Balancing out commanders and the commanded

[Battlefront.com]

Since the beginning of wargaming there has been a lively debate about how to simulate command decisions and command levels. Unfortunately, many wargamers do not really understand how simulations work and what impact one part has on another. While it is certainly true there should always be ways of improving a given design, it is also true that at some point there will come a point of diminishing returns. It can also happen that an ideal can not ever be reached unless certain other elements are changed, and therefore diminishing returns start pretty quickly if the other stuff is left largely unchanged. What I'd like to do is get a bit philosophical for a bit in order to try and get everybody on the same page. It's not going to work, but it's always fun to try

A simulation basically contains four elements which represent:

1. Entities

2. Environments

3. Interactions

4. Decisions

I don't care what kind of simulation it is, it contains these four concepts in some shape or form. The more simplistic each element is, the easier the sim is to make accurate. For example, to simulate a ball rolling down a playground slide you have the ball (entity), the slide (environment), physics of ball on a slanted surface (interaction), and how the ball is released (decision). If you have only one type of ball, one model of slide, and one way of releasing it things are a piece of cake because they are all pretty much in balance with each other. But change one part of the sim and you will either have to change others or be faced with an inaccurate model.

For example, having different balls of different weights or textures, possibly even imperfect roundness, would mean simulating the other things in more detail. The slide's surface would have to be simulated to interact with the different textures and the physics would need to account for the various properties and their values. The decision part, how the ball is released down the slide, could probably remain the same. But what if you introduced different ways of releasing the ball? Now you have to make the various decisions have different effects based on the type of ball, type of slide, and method of release.

The point I am trying to make here is that all elements must be kept in balance for a sim to function realistically. The more you deviate from balance between the elements, the less realistic the simulation is.

In a military sim there is a special dynamic between decisions and the other elements. The fewer entities the fewer decision makers need to be simulated. The more diverse the entities the more decision makers are needed. The less detailed the entities the less detailed the decision making. The opposites of each are also true. It is also true that the more detailed the entities, environments, and interactions the more decisions there are to make.

What this all means is that the smaller scope, lower level games (first person shooters for example) are better suited to single players. Ironically the opposite end of the spectrum is also better suited for single players, but for different reasons. Commanding a Platoon of soldiers is more difficult than commanding a Squad which is more difficult than commanding a single soldier. Commanding divisional sized units for a section of front is easier than commanding regimental sized units for a large operation, which is easier than commanding battalion sized units for a big battle. In between is the apex where things switch from becoming harder to becoming easier. Unfortunately for us, Battalion level is that apex.

At the Battalion level you have the best and worst of everything. You have all the detail of a first person battle sim with significant tactical and strategic elements. Pretty much everything has even weight from a sim standpoint. And that is where the trouble lies if the decision element is tied to one individual decision maker. In order to balance out the other elements there needs to be multiple decision makers. The more the better from a sim standpoint, the less the better from a game standpoint.

See the inherent problem we have as designers of the simulation? Arbitrarily limiting the decision makers to one entity, without adjusting the other elements, means that the system as a whole is inherently less realistic than it should be. The solution is to allow more decision makers or to reduce the other elements. Since the latter is impractical to do for a Battalion level sim, we're stuck with needing more decision makers if we want to get the maximum realism possible.

The possibilities open to us are mandatory multi-multi player or mandatory control of most forces by AI players or a combo of both. In all cases a single player would have a restricted role in the game. Additional players and extensive AI are both impractical (especially the latter), which means even if a player agreed to play in a highly restricted capacity there would still be significant problems for the sim (i.e. a shortage of players means no game, less than near Human competent AI means frustration). However, few players want that type of restricted environment in the first place so it makes the rest rather moot.

And so there we have it. The level and scope of CM's combat environment requires many commanders in order to be more realistic than it is right now. But players by and large don't want this. Therefore, players are going to have to accept that they are getting what they asked for, limitations and all. The only thing we, the designers and developers, can do is attempt to minimize the negatives of having a single player in command. And that means approaching designs that recognize the limitations of what can be achieved so as to not get into a situation of spending time on ideas that offer diminishing returns on development resources.

