Change to the Interface

[dalem]

Originally posted by Andreas:

I would however like to understand a bit more about how squads/platoons/companies move in various situations before I accept the reasoning that short movements in the presence of the enemy are less complex to execute than long movements in the absence of the enemy. I would e.g. think that moving around an obstacle is very easy in a safe situation, but if enemy presence is suspected, it is very complicated, because you have to assume the enemy has the obstacle covered somehow - check for mines, closely observe potential ambush locations, send a scout forward to see if a machine-gun is trained on the 'easy' path around the obstacle. All these things cost time, and we have to assume they are currently abstracted into the movement somewhat.

Hmm. Is it safe to assume that at the scale of CM, every minute of every scenario is assumed to be in the presence of the enemy, therefore every path must be suspect?

-dale

[Andreas]

No, but I don't assume that in my post, or anywhere else.

[Vanir Ausf B]

Originally posted by Andreas:

</font><blockquote>quote:</font><hr />Originally posted by Vanir Ausf B:

In comparing complexity of long and short movement I did not specify whether it was in the presence of the enemy or not so I'm not sure where you got that.

Here:

Originally posted by Vanir Ausf B:

It could just be my own play style, but longish movement orders become unusual after contact with the enemy is made and if large numbers of movement points are used they are bunched together around an obstacle.

</font>

[Battlefront.com]

Sergei,

No, I read what you wrote very clearly. You said that 2 extra waypoints for a Conscript, out of C&C, added 45 seconds. You said it shouldn't take that long to figure out how to walk around something. I said they shouldn't even be able to move, at all, in 45 seconds. Nor in 2 minutes, nor possibly even in 10. You guys keep dancing around this. What you are asking for is unrealistic responsiveness to the player's whim. We feel there is already way, way too much of that as is, so there has to be some attempt to reign it in.

And I disagree completely with Vanir that the system, as is, is less realistic overall. Although I don't think the system we have is perfect, it is far better than the CMBO system where movement wasn't punished hardly at all. Going back to the CMBO system is, therefore, not an option. Going to something else better would be. But still... nobody has come up with an alternative that is any better (doing nothing is worse, so that is out).

Sergei said,

Well, for instance, if the order to move five meters to left and six meters forward wasn't treated with the same delays as orders to move five hundred meters to left and six hundred meters forward, I think it would be a start.

I'd agree with Andreas that you'd first have to show me that small tactical movements would be more time consuming to start compared to longer ones of equal complexity. In fact, I'd say that tactical moves would have more hesitation and less responsiveness to complexity than larger moves issued all at the same time. Reasoning is that the majority of the delay is getting the guys going and following some sort of generalized instruction. For large moves the instructions would be less inclined to be complicated by enemy fire, tactical cover, and other split second life or death considerations. In a tactical environment there are a lot more small things to consider, a lot more riding on each decision, and a lot more reason for Self Preservation Instincts to kick in and slow down the whole process.

Therefore, if we were to do something distance based, I'd say penalize the Hell out of frequent, small tactical moves and leave the larger "strategic" moves alone. But actually there is no need to do that because it is already built into the game as it is. Try covering the same distance using three single waypoint hops vs. doing all three waypoints at one time. The one time delay is less than the three individual hops. That is realistic.

So there we are, right back where we started from Some people are advocating making the God problem worse, the game more unrealistically paced, and rewarding players who don't plan well at the expense of those who do. That's what removing the current waypoint delay system would mean unless there was something to stick in its place. The problem with that is nobody has come up with a viable alternative. Except to do a far more detailed simulation of the command structure in terns of defining the levels of decision making quite explicitly. Since that is the obvious answer to this perplexing issue, that is the direction we are headed. It's just that we won't get there all at once. It's going to be an evolving thing.

What I will say is that the CMx2 system, overall, should slow down the game substantially. Therefore, it is possible that we may tweak the waypoint/delay system because other parts of the simulation solve a portion of the problem. Specifically the part about units moving around way too quickly with far too much certainty of purpose. This means that the system for simulating C&C issues is still somewhat undecided at this point. But no way, no how are we going back to the CMBO system of puny penalties for poor planning ( for a peck of pickled peppers )

Steve

[Andreas]

Originally posted by Vanir Ausf B:

In that snippet I'm clearly (I thought) comparing the frequency with which I give long and short orders, and at what point in the game I'm more likely to give long ones. I don't even mention complexity there.

