Auto-Resolve Battles: Changed, still the same?

[WineCape]

Would like to know if there is any difference regarding the logic of Auto-Resolve battles.

As far as I can recall from RTW, you would almost always lose the auto-rersolve battle if force parity/General's experience is roughly equal. Even if the AI is slightly outnumbered.

Is it still the case?

[FesterShinetop]

No, I don't think so. I've seen a few wins and draws when the stats were roughly the same. Still better to fight yourself though as it can also turn out the other way round....
In one of the sea battles I was facing a much stronger force of pirates (man, there are a lot of those around) and I was amazed to see I actually won!!! Against all odds!

[Stolpmeister]

The autocalc seems to have been buffed considerably, almost to the point of being overpowered. Especially on sieges I've gotten some really good results. On one battle where I invaded a citadel with about 500 troops against 200 defenders, I autocalced a victory with only 30 casualties! If I'd played it out I would have lost AT LEAST that many just walking up to the first layer of walls, then probably double that fighting on the walls. Almost seems a bit buggy.

At least that's my impression.

[Merlin's Apprentice]

Ive seen a ton of draws in siege autcalcs that end with a win anyways

[Fridgebadger]

In Rome, didn't it depend on Campaign difficulty, not battle difficulty? That said, my first campaign was hard and the autocalc seemed to be quite friendly...

[Ciddler]

I only auto-resolve battles against rebels. Sometimes I don't even resolve those, I fight every battle which is why it takes so long for me to finish. I'd rather put my men's fate in my hands than in their own ;)

[darsalon]

I think the auto resolve ismore weighted in your favour than it was in original Medieval for instance. I'm making far too good progress as a result I think. For instance in a game as the Byzantines last night I was besieging a Viennese fortress holding foot knights and mailed knights (I think) with spearmen and horse archers. Did an autoresolve and I lost only 40 men. Should have lost far more than that.

Really must try and avoid taking the easy option but if I'm losing, as I was in that game fairly heniously, then it was far too tempting to at least get a small bit of revenge.

[PseRamesses]

In previous titles autocalc has been a debatable thing to say the least but now (M2) it seems to be too good to the player. In numerous cases I´ve autocalced battles, reloaded and fought them manually and in most cases I get some 3-5 times higher casualities than when I autocalced the same battle. Maybee I´m just a crap-general or is the autocalc function too good in favour of the human player?

[Ciddler]

I only auto-resolve battles against rebels. Sometimes I don't even resolve those, I fight every battle which is why it takes so long for me to finish. I'd rather put my men's fate in my hands than in their own ;)

I was going to bed yesterday, saved and still had a battle and a siege open, I thought I would just autocalc those then go to bed (1 more round, 1 more round...). Well I lost the siege even though I outnumbered them 2:1 and I lost the battle with 1:1. So I don't think it is always in favour of the player. I do not recall the losses but Milan asked for a ransom.

[tonedog]

how realistic does auto resolve work, does it take into account general skill, troop type and terrain etc?

[GeorgeBush]

Supposdely the auto resolve does take into account all of those factors

[tonedog]

supposedly? what r peoples experiences of it?

[Carl]

It takes the nto account, trust me, the bar also lies badly, that just uses attack, defence, and numbers, it dosen't take into account any unit attribbutes that might tip things. For example i've had 3 units of mailed knights try to take on 2 units of italian Spear militia in Auto-calc and accordding to the bar it's well in my favour, but i've lost with 2:1 losses (i.e. i lose 2 knights for every 1 spearmen I kill).

Likewise it dosen't take animations into account so 2-handers are not affected by the animation bug and the sheild bug also dosen't seem to have an effect, neithier does the pike bug, needless to say this makes auto-calc VERY EASY when it would be VERY HARD on the battle map itself if you fought the battle.

[Midnight]

In my experience it doesn't take into account castles or cities! I've fought some battles through auto-resolve that have led to my winning with very low casualties (given walls with defender fire, defender wall bonuses, narrow street fighting, etc) and with the enemy having men left at the end (not possible as they should run to the centre and rally, then fight to the death)! These men subsequently disappear as I capture the city\castle.

