Number of Battles and Strategic AI

[Red Harvest]

The AI really cuts its own throat and makes the campaign game tedious by sending out so many understrength scraps of units at the player. When you are sitting in a town with a half stack or a full stack, why will the AI trickle in 1, 2 or 3 units every turn? Or worse yet, a single unit, plus another army of a couple, plus a third of a few more. You end up having to fight 2 or 3 battlefield map actions to clear them. Plus you have 2 or 3 "AI withdraws" type strategic map events giving you victories.

All of this gives the human player a lot of easy victories and command stars out the wazoo. My generals are going from 1 up to 6 or 10 stars in only a few years.

Unfortunately, this means you have to fight many small battles that are simply a waste of time (but auto resolve is bizarre and untrustworthy.) CA could easily cut the number of battles in half or even a quarter, by having the AI merge its armies and fight in strength, rather than being killed in detail. It would make sense for it to skip turns and merge big armies rather than trickle in a few at a time. The Gauls seem to do this some...but they also like to send family members and faction leaders running about by their lonesome.

The one advantage of "nuisance armies" would be to lure you out into a trap or to lure your general out so far that he could not get back to the city on the same tuern. But usually the AI sends the little armies right up to your city wall, but doesn't even start a siege on that turn. That's really a dumb move.

In a related area: I've been thinking about sieges and blockades (because if the AI is fixed it is going to get truly nuts with them.) I don't think units that fall under some relative or absolute strength threshhold should be allowed to initiate a siege or blockade. Perhaps it would take X + 1 units to initiate a siege if the settlement was at X level (so higher level cities would require more to initiate siege action.) Same for blockades. Perhaps it might take 2 boats for the base level port, 3 or 4 for the next size, and so on. This type restriction should apply to moving armies as well...

Any thoughts?

[Morindin]

The AI really cuts its own throat and makes the campaign game tedious by sending out so many understrength scraps of units at the player. When you are sitting in a town with a half stack or a full stack, why will the AI trickle in 1, 2 or 3 units every turn? Or worse yet, a single unit, plus another army of a couple, plus a third of a few more. You end up having to fight 2 or 3 battlefield map actions to clear them. Plus you have 2 or 3 "AI withdraws" type strategic map events giving you victories.

All of this gives the human player a lot of easy victories and command stars out the wazoo. My generals are going from 1 up to 6 or 10 stars in only a few years.

Unfortunately, this means you have to fight many small battles that are simply a waste of time (but auto resolve is bizarre and untrustworthy.) CA could easily cut the number of battles in half or even a quarter, by having the AI merge its armies and fight in strength, rather than being killed in detail. It would make sense for it to skip turns and merge big armies rather than trickle in a few at a time. The Gauls seem to do this some...but they also like to send family members and faction leaders running about by their lonesome.

The one advantage of "nuisance armies" would be to lure you out into a trap or to lure your general out so far that he could not get back to the city on the same tuern. But usually the AI sends the little armies right up to your city wall, but doesn't even start a siege on that turn. That's really a dumb move.

In a related area: I've been thinking about sieges and blockades (because if the AI is fixed it is going to get truly nuts with them.) I don't think units that fall under some relative or absolute strength threshhold should be allowed to initiate a siege or blockade. Perhaps it would take X + 1 units to initiate a siege if the settlement was at X level (so higher level cities would require more to initiate siege action.) Same for blockades. Perhaps it might take 2 boats for the base level port, 3 or 4 for the next size, and so on. This type restriction should apply to moving armies as well...

Any thoughts?

I pretty much disagree with everything you've said so far about RTW apart from this.

This is a real downer for me, especially getting into the late game now and beginning some scraps against superpowers.
One thing I think the strategic map should do when a battle is joined, is merge all available armies together who have movement points remaining.
That way you'd get some of the truely large scale MTW battles that we all are sorely missing.

The command star thing is pretty ridiculous too, especially since the AI rarely has any.

However, at least the AI doesnt retreat and retreat and retreat out of all its provinces not even GIVING you a fight like it did in MTW when you had the stronger army, but still, im not sure the RTW system is more fun or realistic to be honest.

[Colovion]

I've had 3 Full Scale battles and I'm half done my second small campaign. I've fought well over 100 battles - most of them being my big stack armies against at most 6 or 7 units. There have only been a few battles in which I an the AI both have 1000 troops (or close to that number). I don't mind playing 500 vs 500 or something - but once you finish the beginning game and get into the Middlegame you'd expect the AI to be able to put up a little more of a strong front instead of wandering armies around everywhere just waiting to get picked off.

