Why infantry combat in CMx2 is so different

[DaveDash]

One thing that niggles with me sometimes is the automated bit of the action spots for multi section squads. For example if you want a squad to move often one of the action spots alway highlighting where you don't want a team to go.

This isn't a problem because we can split squads but the then they go out of C2.

I'd like to see the C2 taken another level lower to let split sections get some C2 so long as the squad leader can communicate with his split team leader be that by sight, audible, or in modern electronic.

If you give the guys a FACE command at the end of the waypoint, they all get into a good position. And as mentioned, they move around to get good firing positions on their own accord.

Examples: Move infantry to a berm, give them a face command, they will all line up on the crest of the berm. Same with a wall. In forests and other terrain features, they will automatically find the points of best cover in the action spot.

I have very rarely seen units not being able to get LOS due to positioning these days. The ATGM team examples just don't happen in the game much anymore. Your guys move around by themselves in an intelligent manner to get LOS based on their weapon system. The only major problem I have witnessed is vehicles not having LOS to infantry units inside buildings (at the corners) because the vehicle cannot see the centre of the building.

Also AFAIK units share C2 in a limited radius.

[Michael Emrys]

Sometimes giving the player more control is a good idea, but generally speaking it is not. As the amount of interface increases the interest level in playing drops.

I'll try to keep this short because it is not directly related to CM, but recently I experienced a good example of that. In the last month or two I developed a mild hankering to play HoI2 again, which I eventually succumbed to. So I moved the icon back up onto my desktop and broke out the manual to get back up to speed on it. About a third of the way through the manual I realized that I just didn't want to play that badly. It didn't help that the manual is written in a somewhat verbose style that is not organized to help the reader find the information he is looking for. But the bottom line is that there was just too much that the player has to do in order to play this game. Now I've never been a beer & pretzel gamer by a long shot, but I find myself wishing that the designers of HoI could have found a way to streamline it while keeping in the essential parts that could make for interesting play. Instead of it being entertaining or fun, it was a toilsome chore to try to get through it.

That's in addition to many other things that can be criticized about the game, but none of that belongs in this thread.

Michael

[DaveDash]

I'll try to keep this short because it is not directly related to CM, but recently I experienced a good example of that. In the last month or two I developed a mild hankering to play HoI2 again, which I eventually succumbed to. So I moved the icon back up onto my desktop and broke out the manual to get back up to speed on it. About a third of the way through the manual I realized that I just didn't want to play that badly. It didn't help that the manual is written in a somewhat verbose style that is not organized to help the reader find the information he is looking for. But the bottom line is that there was just too much that the player has to do in order to play this game. Now I've never been a beer & pretzel gamer by a long shot, but I find myself wishing that the designers of HoI could have found a way to streamline it while keeping in the essential parts that could make for interesting play. Instead of it being entertaining or fun, it was a toilsome chore to try to get through it.

That's in addition to many other things that can be criticized about the game, but none of that belongs in this thread.

Michael

HoI3 you can put various parts of your country under AI control for this very reason (trade, research, production, etc), and your can put formations under AI control to, from the Theatre level down to the Corp level. In fact, if you wanted to, which you dont, the AI could control pretty much everything.

Illustrates the point, I think.

I find CMx2 is pretty much at the limit of how much micromanagement I want to do. I'd like to see more intelligent behaviour around certain things, to take away some "area target" micromanagement, but adding any more level of player control would just bog it down. Having full individual squad control would be a nightmare in micromanagement, because I doubt the AI would get itself right. Have a look at how retarded the AI is in games like OFP/ArmA etc.

So basically IMO CMx2 is in the sweet spot of fidelity. The CD/Vinyl record analogy is rather amusing (being a part time DJ and dealing with oldschool elitists). Fact of the matter in my opinion is that CMx2 in its current state is a FAR superior combat simulation to CMx1, so much so I cannot be bothered with CMx1 now due to its flaws. CMx1 has a old "wargamers" feel to it though, that like an old Vinyl record, has a lot of nostalga attached.

[Battlefront.com]

Thanks for the HoI2 example, Michael. I'm sure there are people out there who love the features that caused you to give up on it (again), but that's exactly my point. The audience who loves, really loves, that sort of level of control is extremely small. Micromanagement controls are, in our opinion, a lazy solution that eventually crushes the fun right out of the game. This isn't to say that CMx1 or CMx2 had perfect interfaces, because neither did. Difference is that CMx2 will be improved over time in that area.

