Obsession with Rock-Paper-Scissors

[grinningman]

Ok, this is a bit of a rant. Why are people constantly talking about a Rock-Paper-Scissors balance between spearmen/cavalry/swordsmen like it's a magic cure? All other things being equal, RPS gives the minimum amount of diversity that will create non-boring tactics. I think that's why it's used in a lot of RTS games. It's an easy way to create a tactical situation that's reasonably interesting.

The Total War games have much more complexity than this. For one thing, the cav/spear/sword RPS ignores missile troops altogether. Then there is terrain, unit facing, experience, speed and morale - all as important or more important than any RPS dynamic in melee combat. I think this is a very good thing.

Of course every unit should have strengths and weaknesses, otherwise people will always or never use certain units. And if people think spearmen are underpowered, boosting their strength vs cavalry might be the answer to make them more useful. But increasing their anti-cav abilities isn't the only way - one of the most useful abilities of spearmen in MTW was that rear ranks could also fight and multiple ranks gave a defensive bonus. I'm not sure how important these abilities are in M2TW, but maybe adjusting them is another way to improve the usefulness of spearmen.

There seems to be an idea floating around that if only we can get a good cav/spear/sword RPS system then combat will be balanced. I think people should keep in mind that if a RPS system becomes the most important factor in unit match-ups, overshadowing the other factors above, Total War battles will become much less interesting, and all armies will end up having a similar composition.

[Booga]

I have to agree with grin here. Real-world battles are not governed by rock/paper/scissors, and there's enough games out there that worship it that I'd rather M2TW blaze its own path. Sure, you can say that some part is unbalanced. Some things are undeniably buggy, like the shields and the 2h units. Beyond that I think it's way too fuzzy to know for sure.

Personally I think the game is far from unplayable - to the contrary, I enjoy it immensely. I think more basic issues with imperfect AI are far more important than whether my town militia got beat up by some peasants. My own thoughts on that whole mess - aren't town militia just peasants with an anti-cavalry weapon and a shield? They're not even a step up on the ladder, just another brand of crap. So I don't find it so hard to stomach.

[Goofball]

There seems to be an idea floating around that if only we can get a good cav/spear/sword RPS system then combat will be balanced. I think people should keep in mind that if a RPS system becomes the most important factor in unit match-ups, overshadowing the other factors above, Total War battles will become much less interesting, and all armies will end up having a similar composition.

But the opposite can also be said to be true. If some sort of RPS system isn't used, then you will certainly have armies compased of the same unit all the time.

Best example of this is STW. I don't know why everybody always waxes so fondly and poetic about this game. As soon as I realised that all I had to do was build a full stack army of Warrior Monks and I would be able to cut through anything thrown against me like a hot knife through butter, I completely lost interest in the game.

Every other TW game since STW has been better than STW, and it's precisely because of the variation of units and complexity that you are talking about.

So I mostly agree with what you are saying.

[Rhyfelwyr]

I don't think a RPS system would limit the use of tactics in TW games. Its pretty historically accurate. Its easier to fight a man on a big warhorse when you have a big spear to poke him with. At the same time, if you break a spearman units line (long spears only, not Town Militia type), then you would find things easier in such close quarters with a sword. If your on a big horse, its easy to cut down men with short weapons such as swords below you.

I think the best system in terms of tactics would be to have a RPS system as the backbone. Then add special functions for every type of unit, to give them a real sense of purpose, something I think units lack in M2TW, although I still love the game. Use long spears or pikes to hold a defensive line or choke points, and defend against cavalry charges. Use two-handed weapons to charge at an enemy line and break it up, then withdraw them as their charge bonus wears off. Try to use halbers, billmen, axes etc as anti-armour specialists. Swordsmen and shortspearmen will form the backbone of an army, and be part of the RPS system in terms of their tactics. Cavalry are used to counter non-spear infantry, and break enemy formations with deadly charges, before withdrawing as the lose their charge bonus, perhaps repeat the process. Lighter cavalry are good for chasing away missile units and catching routers. Archers would be basic support troops. Crossbowmen and javelinmen should be short range anti-armour specialists.

