more dead than in real life, due to unrelated indirect fire dropped there and unrelated units directed there.

The next step to a solution would be not to show the position of a spotted gun to the player. The unit which spotted the gun gets a marker "I am tangling with a gun", and that's it. But such a systems would not fit CMBO's play style, it would rather fit into a wargame simulating a real commander position by way of maps and chatting staff, but no own observation.

[Herder]

I wouldn't mind seeing an option to use Marco's rules for single player engagements where CM forced the player to use the rules.

I think it would add a lot.

Basically marco's rules (can't remember where I saw them) were that you could only use the +/- keys to select units, that you could only use levels 1/2/3 view mode basically, weather options forced on, etc, etc.

This would go a long way to making the battlefield more "mysterious".

[zahl]

<blockquote>quote:</font><hr>Originally posted by redwolf:

Even the relative spotting in the engine rewrite as describben by Steve in some post will only be a partial solution. Once somebody spotted the gun, the player will still be able to drop area fire on it or direct a platoon there which in real life would have no idea that easy prey is waiting.<hr></blockquote>

The player should not be aware of the gun just because somebody has spotted it. He would only know that the unit is firing at or can see something, or not even that. This would be confusing and make coordination incredibly hard, as it should be.

The gun would be identified as Gun? only after more troops have relatively spotted it on their own.

[Ligur]

I don't know about you guys, and I am pretty much of newbie with only half a year experience of this game, but damnit if I always spot the guns. I really don't. My infantry will have to be a lot further forward then my armor is if I want to make sure some of my nastier TCP/IP buddies don't have a Pak75 somewhere.

Example: In a recent fight a Pak75 took out a flanking tank, and two armor pieces fighting in my main line before finally opening up on my infantry and being spotted. It shot my flanking light tank to the side from under 100 meters and the surviving crew didn't spot it at any point... Wow? I guess "panic" status makes it harder to spot stuff, is there any truth in this?

But, I'm only a poor newbie. This far I've thought the guns are pretty hard to spot. Tho, playing big games against the computer, I can remember how taking a gun from hiding usually results in 5 PanzerIV's opening on them the second they fire off the first shot only too well... Guess that is because of so many more eyes to spot them, like someone said earlier.

[Redwolf]

<blockquote>quote:</font><hr>Originally posted by zahl:

The player should not be aware of the gun just because somebody has spotted it. He would only know that the unit is firing at or can see something, or not even that. This would be confusing and make coordination incredibly hard, as it should be.

The gun would be identified as Gun? only after more troops have relatively spotted it on their own.<hr></blockquote>

That still returns us to the original problem, that it is unfair for a single tank, or a tight pair of tanks being shot at by a gun. Those tank(s) need a fair chance to spot and engage the AT gun. What you say is more like the sound contact now, but nobody can shoot at it.

What we could need is a unit shooting at a gun, so that the unit knows where the gun is, but the player gets only told that there is a gun, but not where. That would be a dramatic break in the 3D environment of CMBO, as you basically had to have unspecified markers for the own vehicles turret position and hull rotation. Think about it, the tank which fires on the gun will rotate towrads it, but the player must not know where the tank is facing. That is a dramatic change from now.

[Mud]

Ligur --

Crews have artificially reduced spotting capabilities. This, as well as high point costs, should encourage players moving crews to the rear ASAP, rather than using them to spot for other troops which can still fight effectively.

Redwolf --

On confusion... does separation from other troops (not just HQs) via such elements as smoke and distraction (e.g. the other friendly squads being busy avoiding arty, or intervening fire even?) affect morale and command delay?

For instance, if one squad isn't under fire itself; and rather than isolated, it's near one other squad -- but the second squad is currently fleeing because it's being hammered -- would the first squad suffer any penalties versus being a squad that is nearer the rest of the platoon? Should it?

