Selective Historical accuracy

[Louis de la Ferte Ste Colombe]

I have been reading this forum for some times now, but I have not posted much.

The main reason being; I have a very hard time relating to the concern of many people here. To the point I wonder if we play the same game with MTW, or will play the same game in RTW.

Most of the discussion seems to focus on unit description in quite some detail, is the helmet correct or not... or the shield...
Most of the time, when in battle, I try to stay as high as possible to get the best view possible; I seldom have the chance to be close enough to the guys to see if they really look good or not. On top of that, the game plays so fast that I really have no or little chance to pay attention to those details.

But, I can understand that some people are interested in those topics and would like to know how accurate units will be.

Personally, I am also interested in historical accuracy, although I am not an expert in history, but from a rather different standpoint.
Given the little experience I have in MTW and its historical accuracy, my concern are focused on

1/ on tactical battle;
- behaviour of different kinds of units is right; no more cavalry not able to just run through missile like in MTW(there is hope for that... give us a RPS that works for MP too).

2/ on campaign
- how RTW will handle tech and tech evolution; I don't like much MTW system with building to get access to new units... How would army evolve from a phalanx model to a legion model?
- how army get raised, and their composition; what's the composition of a legion is going to be? Is there going to be such a thing as a legion? Or armies are just going to be the sum of everything and anything built by players? How long does it take to raise a complete legion in Rome? How does that compare to reality? Are we still in a 1 unit / turn kind of system?
- is there any chance that provinces and armies composition will be balanced so as to create a model of the military / economic strenght of each faction? Or is it going to be like MTW; very fast you can't make a difference between a French and an HRE army...
- strategic agent? any news? What is possible in Diplomacy? Will it be possible to have allies paying tribute and sending auxilaries troops? Will there be any colonizing option?

I found MTW lacking in the medieval spirit and theme on most of those points. It's great and fun battle simulation, but there were no medievial feeling to the campaign, because most of the feature of the campaign were 'unmedieval'.

So sure, I'd like all units to get the right helmet, and the right shield. But the overall feel of historical accuracy may not depend on that. It's also a matter of getting campaign features right. It's my understanding STW did a good job at that, but not MTW. What do we know in that regard for RTW?

Louis,

[ziphnor]

I must say i agree with you. As long as the units arent completely silly and the combat engine is superb( im pretty sure it will be), i would rather see a lot of improvements in the strategy game, for example, like you say, a better tech system, and a more realistic approach to raising armies.

[Barkhorn1x]

Quote[/b] (Louis de la Ferte Ste Colombe @ May 20 2004,16:40)]What do we know in that regard for RTW?
Well, that's just it - isn't it?

We don't know very much concrete information about many of the aspects of the game at this point. But we get a new unit shown to us each week. Of course the comments will be focused on these units and their fidelty to the historical record (or even just plausibility - oink, oink).

I for one am VERY concerned about the tactical applications of the pachyderm on the field of battle.

As far as your concerns go - from what we DO know now - methinks that you will be disappointed in some areas.
- tech evolution is basically the same model as M:TW - as far as we know.
- the concept of a legion being composed of maniples based on the Triplex Axces (sp??) or 10 cohorts is not modeled. Sure, you can raise the appropriate troop types but I havn't heard of (and I doubt there will be) any mechanism for raising complete Camillan or Marian legions.
- There will be more civilzations than in M:TW - and the divisions are appropriate, IMO.
- We know the diplomatic model will be much deeper.

But, hey guess what?? Tomorrow brings another new unit and another 3 page thread that comments on headgear and footware.

http://www.totalwar.org/forum/non-cgi/emoticons/wink.gif

Barkhorn.

[Kraxis]

I too felt that MTW lacked in the department of 'feeling'. But I also think that a part of my feeling in the game will be derived from the units. The 'biker'-barbarians didn't add any great feeling to them. And discriptive names don't either, the units will just becomer 'another' one, instead of this specific unit it really should be.

When it comes to nitpicking about the helmet for instance, then I'm more leenient, the shields are perhaps more important. I wouldn't like to see the Gauls and Celts without their unavoidable oval or hexagonal shields. Oter shields are ok, but in lesser number than the oval or hexagonal ones.

[The Wizard]

I, for one, am much more concerned about correct gameplay which is fun and at the same time realistic. That will be far more important than checking if a Briton warrior has the right amount of facial hair, especially in the long run.