If this sort of system could be implemented, it would really add a whole new depth to M2TW battlefield tactics. In fact, I plan on making a basic mod to do this as far as possible.

[Bijo]

Best example of this is STW. I don't know why everybody always waxes so fondly and poetic about this game. [......

...]Every other TW game since STW has been better than STW, and it's precisely because of the variation of units and complexity that you are talking about.

Welll, I don't entirely agree with the Every other TW game since STW has been better than STW, ..., since I know M2TW isn't as immersing as STW, even when I've checked it out first time recently, STW. And immersion is a great factor to consider, something M2TW lacks.


But I do agree with the battle complexity of units whereof you've all spoken. I think the currently incorporated complexity's good enough, but there's always room to improve, which is what I hope they'll do.

[grinningman]

If you equate 'RPS system' with 'there shouldn't be a single strongest unit', then of course all balancing should aim for a RPS system. I just wanted to point out that a sword-beats-spear-beats-cavalry-beats-sword system is not the only balance aspect going on, and it's not required or even desirable for an interesting, balanced combat system.

Out of interest, does anyone know if spear units do get a defence bonus in M2TW based on ranks as they did in MTW? I haven't noticed. In MTW it was important to keep your spear units in hold formation mode for this reason.

[Carl]

No they don't. they just get 2 attack bonus vs. cav and an attack pealty vs infantry, (the penalty being part of the same thing that gives spears 1 of the attack bounses against cav), and also a charge resistance bonus vs cav too, again this is tied to the same thing that kils their infantry killing powor.

[econ21]

But the opposite can also be said to be true. If some sort of RPS system isn't used, then you will certainly have armies compased of the same unit all the time.

That's a danger, especially in SP where the budget constraint typically does not stop you maxing out on elites. Some mods, or just self-discipline, can overcome it, however.

But I am not sure RPS is necessary to obtain variety of army composition, or interesting tactics. If you think about real military operations, they are not necessarily RPS. For example, for a long time, a tank was the best way to kill another tank. The best way to get air superiority is to have better or more fighter aircraft etc. But you don't see armies composed just of tanks or fighter aircraft. It's not that different arms behave in an RPS way, but they have different comparative advantages or functions.

For example, in the TW period it's not that missiles were good against spears per se (as one RPS interpretation of STW implies: archers beat spears, spears beat cav, cav beat archers). Rather it is just they were useful for softening up an enemy and lowering morale so that your melee units can win with less loss. The STW tutorial, with an archer unit against an attacking spear unit illustrated the edge from a little preparatory missile fire rather vividly. But in general, missiles were pretty useless at holding ground.

Cavalry was good for flanking, due to its speed; for pursuit; for scouting; and for fighting other cavalry. And for shock combat against inferior infantry.

Heavy infantry was good for holding ground and taking well defended ground. In the Medieval period lower quality infantry (spears in TW) might only be useful for defense; higher quality stuff (swords in TW) might be necessary for taking the offense.

I don't think any of the above is RPS per se. I guess it's combined arms - typically having all three arms would generate synergies. Most of it, probably all of it, is pretty well captured by TW - which is why, despite its broader appeal, it attracts fans of historical wargamers like me.

[Lusted]

Rock, paper, scissor balance is not a magical cure, but it is a great backbone on which to build unit balance. There will be exceptions to the rule, and unit types that do not fit the rule, but a rock, papaer, scissor approach gives each units its purporse and leads for more challenging battles, without reducing all armeis to the same composition because of the variety around a rock, paper, scissor approach.

[grinningman]

No they don't. they just get 2 attack bonus vs. cav and an attack pealty vs infantry, (the penalty being part of the same thing that gives spears 1 of the attack bounses against cav), and also a charge resistance bonus vs cav too, again this is tied to the same thing that kils their infantry killing powor.

