1-1 Infantry - Improvements

[Battlefront.com]

akd,

I'm sure you were not being entirely serious, but I find this a poor argument. Photographers (and videographers) will gravitate towards close-up photos in which individual detail is discernable. This makes any clump of soldiers a magnet for the camera man. Also, because of the urban nature of much recent combat footage, tighter groupings of individuals are much more likely to be encountered and filmed/photographed.

Oh, I know that pictures (and video) aren't the be-all-end-all to understanding combat. My point is that some people think that spread out is the norm in combat, but for various tactical reasons that is not the case. The pictures and videos are just one way of verifying that. And the denser the terrain, and the closer the enemy is, the more bunched up the forces become. All the way to the extreme of entering a house "stacked" with each soldier LITERALLY touching the back of the guy in front of him.

Again, my primary point is that bunching is more of a perceived problem than an actual one. Those pictures of soldiers walking out in the open very spread out, that's actually very possible in CM right now. That top picture looks to have about 7 or 8 guys in the same footprint of a CM Action Spot. At most that's two Action Spots, which would be what a Squad of that many men occupies, by default, in CM right now.

Also note that in both of those sparsely marching shots they are obviously on a low-intensity march. If you do that in CM you will get men strung out over a much, much wider area than just a couple Action Spots. So again, comparing apples to apples CM can do what is in those pictures.

The second sparse picture isn't all that sparse You've got 4 guys no more than 1-2 meters apart, which is definitely something CM currently does. I have no idea what is off to the right of the picture, so it's impossible to say if the density is better/worse for the others.

Regardless, when contact is made soldiers tend to cluster. The trick is to cluster wisely, which is what a lot of drill is all about. Clustering allows for greater supporting fire, greater coordination, and increased tactical options. Clustering the wrong way, however, can do exactly the opposite. Generally speaking, being strung out with major distances between soldiers is tactically inadvisable, though sometimes necessary due to too few men covering too much terrain. This is one thing that CM doesn't do very well at the moment. You can divide up Squads only so much.

Steve

[Battlefront.com]

Having said that, I don't disagree that there is room for improvement. What we have to avoid is something that Redwolf put very well two pages back:

The old question whether you rely more on automatic TacAI or whether you make more use of player control via SOPs and formations is always on the table.

The latter approach will seriously clutter up the luser interface, make the learning curve (much) steeper and pretty much kill RT play. Let me guess, it's not your favorite approach?

Exactly And it won't necessarily fix the problems you guys have with CM in its current state.

Coincidentally a few days ago I played a tactical wargame that was being pitched to us for resale. Set on the Eastern Front it has 1:1 depiction and has formation controls. I didn't know this when I booted the game so I was suddenly, and surprisingly (actually, I was thought it was a tank sim ), given an example of this stuff to play and see if my opinions have changed about micromanagement controls for Formation.

Throughout the game I went back and forth on if the Formation controls were a good thing or a bad. In the end I concluded that overall it was a negative. And here's why:

1. The UI was quite simple for a hardcore wargamer, though a probable turnoff for a casual wargamer. Don't get me wrong, they did a pretty good job with that portion of the UI. It was about as straight forward as it can get.

2. You could choose between three orientations left/right, three orientations front/back, and three densities. That's 9 options which can be combined together. Having things like "echelon right" and "double inverted V", etc. were not possible with this system. If those options were made available the 9 buttons they had would have had to increase dramatically or they'd have to move to a more cumbersome form of control (popups, for example). So while there absolutely is more theoretical control over formations, the options provided were extremely basic and likely not all you guys want them to be.

3. The TacAI, as we conceive of it in CM, stunk. That highlighted the fact that having micromanagement control over your soldiers does not eliminate the need for very good TacAI. The virtual soldiers need to know when your Formation choice is no longer valid and take independent action to correct for that. Otherwise you get what I saw... guys sitting around waiting for you to come down from the heavens and give them a new Formation that is better suited to the situation they are in. And in order to do that you have to notice their needs and then examine what alternative would be a better course of action in time for the choice to be relevant.

