the AI's too small stacks and force reconstruction

[veji1]

Playing the game as the french versus the AI, you know you are going to trash the AI if you have any experience with the AGEOD engine because you know how it works. This is not criticism of the game, it's just the way it is a good player will trash any AI. But I have 2 real worries that could dampen enjoyment of the game :

1/ Suboptimal AI stack size :
When you are fighting against Austria or Prussia in 1805 or 1806 you know they don't have corps and MTSG, but you expect them to at least regroup their troops so that their stacks are strong enough for real battles to occur, not just and endless wiping of lonely divs and brigades. in 1805 the Austrians are sort of ok, but that's more because you just catch them so early after the game begins that they haven't had time to start sending units traveling around stupidly. But for the Prussians it's just horrible because usually they dow you in mid 1806 while they have already dispersed their troops attacking Wurtemberg or whoever else. I mean 2 corps of french troops would probably suffice to win the war because you catch Blucher and Hohenloe and such with stacks of 500/600 power with your corps of 1000ish if they have moved a bit, 1400+ if you have been parked before. I don't expect the AI to be clever, but it should at least bunch its troops to the command limit of its leaders : Have Hohenloe moving slowly with a stack of 1500 combat value, you'll smash him nonetheless probably but at least you'll have a "proper battle" you can remember with the feel of a napoleonic battle where you will mechanically incur losses (maybe even a leader, wouldn't that be great !). Oh and of course the only units which should travel alone are cavalry set on the green battle stance to escape battle if engaged. Arty travelling around is just a shame.

2/ AI force reconstruction :
I don't know exactly how it works but I have big worries on this one : Say you trash Austria properly in 1805, smash its stacks and capture a bunch of cities, basically having the Armée d'Italie going all the way to Innsbruck/Graz/Karlsbad and the Grande Armée to Wien/Pressbur/Prag... By my estimate from last game, you have basically wiped 150 000 austrian soldiers from the board be it casualties or prisonners. How on earth is the AI suppose to rebuild its forces ? Can it only count on the slow organic unit building engine ? Is there some sort of mechanism ? To me there should be something a bit like when the first armies popped up in AACW/CW2 : you would get a message saying that your armies had been mustered and they would fill out with replacements. Because otherwise the AI is completely doomed it will never manage to : build enough units (mainly because the money will be the blocking issue) and secondly will never manage to organise them well enough. Each major faction outside of France (because if France gets trashed the game is over for them, it's restauration, etc...) should have some sort of "armies being reconstructed after the peace treaty" event firing which would see the AI been given a new STRUCTURED set of forces locked in a few strategic places and only partially manned, needing filling out with replacements. It should take a long time, say 20/24 turns so the units should come really close to empty, at about 20% strength and need a big investment to be filled ou, but this the AI should be able to do.

I say this because stupid Austria dowed Bavaria in mid 1806 and I had to wack their face and I only needed one corps in south germany to do the work, since the Austrian army was just a bunch of disjointed units.

This type of event, for AI only use or for player use too would also be a good way of implementing army reforms. AND it could help alleviate one of the AI's structural weaknesses since AACW : it does use the troops it has decently enough, giving you a good ride, but it does not know how to structure its forces well. Give it a well structured army (the 1862 union for example) and you'll have fun. Leave it to poor Athena to try and build an army and it will just not know what to do and you'll have disjointed units here and there and you'll be playing whack a mole (1861 GC in AACW and CW2 in the west and transmississipi for example).

[veji1]

to pursue on this idea, each of the 6 non french big factions (at least) should get an event that checks, after a peace treaty is signed with another major power, for the number of combat elements in its land forces. For each country there are thresholds such that if (i'll completely make up the numbers here) for example Austria starts the Augus 1805 campaign with say about 30 divs (moving troops plus all the garrisons, etc) equivalent of elements which would be around 12*30 so say 360 elements.

When the Pressburg treaty is signed the computer checks for how many austrian elements there are left on the map. In my last game I wiped about 150 000 austrians off the map so I'll say this is about 200 elements worth, so let's say that the engine finds that there are only 160 austrian elements "left" on the map (mainly garrison units here and there, lonley leaders and lonely arty stacks moving around....). It should give Austria around 200 20% full elements in organised troops (if not all in Vienna, in only a couple of cities say Vienna, Prag and Buda) which would be locked for 16 to 20 weks and slowly gaining strength through replacement shits.

I don't know how practical this would be to do, but my worse fear is that basically once you have trashed a big nation once, it's done for in the game, its back broken and never able to rebuild enough forces AND structured enought to give you a semblance of fun.