Ok, thanks. I guess this has been the case since RTW, as I don't remember ranks giving a defensive bonus there either. It's a shame though, that was an interesting ability for spearmen. I suppose the multiple ranks fighting from MTW has been accounted for by spears in M2TW having a longer range.

[Carl]

Not really. Spears only fight in 1 rank in M2TW.

[grinningman]

@lusted

Fair enough, but I think the backbone of tactics in TW games is based around flanking and morale, not a RPS system (despite what the STW tutorial says ), and I really like it that way.

@carl

Well then scratch any possible modifcations to multiple ranks fighting/rank defensive bonus to improve spearmen. D'oh. I suppose pike spearwalls are how they simulate many ranks fighting at once in RTW and M2TW and spears have lost that ability.

[Rhyfelwyr]

But a RPS element is good to have, and realistic as long as it doesn't dominate tactics.

[hoof]

RPS can be a basic balancing system, but it takes the character out of the game, and makes it feel like each unit type is no different than the next, just substitute what it's "good" vs and what it's "bad" vs are, and the 3d shape it uses.

I worked once on WW2 air combat simulators (online), and learned all about how balance and competitiveness can be non-RPS style and work really well. Sure a Spitfire can outturn a FW190A4, but that FW190A4 is faster, and handles better at high speeds. A P38 is also fast, but rolls poorly (except for later models at high speed), and cant follow FW190A4's through high-speed maneuvers due to nimbleness factors. OTOH, the P38 has all its guns in the nose and lots of ammo, so it can spray 'n' pray, while the FW190 devastates on snapshots at short-range. In other words, like the "real thing", air combat doesn't have an RPS style to it, even if in one area (like level turnfights), a plane might have a big advantage over the other. There are definitely certain designs that do better at certain air tactics, and technology did improve (meaning a late-war plane had a big advantage over early war planes due to more engine power, better guns, better armor, etc), but on the whole, similar-era planes were competitive with each other even if one did better at activity A, and the other was better at activity B.

My point is that if you add character to the units, where some things are better than others with different units, then you don't need a RPS style. For example, take RTW's Hastati vs Hoplies. Hastati on the flank or rear chew up Hoplites, but Hastati do much worse head-on (esp. if they can't flank). OTOH, the Hoplies don't do nearly as well on their flank as Hastati do (Hastati, by their flexible nature, can fend off flanking attacks w/o losing significant fighting capability, while Hoplie formations really suffer when an enemy is on the "side" of the unit). This adds character.

Blind "A beats B" and "B beats C" style games bore me because once you figure out the RPS setup, unit-vs-unit fights are foregone conclusions. OTOH, a TW-style game means that isn't necessarily so, you can adjust your tactics and use multiple-unit tactics to overcome a unit-vs-unit deficiency, or use some other method to nullify their advantage (like hurting their morale or fatigue setup).

This variety and character that a non-RPS system gives you is why I keep playing TW-style games. Sure eventually you find combined arms tactics that works very well in most of the situations, but rarely it's a homogenous army setup (with the exception of the armored horse-archers in BI that mowed down East Empire footsoldiers like butter). If your guys are better trained and armored than the enemy, then a single-unit-type army can also work, but it *should* in that situation (you're fielding a superior army, thus inter-unit differences matter less).

RPS systems are usually the first balanced combat design one learns about when learning about game design. However, IMO, it's too simple for anything but a very short game, and isn't necessary if the game has enough competing dynamics to mean that A doesn't always beat B in various situations.

[Carl]

RPS systems are usually the first balanced combat design one learns about when learning about game design. However, IMO, it's too simple for anything but a very short game, and isn't necessary if the game has enough competing dynamics to mean that A doesn't always beat B in various situations.

I think your misunderstanding lusted here.

What he and others are saying is that the base point should be RPS, i.e. you build it on RPS. Then start adding things like the Hoplite/Hastati thing you mentioned. Add enough littile extras like that on and eventually they become more important than the RS balance which quickly ceases to have more than minor effects any more.