I suppose this is something that wasn't predominant in MTW because the AI only had a certain few amounts of places they could put their troops - and generally they were on the front lines and would by default merge into one big stack in that one Province - now they do that, but not in one army, just in the one province. My Generals when I was Julii got to more stars in 10 game years than I had ever had in MTW (unless I was playing with the Byz)

[The_Emperor]

I have had some very large scale battles in the game, so I do not know what you are on about. One was against a full Gallic army (that had Naked fanatics and druids and swordsmen) that tried to besiege a settlement of mine that was taken form them years before (it only had wooden walls).

I sallied and opened the gate, the Gauls rushed for it and straight into a waiting Phalanx of Merc Hoplites backed up by Hastaati, My General sallied from a side gate with some cavalry and rode around and charged into the rear of the mob of Gauls piled up around the Gate.

I only had around 700 men but I killed over 1500 Gauls in the ensuing chaos and mass-rout, they won't try that again in a hurry!

But the one thing that really gets on my nerves is how the AI builds up seemingly limitless fleets while I am struggling to pump out a navy and manage my upkeep costs...

My Brutii Campaign I have just walked over Macedon, and now I am fighting the Greeks again. They only have Sparta, that city in Asia minor and Rhodes, and a territory in Thrace, yet somehow the sea lanes are swamped by full stacks of fleets!!

Two family members (and their armies) were killed when their fleets were sunk (wich no hope of escape, because they got caught between stacks)... My revenge shall be brutal.

[Lord of the Isles]

Main point: when posting in threads like this, can you give the difficulty levels you are playing at? It doesn't take long to add (Hard/V Hard) or whatever to a post and greatly helps the debate if we know what level your comments apply to.

Lesser point 1: The AI does use lots of small armies (at Normal/Normal & Hard/Hard) in my experience. I wonder if there is a reason: they are trying to cut down your trade? Unlike Red Harvest though, I've got used to the autoresolve for these small battles and find the results ok as long as I outnumber the AI army.

Lesser point 2: Red & Morindin agreeing ... can this be stickied...

[Jambo]

I quite like the current mixture of some small skirmishes and some huge battles. The one thing I hated with MTW was the having to fight 4 or 5 2000+ battles every turn. Whilst the odd one was a memorable and fantastic struggle the majority were rout and reinforcement fests, which swiftly became very tedious. If the odds are so much in your favour then simply autoresolve them.

[ToranagaSama]

The AI really cuts its own throat and makes the campaign game tedious by sending out so many understrength scraps of units at the player. When you are sitting in a town with a half stack or a full stack, why will the AI trickle in 1, 2 or 3 units every turn? Or worse yet, a single unit, plus another army of a couple, plus a third of a few more. You end up having to fight 2 or 3 battlefield map actions to clear them. Plus you have 2 or 3 "AI withdraws" type strategic map events giving you victories.

All of this gives the human player a lot of easy victories and command stars out the wazoo. My generals are going from 1 up to 6 or 10 stars in only a few years.

Unfortunately, this means you have to fight many small battles that are simply a waste of time (but auto resolve is bizarre and untrustworthy.) CA could easily cut the number of battles in half or even a quarter, by having the AI merge its armies and fight in strength, rather than being killed in detail. It would make sense for it to skip turns and merge big armies rather than trickle in a few at a time. The Gauls seem to do this some...but they also like to send family members and faction leaders running about by their lonesome.

The one advantage of "nuisance armies" would be to lure you out into a trap or to lure your general out so far that he could not get back to the city on the same tuern. But usually the AI sends the little armies right up to your city wall, but doesn't even start a siege on that turn. That's really a dumb move.

In a related area: I've been thinking about sieges and blockades (because if the AI is fixed it is going to get truly nuts with them.) I don't think units that fall under some relative or absolute strength threshhold should be allowed to initiate a siege or blockade. Perhaps it would take X + 1 units to initiate a siege if the settlement was at X level (so higher level cities would require more to initiate siege action.) Same for blockades. Perhaps it might take 2 boats for the base level port, 3 or 4 for the next size, and so on. This type restriction should apply to moving armies as well...

Any thoughts?

My Friend, I hope you don't mind my calling you that, because we are Kith (sp?) and Kin on the subject of RTW.

I spent all day yesterday, unintentionally getting sucked in by the game, and am still in the beginning game.

Your comments above, describe my experience and thinking accurately. There are TOO many battles!! dragging the game out unnecessarily. All of the battles were against the barbarian Gauls, and they were vastly outnumber in all but 2 battles. I must have fought a couple of dozen. The latter third of the battles, my troops outmatched them in *Quality* by a couple of degrees. So even in the *large* battles, the Gaul amry routed without much fuss with my losses less that 100 men.