Stragetical movements; diplomacy; tactical combat; city sieging; faction attributes; etc.

This game is a model of the time itself, as Cpt. Fishpants stated. Therefore, be happy we have this peek in the ways of war of the peoples of the timespan of RTW. The nitpicking on Celtic moustaches, Germanic beards, and Roman sandals is for later.



~Wiz

[andrewt]

Same here. I only want the units to be somewhat realistic not totally realistic. Facial hair, what kind of sandal the other guy is wearing, exact shape of shield and helmet are really stuff I don't care about.

The only major concern I have right now is the Egyptian faction. They make me think that I might be able to hire a "Moses" mercenary to unleash 10 plagues on them.

[Leet Eriksson]

Right now i care more for balance than how units look,or what they are called or if they are fictional.it will be fun discovering tactics and strategies for the units.

[econ21]

Good post, Louis - where MTW and STW scored as a historical wargame was in creating fun and atmospheric tactical battles that "felt" rather authentic. Other more hardcore computer wargames may simulate battlefield outcomes better (although there are no better ones for the period), but don't make you feel like you are there as much and frankly are not so much fun. Daft units - Jomsviking types - do threaten to undermine that "feels right" aspect of the battle, so maybe that is why people focus on them. However, I agree there is scope for improving the accuracy of TW, especially in terms of some of the factors you mention - primarily relating to the strategy map.

For example, I suspect a major strength of the Roman army was that it was possible for it to move legions across the known world with formiddable speed - thanks to communications, roads, transport, supply chains etc. Thus a mobile smaller Roman army could defeat in detail slower and more dispersed forces of other factions. But I doubt we will see that kind of factor being modelled.

[Mr Durian]

This game isnt going to perfect. CA has to little time to make it 80% historically accurate.

[Nowake]

Louis de la Ferte


It seems indeed that most threads are devoted to unit descriptions etc. But you have to bear in mind that those are the single things of which we are well informed. If they gave us the description for the thracian, it means that that's how is going to look, so I can discuss the minuses and pluses. What broad info do we have about the campaign map? People speculated for almost an year, with not concrete result. They just got tired, esspecially after the news about cm MP was proved to be a hoax. So they learned their lesson and disscus more on what is officially pinpointed.

[SwordsMaster]

Guys, dont worry. Modding is your answer.All the unit inaccuracy will be solved in less than weeks time by new mods. In the other hand, the hardcoded diplomacy/technology/ingame mechanics wont be moddable with such ease, so there is where the accuracy and effort should go.

[Aymar de Bois Mauri]

I agree that hardcoded issues are more important than modable ones. But do we have any idea of the factors that will be modable? We don't. Accordingly, everyone feels they have something to say about what needs to be changed first. For people to exchange info and make serious recomendations, we all must know what are the hardcoded factors.

My question is:

What factors, in the game, will be hardcoded?

Only after knowing it, can priorities be made to allow for meaningfull discussions and recomendations.

[shingenmitch2]

HEY, Smurfy lives http://www.totalwar.org/forum/non-cgi/emoticons/wave.gif

Well, u know me Louis, I'm am very concerned about the multi-play of the game and how the campaign is modeled.

I guess I've expressed most of my concerns on those all a long time ago so I haven't felt the need to rehash.

I fixate on the units at this point because:
1. we're getting to see them, and
2. It seems like the easiest thing in the world to get correct and/or fix.

hope to see you on the field

[Louis de la Ferte Ste Colombe]

I thinks Swordmaster got my concern right. If in game hardcoded mechanics are as they are for MTW, I don't quite think we'll have a 'Roman' feeling.

Even if we assume CA is going to get all the units right (quite a strong assumption...), the campaign in game mechanics, such as building based tech trees, and units being built 1/ turn / prov will probably prevent players and AI to build historic armies.
OK, maybe players will be able to do it, probably strongly restricting themselves in the process, but AI is likely to get into trouble getting good and historic armies.

With the current system, honestly, I don't expect AI roman to come up with legion-like-armies.
I find it a little sad, and the first time my parthes will face a rag tag roman army of ballista, cav auxilaries and hastatii, but not a real roman legion, it'll probably break the 'authentic' feeling. Who cares if principes are right, if they never get built in the right army and the right context?