4. When a unit ran out of movement orders it would always sit down in Formation. This sucks because you can't stack Formation commands (see next point). When you move your guys in the open and into a copse of trees you have to wait for them to get into the trees before you can tell them to line up abreast or single file (depending on tactical conditions, obviously!). If you don't then they'll march into the forest all spread out and that is almost a sure recipe for getting them slaughtered. So not only doesn't the TacAI have the ability to smartly override your instructions, but there's no way to easily coordinate movement and spacing depending on conditions. Which means you have to butcher your tactical planning and execution cycles or accept that your forces will move into the wrong situations with the wrong Formation. The latter gives you the worst of both worlds since a bad Formation choice can't be made good on by the TacAI, which in effect produces results far worse than what you see in CM without any Formation controls.

5. The big point, however, is not what the game included that CM lacks, but what the game excluded that CM has. In order to make the game manageable from a player's standpoint they had to over simplify the rest of the game. As we've said for eons now, there are only so many commands, buttons, and other forms of options that a player can handle. The developer of this game also understands this and so there are 9 Formation buttons but there is only *one* movement button with the option to go faster or to stick to roads. Combat instructions amounted to two different types, which were to target a specific unit or area fire. There were very few other options. Not even Hide/Button Unhide/Unbutton commands, though interestingly there were commands explicitly for Mount and Dismount.

6. In addition to the vastly simplified tactical movement and general behavior options there were no waypoints just like Close Combat. My guess is that once people get waypoints they want to be able to specify different behavior for each waypoint, be it a different speed or Formation. I can assure you that this vastly complicates the game programming and overall UI design. So they just left it out completely.

CONCLUSION...

As with most things, there are tradeoffs. In my opinion was that the net effect was a major reduction in both tactical flexibility and enjoyment. My frustration level wasn't extremely high, mind you, but my enjoyment level was quite low. I thought long and hard about what minor things they could do to fix the problems and not lose the Formational functionality. I wound up the same place where I've always wound up when pondering this... there's no good answer.

Our plans is to continue to keep the player's fingers out of the micromanagement business for their own protection. You guys definitely need protecting even if you don't think you do We'll keep improving the TacAI over time and get far, far better overall results from doing that than we would by trying to cobble in a micromanagement system or gutting the existing system to focus on micromanagement designs. After all, without excellent TacAI the Formation thing sucks anyway. Which may be counter intuitive to you guys since some of you think that this will fix all of the current problems... I say it will do the opposite and this game helped reinforce that.

BTW, the game had a bunch of problems which I don't want to get into. Therefore, we decided to not publish the game I've mentioned. And no, I won't tell you what it's name is

Back to Redwolf's other point:

Overall I think there are two separate issues getting mixed up here, which is the somewhat academic question whether some things are easier with point-like squads (assuming the play can only control squads) and on the other hand what to do once you decided to go 1:1 and have to deal with it.

Indeed! And my experiences with this other game, and those before it, reinforce my feelings as someone who has been doing game development for quite some time... abstracted systems are easier to make and maintain, but they aren't necessarily better when compared to a less abstracted system. As I've already said, having worked for years on both CM systems I can say, without even a hint of doubt, that we prefer the challenges which come from the increased fidelity to the frustrations which come with a far more abstracted one. CMx1's code was holding us back, not helping us move forward. CMx2 is the opposite.

Plus, as a game player I'm "done" with overly abstract portrayals of infantry such as CMx1 had. It doesn't hold my interest at the tactical level any more. Higher up, when individuals don't really matter... different game and different story.

Steve

[dan/california]

Allow me to throw in a theoretically less complicated idea then. Would it possible to allow the player to adjust the AIs balance between self preservation, and order completion. So that with self preservation set very high a unit would be much more likely to seek cover/retreat, and with aggression set to max they would carry on to limits of there morale as currently defined. It could be set for the players entire force or a slider on each individual unit. It would make it easier to have a unit go fast until first contact without getting obliterated for instance.

[Other Means]

I just hate them bumping into each other as they do now.

"Sorry!"

"Oof!"

"My fault!"

"Hey!"

"Why stop there? Come onnn!"

"Ak!"

It's like a CM1 traffic jam in miniature.

[MarkEzra]

I just hate them bumping into each other as they do now.

"Sorry!"

"Oof!"

"My fault!"

"Hey!"

"Why stop there? Come onnn!"

"Ak!"

It's like a CM1 traffic jam in miniature.