Hope this all makes sense to you guys

Steve

[Michael Dorosh]

Unlike wbs, I claim the pole position - reply 1.

Unfortunately, I have nothing of substance to say, but good luck.

[Tero]

If I'm reading this right you are talking what amounts to a RPG with X number of commanders in multiple levels of command and at least one of them is the human player.

Where does the human player sit ? On the top or anywhere in the chain of command ?

BTW: where, in your elements hierarchy, does the randomness (for example the mechanical functionality/reliability of the element, random changes in the environment, random factors in the interaction and timing of the decision) fit in ?

[Battlefront.com]

I'm not talking about anything specific to CMx2. I am instead simply talking about the dynamics of simulations in general. But yes, if you want a "perfect" Battalion level simulation the player would be limited to ONE role and the hundreds of others would be handled individually by other players, specialized AIs, or a combo of both. The more one tries for perfection, the more one has to accept the necessity to restrict the player. The less restriction inherently means less realisim. It is a dynamic that can not be worked around.

Don't confuse how a sim is designed from how it works once it is "played". In a sim design there is nothing that is random. Randomness is nothing more than a design with a wider array of possibilities.

Steve

[Runyan99]

Originally posted by Battlefront.com:

See the inherent problem we have as designers of the simulation?

Yes. At the battalion level, the player is:

1) The battalion commander

2) Every company commander

3) Every platoon commander

4) Every squad commander

5) Every NCO in charge of a weapons team or vehicle.

all at the same time. Thus introducing the problem of the Omniscient All Guiding Hand.

By the way, I hear the Emperor in the Star Wars mythology was able to use the force to achieve this kind of control over his forces in battle. Or maybe the Borg. Or maybe Sauron. Sorry, my geekdom is showing.

Anyway, as far as real life war goes, nobody else can realistically achieve this kind of cooridination on the battlefield. So yes, there is a problem as far as simulation goes.

The only half-solution as I see it is to devise some ingenious system of handicaps on lower level commanders based on chain of command limitations. What does that mean exactly? I'm not sure. It's very complicated actually, but you make some effort to simulate the speed of communications and command down the chain, while at the same time allowing split second decision ability at the squad and tank level. In other words, you weight the simulation somewhat more to the battalion level than to the squad or platoon level.

Good luck.

[Tero]

Originally posted by Battlefront.com:

But yes, if you want a "perfect" Battalion level simulation the player would be limited to ONE role and the hundreds of others would be handled individually by other players, specialized AIs, or a combo of both.

Can the human player, in your opinion, sit anywhere else than on the top if playing alone ?

The more one tries for perfection, the more one has to accept the necessity to restrict the player. The less restriction inherently means less realisim. It is a dynamic that can not be worked around.

Could you envision a simulation where the parameters (like lattitude in making decisions and how they take effect) the human player is playing along would be controlled dynamically by the AI ? Would that work in a commercial strategy/tactical simulation ?

Don't confuse how a sim is designed from how it works once it is "played". In a sim design there is nothing that is random.

From the design point of view, yes.

Being a simulation it would have to mimic RL where the chances for (seeming or genuine) random occurances are infinite.

Randomness is nothing more than a design with a wider array of possibilities.

The simulated world is perfect.

[dalem]

[Lloyd Bridges]

I picked the wrong night to START drinking whisky[/Lloyd Bridges]

-le dale

[Blutzeit]

The level and scope of CM's combat environment requires many commanders in order to be more realistic than it is right now. But players by and large don't want this.

WHAT?!!! It's the top 1 missing feature on my list. In CMx1 this is only possible (but cumbersome) in PBEMS. But it would be most advantagous in TCP play, having a team of 3 or 4 or more players controlling a company each, you would be able to play fairly large battles running at the fast pace only seen now in smaller battles.