I don't for one second believe CA when they say auto-resolve takes everything into account - something funky's going on here with settlements.

[JCoyote]

My only thing is, they said it takes into account "maneuver", but I don't think that means enfilade fire or things like that. It seems more like it takes into account maneuverability, and gives a bonus to attack or defense depending on what kind of battle it is.

I'm pretty sure autocalc uses a set of rules closer to a tabletop strategy game... terrain probably means an extra point in defense for the defenders, etc. I seriously doubt it actually simulates the armies actually moving around each other in any meaningful way.

And apparently having any siege equipment whatsoever eliminates the castle from consideration for the defenders. If you want to win, autocalc taking settlements, but defend settlements yourself.

I just had one exception though... FYI elephants are insanely overvalued in autocalc. Always best to fight them yourself.

[Ironside]

In my experience it doesn't take into account castles or cities! I've fought some battles through auto-resolve that have led to my winning with very low casualties (given walls with defender fire, defender wall bonuses, narrow street fighting, etc) and with the enemy having men left at the end (not possible as they should run to the centre and rally, then fight to the death)! These men subsequently disappear as I capture the city\castle.

I don't for one second believe CA when they say auto-resolve takes everything into account - something funky's going on here with settlements.

I suspect that settlements is back to the old MTW style. A force multiplier instead of a multiplier+defence structure by itself.

Playing a H/VH (correct order? VH on battles, H on campaign) campagin, I've seen some oddities in some cases.
Elite troops seems to be valued very highly (I lost a battle vs 2 foot knight units, whom I could win against without barely any casualities the turn after with weaker troops than the first time), pure cav armies are weak vs spears (3 feudal cav lost vs one unit of spear militia and 1 peasant archers).

But compared to RTW, the results are much better on average. Auto-calcing is viable again.

[Carl]

I seriously doubt it actually simulates the armies actually moving around each other in any meaningful way.

Whilst CA's statment isn't absolutly clear on this, it's strongly implied that it actually DOES fight the battle out. I'm also pretty sure it takes things like enfidle fire into account when it happens, but genrally the AI just charges straight aheacjh and never tries for enfidle, and it's A's controlling both sides so...

Lastly, it's been confirmed that Auto-Calc simply dosen't see Wall/Towers and as we know the AI has no choeasion issues in cities so the rest of the building related things go out the window.

[TevashSzat]

I believe midnight is right, i always autocalc with sieges since you always get suspusiously low casualties through auto calc compared to fighting it yourself

[JCoyote]

@ Carl: I really don't think it simulates them battling it out, I think it's more a tabletop or RPG sort of rough math. Because, when you look at the casualties afterward, there isn't rhyme or reason to them. Afterward it almost looks like someone rolled a dice about which units took casualties. They usually are almost evenly distributed across units. Artillery is notorious for taking small numbers of casualties. Why is that? When battles are fought out, artillery surviving is usually almost an all or nothing thing... not a quarter or half the unit. The way artillery takes losses has always made me think its a rough math on the stats that's used.

[General Zhukov]

I believe midnight is right, i always autocalc with sieges since you always get suspusiously low casualties through auto calc compared to fighting it yourself

I knew something odd was going on with those. Gonna fight 'em out and suck up the losses from now on.

[Jonathan_Thompson]

Auto battle is kind but its unpredictable and loves to throw curveballs like this

H/H campaign as venice. Sieging Florence, Milan's already attacked it and reduced garrison. I know its going to be a pain to hold so I send a large army.

845 men vs 89 men

15:1 odds

I lose 845 men and he loses 6:furious3:

Sometimes I wonder

[Skott]

My experiences is that you are more likely to win using the Auto Calculator than doing the fight yourself. Its basically the total opposite of RTW. This is in the vanilla game. Not sure if this changes using the various mods availble.