Triumvir

OT: Incidentally, has anyone seen Reynolds' OTHER book, _Men of Steel:1 SS Pz Corps in the Ardennes & the Eastern Front_? It's been rumoured to be out in paperback, but I couldn't find it on the web. ObCM: Playing with hordes of Conscripts, surprisingly, can be more effective than playing with Regulars. I think Steve is right when he says that QBs can come close to simulating historical battles. I once played a random expierience attack/defend QB against an unnamed opponent, with me having hordes of conscript Germans and him having elite Brits. The details are too gory to go into, but once my opponent ran out of ammo, even shaken conscripts could mop him up. I'd like to see an addition to CMBB whereby you could select from the entire experience range of troops. This would not be the default selection, of course, but would be analogous to the Unrestricted option we got for formations in the 1.12 patch. This would let us simulate the gamey Russian tactic of flinging line units at the enemy till they were exhausted, then break through with a fresh group of veteran soldiers i.e. Guards units. I call this the Zhukov gambit. I recognise that there'll always be people who take all elite units when they get the chance, but again I stress that I'd like to see this not as the default option. NB I think I remember something of the sort being posted some time back, but again, no search ran in reasonable time... and I can't remember the outcome of that request. (Besides, minds can change...) [ 07-11-2001: Message edited by: Triumvir ]

I'm happy that my question has been answered. I was quite wrong in my assumptions; nonetheless, je ne regrette rien.

Slapdragon, read my edited post as for why "clicking go" is not an acceptable solution. As for a turnbased wego system for Quake 2, I refer you to Counterstrike, which has a detailed body engine system, relatively accurate ammunition modelling, and a millisecond granular turn. And yes, Counterstrike is a helluva game. I'd also point you at, for your happy belittlement of RTS, Fighter Command, one of the best wargames I've ever played on the Apple II. Note that I have answered all of your snide remarks as politely as I can to be best of my ability. Your refusal to extend that same courtesy, as ever, is entirely your decision. <edited for stupidity> [ 07-09-2001: Message edited by: Triumvir ]