[Kanonier Reichmann]

The way I see around the problem is to have the battallion commander (the player) being forced to assign general objectives to say 2 out of 3 of the companies within the battallion (assuming no further attached comapnies) while still allowing the player to fully control the remaining company. It's a bit of a compromise but allows for a certain level of scripting to be inputted by the battallion commander to set specific objectives to be achieved by a certain time for subordinate comapanies and then relying on the AI to come up with a decent enough plan to at least make a sensible go at succesfully achieving the objective(s).

If such a level of scripting can be allowed in the program and a decent enough AI can be relied on to carry out your orders then the player can have the best of both worlds. (S) he'll be able to see if his basic plan was workable from the end result of the battle while also having the hands on enjoyment of commanding one company (no doubt the best one) to try and get the most out of it to achieve final victory. You'll know the timetable that the other companies have been set to get where by when and can therefore personally issue orders to the units under your company command with this in mind.

Just a thought anyway. Might as well throw it up in the air to see how well it gets shot down.

Regards

Jim R.

[hoolaman]

I can see where Steve is coming from, especially regarding the degree of abstraction vs realism. I don't envy the guys working on a new game where each man is accounted for. When you account for each man, you then have to calculate where each tree or piece of cover is, exactly where bullets are going etc. etc. In current CM the abstraction of squads and teams and cover means that although the simulation is less thorough, it is more likely to coincide with reality and no doubt far easier to program. Making the environment smaller makes for more nit-sized pieces for people to pick.

The possibilities open to us are mandatory multi-multi player or mandatory control of most forces by AI players or a combo of both. In all cases a single player would have a restricted role in the game.

I would put forward another option, whereby a single player is able to take on the role of the several command levels sequentially. As in on every turn, firstly the player plays battalion CO and gives orders to formations XYZ. Next the player plays company CO, and gives orders to formations ABC but the game system binds him in some way to the orders that he gave (himself) at battalion level a second ago. And so on down to individual men if you are that dedicated/anal. This seems to me the only way to simulate a chain of command which should have twenty different decision makers and yet still allowing a single player per side, which is what a PC game must primarily be. I sure don't want to have to find six other players to be able to play a good PC game. It doesn't *need* AI to control friendly forces, and it allows the player to take on several roles which are separated by the game design.

You restrict the player, but don't remove him from any of the decision making process. Tedious and cumbersome? Maybe, but I think it *could* be made to work. I suppose you could use a "broad strokes" AI in the initial stages to plot the micromanagement of squads etc. So you may only use above company level orders in the the first few turns of the initial advance, relying on an AI to plot the moves of lower level elements just like a group move in the current CM, but when the action starts you may want to get down in the dirt to squad level.

What does this achieve you may ask? Well when the fighting starts, making batallion level adjustments stick would get much harder and slower, and squads at the front would be the most responsive element to adjust to new situations.

[ January 17, 2005, 03:41 AM: Message edited by: Hoolaman ]

[kipanderson]

Steve,

Your last posts on the old thread, plus what you have posted here do sound great. And very reassuring.

That the focus has to be narrowed to the exact scale you deicide to go for, and that what works for one scale, will not for others, is very true. In my teens, just with cardboard and the good old CRTs, Combat Results Tables, I and my chums used to modify all the wargames we played. What works for the operational level, battalions as the manoeuvre units, and what works for the CM level are very different.

“The level and scope of CM's combat environment requires many commanders in order to be more realistic than it is right now. But players by and large don't want this. “

This may be an example of the narrowing of focus, “if” it means that there will be no multi-player feature in CMX2. If it does mean that it may be that the C&C system you have developed, to optimise play for a single player, is just unworkable for multi-player games. If this is the case I can well understand it.

