Fixing the old Shogun: Total War morale system

[Tangent]

Perhaps the discussion of the morale problem in Shogun: the Warlord Edition/Mongol Invasion (WE/MI) should be approached from a different perspective.

Currently the debate is reduced to one between 1) players who either don't care as much about military history or who have not practised the old Shogun: Total War (STW) morale system and 2) those whose interest in military history and experience with the old STW system has led them to conclude that the WE/MI morale system is broken.

There is a risk that such a debate will encourage a compromise that will allow either a "dial-a-morale" solution or set up some "happy medium" between the old and the new systems. In my opinion such solutions will result in a further departure of the Total War system from being a wargame to becoming a conventional strategy game. Such solutions will instead end up dividing the community further. Experienced wargamers will drift away as they see more units fighting to the death and experienced strategy players will become bored more quickly with another attrition game. Furthermore, I believe that such solutions do not get to the heart of the problems that players experience.

Instead I propose that this debate be turned into a detailed constructive criticism of the Total War morale system. Since that system shows up most clearly in the original Shogun: Total War, I propose that the constructive criticism be directed on that game's morale system.

What is right with the morale system in Shogun: Total War?

What is wrong with the morale system in Shogun: Total War?

What are your proposed corrections?

[Tangent]

While I am low on time, I will take a stab at my own questions.

What is right about the old Shogun: Total War (STW) morale system?

As Napoleon is reputed to have said, "The moral is to the physical as three is to one." In the broadest sense, the STW system rewards attacking the morale of the enemy army more so than almost any computer wargame I can think of. This is good since the overwhelming historical evidence, even in Japan, is that armies and units generally did not fight to the death. There are of course exceptions to this generalism, which any good system needs to allow to happen.

Furthermore, the STW morale system does have morale modifiers for most effects that one would expect to have an impact on morale. These include being flanked from the side or rear, casualties on individual units, seeing friendly units fleeing nearby, seeing one's leader fleeing, having enemy units in the strategic rear of one's units, seeing one's leader killed, the unit having previously fled in the battle, and overall casualties on the army. These are morale effects that have been seen in other games by any experienced wargamer. The game also rightly gives a morale bonus to samurai versus ashigaru. Perhaps somone could correct me on this but I believe that STW also has morale effects for the rate of casualties inflicted in a given period of time on a unit; i.e. the more rapidly the same casualties are inflcted the more likely the unit will take a morale check (if it doesn't it should have such an effect, IMHO). Fatigue also plays a big role in STW morale, if I recall. Finally units can be rallied.

The fact that all of these effects are present is a good thing and it allows us to identify their interactions for improving the game.

What does not work in the old Shogun: Total War morale system? I will put suggested fixes in some of these comments.

In the broadest sense, while morale is of decisive importance at lower honor levels, the casualty rates to achieve those results seem to be actually too high. Many STW games end with the defeated and the victorious armies suffering well over 50% casualties, indeed often over 70% in my experience. This does not jibe with historical experience. While STW comes closer than most computer games, its casualty rates are still too high for most battle situations. This suggests to me that it should be easier to cause the flight of individual units at a given honor threshold. I know that many players will be surprised at this opinion.

Furthermore, most routs start in the rear, not at the front. I am not certain what do about this is in STW. It may be abstractly handled within an individual unit but it may require a separate treatment for an army as a whole.

In addition there are some specific morale interactions that I have some problems with.

The effect of having the enemy in the strategic rear of a unit (i.e. between the unit and the map edge of its original deployment side of the map) appears to me to be exaggerated. Please note that I am not talking about the effect of the enemy being in the tactical rear of a unit (i.e. engaging the friendly unit on a flank in some fashion); this is a separate effect. Too many times, I have deliberately sent units to get in the rear of the enemy only to have them flee at the slightest provocation (into the enemy no less, but the flight direction is a separate problem). In my opinion, having the enemy across the strategic rear of a unit should not prohibit that unit's mission to get in the rear of its enemies. I believe that the effect of having enemy units in the strategic rear of a unit should have its effect only after the unit is broken, making it more difficult to rally so long as the enemy is between it and the supposedly safe rear.

I have some concerns about the effect of seeing friendly troops fleeing. For one thing, the farther away friendly fleeing troops are, the less effect that should have.

While chain routs are a valid phenomonon, perhaps this effect should be more dependent upon whether or not the subsequent members of the chain are heavily engaged in melee as opposed to being sniped upon or not being engaged at all.

Units that are surrounded, particularly in a melee, with no realistic escape route are probably more inclined to fight to the death. The old Roman proverb of providing one's enemy a golden path to retreat probably applies.

Finally while I think units should break more frequently, I also think that they are probably somewhat easier to rally than in STW, until the army has broken that is.

-Tangent

[KumaRatta Yamamoto]

Tangent, with your permission (in fact without your permission ), i will use your opinion as a base for discussion (no sense of repeating the same thing over and over) SO:


What is right about the old Shogun: Total War (STW) morale system?

"...the STW morale system does have morale modifiers for most effects that one would expect to have an impact on morale. These include being flanked from the side or rear, casualties on individual units, seeing friendly units fleeing nearby, seeing one's leader fleeing, having enemy units in the strategic rear of one's units, seeing one's leader killed, the unit having previously fled in the battle, and overall casualties on the army. These are morale effects that have been seen in other games by any experienced wargamer. The game also rightly gives a morale bonus to samurai versus ashigaru. Perhaps somone could correct me on this but I believe that STW also has morale effects for the rate of casualties inflicted in a given period of time on a unit; i.e. the more rapidly the same casualties are inflcted the more likely the unit will take a morale check (if it doesn't it should have such an effect, IMHO). Fatigue also plays a big role in STW morale, if I recall. Finally units can be rallied..."

