I have played three games through to mid 1863 or later, two of the three starting with 1862, and two of the three as CSA. Settings normal aggressiveness (turned down for the winter months), modest FOW advantage.
In the USA game, the AI was interesting enough to be a good sparring partner, which is about all I ever ask of an AI. If I accidentally left anything undefended, the CSA AI jumped on it, in one case taking a depot I had only recently built. (It would do better to destroy those depots, since they are always intact when I reclaim them.) If I left an opening to my rear, the AI was right on it, occasionally to its regret but more often to mine. And, of course, the endless raiders!
In the CSA games, by contrast, the Union AI did virtually nothing. It had been so inert in the first game, that last time I decided to play entirely passively and just see what the AI would do. I build no ships or troops, except for militia, which I sprinkled around to various cities. I organized my forces and planted them in defensive locations -- Winchester, Manassas, and Fred in the east; Chat, Corinth, and Granada in the West, Ft Smith and Little Rock and Madison in Ark. At the end of 1863, I still held all of the above except for Winchester. None of the others had even been attacked. In fact, they left the towns in Western Tenn ungarrisoned, and my guys rebelled and took by Memphis, as well as others, to which I moved the corps that had been in Granada, and there it sat, unmolested, for the rest of the war.
The AI handles the very complex functional tasks quite well, as far as I can tell, and has a kind of operational canniness, but it doesn't want to fight the enemy armies. It didn't when I played the Union either, but there it didn't matter so much, since I was strung out all over the place, with plenty of vulnerabilities that it could take advantage of. I assume it would be equally adroit against an aggressive CSA player who invades the North.
The scoring also seemed odd. Without taking any significant Southern cities or even fighting, much less defeating, any rebel armies, the Union AI had a comfortable National Morale of about 110. A very easy-going public indeed!
I am wondering if this matches the experience of other players. Thankfully, Pocus is working away on these things and, hopefully, this feedback is helpful.
-
General Quarters
- Private
- Posts: 24
- Joined: Thu Jan 25, 2007 2:35 pm
- Location: Bucks County Pennsylvania
-
General Quarters
- Private
- Posts: 24
- Joined: Thu Jan 25, 2007 2:35 pm
- Location: Bucks County Pennsylvania
support@ageod.com , that's me.
Hofstadter's Law: "It always takes longer than you expect, even when you take into account Hofstadter's law."
-
Zoetermeer
- Sergeant
- Posts: 71
- Joined: Sun Jun 03, 2007 4:08 pm
- Location: Memphis, TN
I've had similar experiences with both sides. As the CSA, the Union AI can be really puzzling. In the east, it hardly ever does anything - usually it just shifts units back and forth between Washington and Harper's Ferry. In the west, usually it holds back also - except with Grant. In my recent game, Grant came directly south with one corps - so far that he was cut off from the rest of his forces. I surrounded him and wiped him out. Meanwhile, Pope, Buell, and the other notables in the west just hovered around their objectives but never did anything.
What seems odd, especially in the case of the Fort Henry battle, is that the AI never really tries to concentrate its forces against me in the west. I know I'm outnumbered there, but I always manage to engage forces smaller than mine.
It also doesn't really respond to me taking the initiative in some places - for example, I used the Virginia to clear the waters around Fortress Monroe, and then sent Huger to lay siege to it. The Virginia still sits there waiting for the Union to come after it, a month and a half later.
The only kind of neat thing the AI did was in the Trans-Mississippi, where Curtis' Union force, instead of trying to attack Van Dorn head-on, moved past him to the east and toward Ft. Smith, forcing me to leave my fortifications and give chase. Then he turned and fought when I showed up.
What seems odd, especially in the case of the Fort Henry battle, is that the AI never really tries to concentrate its forces against me in the west. I know I'm outnumbered there, but I always manage to engage forces smaller than mine.
It also doesn't really respond to me taking the initiative in some places - for example, I used the Virginia to clear the waters around Fortress Monroe, and then sent Huger to lay siege to it. The Virginia still sits there waiting for the Union to come after it, a month and a half later.