Sixty posts have hit this thread since I last checked the board. I had a nice long post half written before I accidentally erased it; therefore upon completing a section I will save and edit my reply. I will state my point and let the matter stand till the furor cools. Ten years on Usenet has taught me -- but not well enough - that posting in heat cheapens my point and is as contemptible as imputing opinions to others that they do not hold. As to the last, I reproduce Slapdragon's comment: <BLOCKQUOTE>quote:</font><HR>In a nutshell it was "you guys have mo (sic) right to critique our idea, only BTS does." <HR></BLOCKQUOTE> I reproduce my statement: <BLOCKQUOTE>quote:</font><HR>I moot that changing a turn length is _not_ a "difficult and time consuming code change" and frankly, unless Charles speaks up for himself or through Steve, I don't think you can say with any certainty that it is or isn't.<HR></BLOCKQUOTE> I still stand by saying that unless Charles -- the actual programmer -- says that something is doable or not doable, one cannot say with any certainty that something is or isn't doable. <credo> I believe that Combat Mission II's engine -- that same engine which will be rewritten after CM2, 3 and 4 -- will materially benefit from additional turn lengths which are more or less than one minute -- henceforth referred to as variable length turns. Specifically, I believe that turn lengths ranging from 30 seconds to 10 minutes, at minute intervals after the first minute will benefit CM II players. </credo> The case for variable length turns is control over player control of troops. For turns shorter than a minute -- 30 seconds in my proposal -- novice players will have more control over their troops. This will affect their gameplay as it will let them more easily recover from poor planning by managing their troops. For turns longer than a minute, each incremental minute forces the player to plan for contingencies, and forces the player to bear the consequences of improper planning. Conversely, this proposal rewards careful planning and execution, and makes its payoffs much greater. This can be duplicated in the current engine against the AI by clicking go after each turn, since the AI in the present engine apparently issues orders that end at the end of each turn and issue new orders at the start of each turn. This explains the strange phenomenon of an entire rifle company sitting in the open three hundred yards from the ambush zones at the start of every turn, and rising en-masse at the sixteenth, sixth, or sixtieth second to being their move. It can not be duplicated against a human opponent because of the inherent moral hazard problem of an opponent agreeing but reneging on their agreement. This moral hazard can be negated by forcing the game to play in n-minute long turns. The case against variable length turns can be reduced to three objections. The first is that the current game AI is unable to handle turns longer than one minute. Erratic results apparently occur which destroy the illusion that one is fighting a genuine AI. This objection is not easily disposed of. Saying that "well, an improved AI should be part of the rewrite" is a glib and meaningless answer. Nonetheless, I really have no choice but to say "well, an improved AI should be part of the rewrite". 8) Specifically, an AI should be improved such that it is scalable beyond a minute. I haven't worked with AI algorithms since high school, and thus am not qualified to make comments about what should and shouldn't be used for the AI. Only allowing AI play at one minute or less, while reserving longer turns for multiplayer games may offer a solution, but the real answer is, again, "well, an improved AI should be part of the rewrite." The second is that the amount of effort required outweighs the gain from the increased control. This is a very reasonable argument to make, and is at the core of just about every human decision. I moot that, assuming all else stays the same, the amount of effort required to increase the calculation phase from 60 seconds to 120 or 30 seconds is relatively small. As I understand it, the AI sets goals before the calculation phase, then the game engine takes over and actually performs calculations. I obviously do not know how BTS resolves their calculation phase. Making an educated guess, however, based on approximately fifteen years of programming experience, they use some form of counter to calculate at each discrete time interval -- probably no less than half a second and almost certainly no more than one second -- what has happened in game reality. Increasing either the start or end condition to a variable amount does not seem like an inordinate amount of effort. Again, I stress that the above is an educated guess, but I would be extremely surprised if it did not in general describe their resolution. The third argument is that changing the turn length -- more to the point, shortening the turn length -- somehow makes CM "not CM". This argument is mere sophistry -- a wego system at 15 second intervals bears an obvious family resemblance to a wego system at 150 second intervals. Arguments about CM scope can be disposed of by Steve's statement nearly a year back that CM was not really intended for huge games -- CM has exceeded its scope; does this not imply that scopes can change? I don't intend to dive into the murky depths of CM being the pinnacle of achievement -- to mangle Churchill, CM is the worst wargame ever created, except for everything that came before it. And on that note, I leave this topic for at least another two weeks. [ 07-09-2001: Message edited by: Triumvir ]

Thanks, John; I was beginning to feel a bit pressed by Slapdragon. BTS, if any of you guys are reading this, I would be more than happy to devote my time, gratis, to helping you code any portion of a new CM engine. I have and am running a programming team (and one that telecommutes, no less), so I do understand how difficult it is to work with multiple programmers -- a single programmer can tie things together far more efficiently than multiple programmers because he has the vision of what his code should be. Nonetheless, I'd be more than glad to help work on any code; I've been coding off and on for about the past fifteen years, starting with Basic on the Trash-80, through to 8080 assembler on up. Most of my work in the past three years has been finance related, but I'm more than willing to do work for you if you're so inclined. Slapdragon: It would be pointless for me to slap together pseudocode for what I think a turn resolution should be without seeing what BTS's approach currently is. If I don't know their approach, there is no way I can usefully integrate my ideas with their code -- assuming they're willing to let me do so. I suspect you have this well in mind, thus neatly solving an awkward situation by redirection. I've already laid out my rationale for adjustable length turns: turns that take less than a minute allow the player more micromanagement and those that take more than a minute force more thought. At no time should these adjustable length turns supplant the current minute long turns; they are wholly supplementary. Again, you continue to exercise hyperbole; I _strongly_ doubt -- again, I stress that I have been reading the board this past year -- that anyone has raised longer turn lengths in the past 18 days, or any multiple thereof less than 10. As for changes to CM, is it not evident that any changes made to CM will make CM be something else? Which implies that CM is at the peak of perfection, which is obviously false, since CMBB will be different from CMBO. Sophistry is a good skill, but you might want to brush up on it a bit more.