_____________________________________________


I FULLY AGREE WITH THIS STATEMENT.

_____________________________________________


What does not work in the old Shogun: Total War morale system? I will put suggested fixes in some of these comments.

"In the broadest sense, while morale is of decisive importance at lower honor levels, the casualty rates to achieve those results seem to be actually too high. Many STW games end with the defeated and the victorious armies suffering well over 50% casualties, indeed often over 70% in my experience. This does not jibe with historical experience. While STW comes closer than most computer games, its casualty rates are still too high for most battle situations. This suggests to me that it should be easier to cause the flight of individual units at a given honor threshold. I know that many players will be surprised at this opinion."

_____________________________________________

I PARTLY AGREE, YES THE PEASANTS (YARI ASHIS AND MORE IMPORTANTLY GUN TOTING PEASANTS) SHOULD FLEE EVEN FASTER/EASIER, EXCEPT IF THEY ARE OF HIGH HONOR (+4 and up).

NOW AS FOR THE PROFESSIONAL ARMY (SAMURAI) PLEASE NO. THIS IS A STRATEGY GAME NOT AN HISTORICAL SIMULATION. I WANT TO PLAY OUT SOME BATTLES AND HAVE SOME FUN HERE.
_____________________________________________

"Furthermore, most routs start in the rear, not at the front."
_____________________________________________

I AGREE.
_____________________________________________

"The effect of having the enemy in the strategic rear of a unit (i.e. between the unit and the map edge of its original deployment side of the map) appears to me to be exaggerated. Please note that I am not talking about the effect of the enemy being in the tactical rear of a unit (i.e. engaging the friendly unit on a flank in some fashion); this is a separate effect. Too many times, I have deliberately sent units to get in the rear of the enemy only to have them flee at the slightest provocation (into the enemy no less, but the flight direction is a separate problem). In my opinion, having the enemy across the strategic rear of a unit should not prohibit that unit's mission to get in the rear of its enemies. I believe that the effect of having enemy units in the strategic rear of a unit should have its effect only after the unit is broken, making it more difficult to rally so long as the enemy is between it and the supposedly safe rear. "
_____________________________________________

I AGREE
_____________________________________________

"I have some concerns about the effect of seeing friendly troops fleeing. For one thing, the farther away friendly fleeing troops are, the less effect that should have.

While chain routs are a valid phenomonon, perhaps this effect should be more dependent upon whether or not the subsequent members of the chain are heavily engaged in melee as opposed to being sniped upon or not being engaged at all."
_____________________________________________

YES THIS IS, WAS, THE MAIN CONCERN I HAD WITH THE MORALE SETTINGS IN STW THAT IN MY VIEW, INVALIDATED, ALL THE OPINIONS STATED IN THIS FORUM, PERTAINING TO HOW "REALIST" THE ORIGINAL MORALE SETTINGS WE'RE. THEY WE'RE NOT. THIS IS WE'RE MI IN MP DOES A BETTER JOB, BECAUSE THE UNITS HAVE TO ACTUALLY HAVE BEEN FIGHTHING AND TIRED BEFORE ROUTING. I NEVER UNDERSTOOD HOW A FULL ARMY CAN ROUT AND FLEE WITHOUT RALLYING BEING PURSUID BY ONLY ONE OR TWO UNITS OF THE OTHER ARMY LIKE PEASANTS ASHIS, THIS EVEN BEFORE THE UNITS HAVE ENGAGED IN ANY COMBAT. THE UNITS NOT BEING DIRECTLY RUNNED AFTER BY' SHOULD BE ABLE TO RALLY EXCEPT IF EXHAUSTED. THE REAR GUARD THAT IS FAR FROM THE ACTION SHOULD BE ABLE TO OFFER SOME RESISTANCE, AND CONFRONT THE PURSUERS, ESPECIALLY IF THEY HAVE HIGH HONOR AND ARE THE TAISHO'S UNIT. RIGHT NOW THEY ALWAYS FLEE, AUTOMATICLY. IT'S A GIVEN THAT WHEN ONLY ONE UNIT OF SAMURAIS STARTS FLEEING ALL THE REST OF THE ARMY WILL FOLD AND ROUT WITHOUT EXCEPTION. I DO NOT THINK THIS IS REALIST AT ALL.

_____________________________________________


"Units that are surrounded, particularly in a melee, with no realistic escape route are probably more inclined to fight to the death..."

_____________________________________________

I FULLY AGREE, THIS SHOULD BE A FACTOR IF POSSIBLE TO IMPLEMENT.
_____________________________________________

"while I think units should break more frequently, I also think that they are probably somewhat easier to rally than in STW..."

_____________________________________________

I AGREE. THIS IS WELL DONE IN MI MP BATTLES. INDIDUAL UNITS RALLY WITH MORE SUCCESS IF THE OTHER ARMY IS NOT IN HOT PURSUIT
_____________________________________________


Thank you Tangent for your well balanced post. I do Hope that a lot of patrons will reply the same way i did.



[This message has been edited by KumaRatta Yamamoto (edited 09-02-2001).]

[DragonCat]

Try having Cavalry charge at some guns from the flank and then stop and return after they have panicked the gunners into disarray. Like as not, the cavalry will rout, even without taking any casualties from the maneuver. This will also now cause the rest of your army to be less stable. This is a big problem. The morale system doesn't take into account feints, sacrifices, etc. It only knows, you showed your back so you have to run!