Me, too... Like a drunken dancer in a conga line.:eek: graphically silly to look at

[Pandur]

since iam a WEGO only player, i myself would opt for any form of standing orders. most basic formations, and 3 options for density of the squad would be great.

the UI limitations do not count in my view, as everyone who sticks with the "standard" UI(relative controls) is certainly not interessted in a optimal playing expirience.

imagine if we had 3 buttons for formations, and 3 for density. all you do with the "absolute" system is maping 6 new buttons, thats easy as throwing stones. to switch this SOPs is a matter of split seconds, as is anything else UI related with the absolute contols system used.

so saying that SOPs would be horribly cluster the player input is plain nonsense. remembering 6 buttons for 6 different fuctions? mokeys can do that

and steve, you said you looked up some vids on youtube about unit spacing? did you by chance see the few clips there are about iraq-iran war and similar conflicts? the arabs do spread out pretty much like any other force so i dont see why there is constant talk about blue force wich is able to spread out in a pretty extreme way allready.

4. When a unit ran out of movement orders it would always sit down in Formation. This sucks because you can't stack Formation commands (see next point). When you move your guys in the open and into a copse of trees you have to wait for them to get into the trees before you can tell them to line up abreast or single file (depending on tactical conditions, obviously!). If you don't then they'll march into the forest all spread out and that is almost a sure recipe for getting them slaughtered. So not only doesn't the TacAI have the ability to smartly override your instructions, but there's no way to easily coordinate movement and spacing depending on conditions. Which means you have to butcher your tactical planning and execution cycles or accept that your forces will move into the wrong situations with the wrong Formation. The latter gives you the worst of both worlds since a bad Formation choice can't be made good on by the TacAI, which in effect produces results far worse than what you see in CM without any Formation controls.

5. The big point, however, is not what the game included that CM lacks, but what the game excluded that CM has. In order to make the game manageable from a player's standpoint they had to over simplify the rest of the game. As we've said for eons now, there are only so many commands, buttons, and other forms of options that a player can handle. The developer of this game also understands this and so there are 9 Formation buttons but there is only *one* movement button with the option to go faster or to stick to roads. Combat instructions amounted to two different types, which were to target a specific unit or area fire. There were very few other options. Not even Hide/Button Unhide/Unbutton commands, though interestingly there were commands explicitly for Mount and Dismount.

about point 3 and 4;

well thats the "other" game isnt it? that doesnt mean you have to copy its failure. just add the best of SOPs to the allready best of CM and there you go, you have the best of the best

I just hate them bumping into each other as they do now.

"Sorry!"

"Oof!"

"My fault!"

"Hey!"

"Why stop there? Come onnn!"

"Ak!"

yea i hate this soo much. i even did a thread or a post about this quiet some time ago, cant remember where, noone seemed to care about it.

i even suggested to make soldiers able to "pass through" other soldiers, if there is no other way to avoid this mess.

i guess a full squad ordered to quickmove over 100m open ground would be around 5 to 10 seconds faster when they wouldnt bump into each other ever few meters.

[GSX]

Surely this from Other Means

I just hate them bumping into each other as they do now.

"Sorry!"

"Oof!"

"My fault!"

"Hey!"

"Why stop there? Come onnn!"

"Ak!"

Doesnt really equate with this

Plus, as a game player I'm "done" with overly abstract portrayals of infantry such as CMx1 had. It doesn't hold my interest at the tactical level any more. Higher up, when individuals don't really matter... different game and different story.

Steve

If what the infantry do now is abstract anyway, even though you can see them all, then whats the real difference. It could be argued that the 3 man Block style of CM-1 served the purpose better as it didnt pretend to be anything else but abstract. Now I wouldnt go that far, as I like the 1-1 representation and think its absolutley the right way to go.

That said though, what we have infantry wise now is still to an extend abstracted. I dont mean in their numbers, but in the way they occupy the battlespace. It just does not feel right. A few simple formation commands would be a start.

As I keep saying, I dont know if its possible at all or even fits into the way CM-2 is designed, I dont know if its a feature you can build in or not. Im not really bothered about how other games play or feel either and Im not sure you can use the formations of one game to explain why CMSF has none.

SF has been moved to a scale where infantry has become more dominant than previous CMs and so infantry should feel right. Get that right and everything else should follow I would hope.

Real soldiers use formations because they work. Can I ask:

1. If every round is tracked ballistically in the game. Is the position of every soldier too?

2. Does the PC see these positions as I do,?

3. If the above 2 statements are a yes, then formations would make a big difference to infantry survivability on initial contact, as in real life thats why they are there.