The Vinyl vs. CD argument is, as DaveDash said, "amusing". When I bought a vinyl record I also bought a high end audio tape. The first time I ever played the record I also recorded it. Then I put it away and didn't touch it again unless I had to replace the tape. Why? Two reasons... 9 times out of 10 I wanted to listen to my music in my car, in my Walkman (heh... remember those?), in my room doing homework, or someplace where my turntable wasn't. I also did it because I wanted the crispest, cleanest recording possible. No matter how careful you are with vinyl the mere act of playing it causes wear and the potential for damage.

As soon as CDs became somewhat affordable I jumped to it with gusto. And with digital music I haven't bought a CD in several years with one exception (wasn't on iTunes at the time). As I listen to randomly mixed music, within a particular genre, sometimes 10 hours a day... the last thing I think about is "boy, I really miss my vinyl".

Steve

[DaveDash]

Thanks for the HoI2 example, Michael. I'm sure there are people out there who love the features that caused you to give up on it (again), but that's exactly my point. The audience who loves, really loves, that sort of level of control is extremely small. Micromanagement controls are, in our opinion, a lazy solution that eventually crushes the fun right out of the game. This isn't to say that CMx1 or CMx2 had perfect interfaces, because neither did. Difference is that CMx2 will be improved over time in that area.

The Vinyl vs. CD argument is, as DaveDash said, "amusing". When I bought a vinyl record I also bought a high end audio tape. The first time I ever played the record I also recorded it. Then I put it away and didn't touch it again unless I had to replace the tape. Why? Two reasons... 9 times out of 10 I wanted to listen to my music in my car, in my Walkman (heh... remember those?), in my room doing homework, or someplace where my turntable wasn't. I also did it because I wanted the crispest, cleanest recording possible. No matter how careful you are with vinyl the mere act of playing it causes wear and the potential for damage.

As soon as CDs became somewhat affordable I jumped to it with gusto. And with digital music I haven't bought a CD in several years with one exception (wasn't on iTunes at the time). As I listen to randomly mixed music, within a particular genre, sometimes 10 hours a day... the last thing I think about is "boy, I really miss my vinyl".

Steve

Well the other reason I found it amusing is because Vinyl has a more "1:1" representation of music (being analog) ie, CMx2.

[panzermartin]

Yeah but I expect a vinyl version of CM:BN, so I can play it on my WW2 gramophone.

[Battlefront.com]

I still have access to a hand crank Victrola with a stack of 78s from the 1920s and 1930s. I'm sure folks back then thought the 33 1/3rd format was a step backwards, what with all those vacuum tube packed recording studios As for me, while it is fun to crank that sucker up every now and again, I'll stick with my iPod thanks very much.

Reminds me of a PBS documentary I saw a few days ago about the observatory built in Palomar, California in the 1st quarter of the 20th Century. One of the great astronomers of the day said he was sure it would never be equalled again, not to mention surpassed. You'd think scientists would learn not to make such prognostications... they're never right

Steve

[Sakai007]

Ahh, but thats the beauty of the human race in general. Why else would we build giant monuments to ourselves? I used to think it was to leave a mark on the rest of history, but my view is different now. They built them because the have the knowledge, the materials, and the man power, and want to show off how great they are. That is why we are constantly trying to build a bigger and better everything, since we have the knowledge, materials, and personnel. Tell me the Hubble is there just to get pretty pictures? Or that sky scrapers are built just a story or two above the next tallest one? All those things exist because we can build them, and we love to show off! Human nature, it's what we do

BTW - All random rants and raves are strictly IMO, just FYI

[Michael Emrys]

Reminds me of a PBS documentary I saw a few days ago about the observatory built in Palomar, California in the 1st quarter of the 20th Century. One of the great astronomers of the day said he was sure it would never be equalled again, not to mention surpassed. You'd think scientists would learn not to make such prognostications... they're never right

Which in turn reminds me of something I read years ago (maybe by Isaac Asimov?), to wit, "When a white haired scientist says that something is possible, he is nearly always right. When he says that something is impossible, he is nearly always wrong."

Michael

[JasonC]

I'm pretty sure that pointing to the strengths for design for effect vs. engineering literalism, and specifically with a focus on the dominant force of various forms of "soft failure" in infantry combat, is the polar opposite of players demanding more micromanagement to fix underperformance in the pursuit of perfect control.

One poster glossed my earlier comments this way - "in CM the soldiers are not reluctant enough to go into their death". X-zactly. Very big part of it.