Granted I have stated this is the baldest most simplistic terms. This issue is that different units should react differently to situations. Cavalry should be able to perform these tasks without routing. I am sure there are other examples as well.

So if we are going to fix morale, we need to consider more than simple settings for the army at large.

I'm at work and am unable to go into a lot of detail on this, but heaven knows, I've posted about it enough times ;-)

Good thread, by the way. Glad to see people thinking about the issues.

------------------
DragonCat
"On the prowl . . . ."

[Kraellin]

there were 3 things that bugged me with the stw morale system:

1. the auto-rout bug. ok, that one's a given. we all hate that one.

2. the massive chain/cascade routs. true, there are examples of this in history. there are prolly even examples of an enemy saying 'boo' before a battle and having the opponent run off, and i'm even fairly sure there are historical examples of one army seeing another and going, 'no way. i'm going home' and never even fighting. but i think these are the exception rather than the rule, just like the fight-to-the-bitter-end battles were prolly the exception. and, regardless of whether any of those are exceptions or the rule, i too want a bit of a battle before my army turns tail and runs home for some fresh sake.

3. harrassment tactics, ambushes, feignts and other somewhat guerrilla type tactics simply cant be done very effectively because of the isolated unit morale thing. and again, while this may not reflect traditional historical line tactics, i tend to prefer the online what-if games more than historical hard liners.

and one other small note that's always somewhat baffled me in stw and we/mi...where did all these daimyos and taishos get this incredibly sophisticated ground radar that seems to be in use ;) ...can you say, 'unreal'? and we/mi is even worse, cause it highlights the generals. i'll bet half the folks playing dont even realize they can manuever their troops using this little high tech device and in some cases it's far better to do so.

other than those items, i think they did an incredible job of reflecting morale. there are some things still left out, like loyalty, fear of reprisal by your own army for being 'cowardly'. adrenaline pumping while in an actual melee. esprit de corps. comraderie...this would particularly apply to units in single player that travel around a lot together. i know a lot of that would be hard to represent in a piece of game code, so that's prolly the reason you dont see this sort of thing represented in games too much, but they were factors in battles and wars. confidence was another. a unit that won and then won again and again, tended to build up a confidence that portrayed itself both to the enemy before a battle and during. you currently see this in professional sports folk. they believe they can win and it carries over into the next game as another of those x qualities.

some folks may believe i'm all wet on this, but there are examples of the 1000 beating the 10,000, of the david beating the goliath, and it wasnt always through sheer skill or technology of arms and weapons. or superior tactics.

K.


------------------
I'm sorry, but i never apologize.

[Tangent]

"In the broadest sense, while morale is of decisive importance at lower honor levels, the casualty rates to achieve those results seem to be actually too high…."
_____________________________________________
I PARTLY AGREE, YES THE PEASANTS (YARI ASHIS AND MORE IMPORTANTLY GUN TOTING PEASANTS) SHOULD FLEE EVEN FASTER/EASIER, EXCEPT IF THEY ARE OF HIGH HONOR (+4 and up).
NOW AS FOR THE PROFESSIONAL ARMY (SAMURAI) PLEASE NO. THIS IS A STRATEGY GAME NOT AN HISTORICAL SIMULATION. I WANT TO PLAY OUT SOME BATTLES AND HAVE SOME FUN HERE.
____________________________________________

KumaRatta -

I am curious as to the details of your position on this issue. Is there anything that could be added to a morale system that would leave you feeling satisfied while moving the casualties closer to a hstorical level or does having casualties outweigh all other options?

[Tangent]

Regarding chain routs:

Perhaps someone who is better read than I am could help me on this but I think that they happened fairly frequently until the gunpowder age.

In most circumstances, I agree with KumaRatta that chain routs are dependent upon whether the mass of multiple units perceive themselves to be in the same fight, for the reasons that he has stated. Only at fairly low morales do I see units in the vicinity but not engaged routing at the same time.

It occurred to me that it might be possible to incorporate the rout from the rear effect into a modified chain rout system. For example, an unengaged unit unit that has been ordered to join a wavering mass in a melee might be the unit that breaks first. As to why it would break as opposed to jumping in to save their fellows strikes me as a very tricky question, however. Of course at the intra-unit level, the rout from the rear effect would have to remain abstracted.

Having said this however leads me to a question: What exactly are STW's criteria for turning a multiple unit rout into an army rout? It appears to be a combination of total casualties on the army and whether or not all of the individual units are routing but there is difference between routs that will end an SP scenario immediately and others where all of the losing army's units are fleeing but the game continues. What is the cause of this difference?

[KumaRatta Yamamoto]

Tangent: "I am curious as to the details of your position on this issue. Is there anything that could be added to a morale system that would leave you feeling satisfied while moving the casualties closer to a hstorical level or does having casualties outweigh all other options?"

_____________________________________________

What outweighs all other options for me is a factor called "fun". I do not have a problem with a strategy game that will tend towards historicall accuracy as to the "feel" of a real battlefield. In fact i don't know of any other game that even comes close to the STW series in that regard.

I am for suggestions that will help improve the game without going overboard in one direction or another. Let me try to be explain clearly what i mean.

In my view, a balanced approach between playability and accuracy is what i want of this type a strategy game. Because after all this is a game, thank god i am not in a real battefield.....

My fear about lowering the casualty rate all across the board, for every type of unit, is even faster routing of whole armies, thus less actual playing time, thus less fun.

