1-1 Infantry - Improvements

[Battlefront.com]

I would much rather spend my time making noticeable improvements to an overall good system than be stuck with something that basically can't be improved and yet needs it. As I've said for eons now... we're taking the long view on Combat Mission. We don't expect, nor need, the game to do everything right. CMAK had many of the same flaws that CMBB had and CMBB had many that CMBO had because they were inherent in the system, hardware limitations, or lack of development resources.

As the game stands now I think its depiction of infantry combat is an order of magnitude more realistic than CMx1's. Yes, sometimes the TacAI does some stupid stuff and you lose a guy or more. That happened in CMx1 all the time as well. The CMx1 archives of this Forum are littered with threads complaining about dumb-assed behavior by infantry, guns, and vehicles. Check if you don't believe me So if the best argument against the 1:1 depiction is "it isn't perfect" then that's an argument that we basically don't find any value in.

If statistics were the most important part of a simulation, well, we're wasting our time with all of this stuff. Just tell me what force you have in an email and I'll email you my force. Then I'll use a spreadsheet to derive the statistical outcome of the battle. Since I'm going to guess that doesn't excite anybody, I'm going to suggest that focusing on statistical outcomes as the basis for what makes a good game, or sim, is just wrong headed. Plus, as any statistician knows, the narrower the range of data and the more simplistic the evaluations are, the less likely individual outcomes can be predicted accurately. Conversely, more data without analysis that are up to that task also fail. CMx1 was really well balanced and produced generally good results. CMx2 is also very well balanced, but since it is more detailed it produces generally better results.

I also take issue with people complaining about CM's infantry spread. Sure, it definitely is too close together sometimes, but generally speaking I think it is about right. Look at combat footage on YouTube sometime. Bunching seems to be worse in real life than in CM. Plus, in real life there isn't a modifier to lessen the effects of indirect fire like there is in CM

Spreading units out over more Action Spots is probably something that we'll eventually be able to do, but it's not on the immediate agenda. It not only increases the load on the LOS and Spotting calcs, but it also means a bigger RAM footprint for all kind of book keeping information. On top of that if you, say, triple the number of units by splitting them up three times as much, then you're effectively tripling the burden on the various AIs. It also requires brand new code, and a lot of it, to get the pieces to coordinate together realistically. Even if we provided players with micromanagement controls this would still all be necessary if for no reason other than the AI Player has to know how to do things like keep units from getting too spread out, covering each other, etc.

It's definitely possible to make improvements, don't get me wrong. I'm just saying that there is a general under-appreciation for how practical it is. CM is already asking a lot from your hardware and we have to make sure we don't ask too much of it.

Even if we could turn the clock back and start CMx2 all over again I don't think we'd change much about the core elements. In terms of the game mechanics the design is solid and the current state of its execution better than any of the CMx1 states. Some of that execution is invisible to you when playing the game, but will be visible to you over time. We, however, see it every single day. Which is important

Steve

[MarkEzra]

Btw, my pop just had a stroke this morning, so any prayers would be appreciated.

Of course...Please if you could tell me his 1st name.

[Battlefront.com]

Redwolf,

The difference is that in the CMx1 you can directly coerce the game into giving less or more casualties by just playing with the casualty probability or the firepower rating or by putting in new modifiers.

Which are just numbers pulled out of our butts and then we get into long arguments with you guys when you feel they aren't right. I see no advantage from where I sit.

If you have a 1:1 game where people become casualties not directly because of probabilities you control, but instead as a result of the TacAI putting them where they are not supposed to be then you have to have a much more difficult task at your hands. If you decide casualties are too high you have to program the TacAI to not move them there and we all know that this is much different than tuning a couple variables.

The same situation exists in CMx1. The TacAI was one of our huge resource drains for that game as it is for CMx2. The difference is, from our perspective, that it's much easier to get nuanced behavior out of CMx2 than CMx1. With CMx1 we often ran into problems where a tweak here to fix Problem A resulted in Problem B getting worse *or* Problem C being created.

Or in other words: the difference between the two methods is not that one is more or less likely to give realistic results in the first implementation. Both will be off, there's no chance to get it right out of the blue. The difference is what happens after you decide which way to tune the initial parameters. In the CMx1 method you can directly adjust. In CMx2? You have to do real programming. And as with all programming, if you fix X you break Y.

If CMx1 were a fully abstracted game, say in 2D and with a very limited amount of variables, I would agree with this. However, CMx1 was sufficiently complex that we had to do "real programming" for CMx1 as well. And because the system itself was much more rudimentary there was a far greater chance of breaking Y when we went to go fix X. We actually gave up trying to improve CMx1's TacAI because it had already hit the point of diminishing returns for us sometime during CMBB development.