Also, it would be interesting to see if the campaign game is tweaked a bit to reflect the social and political life of Rome (and other civilisation). MTW was very disappointing for that, hardly taking into account the feudal nature of most of the European kingdoms.
We already know that some effort are made in that direction, but I'd appreciate more information on that than yet another unit description.

Louis,

[econ21]

Quote[/b] (Louis de la Ferte Ste Colombe @ May 26 2004,08:44)]With the current system, honestly, I don't expect AI roman to come up with legion-like-armies.
I find it a little sad, and the first time my parthes will face a rag tag roman army of ballista, cav auxilaries and hastatii, but not a real roman legion, it'll probably break the 'authentic' feeling. Who cares if principes are right, if they never get built in the right army and the right context?
Interesting observation - I never thought of that. Isn't unit choice partly moddable as build priorities? I think WesW played around with it early on with his modding, to deal with the surfeit of peasants and dearth of ships, although other solutions were ultimately required.

The more I think of it, the more I would like to see some "quotas" or other restrictions on unit selection alongside point-costs. I'm thinking of something like Terry Gore's (and other minature gaming systems) army lists, where you can have 1-3 elites, 4-12 knights, 4-6 archers or whatever. By differing the restrictions you can get factions to field roughly historical armies and also give them more distinct character. Originally, I thought it would be required to avoid having too many elite types (elephants, praetorians etc) but I can see it would be more generally useful.

[Kraxis]

Louis, I understand you. I fooled around with the files as well and ended up with a good game where the enemies would train better troops faster (since I put more value on building ports so they gained more trade).

But I remember a chat I had with MikeB (Captain Fishpants) quite some time ago. I can't remember what it was we talked about but he mentioned that right at the moment the Numidians were overpowering the entire map since they had access to late legionaries and other Roman techs.

They had access to late techs, and used them. That must mean they actually advanced in tech to reach them.

[Aymar de Bois Mauri]

Quote[/b] (Simon Appleton @ May 26 2004,10:32)]
Quote[/b] (Louis de la Ferte Ste Colombe @ May 26 2004,08:44)]With the current system, honestly, I don't expect AI roman to come up with legion-like-armies.
I find it a little sad, and the first time my parthes will face a rag tag roman army of ballista, cav auxilaries and hastatii, but not a real roman legion, it'll probably break the 'authentic' feeling. Who cares if principes are right, if they never get built in the right army and the right context?
Interesting observation - I never thought of that. Isn't unit choice partly moddable as build priorities? I think WesW played around with it early on with his modding, to deal with the surfeit of peasants and dearth of ships, although other solutions were ultimately required.
Correct. WesW has made the AI trainning much more efficient and diverse, fielding quite able combined-arms armies. By tweaking priorities, Rebellions also became something to be fearfull of.

[Louis de la Ferte Ste Colombe]

Quote[/b] (Kraxis @ May 26 2004,12:20)]Louis, I understand you. I fooled around with the files as well and ended up with a good game where the enemies would train better troops faster (since I put more value on building ports so they gained more trade).

But I remember a chat I had with MikeB (Captain Fishpants) quite some time ago. I can't remember what it was we talked about but he mentioned that right at the moment the Numidians were overpowering the entire map since they had access to late legionaries and other Roman techs.

They had access to late techs, and used them. That must mean they actually advanced in tech to reach them.
Kraxis, there are different issues at play here... One of them being the capacity of the AI to tech up and build competitive armies.

But the one I am really concerned with is how appropriate those armies will be.

Sure, Numidian AI seems to be good at teching up and building plenty of good troops. That does not mean that numidian army will have a numidian 'feel'. Just that they build powerfull troops.

It's a little like saying that, in MTW, the Spanish AI is good because it knows how to to tech up and build armies full of lancers, and lancers, and more lancers.
On one hand, it's good, because it makes the game challenging.
On the other hand, full stack of lancers is not my 'vision' of a spanish medieval army. It fails to give me 'suspension of disbielief' required to make the campaign interesting.

It's not so much about teching up than about making the AI build realistic armies.

In my opinion, the tech and buying troops system TW is build around make that very difficult. I wonder if CA has considered alternatives to the "building / tech" system, and to the "buying army 1 unit by 1 unit / province / turn" system and their assessment of why the current system is better to give us a better roman feel experience.
Other games got other ways of achieving this...