[Battlefront.com]

Having guys run through each other is a very bad idea. In real life guys neither run through each other nor do they bump into each other (as common practice). Instead, guys don't start their moves until they can see a clear path to where they want to go. If they are correctly aware of others then they don't stand up and get in the way of the guy running up behind them. And if the guy in front of him stops he doesn't start his motion until the guy continues on.

Therefore, although the bumping of guys is somewhat distracting, overall it's more realistic than having them pass through each other. Passing through each other would increase the already existing artificial "time compression" problem by further increasing the speed of your units to do things which, in real life, usually take a lot longer to initiate. This is yet another example of us having to protect you guys from yourself, because an idea like "let guys pass through each other" has negative consequences which obviously haven't been considered. Which is understandable... I get paid to sit around and think of these things and have a direct financial disincentive for putting in features which make a game worse

What is needed is more AI for soldiers to anticipate, ahead of time, where they should go in order to avoid bumping into each other or getting to cover more efficiently. This is "expensive" from a processor standpoint, but definitely possible with time.

Steve

[Battlefront.com]

GSX,

Doesnt really equate with this

That's because the world isn't black and white. Everything is on a continuum. When we designed CMBO we were "done" with 2D abstractions. But that didn't mean we were "done" with abstractions completely. With CMx2 we are "done" with massive abstractions of individual soldiers, but we're still not "done" with abstractions completely. Because being "done" with abstractions is impossible. We'll NEVER be done with them. So it's all relative.

If what the infantry do now is abstract anyway, even though you can see them all, then whats the real difference. It could be argued that the 3 man Block style of CM-1 served the purpose better as it didnt pretend to be anything else but abstract.

To counter this point (even though you aren't in support of it), the truth is that CMx1 was pretending to be all kinds of things. It was pretending that there actually were 12 individuals doing individual drills, which in turn allowed you to pretend that that Moe, Larry, and Curley were representing actual soldiers in an actual battlefield environment. Many, including myself, found that degree of pretending to be a problem. An interference, not a benefit, to the overall game experience. From a simulation experience there's absolutely no question about it... the gross abstractions lead to far less realistic depictions of warfare and a Hell of a lot of limitations on us in terms of our ability to improve the fundamental simulation.

That said though, what we have infantry wise now is still to an extend abstracted. I dont mean in their numbers, but in the way they occupy the battlespace. It just does not feel right. [/qutoe]

It's all relative. To me it feels more "right" than any other wargame I've ever played in my entire lifetime. It feels a lot more "right" than CMx1, even though the scope and scale of the battles are about the same. It feels far more "right" than looking at 2D chits on a mapboard. It even feels more "right" to me than Close Combat, which I was quite critical of at the time for many reasons, but 1:1 wasn't one of them.

But does CM feel "perfect" to me? Absolutely not. Does it feel "right enough" that I think further change is unnecessary? Nope. But compared to any other game I've ever played, overall, do I think CMx2 is feels more "right"? Yup! And I think that's a pretty good place for us to be.

I disagree. More case sensitive TacAI would be the place to start. Far better bang for the buck and far less risk of the game system becoming a cumbersome experience which only a few players would enjoy (which greatly reduces the bang for that buck).

While there is huge amounts of room for flexibility in game designs, there are fundamental boundaries and rules which game developers must adhere to. One of them is marketing, another is how many things the average Human wants to be responsible for, another is how much a given computer system can handle, yet another is how much cost is involved for return, etc. My analysis of that other game was with these things in mind. It's what I do for a living and a lot of it is intuitive rather than learned. And it is with that in mind that I point out things like what the other game I mentioned had to give up in order to accommodate the formation stuff. It is with that base of knowledge that I point out what things within that game would not be acceptable to people who like Combat Mission. In short, I don't think average gamers are in a good position to understand the limitations and tradeoffs of one game vs. another, but it definitely is within my capacity to do this. It had better be, since the products I make are largely based on this sort of stuff

I think this statement is important. Yes, CMx2 has more emphasis on infantry than CMx1. Why is that so apparent to you? The TO&E is basically the same sort as with any CMx1 game. The numbers of soldiers involved in a typical battle are about the same, the combined arms are similar, and the balance of infantry to vehicles is actually more heavily weighted towards vehicles in CMx2. So why do you feel that infantry is more a dominant part of CMx2 than CMx1? Because of the 1:1 simulation. Therefore, it's obvious that simply going to 1:1 infantry made a huge impact on the game in terms of the importance of infantry. The inverse is that CMx1's infantry simulation was harmed by being so abstracted. A point which, obviously, I agree with completely.