However, on the narrow point of multi-player games not being wanted by most players, I do think you are wrong

It is not at the top of my wish list, but I had always taken for granted the feature would be there. One reason why this puzzles me is that, in my view, it is the single biggest feature change to reduce the Borg Effect, God like view of the battlefield. However, another thing I have got wrong

Steve,

If it is the case that there will not be a multi-player feature, I have noticed that you may generally have underestimated the demand amongst CM fans to use CMX2 for some form of cooperative play. Large numbers of us have by now known each other, whether over net, or down the pub and such, for some years. We tend to congeal… like nagging groups of bugs and plan to use CM for more than just single human v human play. CMBB was already such a fine simulation, it already does single human v single human so well, it’s a form of military history. In many ways CM is already a done deal in single human v human play. Whether in the form of a meta campaign, or at the tactical level three or four players on each side in a live game, I think I speak for more people than you seem to imagine in hoping for cooperative play. Who knows

You need to leave your snow hole.. join us down the pub here in London and we will soon put some beers down your neck and sort you out on what should be in CMX2

All the best,

Kip.

PS. You knew you would not get away without being nagged over features… and you were right

[ January 17, 2005, 07:38 AM: Message edited by: kipanderson ]

[Holien]

Originally posted by kipanderson:

Large numbers of us have by now known each other, whether over net, or down the pup and such, for some years. We tend to congeal… like nagging groups of bugs and plan to use CM for more than just single human v human play.

Hi I can back up Kip on the desire for multi-player and some sort of campaign environment.

The ideal would be as described above having the ability to assign Company Commander roles to other players / the AI.

If however that is too much of a compromise I would not want to lose what you have done with the current engine as the majority of play is single player vs PBEM and that is the majority market.

As for congealing down the pup I have never done that and never will.

Not sure what Kip does down the pup??

H

[aka_tom_w]

I must completely agree with Kip when he suggests Multi-multi-player (mulitple players commanding each sides forces) would be one good way to reduce Borg Like spotting effects and make the game feel more realistic.

I suspect this feature is widely viewed as a good and desireable concept by many players here.

BUT here is the question....

Can the good folks at BFC design CMx2 to work well BOTH for single player head to head action AND multi-multi player??

This quote suggests to us that Steve sees it as an either/or situation....

"The possibilities open to us are mandatory multi-multi player or mandatory control of most forces by AI players or a combo of both. In all cases a single player would have a restricted role in the game. Additional players and extensive AI are both impractical (especially the latter), which means even if a player agreed to play in a highly restricted capacity there would still be significant problems for the sim (i.e. a shortage of players means no game, less than near Human competent AI means frustration). However, few players want that type of restricted environment in the first place so it makes the rest rather moot."

Steve's philosophical treatise above was very interesting and after having only read it once I am not sure the game can be wisely designed or developed to "swing" both ways (GOOD at single player vs single player AND multi-multi-player battles.

I know we are a terribly demanding bunch BUT according to Steve's Philosophical treatise above I think that dual capability is a heck of a lot to ask.

"And so there we have it. The level and scope of CM's combat environment requires many commanders in order to be more realistic than it is right now. But players by and large don't want this."

I think you might accurately conclude that players by and large want this option (mulit-multi player battles).

How and when we all find time to play is our problem

BUT you KNOW we (the faithful) will ALL buy the game anyway even if we only find time to play it in all its true multi-multi player GLORY only occasionally!

Thanks for the post, the new thread and ALL of the clarifications Steve! (I could NOT believe my eyes when I woke up this morning, NOW look ! I am late for work Again!)

-tom w

[ January 17, 2005, 06:07 AM: Message edited by: aka_tom_w ]

[dieseltaylor]

From my reading of WWII tactical histories for ultimate realism we would sit around for days at a time with very minor action at company level. : )

The other thing that comes out in alot of commentaries is how often companies disappeared off into battle and ended up somewhere where they were not meant to be. Or got decimated 9 times over as they tried to carry out orders which were not possible but Battalion said were.

As a gamer I am happy to have a game were my intelligence is reasonably accurate and I have a chance to win. That either side has a chance to win is also perhaps more ahistorical than the "sim realists" would care to admit.

In a sense the answer is already in the game in that if players agree that original orders stand until a command unit ot its subordinate unit/ section etc is fired upon.The officer is then allowed to re-issue orders. For other units they must continue until two turns later from the sound of small arms firing before re-ordering. Units you have established in overwatch would be allowed to unhide.

This could both increase command realism and diminish borg spotting effects.

