Leonidas

ROFL!

An excellent point (and overall a good post). If you were playing a tank simulator, you would be responsible for finding your own HD position. But you would also be given the tools for achieving that position - the exactly visual perspective of the tank and fine control over its movement. In CM, you also have personal responsibility for finding the HD position, but you aren't given the proper tools.

I just saw EATG, and I loved it. I think you guys are missing the point of the love story. It's all summed up by Dainlov in his final scene. The point of the love story is that it shows that Communism can't work, because only one of them can have the girl. Danilov realizes that the Communist system will not create a perfect world, because men are inherently imperfect. It's the same point I made in my post in the now-locked thread about the five year old playing CM. I didn't memorize the exact text of Danilov's final speech, but I thought he laid it out pretty clearly. It wasn't the whole point of the movie, but it was an important part. And my wife reminds me: "You've got to have something to interest the women who are going to the movie." I'm surprised you guys are ragging on the Pearl Harbor movie. I'm just glad that they're making historical movies at all, especially historical movies without a heavy PC slant. Why would a love story be inappropriate in a war movie? Don't the boys at the front think about they girls back at home?

Does the AFV have a turret? If not, the TacAI sometimes has strange attitudes about rotating the hull to bring the gun to bear on a target. I think that if there's not an enemy threatening the tank, the TacAI doesn't engage and won't rotate the hull to bring the gun to bear. I've had this happen with a StuH before. If it's not in the gun's narrow little arc, the area target command is ignored, IIRC.

I've played too many games to count. I'll play just about anything, but I quickly tire of Myst-type puzzle games and vehicle simulations. As for the theory of why other gamers don't play CM, I doubt it's because they don't understand it. Instead, they are looking for something different from their games. Shooters and RTSs give a visceral adrenaline rush. The first person perspective and intuitive controls create a powerful suspension of disbelief, so you feel as if you're really there. The point of FPSs is the emotional rush they produce from being totally immersive. CM isn't that kind of game. It's a thinking game, with no speed requirements and a heavy dose of history. You don't win with reflexes, but with making better decisions than your opponent, and a little bit of luck. Some people like visceral games, some like intellectual games. Some like chess, some like football. And most like different kinds of games at different times. So there's no mystery. It's just a question of what kind of entertainment you prefer.

I'm assuming I would be classed as a gun owner, for my anti-PC comments. Except that I don't own a gun, have never owned a gun, and have only once or twice even fired a gun. But I'm glad that other people own guns, and I wish more of them did. It keeps the crime rate down, and it's vital to maintaining a free society.

I think I understand the idea. And there are lots of games where this kind of approach would be interesting, though implementation can be problematic. I've made a detailed proposal for multiplayer Thief that would involve what you're talking about, where the script files detail guard patrol routes and behavior. But scripts can have unintended consequences. Tribes had an extensive scripting system, but it was a little too powerful, and became a bit of a problem. The only way to be good at the game was to get good scripts, which meant you had to spend hours combing through web sites or personally learning the scripting language. Neither of those activities was much fun, so Tribes quit attracting casual gamers. I hope they eliminate it or cut it way back in the sequel.

That would be too detailed for me, and I thought I was extreme on the detail issue. I want more detailed orders because I think that real soliders get pretty detailed orders. That is, they know whether they are free to use ammo or whether they should be saving it. They know whether they are trying to avoid enemy detection or whether they should engage any targets they see. They know whether they should keep their AT gun hidden until they see armor, or whether the brass has concluded that there is no armor in the area, and they should load HE and fire at infantry. The tanks know whether enemy armor is still considered a threat in the area, or whether the area is considered clear of enemy armor, so they should engage infantry. But even I wouldn't want to go beyond that, to writing actual scripts.

Everyone would love to see this. There are lots of researchers working on these things full time. If it happens, I'm sure the initial applications will be in areas much more lucrative than computer games. But from what I can tell, nobody is anywhere close to dramatically improving AI. In fact, I don't think anyone even has a very good theory as to how AI might be dramatically improved.