I agree WHOLEHEARTEDLY that's theres just too much UNECESSARY fighting. Not only do we have to contend with understrengthed enemies, but, ALSO, the little bands of Rebels causing *Devestation* effects to the Cities. There can be DOZENS of these bands, which haven't the slightest possiblity of defeating the Player's army. Soooo, ahhhhh, what's the point?

Yeah the *Devestation* effect, OK, but cleaning up all the little Rebel armies is a total PITA!

I think the solution is that similar to MTW, small armies of ANY kind when faced with overwhelming odds should RUN AWAY! They should remove themselves to another provice.

In the case of Rebels, the AI s/b able to make an assessment, after once being confronted, and SCRAMMM! Not to be heard from again, until and unless the Player's Troop strength in that Province falls significantly.

Possibily there should be some sort of *Patrol* function for the "Town Watch". You Train a required number of Town Watch units to keep Order, and set them on Patrol. Having these Units should keep *Rebels* away.

The Key is "having the required number". As the City becomes less orderly, just as the game is now, greater numbers of Town Watch (or other troops) are necessary (or possibly limit this function to Town Watch only). The number of units drops below the "required number", and then Rebels begin to show up and harrass the City.

Whaddaya think?

In Regard to all the little Enemy Army, a bit more work on the AI may be necessary.

Prudence, PRUDENCE, the AI shows NO predence!

Particularly, after singificant victories over huge enemey armies. I understand that the barbarians are a conglomeration of smaller bands and tribes (or whatever), with several individual type leaders, but when confronted with a superior and previously dominant force, the AI should retreat the little armies similarly to as I outlined for the Rebels. Except that the little armies should *gradually* over the course of time band together into a force more capable of confronting the KNOWN enemey player.

For example, the armies, being of an individual mind, might enter into my territory, but once I, as the player, make a 'Show of Force', the little army should show Prudence and high-tail it outta there, and seek out some willing comrades to join them. Until such time, the little army should stay out of my Province, as long as I have a, KNOWN, DOMINANT, previously victorious FORCE.

Perhaps this should be tied to the Reputation of particular Generals. A particular General could develop through battle victories a reputation that defeated enemies would respect, and consequently stop sending troops into the General's territory. I like this!!

If the General suffers Defeat, the perception of weakness could be determined, and THEN the enemey *might* seek to Probe and/or Test the General's forces. Similarly, upon the General's death, or his replacement with a new general, the *new* General would have to *establish* Dominance, and then the enemy would respect his territory.

Yeah, I like this ALOT! Helloooooooo CA....

---

Like with STW, and, as carried over into MTW, to some degree, RTW should follow the Tenents of Sun Tzu. Battles should be fought as a LAST resort, and most battles should be won on the Campaign Map, not the battlefield. In this way actual Battles would MOSTLY be Desparate Battles fought under Deparate Circumstances, the result of the Battle have *DIRECT* consquence to the Campaign.

The Campaign should NOT serve as a rather lame execuse to have (endless and pointless) Battles (isn't that what MP is for?). Rather the Battles should be determiners of the Campaign. I feel in love with TW because it gave true *meaning* to the Battles. Yet, with RTW, all these little battles are MEANINGLESS and lessen the greatness of the game. Win or lose, these battles have *virtually* no effect upon the Campaign. QUALIFIER: At least this has been my experience so far.

I'm playing as the SPII (??), you know the RED guys. I've had at least 24 battles (or more). TWENTY out of the 24 have been of the kind described above.

---

[/QUOTE]However, at least the AI doesnt retreat and retreat and retreat out of all its provinces not even GIVING you a fight like it did in MTW when you had the stronger army, but still, im not sure the RTW system is more fun or realistic to be honest.[/quote]

That's PRECISELY what it should do. As the Player continues to advance, and his empire continues to expand, Negative Factors in controlling his Empire should begin to occur, and the game focus shifts from that of warmongering toward *management*. Further testing your skills as a Damiyo or King. This is the whole point! Not only must you be a warlord, but also an administrator, as well.

Either your adminstrating skill will be up to the job or it won't. If not, then your empire will beging to degrade and fall apart, whereupon the Player will either voluntarily or involuntary retreat and consolidate.

Even if you, the Player, are up to the challenge of Adminstration, then one of two things should occur if the AI continues to retreat before your jauggernaut.

The first thing is that your troops will begin to be stretched thin as the Player attempts to keep his newly won empire under control (for the most part preventing Massive Rebellions). In which case, your jauggernaut will be weakened to a degree that the AI will STOP retreating and begin advancing; and/or some other Faction will take advantage of the situation. In any event, the *design* aim is for the Player to have too many *hot potatoes* to keep in hand;

OR, at some point, the AI will be cornered and the determining battle will take place. You know...5000 men on the field......stuff. Often, it will take a bit of SKILL to corner in the least amount of time!