Yes and yes.

No. Spotting and LOS are not drawn from polygon to polygon, so the world you literally see is not how the game engine views things. Seeing 2 polygons poking from around the corner of a building does not trigger a unit to become spotted, for example.

Another abstraction is that for blast type weapons the effects are modified to decrease the lethality based on distance from impact. Not a ton, but some.

No doubt that formations would increase the overall realism of the game. It would be idiotic to suggest otherwise, which is why I've never suggested it However, the degree of improvement would likely be less than you think because there is far more too this than just those first two questions. For example:

1. Is the terrain literally all that you see?

No. We can't simulate every twig, rubbish can, minor mound of dirt, etc. that the same chunk of the real world holds. Therefore, there is a basic amount of abstracted cover and concealment in all forms of terrain. The most obvious example of this are buildings, which are internally devoid of all details.

2. Are the Human figures absolutely 100% accurate representations of real Human motion?

Obviously not All soldiers are the same height and mass, for starters. But more importantly, when 2000 soldiers go down to one knee, all 2000 soldiers are 100% identically orientated in terms of posture. Put 2000 soldiers in identical positions behind a wall, and have them all kneel, and all of them have 100% identical exposure to enemy fire from similarly orientated enemy forces.

3. Does each solider in the game have his own thought process which, in turn, can influence the group's behavior?

No. All soldiers have identical methods of "thinking" and therefore act robotically within their range of preprogrammed behaviors. Group actions are influenced by individuals only indirectly, with a higher level set of rules dispassionately determining how the group should behave according to various preset rules.

And there's probably others, but I have to move onto other things My point is that there are now, and will forever be, some very significant abstractions within the game. These abstractions are not limitations on our creativity or knowledge of how the real world works, but are instead limitations on how the real world of game development works. We can make improvements over time, however the never ending quest to get perfection from a game which simulates something this complex is simply setting oneself up for perpetual disappointment.

CMx2 is light years ahead of CMx1 in terms of its overall game atmosphere and mechanics, but it doesn't mean there isn't room for improvement. There always will be. But no one single thing will "fix" infantry realism when there are so many other things which will remain theoretical problems. I agree that better infantry formational behavior is a very good thing to improve over time, I vehemently disagree that giving the player micromanagement controls of that behavior is a good way forward. Perhaps a few minor things mixed in with major TacAI improvements... sure, but that's not the same thing in our view.

Steve

[Pandur]

artificial time compression, i dont get why this is is a "problem" at all. you can tweak play dynamics in whatever way you guys like by adjusting different parameters and values. may it be speed of movement commands to off set the newly gained speed. adding more so called weapon fudge factor to draw out firefights and modifie spotting and supression accordingly.

this "would" be as possible as you tweaked all the things over the many patches, but ultimatively it was not aimed for this. so the "problem" is or was kept intentionally i feel.

iam ok with that, or i have to in the end, but argumenting that this more effective movement of individual soldiers in a unit, resulting in more speed, is such a huge problem and not possible to counter is...i say odd.

the technical limitation, ok nothing to do here, passing through each other would be cheap on CPU, easy and do , instead of useing CPU power, but if thats the way you guys want to go, so be it. if it gets the job done too iam fine with that, but its needed rather fast

[dan/california]

Stop the bumping first, that is the thing that is just so WRONG visually.

[Guest Guest]

deleted per user request

[Guest]

Stop the bumping first, that is the thing that is just so WRONG visually.

Hmm. And then deal with the running-in-slow-motion thing. Better anticipation could solve both those issues. But that's not an "easy" AI problem. Believe me. Oy. Let alone making it all look good.

[Redwolf]

There is one more thing about formations:

In the real world the major reason to have shaped formations (instead of just spacing) is to bring firepower suddenly to the right point, and to always be in positions where as many guns as possible can fire at a certain point in the suspected enemy positions. This is true for infantry formations and even more so for tank formations. Tank formations are mainly desired because they cannot shoot through friendly tanks and they cannot reposition to shoot around a friendly easily - in the real world.