I think every significant advance in the realism of combat simulations since oh about the mid 1970s, when they were just absymal, has come through innovative design and new mechanics, nearly all of them (I'd stump for literally all but might overlook something) first appearing in low tech board wargames, back when game designers took their craft fairly seriously, and had to in order to stand out. I think the board wargames of today are light years ahead of even the best board wargames of that era, in their ability to recreate the true command dilemmas and actual OR relationships determining the outcomes of battles, operations, and campaigns. And I think they are also light years ahead of computer wargames, on the same subject. Computer wargames excel at immersion and playability, but nearly all of them have had just atrocious game play and OR fidelity. CM stands out by miles ahead of the computer wargame field in that respect - it gets about to the level of good board wargames of an earlier era. It doesn't reach the heights of contemporary wargame design for the board, in my opinion, though those excel at larger scales than it tries to handle.

Infantry combat with morale and a leader-command and control system - boards first. We go - boards first, but much more playable on the computer. Double blind - goes back to original Kriegspiel as played by militaries, again much more playable on the computer. Detailed gun to armor match ups with specific tracking of each shot to different hit locations - boards first (Tobruk system for those unfamiliar). CM did it better but nothing like perfectly. Replayable scenario system with point based force design, boards first. Etc. I understand BTS being proud of what they've achieved on the computer, and bully for all that. But I am not measuring them against video games. I am comparing them to serious board wargames of modern design - which I spend more of my time playing these days, frankly.

Yes you could engineer in lower bravery for the average pixeltruppen. Probably should, but at any rate we can do some of it as players, which is great. But once committed to the engineering solution to everything, first the problems aren't as apparent because the effect being designed for has been tossed - the sim does whatever it does, it is not being measured against a real result. A design for effect designer is trying to recreate a definite thesis about the key OR variables faced, and it is quite clear he hasn't if he misses. Second, because the bits that don't get engineered enough are always those harder to measure or implement, and the whole affair becomes an MM rabbit hole when you try to solve it with the engineered version.

Take cover seeking. What miracles does the tac AI need to be able to perform, to get it perfect, as measured by realistic resulting loss rates and their time profile through the engagement, when using a 1 to 1 engineered approach? The terrain needs to have every wiggle in it. The men need to see every wiggle. The men need to assess every scrap of incoming fire and settle into the appropriate wiggles, some of which are exposed to fire from different locations or to fire of different sorts of weapons, while being vulnerable to others. You get some of the way by those who do living longer and those who don't being wiped out, but this falsifies the experienced loss rate to achieve half the result.

But to design for effect it? Dead simple. Units have some abstract suppression level, taken from the morale system anyway. They have some abstract exposure percentage to incoming fire. Just link the two - more suppressed units fire less often but also effectively have automatically better cover. "Oh but truly billiard table lethality would go wrong" - simple, tie the suppression gain to the terrain exposure level; on 100% exposed train tracks or wire, no improvement is available; in an 11% stone building lots is. "Oh but the high cover would soak up too much ammo and get ammo expenditure per casualty wrong" - so you link that. Every critical variable is directly in the designer's hands. He has to know what key OR relations are involved and what he wants, but if he does he can get it and get it straighforwardly. Without needing to invent lines of code as smart as the self preservation instincts of scores of living men.

I understand the reasons why designers go engineer instead. They think they are going to get gobs to great realism for free. But in fact, they get gobs of lethality and superhuman heroism for free - and no doubt, good immersion too, a fine selling point in a world of first person shooters. But that isn't the metric, "combat realism" or putting the strategy player in the OR dilemma "shoes" of the historic military situation. It is a different one and legit in its own right in business terms, no doubt.

But all the advances in game design, have come from and continue to come from conscious game design. They come relatively slowly and infrequently, and on some platforms can stagnate outright for a decade at times. Basically none have come from engineering realism attempts to avoid the OR driven design problem. And I don't expect any to, pretty much ever.

[JonS]

Jason,

Would you have a word to my soldiers please. They don't seem to understand that they're supposed to be "superhuman heros". Instead they're running away, even though their section hasn't suffered any casualties, they aren't even being shot at, and are in a great position to inflict casualties. It's really mucking up my counter attack.

kthnxbai

Jon

[WillLight]

JasonC, fair points, but do you think it's possible to design a system at this level, i.e. 1:1, etc. and still have it designed for effect? If so, I would genuinely be interested in how you see it done, given Steve's convincing arguments re design for effect not working with the too many possibilities afforded by the higher fidelity. If not, then fair play, design for effect is possible only for by-far-more-abstracted systems. I don't think you'll get much argument there, apart from the one saying this more detailed system is more enjoyable to play.