Some games can work well without RPS as a starting point, and occasionly those that do work best with the RPS starting point can get by without it. Most attempts that don't use RPS as a starting point that i've seen though tend to fail as good games.

Thus thats all i think Lusted means. They should start with RPS and build up from their, give themselves the most basic foundadtion level to work from.

[Censor]

Pure RPS balance is unnecessary, what we need is a price balance. Cavalry in M2TW is simply too cheap, especially on multiplayer. You can spend all of your allowance on a good combined arms force, but with the average 10k, the enemy can still purchase an all cav force that will destroy you completely. Lowering the allowance is problematic because cav and infantry are so closely priced. What singleplayer needs is a lowering of the cav pool in addition to increased recruitment and upkeep costs. These matters aside, the main problem is that the cavalry charge seems to be indiscriminate of unit type in general. A heavy knight charge can almost be matched by a light cavalry or *gasp* HoRsE ArChEr charge.

[Foz]

The term RPS balance gets tossed around an awful lot... but I think largely this label has been incorrectly applied to what we often mention in the forums. At the very least I can speak for myself in this matter, and I've never felt it is necessary at all to adhere to any strict "each unit beats one other type and loses to another type" standard. My feelings can be more correctly summed up by 2 tenets:

1. No faction should be left powerless or without a good tactical option against a unit another faction can field.

2. No unit should be left without a good tactical use.

The actual application of these points is of course open to some interpretation. While I freely admit that it is potentially possible to open viable tactical options against any given enemy unit using a combination of one's own units if they are designed very well, it is in general far easier to assure this happens by just guaranteeing that each faction has at least one unit capable of besting each class of enemy unit, and in general I consider it a good design idea for this to be the case. Likewise I feel that a unit is included in the game for a reason, and therefore should have a tactical use, which in practice often means it again should beat some particular enemy unit in combat. So I would suggest that the easiest way to ensure good tactics in the game is to make sure each unit beats one unit type, and has a counter to it in each enemy faction. I suppose this is a loose application of RPS, but is also distinguished from that most basic idea by not precluding units to have more than one good tactical use, nor precluding them from having a plethora of good counters in enemy factions, and the ability to do both of these things allows a near-infinite set of possibilities and tactical approaches to the game. In M2TW units that represent many tactical possibilities include knights and horse archers, while a good example of a unit with many enemy counters would be spearmen.

Notice I make no mention as to how viable a given unit must be when it is the counter for another... meaning that there is an awful lot of upward playroom to adjust the unit stats to keep things interesting. If cavalry are decided to be the main counter to archers, then European factions may field better heavy cavalry than the Muslim ones do, but the requirement is still met as long as the Muslim cav retain the ability to at least cause MAD against archers. This in fact allows a LOT of room to differentiate units and factions, as units can be way better than simply competent as a counter to a given other unit. This in turn distinguishes units from one another, and allows for factions to retain their identities by being correspondingly weaker or stronger against any given unit type.

The main reason I arrived at the 2 tenets above is in order to achieve the following, which I feel is required for good and fun gameplay in this genre:

The player should never be in a position where it is impossible to field an army that can possibly make him able to win.

This is not to say that in every state the game should be winnable for the player, but what I mean is that at the start of the game, there should be no strategy another faction could employ that absolutely ensures defeat for the player (or any given faction for that matter), as he/it lacks units able to counter the strategy. It's the primary reason I feel every faction must have a capable response to each enemy unit. The game must always be winnable for the player when it starts, and the easiest way to assure this is to make sure he can field at least one unit that counteracts any given enemy unit. Thus the player always has some kind of out, and it is almost a given from this that he is capable of winning (with good play, tactics, and unit selection) any given game with the units at his disposal. Other methods can achieve this guarantee of a fair shake, but it is often very difficult (if even possible at all) to prove that they do in all cases, and it would be a bad enough thing to fail at that the sure bet is IMO the best one.

[grinningman]

@hoof

That's exactly what I was trying to say, thanks.