Desparate battle, desparate circumstances....
(Too bad, it would seem, that in your experience its always the AI that's desparate!!!)

Unfortunately, many players have not and will not experience the above, if they've only played *vanilla* MTW, even if Patched.

To experience Total War in its full glory, one needs to buy and install Viking Invaison, Patch it, Turn on :green_generals; establish and play with something approaching the "Hardcore Rules"; and downloand and install the MedMod. Oh! and *master* it all.

Until experiencing such, how can one agree or disagree? On both sides of the issue, we are talking and comparing different animals!


Vanilla MTW vs RTW
[Vanilla MTW allows the Player to have too much money, and the AI is inept at compensating for this fact, among other things. Player Jaggernaut vs. AI *Peasant* Armies]


Med Modded, Harccore Ruled, VI, patched, green_generaled vs. RTW
[Not enough Money, no Peasant Armies, nuff said]


From many perspectives, the expectation was that RTW would, at least be the ***evolution***, in terms of Gameplay, to the combination of the Med Modded, Harccore Ruled, VI, patched, and green_generaled.

To be fair to CA, in many ways, if one looks closely, RTW is much of just such an evolution, but....somethings are either, missing, hidden, or lost within the amazing Strategy Map (and lack of documentation).

The new Strategy Map is a thing of wonder, and adds a whole new dimension to the Total War experience, in terms of both Strategizing and Tactical manueverings. Yet, with all the evident capabilities inherent to the Strategy Map and the wonders of the 3D Battle Engine, for those who have experienced and/or mastered the combo of the, Med Modded, Harccore Ruled, VIed, patched, green_generaled, MTW, and have spent countless hours playing such, SOMETHING IS MISSING, and one cannot agree or disagree without the experience for comparison.

Guess what? This thread testifies that its not all about Unit speed and kill ratio.

Long Live Shogun: Total War and Long Live the MedMod!!!

~ToranagaSama

[ToranagaSama]

btw, autoresolve is NOT the answer!
I believe I'm playing at the default settings. How do yo check this, and is it possible to adjust the Difficulty during a Campaign?

[Jambo]

If you're bringing an army of 600 against 100 or so rebels, then autoresolving is most definitely the answer.

If you're finding the AI using pitiful armies against your higher quality armies, then crank the campaign difficulty up to VH.

Don't for a minute tell me that STW had a better system than this! That's absurd. I could beat STW campaign on any difficulty without any problem. Sometimes, even the rebels did the job for me which was a real bore.

One thing I do agree with, is the rebels could do with being a little on the tougher side. Nevertheless they can serve for promoting the ranks of yours or the AI's generals.

If any of these assumptions being made here are by people playing the game on lower than Hard (myself I've only played on VH/H), then I suggest playing harder campaign or battle difficulty levels.

[Quietus]

I don't know about that. I sure fought alot of stacks. The only too-easy problems with the AI stack is that I can attack and trap them while they are sleeping in their city. That means I'm now at an advantage and they have to sally forth.

Or sometimes, I can easily lure them to attack me in bridges which mean their army getting "crushed" as the narrator would say.

Also, if the AI had limited provinces, they can't possibly build as fast. Right now, the Brutii have about 12-15 full stacks and they are at war with Dacia and Thrace (so am I).

It's the only reason why I haven't attacked Rome yet (with the Scipii about 6-7 stacks).

I only have 1 and a half stack (reinforcement) fighting in the East. Another half stack I''m building up in Alesia since I'm attacking the Britons or Germania (I'm still planning). The rest of my troops are mostly Town Watch garrison in rebellious towns as well as high valor very depleted units I don't want to disband.

[Duke John]

What about this proposal for a patch? Note that I no idea about the difficulty in programming this.

1. The AI should only besiege a city when it outnumbers the besieged at least by 1.5:1
   
2. When a small AI army is attacked by a larger army it will retreat and will try to merge with another army as soon as possible.
   
3. The AI should only attack with a smaller army if it's defending against invading armies.
   
4. The AI should attack with a larger army (number of units, not total of men) if invading other provinces.
   
5. If the AI is strategically and tactically more able at harder settings then the setting should not have additional benefits for the AI (stat bonus, more income, etc.) Note that I'm not entirely sure about this.

I hope that point 1 to 4 are just easy changes in the code:
IF NumberUnitsInArmy < NumberUnitsInEnemyArmy THEN Retreat AND Merge
or something like that.