[DaveDash]

I'm pretty sure that pointing to the strengths for design for effect vs. engineering literalism, and specifically with a focus on the dominant force of various forms of "soft failure" in infantry combat, is the polar opposite of players demanding more micromanagement to fix underperformance in the pursuit of perfect control.

One poster glossed my earlier comments this way - "in CM the soldiers are not reluctant enough to go into their death". X-zactly. Very big part of it.

I think every significant advance in the realism of combat simulations since oh about the mid 1970s, when they were just absymal, has come through innovative design and new mechanics, nearly all of them (I'd stump for literally all but might overlook something) first appearing in low tech board wargames, back when game designers took their craft fairly seriously, and had to in order to stand out. I think the board wargames of today are light years ahead of even the best board wargames of that era, in their ability to recreate the true command dilemmas and actual OR relationships determining the outcomes of battles, operations, and campaigns. And I think they are also light years ahead of computer wargames, on the same subject. Computer wargames excel at immersion and playability, but nearly all of them have had just atrocious game play and OR fidelity. CM stands out by miles ahead of the computer wargame field in that respect - it gets about to the level of good board wargames of an earlier era. It doesn't reach the heights of contemporary wargame design for the board, in my opinion, though those excel at larger scales than it tries to handle.

Infantry combat with morale and a leader-command and control system - boards first. We go - boards first, but much more playable on the computer. Double blind - goes back to original Kriegspiel as played by militaries, again much more playable on the computer. Detailed gun to armor match ups with specific tracking of each shot to different hit locations - boards first (Tobruk system for those unfamiliar). CM did it better but nothing like perfectly. Replayable scenario system with point based force design, boards first. Etc. I understand BTS being proud of what they've achieved on the computer, and bully for all that. But I am not measuring them against video games. I am comparing them to serious board wargames of modern design - which I spend more of my time playing these days, frankly.

Yes you could engineer in lower bravery for the average pixeltruppen. Probably should, but at any rate we can do some of it as players, which is great. But once committed to the engineering solution to everything, first the problems aren't as apparent because the effect being designed for has been tossed - the sim does whatever it does, it is not being measured against a real result. A design for effect designer is trying to recreate a definite thesis about the key OR variables faced, and it is quite clear he hasn't if he misses. Second, because the bits that don't get engineered enough are always those harder to measure or implement, and the whole affair becomes an MM rabbit hole when you try to solve it with the engineered version.

Take cover seeking. What miracles does the tac AI need to be able to perform, to get it perfect, as measured by realistic resulting loss rates and their time profile through the engagement, when using a 1 to 1 engineered approach? The terrain needs to have every wiggle in it. The men need to see every wiggle. The men need to assess every scrap of incoming fire and settle into the appropriate wiggles, some of which are exposed to fire from different locations or to fire of different sorts of weapons, while being vulnerable to others. You get some of the way by those who do living longer and those who don't being wiped out, but this falsifies the experienced loss rate to achieve half the result.

But to design for effect it? Dead simple. Units have some abstract suppression level, taken from the morale system anyway. They have some abstract exposure percentage to incoming fire. Just link the two - more suppressed units fire less often but also effectively have automatically better cover. "Oh but truly billiard table lethality would go wrong" - simple, tie the suppression gain to the terrain exposure level; on 100% exposed train tracks or wire, no improvement is available; in an 11% stone building lots is. "Oh but the high cover would soak up too much ammo and get ammo expenditure per casualty wrong" - so you link that. Every critical variable is directly in the designer's hands. He has to know what key OR relations are involved and what he wants, but if he does he can get it and get it straighforwardly. Without needing to invent lines of code as smart as the self preservation instincts of scores of living men.

I understand the reasons why designers go engineer instead. They think they are going to get gobs to great realism for free. But in fact, they get gobs of lethality and superhuman heroism for free - and no doubt, good immersion too, a fine selling point in a world of first person shooters. But that isn't the metric, "combat realism" or putting the strategy player in the OR dilemma "shoes" of the historic military situation. It is a different one and legit in its own right in business terms, no doubt.

But all the advances in game design, have come from and continue to come from conscious game design. They come relatively slowly and infrequently, and on some platforms can stagnate outright for a decade at times. Basically none have come from engineering realism attempts to avoid the OR driven design problem. And I don't expect any to, pretty much ever.