You are, of course, entitled to your view, but it makes no sense to me, either theoretically or empirically. Do animals in the wild have to be taught about sex and violence, or do they do those things instinctually? The slightest bit of thought would indicate that of course animals must have instincts towards sex and violence, or they wouldn't survive and reproduce. If there has ever been a species in which the young had no instincts about those things, that species wouldn't survive for long. And then there's the empirical side. Have you ever watched boys and girls at play? I've been watching my girl for the past 13 years, and I guarantee that we didn't have to teach her to be interested in boys. As for Lucke's statement that violence and sex are unrelated, I think they're related in that they are powerful primal animalistic instincts that can be extremely destructive unless they are harnessed and controlled. But a discussion on sex would be off topic for this forum, so I'll stick to violence. It's important whether we view children as pure little angels corrupted by society, or as free moral agents with the power to choose evil, and possibly the inclination to do so. If you believe that children are inherently pure, then you will logically conclude that social engineering can end all violence and produce heaven on earth. These thoughts are dangerous. Remember that WWI was called 'The War to End All Wars.' People really believed in the twenties that if everyone just cared really hard, there would be no more wars. This delusion caused them to ignore Hitler's rise, capped off with that most classic of statements by the liberal Neville Chamberlain after appeasing Hitler in the late thirties: "Peace for our time." If WWII has taught us anything, it should be that the best way to invite war is to refuse to prepare for it. On the other hand, if you accept mankind's natural inclination toward violence you will realize that you must always be ready for war. And once you are ready for war you won't see much of it, because you can stop modern day warmongers (Saddam Hussein, Kim Jong Il) before they powerful enough to fight you. Millions of people have died to teach us that lesson. I hope we can remember it.

This is offensive and insulting. [This message has been edited by Leonidas (edited 03-16-2001).]

I think of maps in terms of infantry saturation. In a QB I like to generate a couple of maps with the parameters that will be used, so that I can get an idea of the tree cover that will be available. My basic rule on defense is that I won't put platoons so close together that the attacker's arty can hit them both at once. And of course, the infantry must have cover and a decent avenue of fire. Under that rule, there are only so many platoons you can fit on a given map. Infantry is better on defense than on attack. Infantry on the attack is trickier, but the basic idea is to keep them moving or hidden until they contact the enemy. [This message has been edited by Leonidas (edited 03-15-2001).]

I've got a 13 year old girl and a son due in August. My experience is that kids are much, much smarter than you think they are. They naturally start thinking about sex and violence at an early age. Don't fall for the PC notion that exposure to violence is a problem for kids. The question is not violence, but moral content. The Book of Joshua is about as violent a tale as you'll find. To me, the really evil stories are those the push moral relativism, regardless of whether they have any violence in them. Given the condition of the schools, the only place your son is likely to learn much history is at home. So take advantage of any interest he shows in anything historical, and slip in a little history lesson whenever possible. The key with children is telling them the truth. You never help a child by lying to him. The child will eventually discover the truth, usually sooner rather than later. And when he does, he will be less inclined to believe you in the future. The guiding principle of child rearing is to always tell the truth to maintain your credibility. You'll need that credibility in the teenage years.

Hooray! Nabla may have won yet another convert to the cause of more detailed orders in CM. My list of detailed orders would be a little different, but at least we can agree on the basic principle more detailed orders are needed.