The auto-rout bug, guns shooting through friendly units or hills, have in my view, a much worse effect on accuracy and playability. These should be top priorities in a future patch.




[This message has been edited by KumaRatta Yamamoto (edited 09-06-2001).]

[longjohn2]

I was responsible for most of the orignal Shogun morale system, although I left CA after the release of STW and didn't work on MI. Anyway having come back to the game after a year myself to play MI, I think it's interesting to discus how the morale system could be improved.

I think you raise some interesting points Tangent, although I don't agree with all of them. Taking them one by one.

I agree with you about casualty rates being too high from a strictly historical standpoint. However, from a game standpoint, I think some blood and guts are required. Maybe we should think of battle casualties as also representing all the casualties from disease and desertion that otherwise aren't modelled.

I agree that routs start from the rear within a unit, and this would be a nice effect to include. I don't think it works that way within an army though. It was pretty common to form up in multiple lines, and commanders didn't expect the rear lines to run away first.

You'll be pleased to know that there is no concept of a unit's strategic rear in the morale system. All determinations of flank and rear are based on the current unit facing. If you are seeing units sent deep into the enemy rear having morale problesms, it is probably due to being heavily outnumbered. They count all the friends and enemies within a certain radius, and the penalty for being outnumbered by 3 to 1 ( I think ) is pretty severe. If you approach the enemy line from the rear, there comes a point where all the enemies count, but your friends don't. Personally I don't really see this as a problem, as otherwise you'd be able to demoralise the whole enemy army by using a cheap YA or CA unit. At least as it stands you need to use a unit with decent morale.

Having surrounded units stand a fight is an interesting idea, although defining what is surrouned on an ever changing battlefield might be difficult. Troops that got surrounded usually got cut down anyway though.

I'll have to disagree with you about units needing to break and rally more readily. It seems to me, from my own reading of history that it was fairly rare for a unit to break, then rally and make a meaningful contribution to the battle.

I've left chain routs until last, as I think this is where some improvement is needed, and I have a couple of suggestions myself.
I think it's important to keep them as a feature of the game, both for historical accuracy, and because they are the mechanism that enables you to win those famous victories where an inferior force defeats a superior one. They also bring battles to a conclusion.
There is no special chain rout mechanism in the game, it's just a natural outcome of the moral system. Each unit is continually assesing morale, and if it sees a friendly unit within a certain radius routing, it gets a morale penalty. If it can see two routing it gets a big morale penalty ( there is no extra effect beyond two ). If the unit has other significant morale penalties, such as being outflanked, losing in melee, or having taken lots of casualties, it will most likely rout itself.

To my mind, the main problem is that it all happens too fast. The chain of routs propegates too fast for players to see, so they don't see where it started, or how it spread, and it all feels very arbitary. I have two suggestions to help this.

i) Reduce the radius a routing unit needs to be within, to cause a morale penalty. This would help troops in long lines survive, but still allow armies in deep columns to be swept away ( as ofter happened in history ).

ii) Introduce a new proviso that a unit must have been at wavering morale for a certain amount of time ( maybe 5 seconds say ) before it can break. This would slow the whole chain down, and allow players to see it develop, and possibly even do something to help. There would of course need to be certain exceptions to allow for sudden panic, such as being charged, or taking a sudden flurry of cassualties.

I think it would also be useful, if players could press a key a see the morale factors affecting a unit. While in some ways I dislike exposing the inner workings of the game too much in play, in this case I think it would be worthwhile, as morale is very hard to understand for people who don't have a wargaming background, or much knowledge of ancient military history.

[clink]

Tangent
I think you've pressed your finger on a button thats annoying a lot of veteran STW gamers out there.

'THIS IS A STATEGEY GAME NOT AN HISTORICAL SIMULATION.I WANT TO PLAY OUT SOME BATTLES AND HAVE SOME FUN HERE'.

This as close as it gets to historical for a very well done strategy war game,the best yet..refering to original STW.
I would rather test myself with tactics and strategey in a war game,than just duke it out with an arcade slug fest.
Some of histories greatest generals won battles not because of the size of thier armies,but because of thier brilliance on the battle field.
I'm hoping MI is not becoming the status quo.

There are plenty of garbage shootem up arcade fantasy games out there to have..'fun'.. with.
Shogun is brilliant fun.

[solypsist]

current morale factors, whilst (whilst?? jeeze..) operate in this regard as far as the mass/cascade rout.

certain "tiers" of troop types are more prone to panic by fleeing fellows than others. A unit of warrior monks almost always never wavers when the rest of the army, composed of (for example) ashigaru starts to flee. But a unit of ashi will almost always turn and run when a unit of anything else routs nearby. You'll notice most armies that mass-rout are made up of the more common types, from ashi/musket/crossbow to archer to yari and so on.

[KukriKhan]

Lo-o-ong time, no see longJohn...Are they bringing you in for Monday's patch meet, as a morale guy? One hopes...and thinks they could use your clear-headed view. Thanks for your clarifications. We (OK, I, and some others) seem to've gotten stumped here in our exercise of moral factors in MI/WE (BTW Tangent, et al, freekin BRILLIANT approach to this, coming at it from the STW angle - sadly,I have to think more to contribute anything valuable).

[TakeshidaSo]

I knew the person responsible for the STW morale system had not worked on MI. STW had H0 units, that would barely stand their ground, to H9 units, who would almost fight to the last man. MI does not.

The problem is MI raised the morale table, that ranged from -4 to 8, with a median unit morale of 2, by TWELVE points. Lowering morale used to be an option, and most games included some H0 troops. Increasing a units morale is just one way to counter natural negative modifiers.