Germanboy: Could you give me a sample search statement which doesn't cause UBB to time out? A simple "ai turn length" and variations forthwith doesn't work; do you have a timeframe by which I can search? As for shortening it, again; it does come down to preferences. I stress -- I'm not looking to _replace_ 1 minute turns, I'm looking to _supplement_ them. If a 30 second or a five minute turn -- which incidentally, will be correspondingly (at least!) longer to process -- fits my needs better than a 1 minute turn, shouldn't I be able to go to them as necessary? (By the by, I still stand by my statement that a CM that's programmed for granular turns can produce CM as we know it as a special case) As for privileged, I meant that -- and should have said, obviously -- in the sense that there is no reason from the code point of view to go with 60 seconds. Apologies for not making myself clear. Slapdragon: <sigh> The point is moot. If you think that I think Charles and Steve are the infinite monkeys, there really isn't much point in talking to you, since what I'm saying is not filtering through. I think that CM is one of the best works that a single programmer has ever made. Still, once more unto the breach. From what I understand from reading the forum, the TacAI has no memory. It decides what happens based on its current state. That explains the occasional "turn away from the Firefly that drove behind that hill and start hammering the PIAT team at 300m" behaviour. I fail to see how this will be affected by a change in the turn length as each TacAI decision is presumably calculated on the fly. For higher-level AI, I can definitely see that more processing is necessary to handle decisions. As I understand it, the AI decides its options based on the conditions at the start of the turn. This is difficult to do when each turn is longer than 1 minute. This is a problem that can be circumvented by playing the higher AI at 1 minute turns and human opponents at 3 minute turns. I would dearly love to see the flow of the AI at the moment, even if in pseudocode (and who uses flow-charts?) I would even more dearly love to work with the code. But as far as I know, BTS does _not_ want more programmers (though I would jump at the chance to help out in the development.) I haven't presented my arguments well. This is obvious, because you're still disagreeing with me. 8) But honestly, from now on, if I refrain from adjectives such as "obviously", "privileged" or "clearly", will you likewise refrain from jumping to either end of the distribution in picking (or creating) datapoints for your argument? For one, I _deeply_ resent your implications about my opinions on Steve and Charles. You sound like Steve when he's been harassed out of his wits, with far less justification. I've been a lurker for the past year and I'm well aware of the arguments for and against making CM an "RTS" game, not to mention the depth of effort that Steve and Charles (and now Kwazydog and Matt) have put into the whole enterprise. Mannheim: I make exactly your point in a previous post -- that a game can only approach command because it's an abstraction. I think, again, that the words RTS are poisoning the discussion because I don't want shorter turns -- I want longer turns.

1)But I _don't_ want a "RTS". 2)I want a CM which lets me plot my orders and stand even further back from the action. If each turn is even two minutes long, let alone five, each player has to really really think about what the objectives are before hitting go. With longer turns, you can't back out of bad orders, and if siht happens, it _happens_. Conversely, with shorter turns, you are forgiven bad orders more easily. Besides, I don't want to replace minute long turns; I want the option of having differing turn lengths. 3) I want the opposite of an adrenalin rush -- or more precisely, I want the option to adjust the amount of adrenalin I want. I think the word RTS is deeply prejudicial to this discussion, because everyone here seems to think that I want to turn CM into some abortion like Sudden Strike. BTS could do worse than taking a look at Steel Beasts' waypoint system which (I know -- it's not the same scale, it's pushing no polygons, it's real-time not CTB, it's blah blah blah not CM blah blah blah really not CM blah blah) has the most beautiful selection of orders that you could want as a wargamer. I don't suggest that they use that system because it's clearly not geared towards CM and it would make programming the AI utter hell. But it does give a glimpse of the granularity possible in giving orders.