The only kind of neat thing the AI did was in the Trans-Mississippi, where Curtis' Union force, instead of trying to attack Van Dorn head-on, moved past him to the east and toward Ft. Smith, forcing me to leave my fortifications and give chase. Then he turned and fought when I showed up.
discussed this problem a time ago
http://www.ageod-forum.com/showthread.php?t=5099
it happens to me in every csa campaign (playing only April 61 campaign)
only in the trans mississippi area the union takes sometimes the offensive
in the east i´m building up my forces, and thats it
the only target is winchester, but it is easy to hold this town (with johnston´s forces)
the problem is, that i can´t promote generals, cause they are not fighting
http://www.ageod-forum.com/showthread.php?t=5099
it happens to me in every csa campaign (playing only April 61 campaign)
only in the trans mississippi area the union takes sometimes the offensive
in the east i´m building up my forces, and thats it
the only target is winchester, but it is easy to hold this town (with johnston´s forces)
the problem is, that i can´t promote generals, cause they are not fighting
-
Zoetermeer
- Sergeant
- Posts: 71
- Joined: Sun Jun 03, 2007 4:08 pm
- Location: Memphis, TN
I agree that this has been discussed before, time and time again. I would also agree that we can't expect the world of the AI, because it's not easy to create a good one. We expect the AI to act just like the Lincoln government would in real life, which is obviously unrealistic.
But I would venture to say that this AI has been way overhyped (if it has a name like "Athena", it's been hyped). It's been said that it wouldn't do anything we wouldn't do, but it doesn't do anything we would do either. I don't know how it manages to command the Union side, with far superior numbers and resources, and still make it so easy for us to win. Maybe it's been made too intelligent or something, because it sure acts like it realizes it's playing against a far superior opponent.
But I would venture to say that this AI has been way overhyped (if it has a name like "Athena", it's been hyped). It's been said that it wouldn't do anything we wouldn't do, but it doesn't do anything we would do either. I don't know how it manages to command the Union side, with far superior numbers and resources, and still make it so easy for us to win. Maybe it's been made too intelligent or something, because it sure acts like it realizes it's playing against a far superior opponent.
Zoetermeer wrote:I agree that this has been discussed before, time and time again. I would also agree that we can't expect the world of the AI, because it's not easy to create a good one. We expect the AI to act just like the Lincoln government would in real life, which is obviously unrealistic.
But I would venture to say that this AI has been way overhyped (if it has a name like "Athena", it's been hyped). It's been said that it wouldn't do anything we wouldn't do, but it doesn't do anything we would do either. I don't know how it manages to command the Union side, with far superior numbers and resources, and still make it so easy for us to win. Maybe it's been made too intelligent or something, because it sure acts like it realizes it's playing against a far superior opponent.
Before drawing general conclusions, it seems this problem is mainly concerning the USA side in the april 1861 scenario ( the most played). So maybe we should investigate some points: is AI able to cope with bad leaders offensively? I suspect AI is seeing McLelland and others as totally unable to manage an offensive and so she chooses to keep the defensive. I guess too AI is yet unable to change army leader ( Lee never being in 1862 replacing Beauregard to command the NVA) and so she keeps bad leader.
So it's maybe just 2 things to learn to the AI: attack with poor leaders and choose best leaders even at NM and VP costs...
Clovis wrote:Before drawing general conclusions, it seems this problem is mainly concerning the USA side in the april 1861 scenario ( the most played). So maybe we should investigate some points: is AI able to cope with bad leaders offensively? I suspect AI is seeing McLelland and others as totally unable to manage an offensive and so she chooses to keep the defensive. I guess too AI is yet unable to change army leader ( Lee never being in 1862 replacing Beauregard to command the NVA) and so she keeps bad leader.
So it's maybe just 2 things to learn to the AI: attack with poor leaders and choose best leaders even at NM and VP costs...
Maybe you should just have it attack somewhat historically.
The AI is never going to be brilliant. That's not possible with a game this complex. By nature you're going to have to rely on difficulty settings or voluntary player restrictions to make it difficult. So instead of having the AI contemplate the best way to attack in Late July 1861 (which is pointless because there is no good way to attack then)....just have it attack in Late July 1861.
But in regards to your specific points, if the problem is mostly in the 1861 campaign you should look at how it's forming it's armies, particularly in the West.