But the complexity of long vs. short movements is the issue here. Happy to accept I misunderstood you. Still, if we go with Sergei's suggestion (AIUI) that movement distance should be penalised more than number of waypoints, it is necessary to understand what the difference really is between a long move across unfamiliar terrain in the absence of the enemy (I am presuming you won't order this if you know the enemy is between your start and end point), or a short move when enemy presence has been established.

The point Dalem makes is an additional complication - what should happen when you are reasonably sure there is no enemy, as opposed to when you don't know? How do you communicate to the AI that 'Avoid obstacle left' should be done in a manner that takes account of potential enemy presence, or in a relaxed manner because you know no enemy is there?

[dalem]

Originally posted by Andreas:

No, but I don't assume that in my post, or anywhere else.

Oh I know, I was asking an honest speculative question. I'm not a smart-ass 100% of the time.

As far as you know.

-dale

[Andreas]

As far as I know - I am not even sure of that.

You raise an important point, and one that further supports Steve's line of reasoning, IMO.

Additionally - when you know there is no enemy, is that legitimate knowledge (realistic), or knowledge gathered through borg-spotting (unrealistic), and can/should these two be handled separately? If not, which one to choose?

[Vanir Ausf B]

Originally posted by Andreas:

But the complexity of long vs. short movements is the issue here. Happy to accept I misunderstood you. Still, if we go with Sergei's suggestion (AIUI) that movement distance should be penalised more than number of waypoints, it is necessary to understand what the difference really is between a long move across unfamiliar terrain in the absence of the enemy (I am presuming you won't order this if you know the enemy is between your start and end point), or a short move when enemy presence has been established.

The point Dalem makes is an additional complication - what should happen when you are reasonably sure there is no enemy, as opposed to when you don't know? How do you communicate to the AI that 'Avoid obstacle left' should be done in a manner that takes account of potential enemy presence, or in a relaxed manner because you know no enemy is there?

I think for gameplay purposes we should always assume that orders are being carried out while in contact with the enemy. This is in fact the case the large majority of the time in a typical game, and as you allude to in your later post this is knowledge that the player has that the troops would not, e.g. the bounderies of the opponent's settup zones.

[Vanir Ausf B]

Originally posted by Battlefront.com:

You said that 2 extra waypoints for a Conscript, out of C&C, added 45 seconds. You said it shouldn't take that long to figure out how to walk around something. I said they shouldn't even be able to move, at all, in 45 seconds. Nor in 2 minutes, nor possibly even in 10. You guys keep dancing around this. What you are asking for is unrealistic responsiveness to the player's whim. We feel there is already way, way too much of that as is, so there has to be some attempt to reign it in.

You guys are talking past each other because you're talking about 2 different issues.

If it should realistically take 10 minutes to start moving around a copse of trees (for example), should it take less than 10 minutes to start moving through it? That's what Sergie is asking, I think. Well, it's what I'm asking in any case

And I disagree completely with Vanir that the system, as is, is less realistic overall.

For some reason, I'm not surprised.

But no way, no how are we going back to the CMBO system of puny penalties for poor planning ( for a peck of pickled peppers )

It is true that the CMBO system did not penalize poor planning. It is also true that the current system penalizes perfectly good planning.

Whether the game is too fast and whether waypoints should add to command delay are tangentially related questions in that one affects the other, but they are not the same issue and there are other ways to slow things down if that is the goal. Using the CMBO system, if it is deemed that a 20 second command delay is way too generous change it to 2 minutes. Or 20 minutes. By simply changing the values you can make the game as slow and ponderous as you wish without penalizing waypoints.

[aka_tom_w]

I sincerely hope that Steve and Charles are considering this:

"What you are asking for is unrealistic responsiveness to the player's whim. We feel there is already way, way too much of that as is, so there has to be some attempt to reign it in. "

There should be no concession to making the game MORE responsive to the whim of the player. BUT going too far the other way could make the game perhaps too frustrating to play and enjoy.

I have EVERY confidence they will find a reasonable compromise that suits their gaming interest and desire, because they say they like to make games that they them selves like to play. So I think we should not worry to much about this issue. I think they are moving in the best possible direction for CMx2 on this one!

The problem with that is nobody has come up with a viable alternative. Except to do a far more detailed simulation of the command structure in terns of defining the levels of decision making quite explicitly. Since that is the obvious answer to this perplexing issue, that is the direction we are headed. It's just that we won't get there all at once. It's going to be an evolving thing.