1). CASCADE ROUT>
Units only react to TWO routing friends MAX. So the cascade occurs naturally, not because so many friends rout, but because each one see's two nearby friends routing. If a unit is close enough to get a boost from a friends presence, then it's close enough to see it rout too.

Solution:
Increase the radius of positive morale factors; like the CNC's presence, or friends on flanks and rear, and seeing enemy rout.
Alternate:
Decrease the radius of negative modifiers.

2). F1 BUTTON>
Of course, the F1 Button needs to show attack, and morale modifiers. The numbers arent where the real art is, but science and art merge in war. Player's should have the chance, to apply the science of the game.

Solution:
Fix the F1 screen features.
Alternate:
Remove the F1 screen option entirely.

[TakeshidaSo]

REINFORCEMENTS
A lot of people are annoyed by the new reinforcements approach. You cant detach units, to act as reinforcements, and reinforcements wont arrive where you want them to anyway.

It's a SP issue, related to the strategy of attrition, and a MP one, as far as KOTH games. It would be good for both SP and MP, if you could tell reinforcements where to arrive. It would help both, SP and MP games, if you could detach forces. It would help MP games, if you could buy more than 16 units, and detach forces.

ROUTERS
If routing units moved toward the nearest mapedge, they would still move thru enemy units, if the enemy were between them and the edge. If they behaved like a skirmisher, while they remain routed, then they could rout more safely. They obviously dont rout logically now, and they do have to rout in some direction.

The question you must answer first, is how should routers behave, while they are routing. If your satisfied that they are acting like they want to survive, then I will disagree. If you think that CA\DT should not be asked to allow routers to behave like skirmishers already do, then I would also strongly disagree.

[This message has been edited by TakeshidaSo (edited 09-09-2001).]

[Tangent]

longjohn2: It is indeed a pleasure to hear from the primary designer of the morale system in Shogun: Total War. I have appreciated the efforts that you put into the STW morale system. More than any computer game that I have experienced, STW gives me a real opportunity to "read the battlefield" and to defeat rather than be forced to annihilate an opposing army. STW is a landmark game in no small part because of your efforts.

There are several issues that people have mentioned recently in this thread:

1) overall casualty levels versus historical results in the game;
2) morale on independent missions;
3) the importance of being able to rally troops;
4) chain or cascade routs;
5) presenting information on morale states using the F1 screen or elsewhere;
6) additional morale factors not included in the game.

1) Overall casualty levels in STW versus historical results.

There does not seem to be any disagreement that the overall casualty percentages generated in the majority (not all) of STW battles are too high from an historical perspective. However, there are strong fears that reducing them may remove too much fun from the game in the form of blood and guts results or in cutting battles too short. I think that it is fair to say that most players who are proponents of morale share this view (as, of course, do those players who don't like using a morale system).

If push comes to shove, I would probably agree with that majority. I just wish that someone had a better idea to improve historicity on this matter while maintaining, even improving, the fun level. Does anyone have a presentable idea on this issue?

2) Morale on independent missions:

Several players, including myself, have commented that units on independent missions seem to be too vulnerable to routing when executing those missions. DragonCat and Kraellin responded by suggesting that the morale system should respond to mission specific circumstances. In responding to these comments, and defending the status quo in STW, longjohn2 stated:

"You'll be pleased to know that there is no concept of a unit's strategic rear in the morale system. All determinations of flank and rear are based on the current unit facing. If you are seeing units sent deep into the enemy rear having morale problems, it is probably due to being heavily outnumbered. They count all the friends and enemies within a certain radius, and the penalty for being outnumbered by 3 to 1 ( I think ) is pretty severe. If you approach the enemy line from the rear, there comes a point where all the enemies count, but your friends don't. Personally I don't really see this as a problem, as otherwise you'd be able to demoralise the whole enemy army by using a cheap YA or CA unit. At least as it stands you need to use a unit with decent morale."

I will refer to this existing modifier as the "intimidation effect." I also infer from longjohn2's comments that, in addition to using higher morale units for independent missions, another tactical solution to deal with the intimidation effect would be to employ a larger mass of troops on the independent mission, thereby improving the local odds ratio. Am I correct in inferring this? When approaching an enemy line from the rear as described above, does more friendly mass coming from the rear still count to reduce the intimidation effect or are friendly units negated regardless of which side of the enemy line they come from?

By the way, while I wonder about its impact on occasions, it is realistic to have an intimidation effect. In retrospect, I have used it on rare occasions to "boo" a demoralized force. So long as such "booing" remains well outside the norm, I like having it as a rare effect. Longjohn2's comments reinforce my impression that the Shogun: Total War morale system is sophisticated. Also note that the implied detail of the morale calculations here could not be easily implemented in a miniatures or board game. The morale system in this game is a unique beast worth preserving.

DragonCat's and Kraellin's comments also intrigued me: they argued that morale effects should reflect the mission that the troops are assigned to. As an inferred example, a unit on a feint would know that its mission is a feint and accordingly be less likely to be routed when a superior force challenges (but not necessarily engages) it. I think that there is some truth to this. Expectations do matter. However, I don't think that an expectations approach is feasible in the short term. One problem that I see is how to define such expectations into the simplified order-giving procedure in Total War. I suggest that this is an issue for consideration in Crusader: Total War.

I am wondering if there might be a simpler short-term fix to this issue. The units most frequently used on independent missions are cavalry, which were in history frequently trained for such missions. The morale of heavy cavalry is already fairly high but that of YC and AC is lower (but YC is not especially low). I don’t know the morale of the NC or Mongol units in WE/MI. Should the base morale of certain cavalry units be increased above their current levels to improve their performance on independent missions, or are the available tactical solutions to the intimidation effect adequate?