I know the devs are in holiday but it'd be nice to have some return from someone in the dev team or Beta team on those issues because they are central to the games long term replayability : you can fix all the Pressburg / Tilsit / Rhine confederation events in the world, if AI Russia/Austria/Prussia are never able to rise again from a 1805/1806 defeat, well that mucks up the game doesn't it ?

In general I am not favorable to an AI cheating as in getting help in battles, not needing supply, etc. but some events to help the AI compensate for its lack of foresight in the force organising department, for example, that'd be handy !

[Montbrun]

1/ If I recall, when I first installed with 1.00 Vanilla, I was impressed with the AI's ability to form coherent stacks, and there were few, if any, "tourists." Something happened with the later patches - "one step forward, two steps back."

2/ I like your idea about mustering new, structured, forces for the major countries after a defeat, although it should take longer than 24 weeks.

I don't mind the AI "cheating" if it leads to a better game for the player.

[veji1]

1/ If I recall, when I first installed with 1.00 Vanilla, I was impressed with the AI's ability to form coherent stacks, and there were few, if any, "tourists." Something happened with the later patches - "one step forward, two steps back."

2/ I like your idea about mustering new, structured, forces for the major countries after a defeat, although it should take longer than 24 weeks.

I don't mind the AI "cheating" if it leads to a better game for the player.

24 weeks would be some sort of compromise but all this could be adjusted, the key question being how does the game organise itself so that it can still be challenging and fun after one has wiped the other nations once ?

I had a similar wish for CW2 : that the AI be given some sort of routine during the winter month when it can "absolve" of transport rules and sort of teleport leaders and troops to behind the front hubs where coherent forces could have been built. I related in the CW2 forum how I would go in the winters of 61 and 62 play the Union as well as the CSA to help it structure its forces for the spring campaigns. I would help it manage its builds, funnel troops and leaders to key hubs (Cincinnati, Washington, St Louis, Cairo) and there build good strong armies with good corps leaders.. And when I went to the other side it made the following campaigning season that much enjoyable because if given proper well organised troops the AI proved actually competent enough to give the player some fun.

[Ace]

My main worry for the AI is its habit to dispers its forces. I wouldn't mind if it had one big properly organized stack of doom if it would bring challenging game. What was the Grande Armee marching to Russia than one big stack of doom. Developers shouldn't be afraid of programing AI to make force concetration. This is not ww1 where you need to form continuos front lines.

[veji1]

My main worry for the AI is its habit to dispers its forces. I wouldn't mind if it had one big properly organized stack of doom if it would bring challenging game. What was the Grande Armee marching to Russia than one big stack of doom. Developers shouldn't be afraid of programing AI to make force concetration. This is not ww1 where you need to form continuos front lines.

Agreed, pre reforms Austrian / Prussian / Russian armies should function with medium sized (ie armies) stacks of doom. After all the Prussian army (and its allies) of 1806 consisted of about 5 big stacks plodding along :
- Brunswick's army of 50/60 000
- Kalkreuth' reserve of 15 000
- Rüchel's corps of 15 000
- Hohenlohe's army of around 35/40 000.
- The Wurttemberg Prussian reserve of around 15 000.

Stacks smaller than that should just be garrison units or cavalry screens (with a blue/green setting)

This is a big worry for me, for the game to be fun the AI must (like it did pretty well in AACW) stay somewhat compact and not disperse its units, otherwise it becomes duck shooting.

The prussian army should look like that in the gam

[veji1]

Hello all, as the Devs are back it'd be nice to have an answer regarding the 2 main points in my original post, as they seem quite central to me in terms of replayability and lasting interest in a GC : you don't want countries like Austria or Prussia to never quite be able to recover from having their forces wiped out in a 1805/1806 french onslaught.

[Ace]

I commented with Pocus about it. It seems part of the AI stack concentration code was disabled because players were often complaining about AI stack of dooms (mostly in TEAW were stack of dooms feel out of place because of another era). It should be back.

[veji1]

I commented with Pocus about it. It seems part of the AI stack concentration code was disabled because players were often complaining about AI stack of dooms (mostly in TEAW were stack of dooms feel out of place because of another era). It should be back.

Good, because although sure in TEAW it doesn't make sense, in WON it is essential, the stack of doom's weakness was its slowness because of supply and congestion issues, not the fact that it did'nt cover a wide front.

EDIT : It'd be also nice of course to know what the devs think of the whole "force regeneration issue" : how can the AI have a viable army again in the game after having had its forces closed to wiped out in a preceding war.

[bjmorgan]

It'd be also nice of course to know what the devs think of the whole "force regeneration issue" : how can the AI have a viable army again in the game after having had its forces closed to wiped out in a preceding war.