Think about it this way. If I'm mixing a drink for you that is straight up, then obviously my ability to produce the results you desire is pretty simple. Give you the correct glass, the right type of beverage, the correct brand, the right amount, etc. It's a piece of piss, as they say But what you really want is a martini. And you have a very, very clear idea of what you want that martini experience to be like. I mix up your martini.

Your first complaint is that I served it in a plastic cup. "What does it matter, it's still the same drink, right? Plus, I don't have any martini glasses on hand". So you grumble, but decide to make do because what's really important is the martini, right? Well, WTF... no olives!

"I didn't think they were important so I didn't put them high up on the shopping list. Sorry, nothing I can do about it now. Just imagine they're there". But a martini without olives is a different thing, so you dig in your heals. "OK, I'll take a look in the back of the fridge. Hey, what do you know... I have an olive jar with some juice in it. I can put that in, however I'm not really sure how much represents what is normally coated on the outside of an olive, so I'll just have to guess. And if you don't like it then I'll have to guess again."

You now ponder if you really want to keep going, but what the heck... you have nothing else to drink and your thirsty. Plus, you already paid me $2.78 to make you your dream martini. You take your first sip and find the initial flavor not all that bad, but the temperature isn't cool enough. So I take it and put it in the freezer for a while while you sit around. It comes out and hey, what do you know, some freezer crud fell into it when I put it in. Really must use defrost more often! Well, no problem... I'll just dig around in there with my fingers and get it out. I wonder if I washed my hands like the sign in the bathroom said? Oh well, you won't notice.

So there it is, cooled down. Now you take another sip and realize it wasn't the temperature. It turns out your favorite vermouth, the one you were picturing in your mind's eye, wasn't what I used. "Hey, I'm not a bar... you get what you get." You grumble and ask me to do something about it. "What am I supposed to do, throw it out and start over again?" You decide at that point that you're going to enjoy it or not, though you will continue to complain regardless of the decision.

And then next week you come back and ask for another martini because we're at a weather research station in the Antarctic and therefore the choices for getting exactly what you want are worse outside with the penguins.

Seriously though, that's what this is like. We can make a better martini now because we have a better setup for it, partly because we know better what people expect. But inherently there is room for improvement in the final product. But we have a far better presentation and the refrigerator has several types of olives and the bar several choices of alcohol. Coupled with more experience, having more options at our finger tips means we have better control over the end product, not less.

Steve

[Battlefront.com]

theFightingSeabee,

Sorry to hear about that. Strokes are a scary thing.

Steve

[dan/california]

Well you did decide to make a living on the worst, most irascible, and generally impossible to please customers on earth. And have done for quite a while now, your surely past being surprised at our lovable natures by now aren't you.

[Battlefront.com]

I'm neither surprised nor upset, nor disappointed nor confused, by this discussion If there's been a discussion worth having on this Forum once, it apparently is worth having a couple dozen more times

This is, BTW, a completely valid and (sometimes) interesting debate to have. Some of the games I think the most fondly of worked on computers with 48k of RAM. Using Redwolf's logic, there was no need to make the improvements in Steel Panthers after Kampfgruppe. There was no need to make Close Combat after Steel Panthers since Steel Panthers' simplified roll the dice system was fine. And of course, Combat Mission Beyond Overlord was just about the dumbest venture of them all because it not only broke with table based results, but it introduced variables into the mix which made the system far more prone to actuarial and graphical error. CMBB was just an insult to an injury. So on and so forth.

But as much as I loved Kampfgruppe, I have zero interest in playing it again. Oh... OK, maybe a few minutes for a chuckle down memory lane, but not beyond that. And I certainly wouldn't pay $45 for that chuckle It's not just the graphics and sound effects that I'd find lacking, it's the overall simulation. It was fantastic in its day, but it just doesn't hold up to current standards.

The truth of the matter is that I found Steel Panthers' abstracted depiction of infantry to be horrid. I found its abstraction of armor to be slightly less so, but not that much. I liked Close Combat's depiction much, much better. But of course I felt it had room to be improved. And by improved I mean a more complex depiction of reality so we could get better overall results from it. CMBO did that and we're still on that path ahead.