What I will say is that the CMx2 system, overall, should slow down the game substantially. Therefore, it is possible that we may tweak the waypoint/delay system because other parts of the simulation solve a portion of the problem. Specifically the part about units moving around way too quickly with far too much certainty of purpose. This means that the system for simulating C&C issues is still somewhat undecided at this point.

-Steve

I LIKE the sound of the direction they are heading in!!!

"Except to do a far more detailed simulation of the command structure in terms of defining the levels of decision making quite explicitly. Since that is the obvious answer to this perplexing issue, that is the direction we are headed."

Thats sounds like the most appropriate way to try to deal with the issue in the future as the code evolves as a "simulation" of combat . (over a period of releases I would assume)

Thanks

-tom w

[ October 01, 2005, 08:22 AM: Message edited by: aka_tom_w ]

[aka_tom_w]

The problem with that is nobody has come up with a viable alternative.

What about a variable or random delay system that the player can't see.

So the idea is this:

Use the current command delay structure as it has been tweaked for CMAK and then take away from the player the little clock that shows how long the delay is for each unit, so the player cannot plan all actions and and moves down to the last second (I know some players like to try to co-ordinate moves and attacks down to the second within a one minute turn)

AND..

AND then add some random variability, to the command delay say somewhere inbetween + or - %50 on the outside. Maybe most random increases or decreases in the delay would be in the %10 - %20 range up or down. Again the game would not show this to the player.

This would for sure make all the units possibility somewhat less responsive to the players comands. (If you would like to throw in the odd chance that a units could actually get lost on the way and not show up at all at the player's desired destination, that could add a whole NEW twist to this whole concept! With of course an added chance the a unit would be more likely to get lost at night in the woods and so on...)

This suggestion might not need to use the number of waypoints as part of the calculation of the delay. I understand some folks here are dead set against a lenghty command delay based somehow on the number of waypoints in the move. Perhaps the new system for CMx2 will slow the game down and add some form of variable command delay (with the EMPHASIS on unit experience as a largely determining factor) for any move order that would result in an approximately similiar length of command delay as what might be expected in CMAK (but with the added variable of random time added or substracted to the delay) that would NOT be dependent on the number of way points in the order? (maybe)

I agree completely that all units should not be so instanteanously responsive to whim of the player.

But lets not go overboard with that thought.

-tom w

[ October 01, 2005, 02:36 PM: Message edited by: aka_tom_w ]

[Battlefront.com]

As I, and others, have pointed out there are several separate issues here. Unfortunately, there is no simple, easy answer and that is why some 6 years after the debut of CMBO Beta Demo we still don't have a viable "easy" solution to the problem.

CM basically has to simulate three fairly different types of movement:

1. Generalized movement as part of a higher level of planning

2. Tactical movement as part of that planning

3. Reactive movement to a specific and immediate threat, which may or may not be a part of the plan

For each type of movement there is a different level of initiative allocated to the lowly Squad/Team:

1. Generalized - very little to none. Things like general direction, boundaries, SOPs, timetables, etc. are all determined by higher commands and are therefore "givens" for the lower commanders. These things tend to be put in place once and followed until superseded by another higher command or self cancelled under extreme conditions.

2. Tactical - some. The Squad/Teams have some degree of flexibility when moving, but they do so under the direction of the most immediate higher commander (usually a Platoon Leader). It is the Platoon Leader that tells one Squad to be on the right side of the street, another on the left, and a third trailing behind. The Squad/Teams don't come up with things like this unless they have been released for independent action. When that happens it is like they are working under a mini Generalized instruction set, for example "1st Squad go left and move around that stand of trees to circle behind that enemy MG position". The Squad determines if it is to stick to the wall or to a ditch that might be there, though it is also just as likely that the Platoon Leader would also have assigned the route if the route was obvious, for example "1st Squad, sneak along that ditch and then circle behind that enemy MG position". The chance of full independent action generally comes up when the situation becomes unclear and the Squad/Team must adapt its orders to the conditions it finds rather than the conditions it was told to expect (if any).

3. Reactive - total. The Squad/Team has the ultimate say on how it moves in order to minimize risk to itself. In theory it shouldn't do anything that is outside the bounds of the plan, but in reality that does/doesn't happen based on conditions such as level of threat, degree of training, how intricate their current orders are (i.e. if they are sitting around vs. actively providing suppressive fire to save another unit's butt), etc.