[zahl]

<blockquote>quote:</font><hr>Originally posted by redwolf:

That still returns us to the original problem, that it is unfair for a single tank, or a tight pair of tanks being shot at by a gun.

<hr></blockquote>

I can't see how.

<blockquote>quote:</font><hr>Those tank(s) need a fair chance to spot and engage the AT gun. What you say is more like the sound contact now, but nobody can shoot at it. <hr></blockquote>

Just because _you_ would not see what your units' can see does not hinder their ability to spot and engage in any way. Tanks engage enemy guns on their own even now.

This suggestion however makes it impossible for the player to manually optimize the unit's reaction and eliminates the two abuses you described: the player dropping area fire on the gun after somebody spotted it or directing a platoon to take it out.

[Redwolf]

<blockquote>quote:</font><hr>Originally posted by Mud:

Crews have artificially reduced spotting capabilities. This, as well as high point costs, should encourage players moving crews to the rear ASAP, rather than using them to spot for other troops which can still fight effectively.

<hr></blockquote>

Not to speak of the points they are worth when killed or even captured.

<blockquote>quote:</font><hr>

On confusion... does separation from other troops (not just HQs) via such elements as smoke and distraction (e.g. the other friendly squads being busy avoiding arty, or intervening fire even?) affect morale and command delay?

For instance, if one squad isn't under fire itself; and rather than isolated, it's near one other squad -- but the second squad is currently fleeing because it's being hammered -- would the first squad suffer any penalties versus being a squad that is nearer the rest of the platoon? Should it?<hr></blockquote>

I don't know that. Morale calculation in CMBO is complex. Obviously, there is some kind of local morale that affects units in one area only, which is used for surrendering or freeing prisoners and probably for flag ownership. I never bothered to find out like I did for victory point calculation, as it is more complex and I think it would be bad for playgame if people knew "if I move just one more meter, that squad will surrender" or if people knew *exactly* when that flag changes ownership.

In practice, I find isolated squads to be in trouble anyway, so it doesn't really matter

[Redwolf]

Zahl, your answer to my posting is ambiguous.

There are two possibilities when you get a generic marker like the current sound contact marker, only that it stands for a unit "slightly" spotted.

1) This marker is not precisie in location and cannot be engaged by anyone, like a sound contact.

2) This marker is being engaged by the unit (and only by the unit) that spotted it, however the firing units doesn't show the player where it is shooting at.

1) has the obvious problem that it it unfair to the tank being shot at

2) has the problem that CMBO generally has player-visible unit facing and turret position. You would have to make that "undefined", which is a pretty big step in the current CMBO logic.

Personally, I like option 2), but I seriously doubt that BTS would accept such a serious breakup of the otherwise coherent game design to partially solve the absolute spotting problem. Mind you, before 2) makes any sense, you will also have to break from reality in that only units in the shot path get a realistic chance of spotting, and that should obviously apply to a different degree if the shooter is a gun, a vehicle or a multi-man infantry unit. This will be quite a mess for a few patches before it is ripe.

[zahl]

<blockquote>quote:</font><hr>Originally posted by redwolf:

2) This marker is being engaged by the unit (and only by the unit) that spotted it, however the firing units doesn't show the player where it is shooting at.<hr></blockquote>

This is more or less what I'm talking about, but there doesn't need to be a marker to start with (that the player can see). Even without the visible-to-the-player marker the unit that has spotted the target is himself aware of it and can engage it.

<blockquote>quote:</font><hr>

2) has the problem that CMBO generally has player-visible unit facing and turret position. You would have to make that "undefined", which is a pretty big step in the current CMBO logic.

<hr></blockquote>

No need to make that undefined. You would quite accurately know the target area, but you wouldn't (yet) know _what_ the unit is firing at. Just observing turret facing and impact locations would give you an idea where the target is. But since you don't yet know the target type, it would be unwise to drop and potentially waste area fire on it. Likewise it could be unwise to distract a platoon to "take it out" because you don't know what it is.