You really have absolutely no means to critize CMx2 games based on factors such as disobeying orders and such. Those factors are up to the PLAYER to dictate as he sees fit.

Im also getting the idea from you post you haven't touched CMSF in a while. CMSF isn't "Super Lethal". Are you even aware some of us are griping over on the CMSF forums because certain aspects of the game (HE) are not lethal enough?

Have you even witnessed the TacAI retreat when overwhelmed? Refused to take your orders when shaken? Anything beyond that is entirely in the hands of the player, you can make the game as realistic or as unrealistic as you like by issuing the orders you like. Are you even reading any of the comments presented so far in this thread about TacAI behaviour? Human behaviour? Look at the tactics and tutorials thread in the strategy forum, where a US army officer goes through various missions applying real life tactics in game - and they work. Can you do that with your boardgame? .

Im glad you're into your boardgames, but they have one huge striking disadvantage that they will never overcome; You can't play boardgames against yourself. And its inevitable that computer games will eventually get far far far closer to reality as time, technology, and experience goes on than boardgames will ever achieve with their abstraction. CMSF isn't there yet, it can feel a little robotic at times. But thank god we have guys like BTS pushing the boundries, and more importantly, listening to their fans and supporting/improving the games well beyond what other developers would do.

[undead reindeer cavalry]

But it is still the SAME TERRAIN. The only difference it might make is in the diminished LOS/LOF when such things become important. But as far as the game is concerned, if you are in one corner of that 20x20m spot it is identical to the other corner. There is *NO* difference in terms of the terrain.

This is not true in CMx2. Each meter of terrain in CMx2 can be different. Trees, for example, are explicitly simulated while in CMx1 they were not. So if you have a tree in one corner of an 8x8m Action Spot, then that corner is different than the other corner. There is absolutely, without any question, nothing similar like this in CMx1. That's because for 20m in any direction the terrain was absolutely identical. Slight exception for the few "linear" type terrain features that were allowed in specific Tiles.

variation in terrain types within the same 20x20m tile does exist in CMx1. one spot might be open, one open woods or scattered trees, one roads, one trench, one crater and so on. this should be obvious to anyone who goes and plays the game.

yeah, there's higher fidely in terrain modelling in CMx2 and it is also obvious for anyone who goes and plays the game.

Control and fidelity are not the same things.

i naturally agree.

the only relevancy is what you simulate and that the simulation replicates the real world aspects relevant for the simulated subject.

what comes to CMBN i don't care if it's accomplished by TaciAI, the human player or abstracted game mechanics, as long as it is accomplished well enough.

the rest is just ideological by nature and everyone is free to form their own opinions.

[BFCElvis]

Have you even witnessed the TacAI retreat when overwhelmed? Refused to take your orders when shaken? Anything beyond that is entirely in the hands of the player, you can make the game as realistic or as unrealistic as you like by issuing the orders you like. Are you even reading any of the comments presented so far in this thread about TacAI behaviour? Human behaviour? Look at the tactics and tutorials thread in the strategy forum, where a US army officer goes through various missions applying real life tactics in game - and they work. Can you do that with your boardgame? .

I have read almost nothing beyond the first page of this thread but popping in just now I noticed this.

I wish there was an engaging way I could show this in my current AAR because it has been what I have spent a good deal of the battle struggling with. There have been times when Jons infantry has come out of the woods right on top of my men, undetected until they were in point blank range and because of the condition of some of my units they disobey any order I have issued and crawl, run, hide, do what ever they can for self preservation. And the men do it independently of each other. 2 guys may crawl for cover while 2 others try to get some shots off. They have at times been issued movement orders but because of their condition immediately go to ground if they hear any type of fire, even mortars falling halfway across the map. The TacAI has performed fantastically.

[undead reindeer cavalry]

yeah it sounds very good. those AARs show amazing stuff.

release the game already!

[panzermartin]

High fidelity 1:1, bullet tracking and finer terrain mesh is a double cut knife imo. It needs a very sophisticated TacAI to produce realsitic results most of the times. Sometimes when playing a wargame you want a limit to unpredictability, otherwise tactical planning becomes a very blurry thing. You want to be rewarded for puting the squad in the right place and not losing because the TacAI couldnt get the grenades behind those walls or trunks, or picking the wrong window. Or because of the placement of the squad on the crest half of your firepower is not fully utilized. These are things that should matter to an FPS and not a wargame. CMSF was awful in that respect when it first came out, since troops didnt take advantage of simple things like low walls for instance. Several patches later its incredibly better but still one doesnt feel 100% confident with units behaviour. You have to constanly check complex LOS/LOF paths. In that regard CMx1 was better, because its simplicity made things clearer to the average player and produced realistic results thanks to the abstractions. CMx2 tries a LOT of harder to achieve the same things, in a different way. When it achieves them though, its a jaw droping wargaming apocalypse.