BTW, if anybody knows about Dupuy's Quantified Judgement Method of Analysis (QJMA) they would know why I don't put much stock in the ability for simplistic systems to adequately cover tactical combat's vast range of possible outcomes. His institution's work on understanding the relationships between reality and probability, however, are quite useful. But for a tactical environment? Big flop

http://www.dupuyinstitute.org/tndm.htm

Steve

[akd]

I also take issue with people complaining about CM's infantry spread. Sure, it definitely is too close together sometimes, but generally speaking I think it is about right. Look at combat footage on YouTube sometime. Bunching seems to be worse in real life than in CM. Plus, in real life there isn't a modifier to lessen the effects of indirect fire like there is in CM

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.

Then add on top of this editors who are going to cut uncompelling, wide shots of the battlefield with abandon and you are not likely to get a good sense of typical dispersement via footage. Pictures like this:

or this seldom make the papers:

but shots like this are very compelling because of the human detail and emotion packed in:

[JonS]

... because of the urban nature of much recent combat footage, tighter groupings of individuals are much more likely to be encountered ...

And therein lies at least part of the problem. What, from a computers POV, is urban, or close, or open, or ... ?

Sure tac fmns would be nice. But they aren't very high on my personal wish list, in part because I can see the AI making a complete hash of it. I can get acceptable - to me - results now by manoovering the elements of a platoon, rather than worrying about manoovering the elements of each section (and anyway I can already do that too, if I want, by splitting the sections).

[Redwolf]

Thanks for the reply, Steve.

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.

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?

[[hirr]Leto]

Thanks for the reply, Steve.

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.

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?

If RT was killed off, would that provide the resources necessary for implementing some of these much needed (but I agree, quite difficult to implement) changes?

(PS, RT fans, no offence of course, but you can send nasty letters run through your crack to me at my beach house address in Hawaii if that sates your hate on...)

: )

Cheers!

Leto

[Ron]

The talk of SOP's and killing RT missed the boat; CMSF is on the other side of the river now. Wake up!

[YankeeDog]

Leto;1144734']If RT was killed off, would that provide the resources necessary for implementing some of these much needed (but I agree, quite difficult to implement) changes?

(PS, RT fans, no offence of course, but you can send nasty letters run through your crack to me at my beach house address in Hawaii if that sates your hate on...)

: )

Cheers!

Leto

Speaking as someone who plays primarily turn-based, as I understand it "killing off" RT wouldn't really free up any resources for anything because the current game engine is pretty much built from the ground-up as a real-time system. In order to gain any "resources" (by which I assume you mean CPU cycles and such) by changing back to a WEGO only system, they'd pretty much have to start again from square 1.

So unless you want to wait at least 3 years for the next CM game, you probably need to reconcile yourself that CM is going to be a RT system, with a WEGO option.

Not that I mind the RT thing at all. I actually like it. The primary reason I stick to turn-based is that I really like the cinematic aspect of CM, and I like to go back and watch the action from many different angles, which if course you can't do in RT. If they ever get around to adding a rewind function in RT, the bulk of my playing will probably be in RT.

Cheers,

YD

[[hirr]Leto]

The talk of SOP's and killing RT missed the boat; CMSF is on the other side of the river now. Wake up!

Well that I kind of knew... I suppose I should have framed the question as hypothetical and not something that I actually expect would be in the domain of contemplation.

I am sure that RT has its charms (such as the option for more detailed unit command), but I've tried it several times with an inability to get past the ADD aspect of my 'grean leaf gummy tree' damaged cognitive abilities. If you don't know what I mean, eat a box of blue whales and you'll find out that a side effect is that you forget math and find Keanu Reeves an accomplished method actor. I really do not know what they put in our food, nor do I wish to know.

Cheers!

Leto

[BFCElvis]

Leto;1144744']Well that I kind of knew... I suppose I should have framed the question as hypothetical and not something that I actually expect would be in the domain of contemplation.

I am sure that RT has its charms (such as the option for more detailed unit command), but I've tried it several times with an inability to get past the ADD aspect of my 'grean leaf gummy tree' damaged cognitive abilities. If you don't know what I mean, eat a box of blue whales and you'll find out that a side effect is that you forget math and find Keanu Reeves an accomplished method actor. I really do not know what they put in our food, nor do I wish to know.

Cheers!

Leto

Don't you owe me a set up or somefink?

[akd]

And therein lies at least part of the problem. What, from a computers POV, is urban, or close, or open, or ... ?

Sure tac fmns would be nice. But they aren't very high on my personal wish list, in part because I can see the AI making a complete hash of it. I can get acceptable - to me - results now by manoovering the elements of a platoon, rather than worrying about manoovering the elements of each section (and anyway I can already do that too, if I want, by splitting the sections).