I'd like to give you a second opinion. The problem is that the game AI won't accept surrender until too much has been lost. It seems that if the recovery rate of men were a little higher, armies could be re-built faster. As it is now, they won't give reasonable terms for surrender until they're totally smashed. (Maybe if they gave up sooner?) I also think that the mechanics of the replacement system is a bit faulty. An example: I pay for a unit, and I get stragglers as I march into Bavaria. Many of those stragglers return to duty, but I have to pay for those again in the form of replacement points or new units. Why can't some portion of those just return to their units once the blisters are healed on their feet? It's not paying a cost that gets me. It's paying the same cost as if these were untrained men with no equipment, that I have a problem with. I know that historically some stragglers never returned, that's true. But the current system already as a portion of them returning, and at the same cost as before. It makes no sense. But, I can live with it. The AI, however, may not be able to live with it.

But, your point on the small stacks is totally valid. I never buy artillery, for example. I just wait for the banzai batteries of my enemy to wander into my range and then let Murat do his thing. The same is true for supply units. I'm not complaining about the fact that I can capture artillery; that happened a lot in the period. It's the AI tactic of making single batteries and supply trains fast food for my troops that is the issue.

Now, about the doomsday stacks. It may not be possible to model very good in the system for the AI, but the strategy of the period, as you know, was to disperse troops (Corps, for the French) as they moved to contact, then concentrate them right before battle. That way, they could at least partially supply themselves off the land until they needed to face a large enemy formation. So, it's not the dispersal of troops that I have a problem with, per se, it's the fact that the smallest unit that's dispersed is a single unit. A battery, a brigade, a supply train. If it were a larger unit, e.g.,a division, or several brigades, etc., it would be much better and closer to reality. Napoleon's greatest operational innovation was the ability to rapidly concentrate and catch unsupported divisions and "corps-sized units," by themselves. That would cost the French quite few more casualties than just snatching a battery every other turn, which very often costs noting in casualties.

All that said, it's still a fun game. I wait for any improvements.

[veji1]

I'd like to give you a second opinion. The problem is that the game AI won't accept surrender until too much has been lost. It seems that if the recovery rate of men were a little higher, armies could be re-built faster. As it is now, they won't give reasonable terms for surrender until they're totally smashed. (Maybe if they gave up sooner?) I also think that the mechanics of the replacement system is a bit faulty. An example: I pay for a unit, and I get stragglers as I march into Bavaria. Many of those stragglers return to duty, but I have to pay for those again in the form of replacement points or new units. Why can't some portion of those just return to their units once the blisters are healed on their feet? It's not paying a cost that gets me. It's paying the same cost as if these were untrained men with no equipment, that I have a problem with. I know that historically some stragglers never returned, that's true. But the current system already as a portion of them returning, and at the same cost as before. It makes no sense. But, I can live with it. The AI, however, may not be able to live with it.

But, your point on the small stacks is totally valid. I never buy artillery, for example. I just wait for the banzai batteries of my enemy to wander into my range and then let Murat do his thing. The same is true for supply units. I'm not complaining about the fact that I can capture artillery; that happened a lot in the period. It's the AI tactic of making single batteries and supply trains fast food for my troops that is the issue.

Now, about the doomsday stacks. It may not be possible to model very good in the system for the AI, but the strategy of the period, as you know, was to disperse troops (Corps, for the French) as they moved to contact, then concentrate them right before battle. That way, they could at least partially supply themselves off the land until they needed to face a large enemy formation. So, it's not the dispersal of troops that I have a problem with, per se, it's the fact that the smallest unit that's dispersed is a single unit. A battery, a brigade, a supply train. If it were a larger unit, e.g.,a division, or several brigades, etc., it would be much better and closer to reality. Napoleon's greatest operational innovation was the ability to rapidly concentrate and catch unsupported divisions and "corps-sized units," by themselves. That would cost the French quite few more casualties than just snatching a battery every other turn, which very often costs noting in casualties.

All that said, it's still a fun game. I wait for any improvements.

The thing is, we as players are excessively "risk/reward" orientated : we think the AI should no when a situation is untenable and it should accept peace asap. To me It's ok for the AI to be stubborn, after all I want war and battles, not a series of skirmishes and peace. But I worry that an utterly trashed AI will never recover because of it's inability to 1/ Pay the price in money/Conscripts/WS to build back its forces and 2/ to structure and build these forces as a coherent army.

Regarding the concentration before battle, this was what made the french better for a few years : Hohenlohe or the duke of Brunswick did march all bunched up during the 1806 campaign. It was just that 45 000 men walking together still takes a good 15/20kms of road so they aren't walking like a meatball, rather a spaghetti, but it still remains just one big long and unwieldy spaghetti. and I am fine with that : make the 1806 slow plodders but with big stacks and at least we will have proper battles, not a series of mopping up operations.