[abneo3sierra]

@Undead reindeer,

I found it amusing that in your replies to Steve, who basically played a large part designing the game, you continue to act as though he didn't understand about the terrain,etc.

I personally like CMX1 very much, but I also understood its abstractions...just because you saw, visually, a tree in one part of the 20 x 20 square, and the rest as clear...the program was treating it as entire trees in that square...I think that is what Steve was saying...because in CM1 games, the program determined terrain as entirely the same, in each square, where in CM2 games, it actually is different in each area...when you see a tree in CM2, it really is A SINGLE tree,where in CM1, it was representing a bunch of trees,etc.

[moneymaxx]

It all boils down to the differences between an old technology that has been honed to perfection where every minor improvement leads to increasing costs and newer technologies in their infancy that almost always start being inferior but have much more room for improvement . (Cmx1->Cmx2, tube television -> flat screen tv, horse cart -> petrol engine car -> electric car?, tape recording -> digital recording...). Many companies that did not recognize these shifts sustained huge losses or went out of business (google Kodak and digital photography).

Is the new technology (Cmx2) perfect, e.g. does the TacAi find every cover and act accordingly? At this moment maybe not but there is still a lot of room for improvement. Some improvements (smaller action spots, more los checks) will come at the cost of more computing resources, but those will become (most probably) available in the future for "free". So if I understood many posts on this forum correctly the new engine scales much better with more computing power. Therefore it has at least the potential to become vastly superior than its predecessor.

A last word on abstractions. Many times they only feel better since they gloss over their inherit shortcomings like low resolution old tv shows look better on an old TV because a high def TV reveals the actual low picture quality. Just one example. In CmX1 you get weak spot penetrations as a certain percentage of all hits. With enough hits you eventually get a statistically correct outcome every time. In CmX2 that weak spot might be blocked by a tree or the corner of a house. That is a much better "high definition" outcome IMHO.

[panzermartin]

Yes but high definition doesnt make necessary something "better" or "more fun". For example chess is the same centuries now, whether its played with plastic pieces or ebony or whatever.It remains a great strategy/mind game.

CMx2 ultimately wants to achieve a self resolving battlefield, with minimal abstractions where everything is taken care of by the capabilities of the units and the TacAI. This is a very, very ambitious plan. Like braking a puzzle down to a million little pieces. Its a great challenge but it might take forever to put together

[undead reindeer cavalry]

@Undead reindeer,

I found it amusing that in your replies to Steve, who basically played a large part designing the game, you continue to act as though he didn't understand about the terrain,etc.

i am well aware of who he is what comes to CM and discussion in these forums. i also think his ability to take contrasting comments and feedback is far underestimated. CMSF and CMBN show that beyond any reasonable doubt.

facts are facts. if one's arguments are against facts one needs to rephrase. Steve can both take and deliver as much bluntness as it takes. and to give him credit it's usually on the delivering side.

[LongLeftFlank]

One quick observation: there is plenty of design for effect and abstraction still in the game, and that's all to the good as far as I'm concerned. An example is the mild nerfing of HE to compensate for the necessary clumping of squads in Action Spots. (or am I behind the times on that?).

[dan/california]

I understand what Jason C is saying about infantry casualty rates and the actual frequency of truly decisive vs attrition based combat at the CM scale. Note I didn't say I agreed with him on either point . What I don't understand is what he wants BFC to do, short of junking the CMX2 engine to write a nice Brigade level simulation where the smallest unit displayed is a platoon or even a company. And somehow I don't see that happening.

Does the way Tac-AI finds cover and responds to incoming fire be improved? Of course it can. Can the fidelity of small cover elements in the terrain be increased, yes, and it will be as soon as the average PC can run it with out smoke pouring out of the vents. I just can't shake the impression that if BFC improved both of these things by a factor of four with CMBN vs CMSF that he would not be any happier than he is now.

I have played some board war games in my time, and I don't need to see another pair of dice or another effects table anytime soon, ever really. Computers were built to deal with that sort of tedium.

[Battlefront.com]

One poster glossed my earlier comments this way - "in CM the soldiers are not reluctant enough to go into their death". X-zactly. Very big part of it.