[Richie]

Hi Steve,

If I understand you correctly, it's either mass multi player or a half decent AI?

I'll take the decent AI and the multiplayer options we already have with CM... if that's possible.

I think being able to play any game vs the AI like you're playing a PBEM with a friend would rock! If the 'near human AI' acts like a 'near human AI', then why would we need all the other human input?

I think most people yearn for greater human involvement because the AI, at present doesn't always reproduce this for them...

But Hey, what would I know?

Thanks for asking anyway Steve!

[aka_tom_w]

Originally posted by Richie:

Hi Steve,

If I understand you correctly, it's either mass multi player or a half decent AI?

I'll take the decent AI and the multiplayer options we already have with CM... if that's possible.

I think being able to play any game vs the AI like you're playing a PBEM with a friend would rock! If the 'near human AI' acts like a 'near human AI', then why would we need all the other human input?

I think most people yearn for greater human involvement because the AI, at present doesn't always reproduce this for them...

But Hey, what would I know?

Thanks for asking anyway Steve!

that sounds good

I think that Steve is saying we need to reduce our expectations about what game designers can feasibly do/design/develop around a single player assuming the role of all commanders....

I have read and re-read this section of the treatise (above) a few times now and thats what it sounds like to me....

We should not waste time trying to discussion solutions for the fact that one player MUST play all the roles in this wargame simulation:

"The possibilities open to us are mandatory multi-multi player or mandatory control of most forces by AI players or a combo of both. In all cases a single player would have a restricted role in the game. Additional players and extensive AI are both impractical (especially the latter), which means even if a player agreed to play in a highly restricted capacity there would still be significant problems for the sim (i.e. a shortage of players means no game, less than near Human competent AI means frustration). However, few players want that type of restricted environment in the first place so it makes the rest rather moot.

And so there we have it. The level and scope of CM's combat environment requires many commanders in order to be more realistic than it is right now. But players by and large don't want this. Therefore, players are going to have to accept that they are getting what they asked for, limitations and all. The only thing we, the designers and developers, can do is attempt to minimize the negatives of having a single player in command. And that means approaching designs that recognize the limitations of what can be achieved so as to not get into a situation of spending time on ideas that offer diminishing returns on development resources."

Note the suggestion in the LAST sentence we do not want to "get into a situation of spending time on ideas that offer diminishing returns on development resources."

so....

where does that leave us?

:confused:

-tom w

[A.E.B]

Steve

I myself have once faced this same issue of multiple command levels for a computerised combat system for resolving battles where the human player had control over a large number of units ranging from combat elements to the CHQ.

It wasn't a WW2 game (arrows, swords, cavalry, pikes & magic) but the issue of limiting the human player's ability to micro manage the whole battle (including BORG) applied.

The solution I came up with was only allowing the human player a limited number of orders per turn (it was a turn-based system).

The way a battle was conducted was as follows.

1. The human player deployed his forces and could issue unlimited orders in the setup phase to simulate pre-battle planning.

2. Each level of command (lieutenant, Captain, Major, etc) had a given number of new orders that they could issue per turn based on their experience/quality (NCOs where assumed to concentrate on carrying out those orders).

3. Therefore, each time a unit was given orders it used up some or all of the immediate commander's order allotment for that turn. Units would require new orders before they would change their current orders.

4. If a unit wanted to report an enemy contact to other units that were unsighted, that also required the consumption of orders.

5. If a unit had used up all of its own order allotment, it could draw on the order allotment of a higher echelon command unit if it was within contact/command.

6. That higher echelon unit would in turn draw on its higher echelon command for its order allotment.

The general rule was that there were never enough orders to change the orders for every sub unit in a turn (beyond maybe a simple single order like "HALT". This became particularly acute if the subunits were out of command or if the higher echelon command units where wiped out.

(In CM you could make a order node (move here) or single action (hide/unhide, set up a cover arc) equal one order).

This forced the human play to decide where to use their limited supply of orders, specially those of higher commanders whose orders could be used by a number of subunits.