@Carl

My point is that I would prefer a factor like unit facing to be more important than a RPS dynamic. e.g. if spearmen charge similar cost swordsmen in the rear, I'd like the spearmen to win. You can start with the RPS dynamic and add to that, but IMO this runs the risk of the RPS dynamic being the dominating factor.

[Carl]

My point is that I would prefer a factor like unit facing to be more important than a RPS dynamic. e.g. if spearmen charge similar cost swordsmen in the rear, I'd like the spearmen to win. You can start with the RPS dynamic and add to that, but IMO this runs the risk of the RPS dynamic being the dominating factor.

Your right about the RPS maybe becoming dominant. But that’s why I want the games to start with that as the base. If they mess up their attempts at other balancing ideas, the RPS system will keep the game balanced 90% of the time. The idea behind the design strategy I described is to ensure the game is balanced if they DON'T successfully implement something else. I've seen too many examples of people trying something different and it going wrong and the game being IMBA. On the other hand, if they implement something else successfully, it will override the RPS 90'% of the time.

As to your thing regarding Flank Charges: NO, NO, NO, NO, NO. It's not remotely balanced or fun. If you want that type of thing I highly advise you to go play WHFB TT. In that game Goblins, (worse than any peasant by quite a margin), can beat Swordsmasters, (Elite Elves who would make fixed DEK look like peasants in comparison), just by getting a flank charge. The Swordsmasters will then rout and get run down 99% of the time.

If that’s the kind of game play you want, then fine, but it's the very fact that the TW series doesn’t have that and has it quite realistic in that elite swordsmen charged by units no better than peasant, (or most missile units in melee), don’t get utterly trounced just because they got rear charged, even though they could have beaten a full stack of said unit in general melee.

It's worth remembering that even Papal Guard, (the best Spearmen), are probably worse in melee against infantry than Yeoman Archers, (who are still pretty weak TBH). The very idea that they could beat units like Dismounted Feudal Knights and Dismounted Chivalric Knights with a rear charge is tot tally absurd without making the entire tactics and strategies of the game revolve around getting rear/flank charges on the enemy and preventing the same on your own units. (Which is what WHFB is at heart).

Foz, actually describes RPS quite nicely for me as it happens. RPS simply means everything in the game has a counter. Cav don't HAVE to be countered by spearmen, but they should be countered by something other than "more/better cav".

[grinningman]

As to your thing regarding Flank Charges: NO, NO, NO, NO, NO. It's not remotely balanced or fun. If you want that type of thing I highly advise you to go play WHFB TT. In that game Goblins, (worse than any peasant by quite a margin), can beat Swordsmasters, (Elite Elves who would make fixed DEK look like peasants in comparison), just by getting a flank charge. The Swordsmasters will then rout and get run down 99% of the time.

If that’s the kind of game play you want, then fine, but it's the very fact that the TW series doesn’t have that and has it quite realistic in that elite swordsmen charged by units no better than peasant, (or most missile units in melee), don’t get utterly trounced just because they got rear charged, even though they could have beaten a full stack of said unit in general melee.

It's not the sort of game I want to play and I was trying to avoid giving that impression by saying a sword and spear unit 'of similar cost'. Of course it would be silly if a unit of spear militia could charge Armoured Swordsmen in the flank and win because it was a flank charge - then unit facing would be much too important.

I think we agree on the basic principles - that no one tactical aspect should become vastly more important that the others. My initial concern was that people were getting too caught up over the sword-spear-cav RPS and not taking into account other important, and to me, more fun, balance aspects.

[Lusted]

My initial concern was that people were getting too caught up over the sword-spear-cav RPS and not taking into account other important, and to me, more fun, balance aspects.

Like i said earlier, and ike Carl elaborated on, RPS should be the base with other variations and dynamics built on it so the game is balanced. It should not dominate the game, but be a part of it to help with balancing.

[Carl]

Ack. I missed the similar cost provision.