In CM friendlies do not block LOF, so this reason for formations is gone. Distance based squad movement would do.

Having said that, I don't think that overall the state of CM:SF's infantry movement within squads is where it should be. Formations aside, the game doesn't seem to try or at least not be good at maintaining distance between soldiers. And that leads to people bumping into each other and extra casualties.

The matter of whole sections wandering into enemy LOF against player's commands is probably more of a bug in other code pieces and not directly attributable to formation or inner-unit distance handling code.

[Redwolf]

Or have I got it wrong and the 1-1 infantry areent tracked in space but are abstracted somehow?

I believe that the CMx2 engine as of today does indeed uses LOS to individual soldiers to determine whether he can be hit. Fire abstractions only apply afterwards, if you are visible you can be hit.

This directly means that code issues that have soldiers wander off into enemy LOS against the player's wishes cause unrealistic casualties. While an experienced player can probably coerce the TacAI into doing it less that would be "gamey", not any less gamey than exploiting the engine to get an unrealistic advantage.

Much of this is unrealistic. In real life, soldiers very rarely wander around the corner of a house when other friendlies have visibly taken position as if that is in enemy LOF. When there is a friend covering himself against direction X you don't expose your full body towards that direction.

Fixing it won't be easy but as is it is a realism problem. I think the individual soldier's position on the map might have to be re-abstracted when it comes to LOF calculations.

[akd]

BTW, the game had a bunch of problems which I don't want to get into. Therefore, we decided to not publish the game I've mentioned. And no, I won't tell you what it's name is

But I think I know anyways:

[commercial link removed]

Regardless, I hope y'all publish their next tank sim. Steel Fury had great potential but was totally effed over by the publishers (not saying the devs didn't play a role).

As for the field photo I posted (the one with the 240 gunner in the foreground, I think your perception that they clustered in a CMSF size grid is an issue of perception of distance in the photograph. Those Marines are maintaining at least 5 meter intervals, as I'm sure they were trained to do.

And yes they are not in contact. This goes back to my point. Photos of troops spread out in combat in oper terrain are rare because 1.) they are often very uncompelling images and 2.) photographers are going to have an even stronger urge than trained fighters to seek larger groupings of individuals when they are under fire, so the most common image is a group shot taken from low behind cover. When such photos do slip by the editors, they are often something like this:

Stuff like that just doesn't get published.

But all in all, I agree that it is not a huge issue in CMSF. I just hope that a future goal would be an allowance for 2 soldiers per 8m grid (or 1 per 5m grid) for, say, movement of a disciplined unit across open ground, but tighter clusterings due to the psychological and terrain factors that CM is capable of simulating.

[BlackMoria]

..... the technical limitation, ok nothing to do here, passing through each other would be cheap on CPU, easy and do , instead of useing CPU power, but if thats the way you guys want to go, so be it. if it gets the job done too iam fine with that, but its needed rather fast

Yes, but having soldiers 'pass through' each other brings up some interesting issues.

Let say that due to whatever circumstances, Soldiers A, B, and C are co-existent in the same spot during the squad move when a small arms round (5.56 for our illustration) hits Soldier A.

What most likely happens is that Soldier A, B and C suffer the same combat result, which results in A,B, and C suffer the same injury or being killed. Not a desirable outcome.

I think most players would prefer soldiers jostling into each other and slowing each other down rather than having a chance of one round = multiple kills when they start clumping together, a regular ocurrance when doing building and vehicle entries and exits.

[c3k]

I think of the jostling and stopping as simulating passing of orders...or cigarettes...or ipods.... Of course, during the assault on an occupied building, I then think of it as untangling some web gear...tying a shoelace...hesitating while looking for a target...

Ken

[Battlefront.com]

Pandur,

artificial time compression, i dont get why this is is a "problem" at all. you can tweak play dynamics in whatever way you guys like by adjusting different parameters and values. may it be speed of movement commands to off set the newly gained speed. adding more so called weapon fudge factor to draw out firefights and modifie spotting and supression accordingly.

Ah, which would be correct if you weren't so very wrong It's a problem that CM is better at handling than most wargames, but we still suffer from it. And there is absolutely nothing we can do about it since the problems are linked to the player and not to the game. The short of it is that since you see everything and control everything you issue more commands in a shorter period of time with more confidence and much less concern than could ever, ever happen in real life. And that means a battle that would take 2-4 hours in real life might take 15 minutes in game time. Doing anything to speed up combat within the game makes this worse, not better. You should check out the threads here about "time compression" since we've had this discussion many times before.