Let me put it another way: if a unit is spread out in urban terrain, it will be very difficult to take a picture of, unlike, say, a squad in a field as above. However, when a squad stacks for an entry, then you have a great photo opportunity. Stacking for entry obviously happens with great frequency in urban terrain.

[JonS]

I can - right now, with no changes to the game - spread a platoon across the breadth and depth of any map - urban or open - with no more than 4 men (and often less) per tile.

Just how spread out do you require?

[[hirr]Leto]

Don't you owe me a set up or somefink?

Yep I do. I've been playing vs AI trying to refamiliarize myself with the game. I will email you to set up a game tonight or tomorrow.

Cheers!

Leto

[Dietrich]

I can - right now, with no changes to the game - spread a platoon across the breadth and depth of any map - urban or open - with no more than 4 men (and often less) per tile.

No more than four men on a tile? Even a fireteam of five men? Is there a fireteam-splitting command I was unaware of all this time? :confused:

[JonS]

Well that depends on the section, now, don't it

[Paper Tiger]

When I play with BLUE forces (or even with Syrian Special Forces/Airborne units), one of the first things I usually* do is split those squads/sections up before I expect to come into contact with the enemy. Naturally, to avoid them forming back up again, I keep them quite well spread apart. Marines and Stryker MOUT Infantry squads even split into 3 groups which is even safer.

Obviously this is NOT what you're looking for. You lose some spotting efficiency but weighing this against the potential for casualties infilcted by a lucky RPG hit, it's a sacrifice I'm willing to make.

* sometimes, I forget...

[BFCElvis]

Leto;1144757']Yep I do. I've been playing vs AI trying to refamiliarize myself with the game. I will email you to set up a game tonight or tomorrow.

Cheers!

Leto

On second thought we might as well wait for the patch.

[GSX]

I can - right now, with no changes to the game - spread a platoon across the breadth and depth of any map - urban or open - with no more than 4 men (and often less) per tile.

Just how spread out do you require?

Yes that is entirely possible. But has nothing to do with realism for infantry.

Even without infantry IA drills having some basic formations would vastly improve the realism of the game.

My whole argument is based solely on the fact that the infantry have been made the focus of the game but dont act the way infantry do. Now I fully appreciate that theres no way in the world that a PC game will ever recreate real life infantry however, a few basic formations would add to the realism.

[Guinnessman]

Even without infantry IA drills having some basic formations would vastly improve the realism of the game.

My whole argument is based solely on the fact that the infantry have been made the focus of the game but dont act the way infantry do. Now I fully appreciate that theres no way in the world that a PC game will ever recreate real life infantry however, a few basic formations would add to the realism.

Agreed. Wee thought that sprang into my head (happens rarely so it's worth mentioning ;-)) - how about having squads/sections that change formation depending on movement orders, direction of travel and fire arcs?

If, for instance, a squad is told to move straight ahead on HUNT, but is allocated a fire sector to it's right flank, the TacAI could maybe then have the squad adopt an echelon right formation? Or if it's the same situation but the fire arc is straight ahead, adopt a line abreast? Or if there's no fire arc and the squad is on a MOVE order, it stays in a column?

Don't know if I explained that very well but hopefully you get the gist...I also have no idea how difficult this would be to program in to the TacAI!

[GSX]

One other thing that struck me is that in CMSF this bunching of men gets a let off as the terrain is generally very open (buildings excepted) and the US is so superior. However, come Normandy, the terrain will be much more dense and the opportunity for infantry ambushes will be much more common. This combined with the fact that both sides will be much more even does not bode well for any attacker.

If every bullet is tracked in the game, and the ballistics model is great, then this is the very reason that infantry formations need to exist, to avoid everyone being hit by those bullets at once. This is after all the reason such formations exist in the real world.

We end up with a fantastic ballistic simulator and a very poor infantry one.

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

[YankeeDog]

@Guinnessman:

Ya, basically the same idea occurred to me. The FACE and/or COVER ARC commands could be used in conjunction with movement orders to tell the squad what the expected threat direction is, and therefore what formation to adopt. No cluttering up the interface with additional commands, minimal additional micromanaging from the player. It would be even better if there was a hotkey you could hold down to automatically bring up the COVER ARC command immediately after plotting a movement order. For example, clicking on/pressing the key for HUNT while holding down the CTRL key (er sumfink) would allow you to plot the HUNT order, and then immediately set the COVER ARC command for that movement leg, without having to go through the interface again to select the COVER ARC command.

Realistically, it's probably still a substantial amount of coding time to get it into the game, but I'd love to see it. Ideally, the formations should change depending on the unit nationality, type and experience. Conscripts, for example, should probably just move as a mob regardless of what expected threat direction the player sets.