Ie. a player would preserve his own observation, but this observation would be delayed, not immediate like now. Like I said, the player would gradually improve his observation when more units become aware of the target due to their own relative spotting. Because of this delay, your two gripes (unrelated indirect fire dropped there and unrelated units directed there) would not work.

<blockquote>quote:</font><hr>

before 2) makes any sense, you will also have to break from reality in that only units in the shot path get a realistic chance of spotting, and that should obviously apply to a different degree if the shooter is a gun, a vehicle or a multi-man infantry unit.<hr></blockquote>

I don't see why these two are necessary if relative spotting is implemented.

[ 12-31-2001: Message edited by: zahl ]</p>

[Redwolf]

Zahl,

so what you want is correct orientation of tank and turret, but no marker for the target and no hint what the tank is shooting at. I don't think that will do any good, the only difference is that you don't know in what distance from the tank the target is and what kind of target. Both will be pretty easy to guess, IMHO.

<blockquote>quote:</font><hr>Originally posted by zahl:

I don't see why these two are necessary if relative spotting is implemented.

<hr></blockquote>

I guess that is what makes our opinions different: non-absolute spotting as implementable in a CMBO-like game and as describben by Steves is not that much different from the absolute spotting we have now. It will make the gun more safe for a short time after it is being spotted (because other units will not shoot on it automatically), but after the next player plot phase everyone and their dog with still rain area fire on the gun's position.

You seem to overestimate what this will achieve.

[Andrew H.]

Certain types of defenses could be represented in CMBB without an engine rewrite by simply adding a new battle type. I brought mentioned this in an earlier discussion of the engineering battle. Look at Breach.

The essence of this battle type is not the points ratio between attacker and defender, but the limitation/restriction of what can be purchased, particularly by the defender. Thus, in a 1000-point Breach type battle, both the attacker and the defender might have 1000 points, but the defender might have to spend 400 points on fortifications. (Of course, for this to work ideally, the game would need to discriminate between manned (pillboxes, bunkers) and unmanned (mines, wire, trenches (?)) fortifications.

Having a battle type like this would not fix difficulties caused by absolute spotting, etc., but it would permit the use in QBs of certain types of common battles, especially on the eastern front.

[Schrullenhaft]

Unfortunately even Option #2 is still not capable of being coded into the current game engine since it is Relative Spotting. The engine is not currently geared to having just one (or a few) units spot and engage a target and not let the whole battlefield know about it. To have spotting and engagement limited to specific units is at the heart of the Relative Spotting model. Steps will be taken to accomodate some of the current spotting models in CMBO (Some of which I mentioned above), but not to the extent that Option #2 would model. Right now any changes to spotting become universal, so any delays in spotting would apply to other spotting issues, which may not be realistic in all situations.

From what I gather of Steve and Charles opinions, not having the turret or hull point in its actual direction is too much of a disconnect with the way things are currently modelled in CM. It goes against the current design where you have accurate info on your individual units. Changing this would have gameplay implications that a number of players wouldn't appreciate (though it might be a bit more realistic, from one viewpoint). Issues like this will probably come up during the attempts to model Relative Spotting in the new engine.

[Specterx]

I don't think the solution to the spotting issue needs to be really fancy. IMHO just make everything a little harder to spot. I don't have a problem with my ATG being knocked out by artillery the turn after opening fire on an enemy armor column. Trading a 50-point ATG for a 150-point armored vehicle is perfectly fine with me. The problem occurs when my ATG ambushes a tank and immediately has everything on the map taking shots at it. Once an ATG becomes PINNED the chances of a hit plummet and the gun has been wasted.

As far as infantry goes, I think spotting needs to be less absolute in the sense that the spotter shouldn't be able to identify infantry with 100% accuracy under all conditions. Infantry moving in open field should be easy to spot and identify, but inf moving around in a forest should be nearly impossible to see and identify.