Excuse me, but we _are_ talking about a rewrite of the engine. I have not asked for any changes to the system apart from varying turn length. Why do you insist on bringing other elements into this? I don't want a WW2 Myth -- there already is a WW2 Myth mod. I also don't see at _all_ how changing the length of a turn makes CM a twitch game. Changing a length of a turn to 2 minutes allows you to think more carefully about what you need to do; to 30 seconds allows you finer control over your men. Incidentally, I built a conversion of first edition Battletech rules to computer in Turbo Pascal as a school-leaving project when I was 16, but have never gone back to game programming since. We're not talking about random line changes of code. We're not asking for graphics capability in word processors. You're clouding the issue with billows of electronic ink because you don't like the idea of changing this. Grud, but your dogmatic wankel rotary engine emulation is annoying me. I moot that changing a turn length is _not_ a "difficult and time consuming code change" and frankly, unless Charles speaks up for himself or through Steve, I don't think you can say with any certainty that it is or isn't. Would you please pull the cork out and stop trying to knee me?

But why not a CTB system like we have now with finer granularity than 1 minute? I agree with you, "Real-Time" is not the way for CM to go. It forces both players to run on the same clock, which is not necessarily good. (Although there's the analogy of speed chess, which I love, where you play a game in under 5 minutes.) Incidentally, one of the oldest ways to make chess continuous turn is to write down moves before-hand and then execute them simultaneously; two units entering the same space simultaneously are both removed.

Datapoint - I've been in an SP 120mm mortar battalion before, firing 120mm mortars from M113s. Yes, this is a huge difference from WW2 usage, but our SOP was to fire from within the track wherever possible. We practiced dismounting and manpacking the 120mms, but it was clearly only to be done in dire situations. Dismounting a mortar requires digging embedding holes, which is not such a problem for an 81, but is a little bit more time consuming for a 120mm. I'd say that about 1 minute would be reasonable for dismounting an 81mm, maybe 1:30 for a 120mm. Rates of fire varied immensely, but I don't think that it's unreasonable to have perhaps a 15RPM sustained fire for 81s, and maybe a 10RPM for 120s. A 120mm bomb is about 13 kilos, and is a real agony to hold in the barrel when waiting for the order to fire. The 120s we used were light mortars, meant specifically for mounted firing, so they weighed altogether about 100 kilos, roughly 40 for the barrel, 40 for the baseplate and 20 for the bipod. Moving them down to the tracks early in the morning was fun... When we (those of us who went for NCO training) first saw 81s during our infantry NCO course, we all laughed at how puny 81s were... (and guns are a whole different ballpark)

Slapdragon, with all due respect, you're talking utter, arrant nonsense. CM's scale was, as far as I recall, originally no larger than Myth -- I remember Steve saying that they never expected battles larger than 1500pts to be fought, which is why there is no vehicle command system. To say that Myth does not attempt to recreate a realistic environment for historical gaming evades the point that even the most realistic environment is an abstraction and both Myth and CM are abstractions. Yes, Myth uses hitpoints etc, so ridiculous results can occur. But that doesn't mean that lessons can't be learnt from it. Apples and oranges are both fruit. Your RTS argument is very different from mine; My argument is that you should be able to take an Explorer and convert it if necessary, at will, into an Escort. You can do that with code. (BTW, isn't comparing code and cars even more outlandish than comparing apples and oranges? You can't change cars on the fly, but you can write code to change itself on the fly.) I've suggested that the player(s) can set the time between order phases _before the game begins_. Hardly different from the current hard-coded 60 seconds. As for 1 minute vs 59 seconds, why not 120 seconds? 30 seconds? Why should you have a single minute turn, in other words, when you can do calculations for _any_ granular amount of seconds? If you set the time interval before starting the game, why not allow any amount? Okay, so you can restrict things to often-used numbers, but the fact remains that there is no privileged reason to stick with 60 seconds a turn. Honestly, I think you're engaging in ridiculous amounts of kneejerk rhetoric. "It works, don't change it! don't change it! or it'll all go to hell!" sounds to me exactly like some of the traders I've worked with who simply cannot function if one of their graph windows is closed. Or are you seriously suggesting that one of BTS's core principles is not to allow any kind of turn resolution less than or more than 60 seconds? Oni like graphics? Ray tracing? Good God in heaven, what is the world coming to, etc. What horrible things for me to ask (not that I've asked for them.) All I've suggested is that instead of setting a single hard coded value of 60 seconds for calculation, before the game starts you allow the player(s) to agree on the number of seconds between each order phase -- and that you allow the player to set a time limit on order phases. Please, I'm not a sixteen year old script kiddie who thinks that RTSs are 1337; I'm a programmer who has made a suggestion which on the face of it seems reasonable for BTS to implement without rewriting their engine. I've played wargames for nigh on fifteen years, from Squad Leader (not ASL, which I could never love) to War In Russia. Please don't try to slap the twitchfest label on me (and if you think that RTSs or FPSs are twitchfests, you obviously haven't played enough of them.)