Louis,

[Barkhorn1x]

Quote[/b] (Louis de la Ferte Ste Colombe @ May 26 2004,12:02)]It's not so much about teching up than about making the AI build realistic armies.
I agree, and I am concerned that as a Roman faction I may be facing Seulucid "Elephant" armies - instead of a more realistic force structure of (at most) 2 Elephant units.

Barkhorn.

[ah_dut]

If anyone from CA is listening Please hear our annoying niggles. The Diplomacy model in MTW is rubbish. The AI can't build a half historical army to save it's life, So help us CA and DO it right plz. http://www.totalwar.org/forum/non-cgi/emoticons/wacko.gif So help us please or else http://www.totalwar.org/forum/non-cgi/emoticons/gc-jester.gif

[econ21]

I suppose unit size is another way (in addition to points and building requirements) in the current TW system to influence army composition. eg 20 royal knights or 100 Feudal spearmen is not a bad way of making sure people don't go overboard on knights.

It would imply making powerful (elite) units - such as elephants, kats, praetorians etc - smaller than normal ones.

I'd still prefer a "quota" system - start with the 16 units per tactical battle constraint, for example, and then say no more than 2 can be elite, arty whatever...

[hundurinn]

It's always fun to have the units historically correct but in my opinion gameplay and AI is much MUCH more important. http://www.totalwar.org/forum/non-cgi/emoticons/gc-clown.gif

[frogbeastegg]

Quote[/b] (Simon Appleton @ May 26 2004,16:32)]The more I think of it, the more I would like to see some "quotas" or other restrictions on unit selection alongside point-costs. I'm thinking of something like Terry Gore's (and other minature gaming systems) army lists, where you can have 1-3 elites, 4-12 knights, 4-6 archers or whatever. By differing the restrictions you can get factions to field roughly historical armies and also give them more distinct character. Originally, I thought it would be required to avoid having too many elite types (elephants, praetorians etc) but I can see it would be more generally useful.
That is a really good idea, I like it a lot. It would force variety and historical troop mixes without being overly narrow in the possibilities it allowed the faction, and because you could choose the units to fill the slots in your broad types you would still be able to customise and use units you like.

For example in MTW if you were allowed 4 knight type units in each army you could use feudal, chivalric, gothic - whichever one you had the technology, money and liking for.

I do find this far better than the idea of making the best units really small in terms of manpower, size should reflect the unit's history. Also small units sometimes have difficulty contributing to the game simply because they are so badly outnumbered. There are units in MTW that lose alarmingly frequently to the types they are supposed to beat, simply because there are too few men in the unit.

I would also like to see the super elite units costing a fortune to build and maintain. Make them so expensive they are only used as the army's elite, instead of as line troops.

[MadKow]

One other thing that could be hardcoded would be a bonus for army composition.
You could build wathever units you wanted but if the army consisted of the necessary units to compose a Legion, units would have a bonus for discipline, or something.
But this (and other suggestions here), may be a bit away from the concept of T.W. ...

[SwordsMaster]

Quote[/b] ]That is a really good idea, I like it a lot. It would force variety and historical troop mixes without being overly narrow in the possibilities it allowed the faction, and because you could choose the units to fill the slots in your broad types you would still be able to customise and use units you like.

For example in MTW if you were allowed 4 knight type units in each army you could use feudal, chivalric, gothic - whichever one you had the technology, money and liking for.

I do find this far better than the idea of making the best units really small in terms of manpower, size should reflect the unit's history. Also small units sometimes have difficulty contributing to the game simply because they are so badly outnumbered. There are units in MTW that lose alarmingly frequently to the types they are supposed to beat, simply because there are too few men in the unit.

I would also like to see the super elite units costing a fortune to build and maintain. Make them so expensive they are only used as the army's elite, instead of as line troops.


Dont quite agree.
Firstly because once the game is a little bit advanced you usually create "patch" armies consisting only of, say spears because you have to reinforce 2-3 armies that stand in the same region + leave some troops for garrison.If something like what you've pointed is implemented then a 10 unit spear army will be horrbly expensive which isnt good nor accurate at all.

2nd I feel that is much more accurate to implement something similar to MadCow's suggestion about a bonus if units are in historical "blocks".

3rd If a legion was composed of some 3000 men, and you are only allowed to have 960 (depending on unit size) in your stack, how in all the (Roman) gods are you going to recreate a legion?