Again, this comes down to the player being the primary point of failure in any system where realism is a desirable end result. It doesn't matter if we're talking about paper and dice or computer. The simple fact that the player doesn't have to worry about his career or life ending based on his play changes everything. The fact he doesn't have to care about losses like a real commander would changes everything. The fact that no simulation, no matter what method of mechanics it uses, can allow one person to control more than a single body without compromises in realism, changes everything.

What I'm getting at here is that you're applying one standard to engineered games and completely taking that standard out of the equation when applying it to design for effect games. That's an unstable position to set up an argument upon.

Further, I've listed extremely specific and tangible failings in CMx1's designed for effect systems and showed how the same situations work far more realistically in CMx2's engineered solutions. You have not challenged these examples in any meaningful way other than saying if you had a 6 sided die, a CRT, and a piece of cardboard sitting on a bit of paper that you would have even better results.

And I think they are also light years ahead of computer wargames, on the same subject. Computer wargames excel at immersion and playability, but nearly all of them have had just atrocious game play and OR fidelity. CM stands out by miles ahead of the computer wargame field in that respect - it gets about to the level of good board wargames of an earlier era. It doesn't reach the heights of contemporary wargame design for the board, in my opinion, though those excel at larger scales than it tries to handle.

I don't think you'll find many people willing to agree with you on this point. The fact that you are even here indicates that you don't fully believe this yourself.

A design for effect designer is trying to recreate a definite thesis about the key OR variables faced, and it is quite clear he hasn't if he misses.

It is quite clear with an engineered solution too since the outcomes are what matters for both systems. The difference is that the more complex the simulation is, the more difficult it is to achieve desirable outcomes with design for effect based systems. I've gone into great detail as to the "science" behind that statement already.

Second, because the bits that don't get engineered enough are always those harder to measure or implement, and the whole affair becomes an MM rabbit hole when you try to solve it with the engineered version.

The same issue exists with design for effect, as I have already shown in direct examples taken from CMx1 and in a more general theoretical standpoint. The difference is, once again, that the more complex the outcome you are trying to achieve, the less likely a design for effect system will lend itself to a successful overall result.

Take cover seeking. What miracles does the tac AI need to be able to perform, to get it perfect, as measured by realistic resulting loss rates and their time profile through the engagement, when using a 1 to 1 engineered approach? The terrain needs to have every wiggle in it. The men need to see every wiggle. The men need to assess every scrap of incoming fire and settle into the appropriate wiggles, some of which are exposed to fire from different locations or to fire of different sorts of weapons, while being vulnerable to others. You get some of the way by those who do living longer and those who don't being wiped out, but this falsifies the experienced loss rate to achieve half the result.

Not true at all. You're mixing apples and oranges again. Plus, you're also ignoring that the direct example I gave in CMx1 produced higher and less realistic casualties than CMx2, despite what you say.

But to design for effect it? Dead simple. Units have some abstract suppression level, taken from the morale system anyway. They have some abstract exposure percentage to incoming fire. Just link the two - more suppressed units fire less often but also effectively have automatically better cover.

Clearly you haven't actually made a wargame if you think things are really that simple. Or you are at least not being honest with your assessment of its capabilities.

He has to know what key OR relations are involved and what he wants, but if he does he can get it and get it straighforwardly. Without needing to invent lines of code as smart as the self preservation instincts of scores of living men.

As you increase the scale this becomes more feasible. At the divisional level there is absolutely no need, and in fact a counter need, to having individuals simulated. But at a tactical level, where the smallest viable engagement (for fun's sake) can be a reinforced platoon, individuals do matter. And they matter a whole lot. Which then gets into a significant issue of how to simulate individuals. Or to at least portray them, as CMx1 did, abstractly.

I understand the reasons why designers go engineer instead.

Actually, based on your arguments I don't think you do.

Through 13 years of direct experience with both approaches, in an environment with more people playing these games than probably all board wargames combined... we've got a pretty good idea of which works better and why.

Steve

[Battlefront.com]

variation in terrain types within the same 20x20m tile does exist in CMx1. one spot might be open, one open woods or scattered trees, one roads, one trench, one crater and so on. this should be obvious to anyone who goes and plays the game.

No, this is not correct. A terrain tile is either OPEN or one of the Forest types, it couldn't be both. The graphics you see meant absolutely nothing in CMx1. As I have said over and over again, a 20x20m tile was a single type of terrain. It was never more than that except in specific, hardcoded, and limited circumstances where there was a "linear" element in that space (road OR wall OR hedge, etc).