3) The importance of being able to rally troops.

Longjohn2 disagrees with my argument that units should be easier to rally. Since I have no doubt that he has researched the matter more than my casual readings, I defer to his judgement.

4) Chain or cascade routs.

I think longjohn2 has hit the nail on the head that most players who have difficulty accepting chain routs do so because the process happens so fast. While I have accepted chain routs as a normal part of play it sometimes happens so fast to me that I can't determine how it happened. While this makes me more determined to figure out a tactical solution, I can see why others would complain of this to the point of hating the entire morale system. A change here may be necessary to enhance the continued use and future development of the morale system in Total War.

For historical and playability reasons, I believe that chain routs should remain a central part of the system. I also agree with KumaRatta that it is currently too easy for non-involved units to rout at the lower honor levels. However, note that I want to make it more difficult for the non-involved units to chain rout, not impossible. Under certain conditions non-involved units will chain rout and this effect deserves to be preserved. This is where I still have some questions: in STW what is the overall morale condition of an army (either the overall casualties to the army or some combination of morale states of its units) that causes an army as a whole to rout? Is this simply a naturally occurring chain rout or is there another process that comes into play? Sometimes an army will be permanently routed leading to the end of the scenario: what are the conditions that trigger that?

While I would like to get some answers to my questions above, without these answers I currently lean toward both decreasing the radii of negative modifiers and slowing down the process of wavering units, along the lines suggested by longjohn2 in his message above.

5) Presenting information on morale states using the F1 screen or elsewhere.

In suggesting a solution to problems involving the chain rout phenomenon, longjohn2 suggested that it would be worthwhile to provide the player with more information on the morale status of the units. Takeshida suggested that the F1 screen would be the logical place for this.

I agree with these proposals to "show more of the science" behind the game. In addition, I propose that the morale system be made more easily understandable to new players in two other areas. The game documentation about morale can be improved both in the standard manual and probably in the official game guides. More important, I believe that it would be a good idea to have an interactive tutorial about using morale where a chain rout is an expected result. This would help to maximize the number of new players who will start thinking of using the morale system as a tool of their play.

6) Additional morale effects.

Kraellin mentioned a number of morale effects that are not currently represented in the game. While I don't entirely agree with his list of missing items, his comments took some intriguing directions. I like the idea of more strategic effects having an impact upon battles. Currently the only strategic morale effect that I know of on battles is when a general is assassinated on the same turn as his army fights. For example, Kraellin mentioned that units in an army that have fought together on campaign would probably have a morale bonus. I like this kind of approach. Perhaps strategically induced fatigue would be a good idea.

Tactically, it would be interesting to see champions challenge each other before a battle as a part of a bid to gain a morale advantage.

I was surprised to learn that having an enemy unit across the strategic rear was not modeled in STW. So much for my relying upon third party game guides! I actually think that such an effect should be modeled, assuming proper testing for play balance. My readings suggest that a threat to the rear, particularly to the camp of the threatened army, does produce a negative morale effect.

-Tangent

[longjohn2]

Another interesting and thoughtful post Tangent. You'll be pleased to know that I'll be returning to CA to work on Crusaders. I'm looking forward to being able to refine the system some more, in both the fun and realism dimensions. Hopefully I'll be able to slip in one or two of the ideas from this thread.

To answer some of the points you raised.

Having more units in your outflanking forces may well reduce the intimidation effect, although as it stands there are set ratios that trigger set penalties, rather than being a continuasly sliding scale. They can also support each other's flank, so removing another penalty ( for each threatened flank ) that a lone unit is likely to incur.

As you noted HC and YC have decent morale and can be used for outflanking, wheras the morale of CA is a little shaky. People expecting to use it for Mongul style tactics may find this a little strange, but should remember that the game attempts to recreate warfare during a period of Japanese history when cavalry archers had ceased to be used. I was aware of their limitations during development, but making them more powerful, or just braver, would have encouraged the use of tactics that are incorrect for this period.

I'm not convinced about having enemy in the strategic rear should be a penalty. The morale system is really concerned with what the soldiers in a unit know at a given time. While they'd certainly be concerned about their camp being looted, they probably wouldn't know about it until after the battle, and even if they did know, I'm not sure it would influence their behaviour on the battlefield. If you know of any examples where units pulled out of a battle of their own accord to defend their camp, please post them ( after battle mutinies don't count ).

There is no special army rout mechanism. If an entire army routs, it is simply because every unit has a lot of morale penalties. The system decides that you've lost the battle when all your units are routing, and it thinks you have no chance of rallying any of them.

Feints are a slightly difficult issue. I don't really agree with Dragoncat's basic assertion that just turning tail will cause an undamaged unit to rout. However, the ability of guns to stop cavalry charges in the open is a little unrealistic. It was done this way, as historically the gunners would have hidden behind obstacles that were difficult to cross for cavalry, but we didn't have time to include these in the game. However allowing cavalry to ride down gunners too easily would have made the gun units too weak.

I'm glad that you broadly agree with my ideas on how to improve chain routs. Hopefully I'll be allowed some time to try them out.

Having a morale tutorial is a good idea too, but making tutorials takes a lot of effort.

I too liked Kraellin's idea that units that had fought together for a long time should get a morale bonus. Having a sense of an army being a permanent thing, not just a temporary collection of units, would definately add something to the campaign game.