4th If you want to recreate a smaller legion (ie composed of all the legions unit types, but with less manpower) then again you cant impose severe restrictions on units, because specially after Marius' reform most of the soldiers were on the same type: legionares.

and how would you restrict flaming pigs? http://www.totalwar.org/forum/non-cgi/emoticons/gc-jester.gif Were they part of a consular legion?

[frogbeastegg]

Quote[/b] (SwordsMaster @ May 27 2004,16:57)]Dont quite agree.
Firstly because once the game is a little bit advanced you usually create "patch" armies consisting only of, say spears because you have to reinforce 2-3 armies that stand in the same region + leave some troops for garrison.If something like what you've pointed is implemented then a 10 unit spear army will be horrbly expensive which isnt good nor accurate at all.
What's a patch army? I have never used an army with that many of one unit type in. Even when I have armies hanging about to provide reinforcements for other armies or on garrison duty I keep a good mix.

A 10 spear army wouldn't be horribly expensive unless you were using elite super spears, the idea for high unit costs is for the elite, super units only and it is something that should already be in the game. Praetorians shoulsd cost a lot more than regular legionaries, for example.


Quote[/b] ]2nd I feel that is much more accurate to implement something similar to MadCow's suggestion about a bonus if units are in historical "blocks".
What idea is that? It's not in this thread.


Quote[/b] ]3rd If a legion was composed of some 3000 men, and you are only allowed to have 960 (depending on unit size) in your stack, how in all the (Roman) gods are you going to recreate a legion?
The idea is to recreate the troop mixes, not the actual numbers.


Quote[/b] ]4th If you want to recreate a smaller legion (ie composed of all the legions unit types, but with less manpower) then again you cant impose severe restrictions on units, because specially after Marius' reform most of the soldiers were on the same type: legionares.
Yes, so the rules on Roman army composition change as the units for the Marian reforms come in. Simple enough. I am guessing you will lose access to the old style Roman troops the moment the new ones arrive, otherwsie the whole reform is a bit pointless.


Quote[/b] ]and how would you restrict flaming pigs? http://www.totalwar.org/forum/non-cgi/emoticons/gc-jester.gif Were they part of a consular legion?
Oddball unit, say 2 (random number grabbed out of the air here) oddballs allowed per army. Ask Simon how this would work, he's the one who suggested this...

[econ21]

I'm not quite getting Swordsmaster's points. The idea of a quota system would be to limit "elite" or "special" units, rather than core. So 10 spearmen would not be a problem.

Take one simple system I know well - the Games Workshop Warhammer Fantasy Battle, WFB, rules[1]. There you have three troop types:

- Core, which have a minimum requirement; for example, they must be 50% of more of your army. Thus Romans would have heavy infantry (eg the three types of legionnaire pre-Marian) as core - you could not have an "all cav" Roman army or a Greek army with no hoplites etc

- Special: these would have a generous maximum, say 50% of less. For Romans, these might be auxiliary troops like velites and cavalry.

- Rare, these have a tighter maximum, say 25% or less of your army. For republican Romans, these might be "allied" troops - Spanish, Greeks, Numidians et. In RTW, these could be Praetorians or (if we must go there) pigs etc.

The result is that the armies have a certain historical character. This is one simple system - some historical wargame rules for minatures have much more detailed army lists that can constrain army choice to be much more realistic.

In principle, it would be easy for CA to add on something like this to TWs game system, rather like the "no more than 16 units on a battlefield" rule is currently imposed. Although I agree it is probably too late to implement this now.

[1] If you want to know more, you can do a google search for "Army Builder" and download a free trial program that can create armies under this system. You'd also need to download the datafile for Warhammer Ancient Battles.

[SwordsMaster]

Quote[/b] ]One other thing that could be hardcoded would be a bonus for army composition.
You could build wathever units you wanted but if the army consisted of the necessary units to compose a Legion, units would have a bonus for discipline, or something.
But this (and other suggestions here), may be a bit away from the concept of T.W. ...

Thats the MadKow´s idea I was talking about.

Get your point about elite units, but if you think about it they are actually 3 or 4 times more expensive than the common ones, so I dont see the point of just forbidding those units to be in a stack.

In the other hand, If your army composition is fixed to some extent, then your tactics will be VERY dependant on the circumstances (terrain, enemy, etc) your army is good when fighting in. Thats the way you should be, you could say, but that mean, that you will always lose against, say cavalry heavy army if your civilization doesnt have generic units with spears (just a hypothetic case).