Also, just an interesting anecdote, the AI is building a lot of "Sailor" units, which seems entirely pointless.
LSSpam wrote:Maybe you should just have it attack somewhat historically.
The AI is never going to be brilliant. That's not possible with a game this complex. By nature you're going to have to rely on difficulty settings or voluntary player restrictions to make it difficult. So instead of having the AI contemplate the best way to attack in Late July 1861 (which is pointless because there is no good way to attack then)....just have it attack in Late July 1861.
But in regards to your specific points, if the problem is mostly in the 1861 campaign you should look at how it's forming it's armies, particularly in the West.
Also, just an interesting anecdote, the AI is building a lot of "Sailor" units, which seems entirely pointless.
I'm doing some experiences with the AI, rather surprising. I suppressed VP values for Winchester and Harper's Ferry, but CSA AI is yet interested to seize Harper's...this AI isn't totally driven by victory locations...
LOL.....not sure why my games are different. Palying CSA/ April / Norm/ low agressivness/ no fow.
USA is attacking all over the map. Both in east and west. Strange because it is winter.
Have noticed maybe two differences.
1) USA seems to attack more w/ no fow. Not sure, but maybe because it can see the disposition of my troops?
2) Have slight randomization of leaders. Could be because AI leaders possibly have better Strat. w/ randomization.
Not sure if these are the reasons. But It's Jan. and McDowel just attacked Me again with upwards of 80k+ troops in the region between Washington and Manassass. ( I have 60k in that region and 20k in region to the west)
Nial
USA is attacking all over the map. Both in east and west. Strange because it is winter.
Have noticed maybe two differences.
1) USA seems to attack more w/ no fow. Not sure, but maybe because it can see the disposition of my troops?
2) Have slight randomization of leaders. Could be because AI leaders possibly have better Strat. w/ randomization.
Not sure if these are the reasons. But It's Jan. and McDowel just attacked Me again with upwards of 80k+ troops in the region between Washington and Manassass. ( I have 60k in that region and 20k in region to the west)
Nial
-
Zoetermeer
- Sergeant
- Posts: 71
- Joined: Sun Jun 03, 2007 4:08 pm
- Location: Memphis, TN
I envy you Nial, because I have never seen anything like this. I think above somebody mentioned that the AI was being strange only in the April '61 scenario, but I get this in the 1862 scenario as well. I've never played the later-year scenarios, but by 1862 the Union AI should be attacking.
I've only played two times as the Union, but the AI there was just as puzzling. I marched directly down to Richmond, and the AI just threw small forces at me piecemeal instead of gathering everything into a large army.
I've only played two times as the Union, but the AI there was just as puzzling. I marched directly down to Richmond, and the AI just threw small forces at me piecemeal instead of gathering everything into a large army.
As CSA, AI isn't so dum in the April 61 scenario, even with my mod 
And I Haven't done any direct changes to the AI as it's impossible. I've just removed some victory locations ( Harper's Ferry, Winchester) and added others( Manassas, Grafton, Harrisonburg).
Game options are set to normal agressiveness, no FOW advantage.
The last turn, McDowell army trid to dislodge Beauregard's one from Manassas. It was a short defeat for the North but McDowell troops remained in Manassas, as Beauregard was entrenched in the town.
Harper's Ferry is possessed by CSA.
So I decided to retreat McDowell and build a force under Banks neighbouring Harper's Ferry.
First, Johston troops attacked Banks victoriously
The forces ratio was about equal.
So, even without VP attached to Harper's Ferry, CSA AI seems to be able to understand somehow the strategical value of this region and choosing attack rather than entrenching the region if a chance of success exists.
At Manassas, Beauregard's force attacked my retreating troops. No surprise here as the ration was strongly favouring CSA side, but yet interesting considering it was an entrenched force.
So I guess there's something wrong with bad generals for AI, which hinders offensive moves .
here's the situation at the end of this turn ( July 1861):
And I Haven't done any direct changes to the AI as it's impossible. I've just removed some victory locations ( Harper's Ferry, Winchester) and added others( Manassas, Grafton, Harrisonburg).
Game options are set to normal agressiveness, no FOW advantage.