Fo what it's worth I partly agree. As I say, I want to ensure theirs some underlying balance their. but pure RPS defintly isn't the way to go. I just see it as a useful starting point and foundation for much more intresting mechanics. Which I think is what CA are intending to achive too.

[Vladimir]

That’s why there should be two major mods like in M1 and RTW. One “realism” and one fun/balanced mod. Then everyone’s happy.

[Von Nanega]

That’s why there should be two major mods like in M1 and RTW. One “realism” and one fun/balanced mod. Then everyone’s happy.

That is a good idea!

[General Zhukov]

That’s why there should be two major mods like in M1 and RTW. One “realism” and one fun/balanced mod. Then everyone’s happy.

Sounds great. That way players could choose between the most "realistic" experience, the Lusted/Foz/Carl Equalizer Fun Experience Extravaganza, and the patched vanilla game as CA's vision of the period. I like it.

[Musashi]

My personal desire is not quite RPS.

I want sword to be offensive infantry and spear to be defensive infantry. I want spear to repel charges and die very slowly. I want their job to be to hold other units, not kill them. "Sword" (By which I mean all heavy shock infantry) I want to be bowled over by charges, but on the offense, to be killing machines.

I want light cav to be great flankers and pursuers but terrible at frontal charges. I want heavy cav to be able to dash heavy infantry to pieces with a charge, but take horrible losses and be held for a long time charging a spear unit head on. I want them to be able to beat, barely, a spear unit of equivalent period in a frontal attack (The idea is to make it monetarily impractical) but lose to pikes every time.

I want archers to annihilate anything if they manage to achieve flanking fire. I want them to generally reduce most units badly when firing frontally as well. Missile troop balance is simple, at range, they win. In close, they die.

That's what I want.

[Vladimir]

The problem with defensive spear stereotypes is that in this game there are several types of spears. I like the variety and each type of unit should be different.

[Puzz3D]

I think we agree on the basic principles - that no one tactical aspect should become vastly more important that the others. My initial concern was that people were getting too caught up over the sword-spear-cav RPS and not taking into account other important, and to me, more fun, balance aspects.

I'm an advocate of RPS (combined arms) gameplay, but not to the extent that it's the only thing of importance. STW was RPS, but it had the minimal amount necessary to ensure that the YS (yari samurai) could beat the HC (heavy cavalry) frontally with no combat bonuses being applied other than the anti-cav bonus which was the only unit-type combat bonus in the game. There was a flaw in STW in that the HC could not reliably beat the WM (warrior monk) which was a consequence of the charge bonus not working. This in itself is interesting because the problem of battle mechanics not working properly which plagues the Total War series goes all the way back to the original game.

Finally, by MTW/VI v2.01, with the help of community beta teams, all the battle mechanics problems were eliminated in the engine, and you have something worth balancing. Those of us that worked on Samurai Wars tried to bring back the original STW gameplay without the flaw of WM beating HC. All other things being equal, the HC can now beat the WM, but it's not black and white. The HC is only 1 combat point (20%) better than the WM, and must rely on a good charge using the 5 combat point charge bonus to win consistently. To put this in perspective, a 4 point charge bonus is equivalent to a 1 point melee advantage. A 1 combat point advantage in melee will generate 6 wins out of 10 fights for these 60 man units. A 2 combat point advantage will generate 9 wins out of 10 fights. A 1 point advantage will probably leave the winning unit with about 50% losses and a 2 point advantage with about 25% losses.

In the case of the extreme infantry matchup of WM (warrior monk) vs YS (yari samurai) you have a 5 point combat differential (250%), but the YS, if put into hold formation, can survive long enough for a second YS to move forward and attack the WM from the rear and the two YS win. As I recall this takes about 50 seconds, and you can't waste any time making the flanking maneuver. The WM does not rout on impact, and has to be reduced to less than 25 men since it's a high morale unit. This maneuver works because the combat bonus for rear attack is large enough for it to work, but not large enough that a single YS hitting a WM in the rear can win. The cost relationships are YS = 400, WM = 1000 and HC = 1200. There are cheaper cav in the game, but they cannot beat the WM, although the medium NC (naginata cav) can win with a charge into the rear.