I guess the question is where the difficulty lies in writing the AI. I assume that that hard part is figuring out how the unit ought to act, rather than making the unit act that way. If my assumption is correct, then more narrow-purpose commands should make the AI's job easier. For example, suppose an infantry squad comes under fire. Should they 1) ignore the fire and advance, 2) charge the firing unit, 3)stop where they are and return fire, 4) retreat to cover and then return fire, or 5) retreat to cover and try to get out of sight of the enemy? Choosing one of those is a complex business, depending on what's firing on them, what the range is, whether the squad can hurt the firing unit, the terrain that the squad is sitting in and the terrain close at hand. Then you've got another set of questions and determinations for infantry that see a target: Move to close combat? Stop and fire? Ignore? Keep moving and fire? Again, that's a difficult decision for the TacAI to make. But it doesn't seem like executing the actual action would be all that difficult to code. In particular, what's complicated about stopping and firing? What's complicated about ignoring the surroundings and keeping moving? Now there would be considerable complexity in finding appropriate nearby cover, but thankfully BTS has already done a fine job there, as we see every time unit morale breaks. Again, I'm no programmer, so this is not based on experience. But it seems that the more detailed the commands, the easier the TacAI decisionmaking becomes. If the player says that the squad is to be cautious, then the TacAI can limit its decisionmaking options to the defensive ones. If the player says that these squads are to boldly charge the enemy position, again the player has done the decisionmaking for the TacAI, eliminating an opportunity for the TacAI to make the wrong decision. This is more of a technical issue. You wouldn't want to junk up the interface or overwhelm new users. My suggestion is to create unit settings separate from the movement commands. That way, you wouldn't have a long list of commands, but rather a grid of intersecting settings and move commands. I agree that CM does some funny things with time. Most things should actually take much longer. Command delays are unrealistically short. But I assume that's because real battle takes a long time and is rather tedious much of the time. CM has to be fun in addition to being realistic. It's definitely too complex for the AI. Whether it's too cumbersome on the player depends on exactly how complex the commands are, how intuitive they are, and how well designed the interface is. My suggestion is an extra setting for each unit: aggressive, normal, and defensive. Those interact with the movement commands. I would eliminate Sneak and only have Run, Move, and Crawl as movement orders, because Sneak is really just a Defensive Move. But then standing still is effectively a move order, so the total possible orders would still be twelve. Out of those twelve orders, I think you could have the player take a lot of the weight off of the TacAI. Does that sound cumbersome, to have three movement orders, and three options on unit aggression?

There was a quote in this thread (though it seems to have been deleted) that claimed to be the BTS position. IIRC, the BTS concern was more practical and technical than theoretical. It sounded like they found it very difficult to get the tank AI working acceptably, so now they are a little wary about launching into more complex infantry commands, which could require similarly complex issues. If this is in fact the BTS view, I find it encouraging. Once the resources become available, perhaps they will tackle the job of refining infantry behavior.

You're not the only one to think of these things. I doubt you would need to go to these lengths to know the opponent's force. Just start a separate game with the same parameters. I don't think the computer's choices vary much from game to game, unless you've got random countries or services.

Doesn't computer purchase give each side a pretty solid idea of what the other guy has? Seems like it would take a lot of suspense out of the game. A feature I'm hoping for in CM2 is more control over QB setups. For example, I'm a fan of Fionn's Short 75 rule. With every opponent I must go over the rule in detail and occasionally quibble about the details - does no arty over 105mm exclude 4.2 in? What about 4.5 in? Then I have to hope that my opponent actually has applies the rule correctly - and my opponent has to worry about the same thing. Wouldn't it be nice if you could write a little file that automatically imposed rules on the QB? The file lists the only units that are permitted, and it is enforced in the unit purchase screen. Then there's no quibbling, no interpretation, and no worrying. Suppose you take that a step further, and a history buff writes a set of rules that push players towards more historical purchases. The little file could be a simple list of the available units, including available quantities: "The battalion has no Tigers available this week. Sorry." Take a step beyond that, and you could write routines that generate availability pools based on the game's parameters, with some randomness thrown in. So a grog could say that tank X was technically available in month Y, but its numbers were tight, so there's only a 40% chance of one being in the pool. In later months, the chance of it appearing and the numbers available might increase. The size of the pool might also depend on the size of the battle: 40% chance of tank X being available per 1000 points in the battle. A somewhat random availability pool could make buying units very interesting. This would be some work for BTS, but I'm not suggesting that they write the actual probabilities. They could just allow QBs to be limited to a force pool list, and then implement the ability to probabilistically describe what should appear in the list. Then respected grogs could fill in the actual numbers. This is probably just a pipe dream. But would you history buffs like it? [This message has been edited by Leonidas (edited 03-14-2001).]