It was also possible to give orders that took more than one turn, with the unit receiving those orders not acting upon them until the order was complete.

Occasionally events in the game would render an order obsolete before it was finished being given. Then the order would have to be scrapped and a new order given.

This system was intended to reflect both the limited ability of commanders to keep issuing more orders once a battle had started, as well as giving the AI more chance against the human opponent (the AI had no order limit).

This solution was by no means perfect, but the testers were routinely sweating and swearing as they tried to manage a swiftly changing battle without the ability to give every unit instant and complex orders each and every turn. And the loss of a senior commander was devastating due to the lost orders.

To give a CMX2 example of how the above could work, in a current PBEM I sent 3 independent tanks to support a platoon of infantry advancing on a small group of buildings surrounded by clumps of woods.

The initial orders were for the three infantry squads and platoon HQ to move fast as they crossed the open ground between clumps of trees, and then change to move to contact as they approached the target. The mortar and piat squads had slighly different orders. All told about 30 nodes for 6 units.

The tanks were to move on hunt to postions where they could cover the first 60% of the infantry advance.

A mortar spotter was targeted on the trees infront of the buildings where I though the enemy may have infantry.

Other platoons and tanks were given different objectives.

The first two turns passed without incident.

Then in turn three an enemy SP KOed one of the tanks and began shooting at the infantry (the other tanks were unsighted).

I wanted to (a) tell the remaining two tanks to move to a position where they could see the SP ( tell the mortar to drop smoke in front on the SP to blind it © get the two infantry squads being shelled to hide (d) order the other squad and the platoon HQ to move fast to cover (e) order the PIAT to sneak to a position where it could whack the SP if it adavnced.

Realistically, how long would it take the only available HQ unit (the platoon HQ) to issue those five orders?

1. Sgt Harris, run and tell those tanks that the SP is on hill 124 to the left of the trees. Tell them that we are dropping smoke to give them cover.

2. Squads 1 and 2 take cover and wait for smoke.

3. Squad 3 occupy those trees front right.

4. Mortar, drop smoke on that damn SP.

5. PIAT, take up a position in those trees by the laneway in case that SP moves forward. Keep Low.

While a trained commander would use hand signals and more concise orders, I doubt all these orders could be issued in the delay time given by CMX2.

Now imagine if that same platoon HQ could only issue orders to a couple of the units in the time frame.

Regards

A.E.B

[Tarkus]

It's very interesting to get insight from professionnal designers about the practical and theorical challenges they face. And I found it quite cool to open the topic for discussion like this.

Originally posted by Battlefront.com:

1. Entities

2. Environments

3. Interactions

4. Decisions

I really get your point of balancing these four. It all makes good sense. But assuming that keeping the best possible balance on these factors is a top priority from a design stand point, don't you feel interactions and decisions is where the commanding issue is most influenced ?

Originally posted by Battlefront.com:

What this all means is that the smaller scope, lower level games (first person shooters for example) are better suited to single players.

In relation to the earlier statement about focusing on element 3 and 4, I would say that lower level games are best suited for single player because of the "limited" interactions and decisions involved. FPS rarely has any form of complex interaction with other entities except in a scripted world, or for annihilating them. CM is quite the contrary, since the player is acting on the battlefield essentially via other entities. That is also why, on the other hand as you point out, battalion level game must really be a nightmare to properly balance since interaction and decisons must present itself almost like a maze. So I guess getting back to defining these elements may perhaps be of help.

From my limited knowledge, I'd be tempted to suggest it boils down to situationnal awareness and command & control features. Situationnal awareness, made from the player own perceptions and what his units sees (and report), as many implications has Steve mentionned in the other thread:

Originally posted by Battlefront.com:

Think about all the assumptions and reassurances one has with the current system. [...] picture all the things decisions you would make and note how many of them involve KNOWING something that shouldn't be known. Now think about the same situation assuming a more realistic knowledge and reaction system, then see how that might affect your decisions.