[Didz]

Have to say I've been playing 4 days now at Normal Difficulty level and I have'nt noticed this problem.

It's certainly a problem I came across with Gates of Troy but so far the AI in RTW has done quite a good job of organising its armies and attacks in my game.

The only real exception was an Egyptian attack on my colony in Phrygia where the AI sent is main army led by a Captain (stupid name for a non-character general) to beseige Sardis and a much smaller force under two of its heirs to attack Pergamum. On the face of it, it was a dumb move but it actually gave me a headache in that my garisson in Pergamum was blocked from moving to assist Sardis and I thought in fact that the AI was trying to be clever.

The same is true with fleets. Both the Greeks and the Egyptians met my ships with massed fleets with almost full stacks. However, I have seen these stack suddenly broken down into multiple small fleets in order to block sea routes or screen a vital troop convoy.

So, at the moment I am quite impressed with the AI and I'm a bit puzzled why the game should play differently for other people.

[Jambo]

1. The AI should only besiege a city when it outnumbers the besieged at least by 1.5:1
   
2. When a small AI army is attacked by a larger army it will retreat and will try to merge with another army as soon as possible.
   
3. The AI should only attack with a smaller army if it's defending against invading armies.
   
4. The AI should attack with a larger army (number of units, not total of men) if invading other provinces.
   
5. If the AI is strategically and tactically more able at harder settings then the setting should not have additional benefits for the AI (stat bonus, more income, etc.) Note that I'm not entirely sure about this.

1. There's more to besieging than simply assaulting. Besieging cuts off trade and severs income and troop movement. If the armies are equal it forces the besiegers then to sally forth to relieve the siege. I'd be more inclined to agree with this ratio if this was for the AI to assault a city with stone walls - that's where the AI sometimes makes daft decisions. If it doesnt have the numbers, it should simply starve the inhabitants out.

4. Disagree. VH has to be as hard as possible. Just look at Civ where some people can beat the AI even with ridiculous cheats. A hUman will always beat an even par AI. Compared to chess this game has too many parameters, and therefore, a human player will always beat an AI on an even footing. If the AI becomes beatable at all levels the game will no longer be interesting to play. I'm happy the AI has economic boosts at the harder campaign levels. It means I will be fighting advance troops and not squadrons of peasants and warbands.

[Duke John]

Point 5 (not 4) doesn't mean that the AI should never get bonuses. I mean that before it gets bonuses there should be a setting in which it has the maximum tactical/strategic capabilities. One setting higher and it gets bonuses.

In M:TW, you had medium in which the AI was not optimal. At hard it was at its best but there were no bonuses. At very hard it would get bonuses for the ultimate challenge.

I have my doubts that in R:TW that the AI is already at its best in medium (taking the method M:TW into consideration). So I guess that is at hard. However at that difficulty the AI already starts to get bonuses; phalanxes are no longer capable of properly holding a cavalry charge. I believe that Red Harvest did some tests and the AI was clearly having bonuses at hard.

[Longasc]

Well, the AI has some problems with Naval Invasions.

The Scipii and Brutii always land on exactly the same spot on Sicily and get slaughtered by the Cav heavy Garrison of Messina.

They do not land in FORCE, the land always at the same place in 2-3 units stacks. (HARD/HARD)

[Mori Gabriel Syme]

In my Scipii campaign (Normal/Normal, I think--whatever the default was, I didn't change it), I frequently have the AI army retreat on the campaign map when I attack with a larger army. Sometimes I pursue it, forcing it to engage, but often not because that takes my army too far from the city it is protecting to return on the same turn. This allowed the Macedonians to build up a few sizable armies to besiege Athens, merging the smaller ones.

[Ktonos]

Well I disagree. The AI use strategic use of its armies depend on the faction. I thing that most of the Barbarian cultures will spread their forces all around, smthing like small annoying raiding armies. Now Romans, Macedonians, Greeks and Egyptians, even Carthage (if she can afford it) have 5-6 medium or large stacks.

I much prefer this. Also MTW could be FULL of many full stacks in the same province (annoying and unrealistic) and apart from that the AI did not took advantage of its advanced units. The only Knights I had to deal with were the bodyguards of princes or Kings. In RTW the AI use its tech tree wise and usually has good quality units to field.

The only AI problem is the tactic use of units in battle map. And that is in certain "buged" occasions (like hopless hastati stand out of walls taking fire while all siege equipment is destroyed and there is no hope to take the city).

[Ulstan]

I have noticed this playing on very hard campaign difficulty.

The gauls have all these small armies running around led by captains - sometimes 3 or 4 in the same province - why don't they merge them together and create an army that can fight mine?