Note that distance and plan complexity is not mentioned in any of the above stuff. Why? Because in terms of authority neither matters. A Platoon might receive an order to "take out that bunker 200m to the left of the farmhouse" or "sit tight until further notice". The instructions themselves are just as simple, but the execution of each is very different. This demonstrates that higher level orders need not be complex or far reaching since they both take roughly the same amount of effort to communicate. At least at the higher level. At lower levels even the most simple of instructions can produce highly time consuming, difficult to execute planning. For example, "take out that bunker" might mean having to contact and arrange for mortar support, or to issue detailed time table based instructions for units to follow. That sort of thing. And there is where we start to run into significant problems.

There is a huge difference between

Move down the main street towards the town hall 1km from current position. 1st Squad take left, 2nd Squad right. 3rd Squad trail behind, MG Team behind 1st Squad and other MG team behind 2nd.

And this:

1st Squad will be the breach team, 2nd Squad will provide covering fire on the bunker, 3rd Squad will put its fire on that house over there where there might be a MG positioned. 1st Squad will wait until the mortars drop a smoke screen to blind the bunker. Then they will move in bounds to the wire, cut it and then move to those rocks over here and here to set up covering fire for 3rd Platoon, which is positioned just behind us. They will take out the bunker.

These two examples show that distance and complexity of Waypoints are both irrelevant. In the first example there are likely to be few Waypoints and large distances for each, whereas the second example requires coordinating Delays, Waypoints, and covering fire of two Platoons for something that is probably 1/5th as far away as the first example. Therefore, distance is a poor indicator of planning complexity. Waypoint complexity is also not a good indicator of planning complexity, but it more often than not does. If the player utilizes jump off positions wisely, anticipates problems ahead of time, and accepts the fact that his units are going to have to pause inbetween executing smaller tactical missions, it works (in general) fairly well and realistically. Where it becomes a problem is when the player does not plan ahead and instead wants to command complicated maneuvers on the fly without penalty. It also, of course, runs into problems when the TacAI doesn't do what the player wants it to do, which may or may not be reasonable of him to expect.

The big problem with CMBB/AK is the TacAI's inability to know what the player's intention is. As I have stated above, the TacAI assumes the player knows what he is doing and therefore follows the player's orders as best it can. When it can not, it does its best to keep to the plan by seeking an alternative path. It generally does this well, but we all know of a few situations where it does not (road behavior and morale/suppression issues). We have fixed both of the latter problems for CMx2 so even if we changed nothing else the overall behavior will be much improved in CMx2.

What we need to do is figure out how to simulate higher level issues better than we do in CMBB/AK. We have a solid design for this, but it is a lot of work and therefore will not be in the first release for sure. Expect it for the 2nd release if all goes well, 3rd release if we feel we can't bite it off for the 2nd. I don't think it will go any later than the 3rd release, and I suspect it will make it in for the 2nd.

Conclusion... distance is not the way to go. Waypoint delays are not the best solution, but they are better than any of the suggested alternatives. They also work far more realistically when the player plans more realistically. The less planning, the more of a problem Waypoints become. The TacAI in CMBB/AK does an adequate job most of the time, but it can not read the player's mind and therefore will not always do what the player wants. However, it should do what the player expects, so a player armed with that knowledge should be able to plan around it.

Steve

[ October 01, 2005, 09:51 AM: Message edited by: Battlefront.com ]

[JonS]

Originally posted by Battlefront.com:

Conclusion... distance is not the way to go. Waypoint delays are not the best solution, but they are better than any of the suggested alternatives.

Has anyone suggested a delay that is derived from some formula based on distance and the number of way points?

The delay is a way of enforcing the friction caused by complex commands, right? Well, the number of waypoints is one metric of complexity, but so is distance. So, why not use both.

It could be something as simple as

(total distance) * (number of waypoints) * random element) * (scaling factor) * (in/out of command) * (quality of unit) = delay

But that wouldn't be right, I think. The importance of the distance factor should be non-linear, so for short moves, it's relatively trivial, while trying to move half way across the map in a single order imposes significant delay, and proportionately more than a series of short orders. So the player could decide to suck up a big delay initially, then have his units move a long distance in a single bound, or get his guys moving faster, but then have a series of delauys as he moves across the map (which would be the unit commander having to re-issue orders every few 10s of metres). Or sumfink.