The Hummel is an awful lot of fun. I don't know of any other way to get control over those massive explosions. If you're in a mood to knock around the AI, try the Hummel on defense. Knock out his armor, then pull out your Hummel and target his infantry huddled in the trees. When you know you'll be kicking your opponent around, the Hummel is great fun. Against a serious opponent, though, the Hummel's weaknesses become a problem. It's not only vulnerable to other tanks, but also to armored cars and any arty, including mortars. The Hummel is also a little slow, and the ROF is miserable. Another problem is the tiny traverse angle for the gun. If the target leaves that little arc, the Hummel will not rotate the hull to find a new target. So your Hummel can spend a lot of time sitting in the open and not firing, unless you're careful. I will occasionally use Hummels on defense, as a mobile infantry gun. Place it in an anticipated line of attack, but not generally visible. Once the enemy inf comes into view, take two or three shots for a turn, really ruining some platoon's day. Next turn, back up and hide while the enemy arty is counting down. Later in the battle you may be able to bring it out again. Another point: The Hummel may be the best way to blow up a buliding before the infantry can escape. With the latest patch, infantry now abandons a building under area fire from a tank, preventing most inf casualties from collapsing buildings. But I bet a pair of Hummels could flatten most buidings on the first shot, before the occupants have time to move out.

Wouldn't that also mean that the squad should gain defensive bonuses under a good leader, because he directs them to superior cover? Reading Black Hawk Down convinced me that infantry deployment is as much about defense as offense. Or is there already a defensive bonus carried in the stealth capability?

I've played both extensively. CM is not quite as pretty, and there's no haunting Japanese music, but in terms of size and replayability CM dwarfs Shogun. Shogun is good for 2-3 weeks, but CM will last you for months if not years.

Note: I ran some experiments last night on Move vs. Sneak. Sneak is 2/3rds the speed of Move, and only slightly faster than Crawl. Let me try to change the structure of the argument. The issue is not control, but rather the amount of detail you can put in your orders. It's always possible that your orders won't be followed, or that they'll be interpreted incorrectly. In that sense, I'm not suggesting more control. In fact, it would probably be more realistic (though maybe less fun) to have less control - longer command delays, less response from units under fire, etc. But in those precious moments when the orders are actually followed, the content of the orders should be more detailed. I'm confused by your assertion that more commands would lead to greater emphasis on the TacAI. Wouldn't it be the other way around? The more detail the player gets in giving commands, the more decisionmaking he is doing and the less there is for the TacAI to do. To put it from another perspective, if something goes wrong for the player now, he looks at his four movement commands and concludes that he must have given the right order, so the problem must be the TacAI's fault. But imagine that the player has a dozen options. Now he's not so sure who to blame. Maybe if he had given a different order, things would have gone better. When you push the decisionmaking onto the player with more complex commands, you also push the blame for mistakes onto him. You say you've led an infantry platoon, so I'm sincerely interested in the realism question: Didn't you give your men detailed briefings before battle? Didn't they have detailed SOPs on how to deal with enemy contact? Or did you give them merely a waypoint and a speed, and assume they'd figure the rest out on their own? Remember, we're not talking about control, but instead the content of the orders you gave.

That's easy: 1) Because it's slow, the attacker is under a clock, and I like to keep my men moving to avoid artillery. 2) Because it's unrealistic, and therefore makes suspension of disbelief more difficult.

I doubt God has much interest in how I play CM. [This message has been edited by Leonidas (edited 03-13-2001).]

Sneak = Slow I haven't tested it, but my impression is that Sneak is much slower than Move. Every time I've used it, my troops start moving so slowly that they rarely make it to the fighting within the time constraints. And Crawl is slower still. In terms of realism, why should infantry have to move so slowly in order to stop and fire on enemy contact? I'll run a test tonight and put some numbers to it, unless someone already has that info on hand.