From this, I understand that the way the information is presented to the player as the basis of his situationnal awareness is where most of the evolution could be done in the great march toward a realistic-yet-fun tactical milsim. If you can successfully simulate uncertainty (while keeping it entertaining ), could it possibly help in tackling the problem ? It would shift some of the game "winning contingency" (forgive my total lack of appropriate wording, by that I mean the average playing conduct to win the game) from proper coordination of offensive ressources (already in CM1 but sometime based on not-so-realistic intelligence) to proper coordination of information gathering means (not quite in as of now). I, of course, loose myself in trying to envision the implication of such a change upon gameplay...

I guess we can then look at control (how the player can act and influence the course of his battle via his orders and own actions) as the other part of the problem. Hoolaman idea was mainly focused on this point IIRC. The idea he suggest on the current post also concern itself mainly with the control issue. But then again, the most challenging aspect of it all is to keep it fun.

Could modelling communications somehow be of help ? Or something as simple as not allowing players to be aware of the exact delay before a unit start implementing its assigned orders could be a step toward realism.

Yet this is anything but a change of paradigm.

Now that I come to think of it, reading Steve's view of the whole question, making a super realistic tactical simulator OR an award wining arcade wargame is a lot "easier" than making a realistic-and-fun sim like CM.

Cheers

[ January 17, 2005, 01:49 PM: Message edited by: Tarkus ]

[kipanderson]

Hi,

Yup…. I agree with others that is does sound as if Steve is saying that BFC can either optimise the game for the single player… or introduce multi-player features, but not both.

My prejudice is that some sacrifice of the ultimate that would be possible for the single player is worth it, if it allows multi-player features. My reasoning is this… not really reasoning at all… just justification of the prejudice

CMBB already does single player v single player so well that if the narrow definition of Borg Spotting were dealt with, all units spotting in their own right as opposed to on each others behalf, the focus of evolution of CM should really be in terms of “how we can use CM” as opposed to just features within CM.

The “how we can use CM” does mean some form of cooperative play. At the operational level I realise a genuine operational level in the form of a near operational game from which we can click down a level to resolve contact battle in CM is years away. For very understandable reasons. Massive amount of work. On this front my oft given request for the ability to fully edit Saved games would be huge help by itself. The CM community can do the rest.

At the tactical level, yes, a multi-player feature would be massive fun, even if we rarely had the time to use it. Two teams playing each other in live CM games would be a huge leap in fun and realism on the Borg Effect front. After all we would still be able to play CM games single player, just not quite optimised for single player as much as you currently plan.

If both fully edited Saved games and a multi-player feature were introduced both meta campaigns and live team play would near become a standard way to play. In my view.

Simply a leap in the fun ratings from single player games. It is a complement to you guys… CMBB is already so good at single player games that no matter how good CMX2 is, the leap would not be as great as to some form of cooperative play.

All very good fun,

All the best,

Kip.

PS. Meta campaigns are rare today because those organising them know only too well that the work load is near undoable. Make them doable.. and watch how they will take off. As for teams of friend wishing to commit live, mass slaughter of other teams of friends… the blood would flow in rivers

[Tarkus]

Kip is right. Option is really a keyword, as there are already many different ways to enjoy CM. An enhanced import troops/map feature could also be of help for a larger contextual vision.

It would be tempting to suggest that allowing co-op multiplaying as an option would tackle the command problem somehow, but this is where the experience and knowledge of how a game like CM is developped and put on the market override everything else. Is it worth the effort, dollar wise ?

I, for one, haven't the faintest idea.

[aka_tom_w]

Hi Kip

If the design choice has to be made for Single player vs multi multi-player and it is a black and white either or choice I am afraid my vote would be for single player with all its limitations based on the single hard wired FACT that one player MUST play all roles.

I must show my own prejudice here as I would like to speak for most Mac users with new computers running OS X. NONE of us are player CMxx games anymore on the computers we use EVERYday. Playing CMBB or CMAK for fun on an OLD OS 9 Mac does not cut it for me and I am not interested in only playing CMx2 when I can join a group and play in huge multi-multi-player scenario or campaign. BUT am of course lobbying for the OPTION to join multi multi-player play when time permits.