Similarly for way points. Non-linear accumulation of delay (which is, I think, what we already have in CMx1).

Also, the proportion of one to the other would be important. So, a move of 1km with 1 waypoint would impose more delay than a move of 20m with 3 waypoints.

Or sumfink.

Jon

[Andreas]

Originally posted by JonS:

So the player could decide to suck up a big delay initially, then have his units move a long distance in a single bound, or get his guys moving faster, but then have a series of delauys as he moves across the map (which would be the unit commander having to re-issue orders every few 10s of metres). Or sumfink.

Still does not sound right - unless the commander knows the way intimately, there should still be short/long delays along the way while the unit is trying to figure out where it goes. In fact, the initial delay can be neglible ("Go to grid reference 8134956, be there in 30 minutes, don't move outside this and this boundary"), but there should be random-length delays at random points along the way (ideally tied to the terrain, they are more likely in a forest depression than on a ridge with a commanding view) where the platoon leader tries to figure out where the hell he is, how he can get to 8134956, and why he ever left Pansyltucky for this horsecrap.

[JonS]

Originally posted by Andreas:

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

[qb] So the player could decide to suck up a big delay initially, then have his units move a long distance in a single bound, or get his guys moving faster, but then have a series of delays as he moves across the map (which would be the unit commander having to re-issue orders every few 10s of metres). Or sumfink.

Still does not sound right</font>

[hoolaman]

We do already have a method of measuring the complexity of moves and expectation of enemy fire.

The player uses MOVE, CONTACT, ASSAULT, based on whether contact is expected. It may be a more transparent sytem if delays for extra waypoints are based on the type of movement order, or more specifically, changes from one to the other. So a series of MOVE points may attract less delay than a MOVE, a CONTACT, and an ASSAULT all plotted in series.

Another design concept I was thinkin of that may be applicable is to make a plotted movement path into a "progress bar". So when the movie starts, the path gradually fills in to symbolise how far ahead the orders have been explained. If the squad catches up with the progress bar, they may halt or slow down. The progress may be slower over rough terrain, but faster for simple move orders like running across a field. The unit might not set out on the next leg of their move unless the subsequent waypoint has been reached by the moving "progress bar".

Not sure if any of this is truly realistic, it's just an idea!

[yuvuphys]

Originally posted by Battlefront.com:

As I, and others, have pointed out there are several separate issues here. Unfortunately, there is no simple, easy answer and that is why some 6 years after the debut of CMBO Beta Demo we still don't have a viable "easy" solution to the problem.

CM basically has to simulate three fairly different types of movement:

1. Generalized movement as part of a higher level of planning

2. Tactical movement as part of that planning

3. Reactive movement to a specific and immediate threat, which may or may not be a part of the plan

For each type of movement there is a different level of initiative allocated to the lowly Squad/Team:

During the scope of a CM battle, orders from #1 are not going to change and aren't under control of the player. So, those orders can be seperated out at the beginning of the game. Now, as a company level game, I'm assuming that the role of the players is of the company commander and partially as the platoon commander (is this correct?). Thus the orders submitted by the player are all tactical, and not reactive. That narrows down significantly the scope of the problem; everything has been seperated out and can be dealt with seperately.

Will it be the case that when I tell a squad to go from point A to point B, that it has some leeway to deviate slightly from the path I pick? It seems to me as this would be a reactive manuever not apart of the player's role, but taken care of by the TacAI as I understand the role of the player.

Now, I don't think you can do away with waypoints until you move to a *much* more complex system of orders (i.e., "storm and clear that building" instead of placing a waypoint in that building)...but that's CMX3 if ever? Within the confines of waypoints, I think the main difficulty would be incorporating tactical planning which occurs before the actual battle. Do you plot expected routes and delays for all your units beforehand, and have the units execute those plans until the player intervenes? How does this work for defenders...do you give units possible escape routes, so that there are preplanned maneuvers that will be automatically executed? I'm sure this poses massive AI problems, mostly in making the AI understand the human's intention. That's the monster to tackle. Once you do that, the tactical orders during the gameplay are all based off of the understanding developed at the beginning of the game, since in general the modifictions to the original tactical plan must be fairly simple.

So, that leads me to my other question; I think you hinted at the ability to give tactical orders at the beginning of a game. Is this going to be included, and if so will it work similarly to the tactical orders given during the game?