Adam,

What are the values? Are they percent based? What amounts of abstract cover does each terrain tile offer?

We have a couple hundred variations of terrain with multiple values and many outside factors which are taken into consideration. Which means you've asked a question that has no possible answer. At least not one of any practical value. Because of the complexity, the only person who can make sense of these values is Charles. And even then only when he's actively watching it execute in the debugger. I have no more idea what the cover or concealment value is for Brush in CMx2 than I did for CMx1, which is absolutely zero idea

What is important to know is that the terrain you see represents the sort of cover and concealment you would expect for that type of terrain. And if it doesn't work as expected, then we can make some tweaks to the correct place (which often has nothing to do with terrain values) and make improvements. However, there is a ton of room for opinion and a LOT of the opinions I've seen expressed are, in our view, erroneous.

Cover provides a greater chance of a direct hit not applying it's damage or full damage. Generally speaking, though, the more solid the hit the less likely damage will be skipped. But you might get a light wound instead of a kill. Concealment simply determines how likely that soldier may be spotted. Some forms of terrain, such as building interiors, are more abstracted than others, such as walls.

All of this is done on an individual soldier basis. If you have one soldier in a portion of an Action Spot which is behind a wall and another who isn't, cover and concealment will be quite different for the two.

Phillip Culliton.

Hmm. And then deal with the running-in-slow-motion thing. Better anticipation could solve both those issues. But that's not an "easy" AI problem. Believe me. Oy. Let alone making it all look good.

Or put another way... collision detection is difficult enough to deal with, while collision avoidance is no fun at all

Redwolf,

In the real world the major reason to have shaped formations (instead of just spacing) is to bring firepower suddenly to the right point, and to always be in positions where as many guns as possible can fire at a certain point in the suspected enemy positions. This is true for infantry formations and even more so for tank formations. Tank formations are mainly desired because they cannot shoot through friendly tanks and they cannot reposition to shoot around a friendly easily - in the real world.

In CM friendlies do not block LOF, so this reason for formations is gone. Distance based squad movement would do.

Correct. And when in complex terrain individual placement to achieve the desired level of outgoing fire on a specific target, or cover from a specific threat, is not formation based. Or at least the theory of formations at this micro scale quickly evaporate once the shooting starts.

Having said that, I don't think that overall the state of CM:SF's infantry movement within squads is where it should be. Formations aside, the game doesn't seem to try or at least not be good at maintaining distance between soldiers. And that leads to people bumping into each other and extra casualties.

Spacing in tight terrain is a problem that even experienced units have trouble countering. As I stated above, bunching up has a lot of benefits aside from psychological ones. So a good amount of what is seen in CM is valid, even if not what is printed in the field manual. In other situations, however, I agree that spacing isn't where it needs to be. Technical limitations are at the heart of that far more than conceptual ones.

The matter of whole sections wandering into enemy LOF against player's commands is probably more of a bug in other code pieces and not directly attributable to formation or inner-unit distance handling code.

There are no bugs remaining, even with pathing, that we are aware of. And that's been the case for months and months. What DOES remain is limitations of how "smart" units can be about moving in relation to threats. It's super, super easy for a player to figure out what is optimal, it is very difficult for the computer to do that. Although if you compare v1.11 (or v1.20 for that matter) to v1.01 you can see radical improvement in this stuff even if you exclude the bugs we had to squash.

I believe that the CMx2 engine as of today does indeed uses LOS to individual soldiers to determine whether he can be hit. Fire abstractions only apply afterwards, if you are visible you can be hit.

Yes, as I've said many times in this thread alone

This directly means that code issues that have soldiers wander off into enemy LOS against the player's wishes cause unrealistic casualties.

This is true, though it's a problem with all games of a tactical nature. CMx1 had a lot of these problems too, but of course they involved the entire Squad instead of just a few individuals.

The crux of the issue is with spacing guys dynamically. We can keep them clustered, and not wandering, by enforcing ridged situational placement. The worst example of this (because it's a BAD thing) is Full Spectrum Warrior. However, in doing this we trade off one problem for probably a half dozen other serious ones. Or at the very least making existing problems worse.