Trenches were not terrain but a type of static unit. Craters and foxholes were similar. Their placement within a terrain tile only mattered in terms of LOS. When in such a "unit" the basic Terrain Tile's spotting and cover variables were modified when combat effects were applied. It didn't matter if the trench or crater was in the NE corner or the SE corner, the modifications were identical.

It is true that the LOS system used a very innovative, but highly abstracted, form of LOS degradation based on distance and what terrain it bisected. But again, there was no differentiating various parts of a 20x20m Tile. It simply counted up how many increments of Heavy Woods it went through and deducted that from the LOS's budget. When the budget was exceeded LOS was blocked. The specific section of that 20x20m Tile that the line passed through was completely irrelevant. The entire tile offered the same, absolutely identical, degradation effect (again, excepting the hardcoded "linear" object possibilities).

Which makes me come back to your original point where you said the fidelity of CMx1 was higher than CMx2. In terms of exact placement of a whole unit, that is true. But the entire environment (as you freely admit) was so abstracted that it's not like this level of fidelity increased the fidelity of the simulation's outcomes.

High fidelity 1:1, bullet tracking and finer terrain mesh is a double cut knife imo.

Yes, this was a well known challenge we accepted when we decided to go 1:1. Just like 3D was a challenge we accepted when we decided to do CMBO. We like challenges. It's how progress is achieved. Playing it safe advances nothing.

In that regard CMx1 was better, because its simplicity made things clearer to the average player and produced realistic results thanks to the abstractions. CMx2 tries a LOT of harder to achieve the same things, in a different way. When it achieves them though, its a jaw droping wargaming apocalypse.

Neither system is perfect. However, I think that most people playing CM:SF now, who have gone back to playing CMx1, have stated that they find the CM:SF modeling of infantry overall more realistic and natural feeling. The reasons for this is that the CMx1 system had quite a lot of modeling issues that weren't necessarily highlighted too much because most wargames suffered from similar problems.

It all boils down to the differences between an old technology that has been honed to perfection where every minor improvement leads to increasing costs and newer technologies in their infancy that almost always start being inferior but have much more room for improvement .

This is a good summary of where we found ourselves part way through CMBB development. We made a couple of improvements to the infantry modeling, to overcome some issues in CMBO, and then found ourselves staring at a brick wall between us and further progress. Whatever faults lie in CMx2 at the moment, there is no brick wall in front of us. That's a very good feeling to have after investing so much time into CMx2. It was totally depressing when we discovered the brick wall so soon into CMx1's life.

So if I understood many posts on this forum correctly the new engine scales much better with more computing power. Therefore it has at least the potential to become vastly superior than its predecessor.

True in many regards. In CMx1 we found the code itself was holding us back. As computing power increased the game didn't fundamentally improve what it did. This is not necessarily true with CMx2, though of course it is a feature by feature sort of thing.

A last word on abstractions. Many times they only feel better since they gloss over their inherit shortcomings like low resolution old tv shows look better on an old TV because a high def TV reveals the actual low picture quality. Just one example. In CmX1 you get weak spot penetrations as a certain percentage of all hits. With enough hits you eventually get a statistically correct outcome every time. In CmX2 that weak spot might be blocked by a tree or the corner of a house. That is a much better "high definition" outcome IMHO.

Good point about the armor penetration issue, and it is true that there are a lot of rose colored glasses being worn by "old schoolers". Same thing happened when CMBO first came out. Some people believed their prior favorite wargame did things better despite evidence to the contrary.

Yes but high definition doesnt make necessary something "better" or "more fun".

True, but often times it does. When we only had CMx1 in front of us people saw Close Combat and FPS games. They said, for them, that the 3 Stooges (infantry representation) of CMx1 was a big detraction from their immersion. Borg spotting was a problem for others. The over simplified terrain options for the Maps bothered still others. Etc.

To solve the issues people raised with CMx1 we had to go with a higher resolution model underneath everything. The sorts of advances people wanted to see simply couldn't be done otherwise (note I reject JasonC's position in advance ).

CMx2 ultimately wants to achieve a self resolving battlefield, with minimal abstractions where everything is taken care of by the capabilities of the units and the TacAI. This is a very, very ambitious plan. Like braking a puzzle down to a million little pieces. Its a great challenge but it might take forever to put together

What can I say, we're ambitious And yes, it will take a very long time to get all the features in and functioning the way we want them to be. But the alternative to this is stagnation and repetition. Stagnation and repetition are death to a game developer, regardless of genre.

Steve