Also, Im pretty sure that a legion fighting somewhere in Germania would have a different compostition than one from the north of Africa.

What if you have elephants AND flaming pigs (or screaming women or something strange like this). What if you need a few "odd" units, but they arent the same type of unit?

I think flexibility is VERY important, BUT you should receive a bonus when gathering a historically meaningful army (the way I would do it is giving a general +1 command if hes in charge of an historical army).

Im not against historical accuracy, but against limiting the players possibilities.


BTW, Simon, I will definitely try that Warhammer thing, thanks.

[a_ver_est]

When I am playing MTW I never think about the accuracy of my armies, I only try to get the most powerful army possible. Am I the only one ?

If we want AI armies accurate, our own armies must be forced to be accurate, if not we will have a great advantage.

The bonus idea isn't bad if (again) the AI can be programed to archive that.

[Nelson]

I like the idea of historically organized armies. It would be good to (optionally) compel Romans to fight with a facsimile of a legion.

I would penalize the commander of an army that was not properly organized rather than enhance the commander of an army that is correct. Subtract a command star or two. Otherwise, a great commander could still be fine with the wrong army style. Even Scipio should suffer with the wrong army composition. Do it right and things are normal. Do it wrong and pay.

[Louis de la Ferte Ste Colombe]

With the current system (as in MTW), it is going to be very difficult to have some good army composition.

Sure, CA can decide to reward players (and AI) with different bonus, or maybe even tactical display for choice of adequate 'historical' army.

A couple of other solutions are possible to somehow 'force' more historical armies.

CA can make the obsolete units impossible to build, and if retrained, they would be automatically upgraded to the new version; a little bit like royals are managed in campaign game today. At least armies would keep some form of historical consistency.

I'd also like a 'raise army' system over a 'buy-units-and-put-together-an-army-with-you-bought' system, a la Crusader King.
A 'raise army' is much more likely to result in an historically correct army than a ragtag of whatever you got in the fridge system.

I don't know how we could implement a quota system per se. How would the check be done? On the whole army? On a stack? On a province?
If we say; 50% max elite troops in a stack, can I get a 100% elite army by just walking two stacks together?

The problem with quota is that a player will resent it as a limitation.
IMO, the 'get rid of obso units' and 'raise army' system can limit a bit (a small bit for obso unit, a large bit for raise army system) army composition, but would be perceived as game feature...


What can be done to avoid facing/playing silly army of elephants and catapults?

Louis,

[econ21]

Quote[/b] (Louis de la Ferte Ste Colombe @ June 01 2004,15:13)]I don't know how we could implement a quota system per se. How would the check be done? On the whole army? On a stack? On a province?
If we say; 50% max elite troops in a stack, can I get a 100% elite army by just walking two stacks together?
Louis, I was thinking of the quota as merely a disaggregation of the current 16 unit limit. Instead of having 16 units max on the field at any one time, you could have no more than 4 elites, 8 specials or whatever. Think of the pre-battle screen as having not 16 undifferentiated boxes, but 4 that could be occupied by any unit (incl. elites), 8 that could be occupied by any unit bar elites and 8 that could only contain "core". Won't happen, I know, but would be workable, IMO.

[SwordsMaster]

ok, developing a little bit what i said in the previous post, this is IMHO the best that could be done:

1. The "check" for the wrong/right army composition should be done to all the armies that are able to arive to the battlefield for a battle.So if you move 3 stacks together, then all 3 are taken into account.

2. The general of a specific nation gets +1 when leading an "accurate" army of his nation. gets +0 when leading an army of the same cultural group (say a greek gen with a roman army), and -2 if leading an army of a different cultural group. (e.g. a roman gen with a persian or briton army)

3. That would lead to new vices and virtues such as Fast learner "cancels the malus for handling an army of a different cultural group", or a vice like Stubborn (or something similar) "-3 for handling an army of another cultural group, -1 for handling an army different from his own nation).

4. For allied nations, the checking is done for the highest ranking general. If both have equal rank, then for the largest army.

5. The malus for commanding the army is done in an "individual" basis, so an army 50% roman 40% briton with a Stubborn roman commander would get an equivalent to -3 command to all briton units.

Thats my vision of how it should be, but I think it might be too late for that...... http://www.totalwar.org/forum/non-cgi/emoticons/gc-computer.gif