[Battlefront.com]

When we were developing CMBB's wapoint delay system. Charles first implemented limited waypoints based on experience, which was quickly rejected for obvious reasons Did wonders to simulate the limitations of crappy units, but it was just too difficult from a game playing standpoint. Then there was delay based on movement, but again it was rejected because it simply didn't feel right (and I explained why not above). So we went with straight time delays per waypoint. That wasn't quite right, so there is a curve where you get a bunch of waypoints fairly "cheaply" and then ones after that are more expensive. I'm sure we thought of waypoint distance plus number of waypoints, but the distance concept simply isn't something that is inherently related to complexity. So we're not going down that path.

Hoolaman's idea is interesting, but in order to make that work we'd have to simulate the higher exchange of commands. We have a system planned for that, but that's the thing I said won't be in the first game. It fixes most of the problems on its own.

The things that add to orders complexity are:

1. Unfamiliarity with the terrain and/or enemy.

2. Orientation problems (not lost, just not sure where located).

3. Necessity to have actions coordinated with others.

4. Inability to act on own initiative because need new orders.

5. Inexperienced leadership.

6. Inexperienced soldiers.

7. Orders that conflict with something, such as another order or the reality on the ground.

And probably a lot more. As far as I am concerned, the two big ones are #1 and #3. ESPECIALLY #3. Trying to clear out a single bunker could take 20 minutes just to organize, and then an hour to execute. This is one reason why it took the Allies months to get get out of Normandy, not days. The more organized the defenses, the less favorable the terrain, and the better the enemy in general... the longer it takes to get anything done because coordination becomes vital, but coordination is time consuming. The Germans obligated the Allies to do things the hard way. And when the Allies tried to shortcut things, they often paid a high price for it.

So I'd say the focus of delays should be on #1 and #3. I think we have #1 covered with Absolute Spotting, but if we could fog out the terrain we'd be even better off (though we can't do that yet). #3 is the real problem. Not really sure what more we can do about that except to make each Squad/Team commanded by a single player. That makes for an entirely different game, but a far more realistic one.

Steve

[Battlefront.com]

yuvuphys,

I'd say that the higher level planning is still relevant, even though it shouldn't change much during the course of a game. The reason it is relevant is because it has an influence on the tatical planning. Right now there is no higher level enforcement of parameters for tactical planning, such as "1st Platoon will provide overwatch while 2nd and 3rd move up routes A and B respectively". Or at a slightly higher level, "A Company will hold while B Company moves long the left flank of the enemy held town". There is no such restriction on the player right know so he can make this plan in his head and then completely change it, without any delay or penalty, as many times as he wants to. That is unrealistic.

The same thing happens at the tatical level, but with greater degree of flexibility in terms of how the plan can be modified. Well, unless you are a Soviet unit, in which case things are pretty much ironclad laid out do or die! So the platoon moving on the left flank can decide to do this or that, but only if it doesn't conflict with the higher level instrutions, such as "be at this spot by this time so you can do that action". The Platoon Leader can't just say "nah... I'm going to go here and take my time about it then figure out what I want to do when I arrive".

Yup, it is a really tough thing to simulate. I think the best we can shoot for is slowing down the ability to change things around. That is what the waypoint delays do in CMx1 and that, plus other things (especially Relative Spotting), should have a pretty good and fairly realistic effect on CMx2's gameplay.

Steve

[stoat]

Won't it have an effect when we play battalion and larger sized scenarios?

[yuvuphys]

Originally posted by Battlefront.com:

yuvuphys,

I'd say that the higher level planning is still relevant, even though it shouldn't change much during the course of a game. The reason it is relevant is because it has an influence on the tatical planning. Right now there is no higher level enforcement of parameters for tactical planning, such as "1st Platoon will provide overwatch while 2nd and 3rd move up routes A and B respectively". Or at a slightly higher level, "A Company will hold while B Company moves long the left flank of the enemy held town". There is no such restriction on the player right know so he can make this plan in his head and then completely change it, without any delay or penalty, as many times as he wants to. That is unrealistic.

The same thing happens at the tatical level, but with greater degree of flexibility in terms of how the plan can be modified. Well, unless you are a Soviet unit, in which case things are pretty much ironclad laid out do or die! So the platoon moving on the left flank can decide to do this or that, but only if it doesn't conflict with the higher level instrutions, such as "be at this spot by this time so you can do that action". The Platoon Leader can't just say "nah... I'm going to go here and take my time about it then figure out what I want to do when I arrive".