Fixing it won't be easy but as is it is a realism problem. I think the individual soldier's position on the map might have to be re-abstracted when it comes to LOF calculations.

Correct that there is no easy fix for this. There was no easy fix for it in CMx1 either, which tells us that reverting to abstractions isn't going to really solve anything. It might nip around the edges in some places, but I bet it will cause problems in others. We'll DEFINITELY see this one in many threads over and over again:

"My guys fired at this dude who was standing out in the open and they wasted 200 rounds on him and the guy DID NOT EVEN FLINCH!!! Then he took aim and threw a grenade at my squad and killed 2 of my guys!!! You have GOT to be kidding me!!!!"

If I know anything really well it's how customers will react. Of course, it's easy. Take whatever suggestion is made here, turn it upside down, and bingo... that's all there is too it

Seriously, abstracting the visuals is not going to solve this. It will, IMHO, make things much worse.

Steve

[Battlefront.com]

akd,

But I think I know anyways:

Could be

Regardless, I hope y'all publish their next tank sim. Steel Fury had great potential but was totally effed over by the publishers (not saying the devs didn't play a role).

I am aware of this and would also like to have a good tank sim in our lineup.

As for the field photo I posted (the one with the 240 gunner in the foreground, I think your perception that they clustered in a CMSF size grid is an issue of perception of distance in the photograph. Those Marines are maintaining at least 5 meter intervals, as I'm sure they were trained to do.

And I am sure you're correct. In WWII march spacing was about the same for a platoon in column. CMx2 can do this a bit better than CMx1, but it's still means clusters instead of massively spaced individuals. CMx2 has the advantage that they do form up into columns and they do spread out quite a bit.

And yes they are not in contact. This goes back to my point. Photos of troops spread out in combat in oper terrain are rare because 1.) they are often very uncompelling images and 2.) photographers are going to have an even stronger urge than trained fighters to seek larger groupings of individuals when they are under fire, so the most common image is a group shot taken from low behind cover. When such photos do slip by the editors, they are often something like this:

Again, I was not saying the photos indicate that everybody is always bunched up. That's trying to prove a negative (or the absence of proof is proof). That's a fool's way of arguing The reason I drew attention to combat videos is to show that CM's spacing isn't nearly as far off from reality as people think because in real life bunching is common. Combat footage is far more useful than still pics, which is why I specifically said to look at videos. And then I backed up my comments by explaining why bunching is not such a bad idea, and even a necessary one (like house clearing), though it does have its drawbacks.

But all in all, I agree that it is not a huge issue in CMSF. I just hope that a future goal would be an allowance for 2 soldiers per 8m grid (or 1 per 5m grid) for, say, movement of a disciplined unit across open ground, but tighter clusterings due to the psychological and terrain factors that CM is capable of simulating.

It's simply a function of processing power and RAM, plus not insignificant TacAI programming. We'll definitely be able to make improvements over time.

Steve

[dan/california]

I admit that I hadn't thought thru the ability to shoot thru your own side in regards to formation mattering less. It is a significant point in the discussion.

[GSX]

Steve, I wont reproduce your huge post here but thanks for clearing a few things up for me there. I guess that what we have now is the best we can have now, if this makes sense?

Suffice to say I can live with it until something better comes along and for street fighting close up works fine anyway, not too good in the open though but as I said, I can live with it. Its just nice to get a handle on the why I need to live with it.

Cheers

[DaveDash]

What I would like to see is some sort of information sharing radius.

So that the 2 man Javelin team sitting RIGHT NEXT to a squad of 12 guys can see the same tank they can.

This would also make splitting squads a lot more viable, which gives you MUCH better formation control.

[Wengart]

I was under the impression that information sharing currently happens, and that is why maintaining C2 is so important.

[DaveDash]

I was under the impression that information sharing currently happens, and that is why maintaining C2 is so important.

The "problem" seems to be that information sharing only happens via C2. I've had numerous times where a machine gun unit or javelin unit cannot see a target that a bunch of infantry, who are right next to them (9 or 12 eyes) can see (yes they both have LOS). In reality, they'd be able to say "hey it's over there".

Implementing a small radius around each unit that could bypass C2 would also mean that one could split squads more often and thus gain a much greater control over formations.

Someone with more knowledge is welcome to prove me wrong, but this is how it seems to be for me.