We Mac users are just dieing to get a NEW game (a NEW GREAT BFC game like Cmx2!) to play on the latest hardware with the latest graphics that will be an experience akin the FUN and excitement of playing CMBO for the first time!

I am REALLY hopeing CMx2 will be available on the MAC (OS X) and it will let me, first and foremost, play strictly vs the AI (for warm up) and it will let me play head to head against at least one other human player. Team play would be GREAT but I am sorry, its not a show stopper for this Mac user.

BUT I totally agree the multi-player team play is ONE good way to reduce the Borg problem.

-tom w

[Sirocco]

The game environment is an abstraction. The goal is to make those abstractions correspond logically to actual events.

When that isn't achieved it breaks the experience, as if you could see the puppeteers strings. The outcome might be the same as before, but the magic has gone.

A designer therefore has to be careful that the solution employed doesn't just work in practical terms, but the mechanism used must be believable.

I look upon the orders issued by the player as battalion commander as a believable abstraction, even when assigning detailed orders to the squad level.

You could argue that a battalion commander might, for example, only order Coy A to take a position while Coy B moves into a flank position to provide supporting fire.

What we miss is the battalion commander suggesting that Coy A attack with two platoons forward, one back as a reserve to exploit. He might also have attached mortars to Coy A with abundant smoke ammo. That would suggest to use smoke to cover the attack.

He might also instruct Coy B to watch his left flank in case of counterattack, and to take up a position to fire on a particular location.

Therefore the nature of the battalion commanders orders might well shape the orders of those commanders, even down to squad level, beneath him.

Without a mechanism to impart those detailed orders reliably, and enjoyably, for the player, the current abstraction is required.

The question is, how much independant action is allowed for subordinate commanders, and how should command and control be implemented.

Independant action is already in CM1, in terms of units going to ground under stress. A leader should also have the awareness to disobey an order. For example, a unit ordered to cross an open field against a spotted and unsurpressed MG should simply not move.

A command system must represent delay. The delay from the issuance of verbal orders, and orders delivered by radio or runner. That also includes orders to stop an action.

Ideally there should be a delay in the reporting of information back up the chain of command. For example, sighting reports.

That would, of course, lead to a more confused, and therefore realistic, representation of combat, which some would appreciate, and some would not.

I think, from the hints given, that we are moving in that general direction. And for that reason I also expect to see a lot of debate in the same context as CMBB was debated after it's initial release.

[Dirtweasle]

Originally posted by A.E.B:

Steve

I myself have once faced this same issue of multiple command levels for a computerised combat system for resolving battles where the human player had control over a large number of units ranging from combat elements to the CHQ....

That was a pretty good solution you came up with.

[kipanderson]

Tom,

I am in no way suggesting a true either or. I am sure that CMX2 will be agreat game with or without multi-player feature. I had just always taken for granted that multi-player features would be there.

If you take the classic three infantry platoons and one tank platoon clearing a village, all commanded by different players, the leap in both fun and realism on the chaos front would be huge. Think of the near perfect timing we can currently arrange in the covering fire tanks but down as infantry advance on buildings. Hugely more challenging if the tanks were commanded by different players to the infantry… also far more realism.

No… all I am saying is do not sacrifice multi-player features for other features in CMX2. It is a matter of priorities that is all.

Wargames of this scale are particularly well suited to the multi-player features because of the added chaos they would cause.

Lead to a strategic leap in fun

All the best,

Kip.

[Jaws]

Ok if I understand this right it is a problem to delegate a certain level of command to an human or AI without negative consequences for 1 of the 4 subjects?

If that is so what technology did the boys from Steel Beast Pro or Steal Panthers (out dated) use? In Steel Beast Pro it is possible to delegate individual vehicles to even a complete company to a human or AI in a multiplayer game. It is even possible to switch platoons between company’s.

All this doesn't effect the

1. Entities.

2. Environments.

3. Interactions.

4. Decisions environment.

The game looks perfect and has the same dept as CM. Only it is not turn based.

Not critical here but just looking for reasons why other games can do this?