[DragonCat]

Tangent and Longjohn,

Thank you both for your excellent posts. I am very happy you are looking at and considering all the issues. This gives the lie to all those that claim that the developers don't care, don't continue to be interested, etc. etc. etc.


As to cavalry routing from merely turning tail, I have had units rout just from using them for "demonstrations" in front of the enemy. I have had battles where there was losts of maneuvering and many feints. If you do one too many feints, or come too close to a flank where YOU become outflanked because of no nearby friendlys, then you WILL see your cavalry rout- this also makes a lot of sense now that you mention that the Cav Archers are more likely to do so. I frequently use them for this task as they can harrass a flank from distance and also make charges on a melee units flank to make them reform. After they start moving I send in my attack units while bringing my cav archers back out (they are in hold formation so they don't get engaged). Many is the time I have seem them continue right on off the board instead of back to my lines and reforming. A few times I have been able to rally them, but frequently they just head out over the hill. ;-)

I look forward to Crusaders being a truly amazing game.

It will be also interesting to see what comes in the patch for Mongols.



------------------
DragonCat
"On the prowl . . . ."

[KumaRatta Yamamoto]

This we're great posts thanks guys. I am please to learn that you will be back to work on the crusaders game Longjohn2.

With your permission, in fact without , i will post this news in the crusaders forum we're i mod (this forum is quite dead and in desperate need of some info).

[Kraellin]

a most interesting thread. and longjohn, i'm very glad to 'meet' you. always nice to have some source folks around.

i very much like the idea of adding in morale factors on a visible display of some sort. however, i would only do this for single player, and perhaps only for custom games, and make THAT the tutorial all by itself. i would be very reluctant to agree with this sort of thing in multiplayer, although i would certainly use it if it were there. from just a pure mechanics reason this could add to lag in online games, which at times are already seriously lagged due to the various relative latentcies of the internet.

as for slowing down the chain routing effect, for both multi and single, i think this is mostly a good idea, but would again recommend caution. it would be nice to be able to see and brace up weakened points, but if done too much, would destroy a feature that needs to remain in place.

one thing i have to say here, is that, at best, i'm an average player in a 1v1 game. and in all fairness to fairness itself, if someone is chain routing me, why isnt it that i chain routed them? well, they just know tactics or something about the game that i dont, or managed to implement it better than i did, or quicker than i did. so any arguement that wants one aspect removed or lessened and another enhanced or added can often just be a cry of 'i dont know how to win'. bearing that in mind, and trying to be as objective as possiible, i do believe that a slowing of a complete collapse might be better. others may argue, 'well, just play higher koku games', but that's not quite the same thing. a higher koku game just delays the when the chain rout will occur, not the actual mechanic of how it works.

one other thing that has always bugged me about stw is the campaign game and the instant knowledge one has of the overall strategic situation. the view taken in the campaign game is basically not of a general on the field or a daimyo sitting in a palace listening to reports, but is a god-like, all knowing sort of view. communications were slow. even a fast rider relaying info took a while to arrive and then return to his unit. the counter-arguement would have to be that a 'turn' in stw takes a whole season, thus allowing for plenty of time in relaying important information. but somehow, this still doesnt sit right with me. i guess it's because there is no mystery of how a given battle turned out, or what the current location of the enemy is and other pertinent information. generals had to predict where an enemy would be at a certain time and could they get their own army there in time to do something about it. i think you get my drift here. it's not that stw is wrong in what it does; it's more in what it doesnt do.

i believe one possible solution for this would be to stretch time out again and thus add a little more mystery to the strategic game. perhaps making a move 1 month, or even 1 week would add some of this, but it would also change the nature of other factors. you wouldnt then be able to take charge of every battle. for if you're taking the view of being the leader of a given clan, then you'd only be able to take charge of those battles where you actually were. you would have to send out your various generals to locations and let them fight those battles themselves, with a delay of information arriving back to yourself and a delay of new orders being sent to those generals as to what to do. this would also allow for the whole port to port issue to be redone with actual sea zones and sea battles and loss of troop transports at sea.

ah well, maybe for crusaders, ay?

when you guys are talking about 'the strategic rear', are you talking about the campaign map or battle mode? a strategic rear effect in battle mode would be, or at least could be, constantly changing during a battle. many are the times that an attacker ends up on the defender's side and the defender on the attacker's, so you certainly couldnt set it per the map you were on, you would have to constantly recalculate this and, in we/mi, with the new skirmishing units, both cav and inf, units can often times get so spread out that there isnt really any front or rear.

if you're talking campaign strat map, then units in a battle could be aware already that they exist in an isolated province and could well suffer a morale penalty, but not necessarily would they be aware of this. timing on this would be important. units in a castle under siege could also be in this sort of situation and may or may not be aware of such.

i tend to agree that there is no way to program 'expectations' of a unit, such as during a feint, without this being highly abused. i suppose the best way to handle this is to designate, while yer buying yer army in multi, a unit that will be used for such and simply 'buy more honor' for that unit so that it can carry out such tasks more easily.

and since we have you here, longjohn, can you give us any idea about the rout bug and what causes it and how to fix it? we know through tests how to re-create at least one instance of it, but i dont think we've found all instances of this. any insights on this?