Also, why only captains? I Have seen some gaulish family members, but they are invariably running all by themselves from one town to another.

I think the AI uses all its family members as governors, and uses none for genearls - thus, they never have any good generals to oppose yours with.

I would like to see an AI tweak that would put lotsof weak rebels popping up in ai lands that the ai generals can clear out for experience and whatnot.

Also, it would be great if an AI army could lose a battle w/out wholesale extermination. In MTW, lots of times I'd win, and a large part of the enemy army would flee away to safety. (Including their good general).

Here, an AI general loses one battle and he is dead.

[Ktonos]

Well, most of Ancient battles with huge armies was decisive. The losser would just fall apart. Either its army would just fell to the last man or the remaining forces would scarce around after the troops lost confidence to their leader. Darius had 3 major battles against Alexander. In Granicus he fielded about 30.000 men, in Issus 55.000 and in Gaugamela about 500.000.After every succesive defeat he had to gather new troops for the next battle. When he lost in Gaugamela about 30 to 40 thousand Persians (and asian allies) were killed in the field. But the bulk of his army scattered around after seeing him fleeing. Only his bodyguards remained by his side.

[Didz]

Its certainly beginning to look like a cultural thing as most of the complaints seem to apply to Gauls.

Now that I cme to think of it I was at war with the Gauls during the early game and never saw anything from them other than small wandering fleets. They sued for peace when I seized and enslaved their settlement at Salona and I haven't had any dealings with them since.

Perhaps its a problem with settlement size, Salona only had a tiny population and was quite incapable of being used to replenish losses amongst my troops.

Maybe they just have problems raising large armies.

[ToranagaSama]

Ulstan, Didz,

It just might be Faction related, don't know yet.

Not only do you have all these little Armies handed by Captains as Ulstan pointed out, but you have "Family Members" wondering around with the smallest of all the Gual armies. I seen 3 or 4 such "Family Members". The AI made no effort to consolidate under with the Family Member.

In fact, I don't believe a single army in battle was headed by a Gual "Family Member". Anyone else experience this? This *might* be the result of Difficulty level. A higher Difficulty level might result in Family Member lead armies. Confirmation?

[Lichgod]

I also find the numerous tiny battles tiresome.

Playing Carthaginian, Hard/Hard.

Auto Calc is a big NO NO if you have elephants. Everytime (about 20 experiments in various situations both battle and siege) where I have overwhelming advantage (5-1 or better) using Auto Calc, the elephants take heavy losses. One case, 600 Carth vs 80 Guals, auto calc, I suffer nine (9) losses – all elephants. I only had 12 in the army. I never autocalc if elephants are in my army. BTW, I like to put one elephant in each of my big armies. Sigh

General gaining stars is way too easy. Chasing bandits (but not killing them) is guaranteed to train your general to three stars quickly. Almost one star a turn.

I did find out that bribing the bandit army might be better. Had a bandit army of a peasent, Iberian Lt Inf, and two Spanish mercs (the pila guys). For 1510 gold, they surrendered with the peasant disbanding and the other three joining me. The Spanish mercs normally cost 750 each and hard to replenish once you buy em, so good deal was had.

Maybe the general gaining stars could use a point system like this:
2 pt for winning on battle/siege map (not where the enemy just runs away on the campaign map)
-2 pt for losing a battle/siege (but your general should be dead anyways, right?)
1 pt per 250 enemy dead after subtracting half of friendly losses, (enemy dead – friendly dead/2) = body count. This would be a running number kept between battles.
-1 pt for 250 friendly dead after subtracting enemy losses, (friendly dead – enemy dead) = body count2. Again, running number kept between battles.
2 pt for killing a enemy faction member (note several may be on the battle field at once)
-3 pt for losing a faction member (AI reinforcements = ouch!)
1 pt for defeating an enemy general per three stars defeated. Running total kept so two battles, one vs 1-star and one vs 2-star = 1 pt.
+3 pt if enemy faction member killed is Heir
+8 pt if enemy faction member killed is Leader

Promotion/Demotion occurs as points accumulate:
3 pt = 1 star
9 pt = 2 star
18 pt = 3 star
30 pt = 4 star
45 pt = 5 star
63 pt = 6 star
84 pt = 7 star
108 pt = 8 star
135 pt = 9 star
165 pt = 10 star

(((Previous Points + (Previous Star x 3) + 3) = new point total


So, over time, you have a leader:
Win 30 battles = 60
Avg 250 kill difference = 30
Kill 6 faction members = 12
Kill an Heir = 5
Kill a Faction Ldr = 10
Avg 1 star enemy ldr per battle = 10
Total 127 = 8 star leader (almost 9 star)

[Red Harvest]

I've only tried to use auto resolve a couple of times (other than naval.) Each time it gave really screwy results. It was on very hard/very hard. It tended to produce casualties in my best units or kill off family members. Being that all of the battles I would autoresolve are very easy wins even on very hard, I have been unwilling to accept the completely random results. Outnumbering 10:1 didn't give good results. Perhaps it is better if you play as Romans. As Carthage, it stinks.