The magnitudes of rate of attrition, flanking combat bonus and flanking morale bonus are well chosen in the engine and optimized for 60 man units. The relatively large number of combat cycles needed to resolve combat ensures a low enough uncertainty in the result that strategic planning and tactical execution of that battle plan is important to achieving a good result. More simply stated, it means you can play the game intelligently because uncertainly is kept to a reasonable level.

There is a weaker spear, the YA (yari ashigaru = 200), in the game, but it's designed for defending ranged units from cav attack and doesn't have high enough morale to operate independently.

The WM has another weakness in that it's lightly armored. A 60 man archer can reduce a 60 man WM to half strength with 10 volleys if the WM is stationary, and that takes 40 seconds (4 seconds reload). The archer carries 36 arrows, so it can be quite potent if the enemy allows it. The archer cost is 400. While the archer is vulnerable to cavalry attack, it doesn't collapse instantly and therefore attacking cavalry can get caught by supporting units if used too aggressively.

The unit selection is small (14 unit types), but the RPS is complex because there are 3 separate RPS systems working simultaneously. Also, the combat advantage between most of the units is so small that situational combat factors are often larger than those differences especially if you accumulate several of them. This introduces the strategic concept of the accumulation of small advantages into the tactical gameplay. Attrition tactics and maneuver tactics are balanced so that they are both important. So, simply making the right matchups is not the whole story and neither is simply flanking.

And finally, the MTW/VI v2.01 battle engine has many features which have been dropped in the M2TW battle engine. So, even if the battle mechanics problems of the new engine are corrected, you still have a more simplistic engine and therefore cannot equal the depth of the tactical gameplay achievable with the older engine.

We see in Samurai Wars the gameplay all come together in very interesting and tactically complex battles vs other human players. There are no unit type restrictions other than the tax on more than 4 of a unit, which I wish we could eliminate because it's nothing more than a hedge against imbalance, and yet we see players taking diverse armies. These armies are suited to their playing style, and I haven't seen any of them conclusively refuted other than the more extreme types which are completely lacking one of the combined arms components. Even then extreme armies can work under the right conditions.

As concerns SP, the AI itself is playing an RPS game. It looks at the combat strength of its units and its opponent's units and makes the best matchups it can. In the older engine, the AI will try to flank if its unit is weaker than the target unit. If its unit is stronger, it makes a direct attack. The cavalry flanks in the MTW/VI engine whereas it didn't in the STW engine. The new engine has incorporated 3D into the combat, but it's not clear whether or not the AI can properly evaluate its winning chances in individual matchups anymore since the numeric combat factors alone no longer determine that.

[kallistus]

There's RPS system buried in real life warfare, and so there has to be one in TW. And there is, and it's a much more complex version than most RTS games have. TW RPS has many kinds of rocks, many different kinds of paper, and many different types of scissors.

That combined with the emphasis on morale and context means that the outcomes are complex, because it adds another branch to the RPS model.

The problem at present is one of those elements - cavalry - is overpowering the other two elements. The issues this has produced shows the problems with not having an effective RPS model underlying the combat system.

The cavalry superiority problem is partially this is due to the extremely powerful charge cavalry possess, but also because they have their historical strengths without their historical limitations. Cavalry are one of those weapon systems that, in the right time and place, are lethal; in the wrong time and place they can be useless.

That is, on a flat grassy plain, they should be deadly. In a wood, on a steep slope, they should be ineffectual. But in this game cavalry can still do a lot of damage when operating in rough terrain. You just have to micromanage them more.

That unbalancing of the general RPS component of the battles means battles become simplistic and much less tactically and intellectually challenging. Lets face it, any tard can win with a cavalry army, right? (Assuming you're not up against the Timurids...) That's sad for the future of the TW series, because it’s supposed to be one of the more historical and tactical games of the RTS genre.