as for units routing from the rear first, i dont know the historic accuracy of this. i can offer this, however. a unit in the front lines is fighting for his life. he's intent on avoiding being killed while at the same time killing others. he's not watching the flanks or action in the rear or very much left or right and so on. his area of attention is primarily on the enemy immediately in front of him. his life depends on this. he TRUSTS that others to his sides are holding up and that the general is still there and that the guns are still firing for support and so on, but he has very little time to be looking around very much or very far. those in the rear tend to be able to look around more. they can see if there are 1000 cav units doing an end-around on the main lines and that things are going well or not so well. thus, they have a broader picture of the battle as a whole and are, at times, more able to predict a final outcome. if that prediction says, 'hey, we're going to get whumped here', then those units might well be the first to rout, seeing an inevitable defeat coming. units in the midst of swinging swords have their adrenalin pumping, are knee deep in blood and guts and are often just too concerned with immediate life and death to be seeing all this other stuff going on. there are instances of units being so caught up in a battle that they never even see that their comrades have deserted them...until it's too late. thus, a unit actually battling will often have a higher morale than a unit in the rear guard that isnt fighting. if you dont believe me, talk to some vets of real wars. the scariest part is the anticipation of battle, not the battle itself.

by the same token, if you are IN the fighting itself and suddenly you do become aware of friendlies running by you in the 'wrong' direction, it is unnerving and demoralizing. it's like a betrayal of that trust i mentioned earlier and few things are more de-moralzing than a betrayal. even seeing comrades falling beside you can act as an impetus to fight harder, but a betrayal makes the heart sink like almost nothing else.

i've been in combat and i've had my share of fights, so this isnt just speculation or 'book learnin'. it's empirical, first hand accounts. ask around. there are other vets on here as well.

stw is to be applauded for adding this most crucial of combat factors, morale. it aint perfect, but it beats anything else out there for this time period. but, it doesnt mean it cant be improved either. and yes, there is the 'play/fun' factor also. wouldnt be much fun if you walk onto the stw battlefield with 8 units and see 16 heavy units on the enemy's side and all your units just said, 'to hell with this', and ran off. might happen in reality, but not much fun in a game. :)

K.


------------------
I'm sorry, but i never apologize.

[Fuh Teng]

Can I get a shorthand view of the differences?

------------------
The Fuh (:

[MagyarKhans Cham]

Chain routs are common, I remember the mongols fighting at Muhi in Hungary against a formidable hungarian army, The best european cav army at that time and after some skirmishing battles a smartly executed flanking attack caused 50000 hungarians to flee.

If they had fought to the death the hungarians could have won but panic caused the defeat.

[Erado San]

Hi LongJohn,

Glad to hear you'll return to CA. You have given some interesting details on Morale that help us understand the mechanism better.

I don't mind the chain routs, although perhaps they do happen too fast and too irreversible. What I am annoyed with is the Auto Rout Bug. Here I will paste the inforation on it that I sent to CA some time ago. I can see where solving this issue could well mean an extensive remake of the AI on this point, but I would appreciate it if you could comment on it.

Quote Symptoms

During battle, especially at the height of hand-to-hand conflict, one or more units may start to rout without apparent reason, while they suffered little or no casualties, aren't tired and are not far away from their friendly units and/or General.This may result in the entire army routing while you appear to be winning, sometimes even when you are at a 75% green status bar.

Cause

The units start routing under three conditions:
The unit moves away from the enenmy units using stacked commands or waypoints.
The unit is ordered to move away from the enemy units, and before it has finished it's current command you issue new commands to it.
The opposing units must be within a certain range from eachother. It appears they must be within a range of about one line of 140 men lined up next to eachother.
How to reproduce

Line up one unit in the middle of the battlefield. The other player lines up one or more units within the 140 men row range, opposed to the enemy unit. Then select one or more units and make them move away from the enemy unit. Use either stacked commands or make sure you issue new commands to the units moving away before they finish their current command. Also make sure that all units cannot engage at will or fire at will, to make sure that no units suffer any casualties. You will find that the units moving away will rout within seconds. This way you can have the entire army rout away from one single enemy unit.

The underlying cause: speculations

We speculated a bit about what in-game mechanics could be causing this, and we agreed that it seemed like the units that are moving away from the enemy are suffering from cumulative honor- and/or morale penalties. As if they get a penalty for their initial move away from the enemy and a penalty added to this with every new command to move away from the enemy. If you let the units move away from the enemy, let them finish their command, then it seems the penalty for the initial move away is cancelled, and when you then issue a new command, the unit gets another penalty, but then the initial penalty is cancelled, so their is no cumulative negative effect.

[/QUOTE]

[This message has been edited by Erado San (edited 09-17-2001).]

[MagyarKhans Cham]

I know my Khan is stunned how many times the morale mechanics and routing works when projected onto is knowing (and feeling) of ancient warfare.
Sadly somethimes the same system is stupid/unexplainable. Perhaps in some cases caused by a autorout by one of the units nearby the battle moving away like erado explained.

the old system is superb but should be refined on some aspects.

many "unexplained" routing happens due to the excessive morale bonusses by the muskets.

i would suggest that people play more 8000-10000 koku games with just 1 muskets max.

[tootee]

Quote Originally posted by TakeshidaSo:

The problem is MI raised the morale table, that ranged from -4 to 8, with a median unit morale of 2, by TWELVE points. Lowering morale used to be an option, and most games included some H0 troops. Increasing a units morale is just one way to counter natural negative modifiers.
[/QUOTE]

Although I have been playing STW for a long time, and just bought MI yesterday, I have long wondered where from can I know more about the scoring/morale mechanism during battle.. you guys mention about morale table of -4 to 8 etc.. is there a guide book that describes all these? Thanks lots!

BTW my email is tootee@iname.com so that replies won't mess-up this discussion.


[This message has been edited by tootee (edited 09-18-2001).]