Honestly, I don't see the problem with making CM continuous time. A WEGO system naturally falls out of a continuous time system. Allow each user to set the time intervals between order phases to 1 minute and hey presto you have CM as it currently is. People who complain that you have to slow down the game to fit each person's needs and then perform a reductio ad absurdum miss the point that most people will be happy at perhaps a 30 second interval between orders. Me, I'd rather 2-5 minutes between orders phase with more elaborate orders (move to that woods, take up ambush positions 30 m into the woods, after 45 seconds, squad c rush to the next woods while a and b give overwatch), but so long as the orders are given, I don't really see why BTS can't change the calculation period. Longer hands-off periods require more thinking, and to me at least show more realism, because they let you do what a real commander does -- issue orders, move on, and pray like hell your people know what they're doing. (And doesn't TacOps use a CTB system? Why not ask the Colonel to help out?) What's so magical about 60 seconds? Why not 45, 59, 132.5? As any good programmer knows, there are only three numbers in programming; zero, one and many (and one is really a subset of many because if you can run something once, you can wrap a loop construct around it.) On the other hand, I don't want the traditional RTS type system where resource management is key. BTS will never do that, thankfully; but think of Myth, which was a _brilliant_ RTS in 3D that had practically everything in the interface line that we have in CM and that forced you to work with what you had, not sit back and harvest your way to victory. As for time limits, well, think TCP/IP and how you can set variable time limits. And above all, the code that runs a CTB system will _naturally_ fall out to run the same CM mechanics that we see now. There really is no doubt about that, since all that we have now is a special case of a CTB system.

Lewis, don't you ever get tired of trolling here? I mean, really, it's like shooting fish in a barrel (though perhaps not so much in this thread) My proposal for nationalistic trait modifiers; since BTS states that nationalistic modifiers applied across the board are unrealistic, how about this: 1) Create behavioural modifiers eg - quickly suppressed - fanatic (very deliberate choice) - better shots - better climbing ability - super-duper ueber Finnish sissuhood plus 50 to stealth and combat in Karelian forests 2) Create a likelihood table for each modifier (eg 5% fanatic, 5% better shots) where each likelihood is the midpoint on a normal distribution. 3) Generate a table listing SDs for each army (eg 1SD for airborne fanatics, 5SD for Finnish ueber-sissu) to decide what the likelihood of a modifier is. 4) Randomly decide if this modifier applies to a particular unit in a game 5) Repeat 4 until all units for all armies in the game have had modifiers assigned (or not) 6) Play game, rinse, repeat. Ideally, you could use some form of trait system to determine ROF, ammo usage, to drive just about everything, really. This would involve quantifying modifiers, so that there would be some Finns who only had plus 50 to stealth and others who had resistance to Russian sniper bullet between eyes. Preferably, all of these should be hidden from the user's view. Ideally, I'd like to see even the command, control and stealth hidden from the user and made continuous (i.e. +1.13 command)

But that's the point; it used a full power rifle cartridge. And presumably it wasn't a spray and pray weapon because of that. In any case, spray and pray is a terrible idea. The CW idea of individual marksmanship is far, _far_ more effective, especially when combined with the USAn idea of massive artillery support.

How exactly do you see the map before purchasing units? Enquiring minds want to know, damnit!