If cavalry was the Medieval superweapon, why on Earth did knights dismount so often?

[dopp]

My definition of an RPS system would be a little different from yours, The_foz. I see RPS at work when units are given combat bonuses against others just on general principles (ie swords ALWAYS beat spears, so spears get attack penalties against swords, spears ALWAYS beat cavalry, so spears get massive bonuses against cavalry to ensure the result).

I would suggest that reliance on RPS is the realm of traditional RTS, where the large scale necessitates some abstractions in the way combat works. Total War has always been a much more sophisticated game. Flanking, terrain, real ranged combat, morale, fatigue; all these make strict RPS balancing unnecessary. A unit doesn't have to 'counter' any particular unit in Total War to have a defined role, it might just have a formation that makes it lose flanks and rear (that spear circle thing), making it a great anchor for your battle line. A cavalry unit that never gets tired becomes a great asset for repeated flank charges, even if its stats are merely average. A unit of fanatics with hopeless stats is worth its weight in gold if it never runs away. What I am suggesting is that each unit can have a role by virtue of its attributes, rather than because of some arbitrarily-assigned bonuses and penalties against particular classes of units.

History part: There was really no such thing as RPS. Units would be designed to fulfill battlefield roles, true, but only rarely would units be intended as specific counters to other units. Infantry carried pikes not only to fend off enemy cavalry, but also to beat other kinds of infantry. Halberdiers carried polearms that were effective against armor, but that didn't prevent them from chopping up less heavily-armored opponents.

More history: The 'historical' method of dealing with opposing troop types was not to create dedicated counters, but to meet them with your own. Cavalry fought other cavalry, not spearmen, who wouldn't have been fast enough to catch them anyway. Heavy infantry proved impervious to attack except against other heavy infantry. Archers shot with impunity unless met by opposing archers. Why then were there so many different units? Because each had a battlefield role that made them useful, and combining their strengths was a key ingredient of success. Cavalry was fast and could exploit openings, outflank and execute devastating charges. They could even threaten a charge and force the enemy to stand immobile while the archers rained arrows down on them. Archers could kill at a distance and force the enemy to break ranks to take cover, making them vulnerable to a cavalry or infantry attack. Infantry could take and hold ground resolutely, pinning the enemy in place for a cavalry charge or for the archers to do their work.

RPS is an artificial system designed to ensure that players make balanced armies. It doesn't always guarantee good gameplay. It can be taken too far, especially when the bonuses are very high, which is usually the case. MAD is usually what results. When your battleline mets the enemy's, each unit seeks out its intended target and annhilates it, while getting annihilated by its own counter on the enemy side. Everything dies, rebuild and repeat. That's RTS, not Total War. RPS often makes all other factors (range, hp armor, damage) irrelevant. I build a balanced army of 20 cuirassiers, fusiliers and musketeers. I run up against a 10 muskeeter army with 2 cuirassiers. His cuirassers charge in. My 20 fusiliers move up and, despite their armor and hp, all die in seconds to the musketeers. My 20 musketeers then all get pwned by the 2 cuirassiers. My cuirassiers kill the 10 enemy musketeers and the 2 heroic cuirassiers. Wow. He lost 12 units and I lost 40, even though I arguably had the balanced army. Fun? I think not. I could have won by simply sending in the cuirassiers only, but that's not a balanced army, is it? The balanced army loses in this case, precisely because of the RPS. RPS is not the magical cure-all for balance, nor is it even the ideal that game design should strive for.

A case could also be made for RPS robbing units of tactical utility rather than adding to their abilities. A unit that is only useful against X is worthless when your opponent doesn't build X at all.

I think CA is wise enough not to rely too heavily on RPS in general. Most units don't get bonuses to anything. Spears is the major one, followed by AP troops and camels. That's really about it.

Gameplay part: I agree with Musashi about spearmen as defensive troops and swords as more mobile attackers. Arbitrarily forcing one into an anti-cavalry robs it of both character and tactical utility.