Yup, it is a really tough thing to simulate. I think the best we can shoot for is slowing down the ability to change things around. That is what the waypoint delays do in CMx1 and that, plus other things (especially Relative Spotting), should have a pretty good and fairly realistic effect on CMx2's gameplay.

Steve

Oh, yes, I wasn't trying to say it wasn't relevant. But it all happens at the beginning of the game. You want company A to do overwatch? The allow comapny A to be placed at the beginning of the game, and not move forward past a certain line.

Well, if you're looking to slow things down and make them harder to change, I'd say one of the best thing to do is make it costly to cancel or change orders. If you want a squad to stop advancing, there should be a command delay associated with making the squad stop. Maybe the squad doesn't get those orders sometimes, etc. Then after the orders are canceled, it takes another command delay to give the squad new orders. That will certainly encourage or force players to make shorter and more realistic movement orders.

[ October 01, 2005, 05:40 PM: Message edited by: yuvuphys ]

[Battlefront.com]

Oh, yes, I wasn't trying to say it wasn't relevant. But it all happens at the beginning of the game. You want company A to do overwatch? The allow comapny A to be placed at the beginning of the game, and not move forward past a certain line.

Which is why this isn't easy What you're talking about is an entirely new layer of commands and consequences. Which we also see as being the only way to do it, so it will happen at some point. Just not sure when, though surely not for the first release.

Well, if you're looking to slow things down and make them harder to change, I'd say one of the best thing to do is make it costly to cancel or change orders. If you want a squad to stop advancing, there should be a command delay associated with making the squad stop.

That's already in the game When you give a unit orders it suffers through a command delay to get moving. If you stop the unit at any time, for any reason, the unit must go through another command delay to start moving again. That is in effect the same thing as you suggest. Smaller moves = better moves.

Or did you mean having a penalty to actually stop the guys? That runs into problems with the Reactive rights of a unit and therefore isn't a good idea. A unit getting shot at in the middle of a field shouldn't have to wait x seconds to stop advancing. It should be able to stop right away. The TacAI usually handles that though.

Steve

[aka_tom_w]

That's already in the game When you give a unit orders it suffers through a command delay to get moving. If you stop the unit at any time, for any reason, the unit must go through another command delay to start moving again. That is in effect the same thing as you suggest. Smaller moves = better moves.

Or did you mean having a penalty to actually stop the guys? That runs into problems with the Reactive rights of a unit and therefore isn't a good idea. A unit getting shot at in the middle of a field shouldn't have to wait x seconds to stop advancing. It should be able to stop right away. The TacAI usually handles that though.

-Steve

Good point

-tom w

[yuvuphys]

Originally posted by Battlefront.com:

</font><blockquote>quote:</font><hr />Oh, yes, I wasn't trying to say it wasn't relevant. But it all happens at the beginning of the game. You want company A to do overwatch? The allow comapny A to be placed at the beginning of the game, and not move forward past a certain line.

Which is why this isn't easy What you're talking about is an entirely new layer of commands and consequences. Which we also see as being the only way to do it, so it will happen at some point. Just not sure when, though surely not for the first release.</font>

[Battlefront.com]

yuvuphys,

Well, it seems at first glace fairly trivial to force a unit to do overwatch. They have all the same commands as any other unit in the game, only they can't advance past a certain line. It'd be a lot more difficult to allow a moving overwatch, or to force company A to assault the left flank while company B assaults the right flank...in which case I see you point

And also that overwatch is just one type of restriction. There are others, such as flank moves, main axis of advance, reserve, phased withdrawals, etc. One interface could cover all of these things, but that interface is not an easy thing to produce.

I mean a penaly to actually stop the unit. Leave the reactive rights of the unit intact; if they get heavy fire, they should stop. But in the absense of a a unit stopping of it's own accord, incurr a penalty of being able to stop the unit.

The problem is there is no mechanism to know which is a tactical decision and which is invoked due to a coordination issue. The reason is there are no explicit simulation of coordination. There is no way to tell if a unit wishes to stop because of a legitimate self determined reason (like suspecting trouble or coming under fire) or if it is "cheating" by allowing events outside of its knowledge to influence its movement decisions. Until we can know the difference there can not be any penalty for tactical decisions like stopping. It would be disasterous to even try. All we can do is assess generalized penalities for starting ations and just live with the fact that they are inherently unrealistically coordinated thanks to the Borg and God issues.

Steve