I have not auto resolved since the "elephant fiasco." I had a nice stack for conquering the small settlements in Numidia. It had one elephant unit for the gates. So I start running into the typical, single unit along the road during hte long march. Usually, I attack, they withdraw, then I attack the 2nd time, and send my cav off to kill it in a pincer losing maybe 4 or 5 men. That clears the road and my army advances until it hits another obstacle or runs out of movement points.

So here I am facing a lone archer unit on the open desert. I had done this a few times before. Race out with cav, run the unit down, its over. Unit is gone from the map. What happens with autoresolve? 60% of the enemy unit survives, and it is able to withdraw...meaning I will have to do this all over again. My elephants lose 10 or 11 out of 12 making them worthless, and I take as many total casualties as the archers. I'll get to face it again on another turn, yay!

While this example sticks out, my other auto resolve experiences were similarly awful. Maybe the AI screws the player in auto resolve when they play as Carthage? I have noticed the naval auto resolve has been surprisingly favorable now that I'm playing as the Brutii on very hard/very hard. It is completely different.

[Red Harvest]

Much of the army size/number fielded seems to depend on what the AI can muster at that point in the campaign. For example, as Carthage the Roman AI sent large stacks for initial battles. Once I won the territories or defended successfully, the stacks became small and piecemeal for many years. As the Julii became stronger, they finally sent some bigger stacks. However, they often landed in three places on Sardinia...what a waste of an army. Had they built up a sizable force over 2 or 3 turns, my small outpost army would have been beaten (as I expected.) Numidia did similar things. The Brutii went off conquering elsewhere, leaving me alone for years, and then sent in large stacks against me. That was a plus, but they had the financial power and settlements that allowed building armies rapidly.

As the Brutii, the Macedonians fought a couple of large battles where I wrested control of a province, then another. The first attacks were large, and my attrition was high, so things nearly went badly for me. Unfortunately, after that they started trickling in disjointed small armies--often three in a province at once. I probably only had one "real battle" every four or five turns.

As Carthage, the Gauls sent in large armies early, but now they send a mix of probably 2 small armies and one large army.

I'll have to keep stats for a "short campaign" but so far I would guess the distribution of battles as:
Paper AI strength/quality greater -- 10% usually the initial key fight, and then the counter strike.
Paper AI strength/quality equal -- 10%
Paper AI strength/quality ~1/2 -- 30%
Paper AI strength/quality much less -- 50%

[Didz]

I must admit I'm also beginning to get the small army/fleet thing now.

At the opening of my invasion of Eygpt I had two massive battles. One at the gates of Sardis and the second at Tarsus. These were both more or less 1:1 and those chariots are a real pain.

Plus two massive naval battles off the coast of Sparta which destroyed the Egyptian fleets.

But since then things have degenerated into a dribble of small Eygptian units (Armies and Fleets) running into my garrisons.

Mind you I have just done something almost as stupid myself. Having taken Tarsus I noticed that Mazaka was held by a rebel faction. Now, so far rebel factions have always been a walkover for my armies so I thought this was good opportunity to snap up another province and give one of my newly adopted generals a bit of combat practice.

So, I divided my Tarsus garrison and leaving Oppius the Mighty defending my Cilicia I sent his subordinate with a medium sized army through the pass to invade Cappadocia. He goes marching up to the city and declares the start of a siege only to discover that the city is brim full of the most viscious looking units he's even seen. Ok! Theres a couple of peasant units but most of the rebel army seems to consist of heavily armourered cavalry the like of which House Brutii have never faced before and my armies still don't have any spear armed infantry.

I am kicking myself for being so complacent and not sending a spy to check things out before I attacked but at least I know I need to run like hell next turn.

[ToranagaSama]

Hey Didz, sounds like you're having a ball!!!

Lichgod, MTW already had a similar such system. I cannot fathom why CA thought a change was necessary. Of course, the obvious answer is to, "...make the game more accessible...."

[Sinner]

It's odd but my auto-resolves have generally been good so far...

I've built up my two main combat stacks to include a general, a first cohort, 9 legionary cohorts/praetorian cohorts/urban cohorts, 2 legionary cavalry/praetorian cavalry, 4 auxiliary archers, & an onager/heavy onager.