A shame... but that would make FJs even more gamey meatgrinders than they already are. I still recall seeing veteran FJ squads with a +2 Combat leader putting nearly 400FP onto some poor British squads in the open... Incidentally, for those of you who collect Dragon miniatures, I saw two new FJ figurines recently; one from KG Hansen (the FJs riding on KTs) and one from the Caen timeframe.

But posession is nine tenths of the law... so it's 90% de jure! A classic case of why conscript armies don't do well in wars of aggression... but will smash regulars in defensive wars (at least ideologically) See Victor Davis Hanson for more details (not that I agree with all of his work, to say the least, but he has a good core idea)

Incidentally, did FG42s make it into CMBO? When looking at FJ squads, MP44s seem to replace FG42s. Since the FG42s fire a 7.92x57 cartridge, I'd think that they'd be slightly better at long distance than the MP44s. Admittedly, there were only about 7000 made, but that should have been enough for at least a few to make it into CMBO timeframes. Or is this an abstraction, since the two were somewhat similar?

No one remembers Uncle Bill because he got things done quickly, efficiently, and above all, quietly. No Patton/Monty/Rommel histrionics, just plain old soldiering in the best Indian Army tradition. Slim was in battle continuously from 1942 to 1945, longer than _any_ other general, against a more fanatic foe than any halfwitted Hitler Youth, without the same ridiculously overwhelming numbers or materiel that turned places like Tarawa or Peleliu into foregone conclusions. Bill Slim, most underrated? By FAR

How the hell do you survive a frontal shot between the eyes without being ubertruppen?

Does anyone know where I nominate units for commendations? Setup: I've been reading books by John Antal and thought it would be interesting to duplicate one of the armoured attacks he sets up (first mission in Armour Attacks, if you have the book.) An M1 company was due to attack a platoon position held by three BMPs, one T72 and a platoon of infantry. To duplicate the M1s, I chose Panthers (natch). For the BMPs, I took Cromwells with 95mm guns (so that the hollow charges would mimic Spandrels/Spigots) and for the T72, I took a Firefly. As for infantry, I had one British platoon with four PIATs. The long and short of it is that one lone Firefly killed seven Panthers, entirely through my stupidity. I have screenshots at home (but no Net connection, sob) which show a nice neat line of four Panthers knocked out from the rear, in perfect convoy formation. The gamey bastard AI had left all the units where I'd placed them, except for the Firefly, which enfiladed the approach I was using to the rear of the ridgeline. The first Panther died at 0:05, and every seventeen seconds thereafter, another Panther died. The bastard AI had even enough sense to spray the advancing Panthers with .50 cal fire to force them to button up (though since they were facing forward, that wasn't much help to him.) A very humbling experience

Or just a muthalover? 8P The CM forum is amazingly well-behaved compared to USENET -- it's nothing to do with age because there are 40 year old wankers running around USENET trolling their hearts out. It's because we all know and love this game and its designers... GROUP HUG!

Not wanting to comment on the SS in any way, but... as for Anglo-American proclivities towards killing prisoners, read any decent book about WW I, or read An Intimate History of Killing by Joanna Bourke. On the Western Front, at least, the Anglo-Americans were as willing to shoot German prisoners as the Germans were Anglo-American prisoners. (and before you ask, there were damn few Anglo-Americans to be taken prisoner on the Eastern Front.)

As for Pacific War campaigns being boring, I agree that most of the island hopping campaigns would be tremendously crap to play as either Japanese or American (unless you get some sadistic/masochistic glee in having/receiving 8" naval fire in _every_ game), Steve et al shouldn't forget about CBI (or Uncle Bill!) or the early war. The early war should be especially interesting because the Japanese were more experienced but more lightly armed than the Allies, and force ratios were in general identical.

Nope, whistles were used by most of the major combatants in WWII because they were unjammable comms. The Soviets used them for infantry work quite a bit, from what I understand; and it's still in their SOP to use them when comms are down. I know that when I did section training, I used whistles because they can be heard even over gunfire.