Despite the posts I'd read about the horrors of autoresolve I decided to give auto-resolve a go when I just couldn't be bothered to command the battle assuming that at worst I'd lose my general (who was old anyway), with the most likely outcome being the inconvenience of having to retrain my onagers and archers whom I otherwise keep safely out of harms way.

So far, at VH/M and VH/H, in both open field battles and siege assaults, against both few and many opponents, I've found that my losses tend to occur among my frontline infantry, with the first cohort perhaps more prone than the others. I occasionally lose a few men from my general's bodyguard and the cavalry do suffer more losses in sieges, but so far my archers and artillery come out of every battle unscathed.

For me the 'feel' is just about right so far, although I think my archers should have lost some men by now, if only to counterfire, plus I would have liked to have seen the occasional outcome where I suffered heavy losses among the cavalry, archers and artillery to account for the odd vagaries of battle, instead of the PBIs always bearing the brunt.

[Ulstan]

The brutii have 6 (1,2,3,4,5,6) captain led stacks (none with more than 3 units) wandering around together right now. They are all in the same province and could *easily* combine into a decent army.

Meanwhile, their general, is off garrisoning a fort in the mountains with only his personal cavalry.

[Lichgod]

Carth large campaign H/H

As I mentioned above, I cannot autoresolve when elephants are present unless I want them to die regardless of the size of the enemy force. I even save prior to a non-elephant autoresolve just in case the results are way out of wack.

Two points:
1) I just exterminated Rome! 28k+ well needed cash. Seems I am always short on cash even with all ports/roads/bazars built. But a problem/bug involving weak armies is showing up. I lay siege with a 1200 man army. Defender has about 600 guys in town. Going to take 9 turns, ugh. Next turn, a relief force shows up and attacks me. Sound ok? Well the relief force is only 81 guys. They force the city garrison to sallie. Garrison is crushed, city surrenders during the enemy's turn. This happened also to both Julii cities north of Rome. But, in one case, the relief force was 1300 guys with 650 in the garrison. I had 1350 troops sandwiched exactly between them. Was a great fight and victory (city surrendered w/loss of garrison). This was with 8 units of Sacred Band Infantry (green), two war elephants, and a full complement of experienced slingers and shield cavalry plus one Sacred Band horse. I saved that battle in case I wanted to fight it again.

The AI should add the Atk + Def + Missle + Charge + Hit Pts of both sides and only attempt a relief if the forces are 1:1 or the city is about to fall.

2) Friendly fire (FF) really is getting to me. My Balaeric slingers have to be micro managed too much. I have experienced too many battles where FF loses exceed enemy kills. I am thinking of a slimy move to make FF work for me. Put new slingers with some Peasents, a general, and some cav. Go find bandits and fight a battle even at 10:1 odds. Put the peasents in front of the slingers to take FF losses. Rest of the army hangs back while the Slingers proceed to kill peasents and bandits. I suspect doing this will make my slingers sliver/gold bar experience quickly.

I have lots of fun with this game, especially with the elephant/cav wave charge. Will have to play Julii next and see if I can get some merc elephants.

Carth Campaign SPOILER (sort of)
*
*
*
*
*
*
As Carthaginians, I experienced early the Julii desperately wave attacking Sardinia until they took it (with about 650 Scipii troops helping - should have saved that AI town assualt with the red and blue troops attacking). Then the Julii left me alone until I took the two gaul towns in Northern Italy. The Scipii, once kicked out of Sicily, only did the tiny army drop offs at Messina . BTW - Let Scipii take Syracruse (Greek town in Sicily) as they seem to go for it first, then take both of the Sicily towns from them. Also, expect war right away with all the Romans, Spanish, and Gauls with the Numedians soon after.

[ChaosLord]

I agree its too easy to get a general his command stars, even if they aren't as useful as in MTW. While not being useful the generals are alot rarer, so given the AIs poor use of them its easy for the player to get the advantage. What we really need back is titles to grant general/governorship. These would not be family members(though you could get the chance to adopt good ones). They would also have half the retinue capacity of normal generals/govenors. Doing this would give the AI an inexhuastible source of generals. It'd help the player too of course, but I think in the end the advantage would tilt towards the AI. It'd mean more AI generals with stars and more armies lead by generals/cities defended by them. It might be more trouble then its worth though.

As for the AI not attacking unless 1:1 I don't think that should be added. The only decent tactic the AI has strategically right now is a war of attrition. They gotta keep hitting your armies/weaking your forces. I've even sacrificed men/family members myself to do this. After losing Cirta I brought a merc army to assault the city, it was underpowered but allowed me to kill another 700 units, meaning next turn the city revolted back to my control.