The last turn, McDowell army trid to dislodge Beauregard's one from Manassas. It was a short defeat for the North but McDowell troops remained in Manassas, as Beauregard was entrenched in the town.
Harper's Ferry is possessed by CSA.
So I decided to retreat McDowell and build a force under Banks neighbouring Harper's Ferry.
First, Johston troops attacked Banks victoriously
The forces ratio was about equal.
So, even without VP attached to Harper's Ferry, CSA AI seems to be able to understand somehow the strategical value of this region and choosing attack rather than entrenching the region if a chance of success exists.
At Manassas, Beauregard's force attacked my retreating troops. No surprise here as the ration was strongly favouring CSA side, but yet interesting considering it was an entrenched force.
So I guess there's something wrong with bad generals for AI, which hinders offensive moves .
here's the situation at the end of this turn ( July 1861):
The next turn, Beauregard started an offensive toward Alexandria
Now, the same intial turn but with USA AI:
McDowell is bypassing Manassas where Beauregard is entrenched and tries to seize Fredericksburg
The battle isn't very bloody but is the result of Mcdowell attack stance:
Mcdowell was active this turn...and due to my mod, it's a highly dizorganized army...
Now, the same intial turn but with USA AI:
McDowell is bypassing Manassas where Beauregard is entrenched and tries to seize Fredericksburg
The battle isn't very bloody but is the result of Mcdowell attack stance:
Mcdowell was active this turn...and due to my mod, it's a highly dizorganized army...
The AI does tend to be a little gun shy, but I've been surprised a few times by its agressiveness.
Playing as the Union, I was besieging Richmond in spring 1863 and Lee rolled up out of Petersburg with an enormous army and essentially wiped out two of my Corps. I was ticked and delighted at the same time because it got really interesting at that point!
Playing as the Union, I was besieging Richmond in spring 1863 and Lee rolled up out of Petersburg with an enormous army and essentially wiped out two of my Corps. I was ticked and delighted at the same time because it got really interesting at that point!
Well, I don't know what it is but my recent game as the CSA (July Campaign, 1861) I set the AI to low agressiveness and small fow bonus. Turned to be my BEST GAME ever. AI made the right moves and its good generals may be the answer.
1. Hooker. The AI created ONE stack with Hooker in command with 3 DIVISIONS, each lead by a * 3-1-1 general(!), fully supplied and with balloons. Strength of about 1200, maybe 5% penalty. On turn one Hooker attacked Fredericksburg, took it on turn two, LEFT a force there under Hunter, and took Manassas on turn 3! BTW, I didn't even manage to destroy the depots in those cities!
2. Butler. Butler attacks in each game from Ft Monroe but it was the FIRST TIME his attack made sense. He was clearly making for Fredericksburg to join with Hunter there! He HAD a chance to establish a corridor from Hampton Roads to Fredericksburg which would supply Union force by the i.e Hampton Roads harbor.
3. Lyon. He came out of nowhere and sieged Columbus, KY with his division. I HAVE NEVER seen such a move from Union AI so far!
4. What did I do? I moved to attack Fredericksburg just to FIND Hunter has slipped out and now is moving towards Charlotsville! Oh, and Hooker with his 1000 strength stack (FULLY cohesive) went back to Washington through Alexandria in ASSAULT posture! The big battle that ensued was a disaster for me and I had to REPLAY it! He mowed my Jackson and Beauregard stacks, losing his 6500 men and killing my 9500 men in the process. Not to mention 14of my units he destroyed (he had LOTS of artillery)! Jackson' stack was down from 350 strength to 10. Basically Hooker assualted so fast that Beauregard couldn't reinforce Jackson!
AND IT IS ONLY TURN 6 !
Now, I DON'T KNOW what happened but this low aggresiveness/small FOW bonus/competenet Union generals in charge (Hooker, Lyon,even Butler) produced the most aggresive and cunning opening I HAVE EVER SEEN!
1. Hooker. The AI created ONE stack with Hooker in command with 3 DIVISIONS, each lead by a * 3-1-1 general(!), fully supplied and with balloons. Strength of about 1200, maybe 5% penalty. On turn one Hooker attacked Fredericksburg, took it on turn two, LEFT a force there under Hunter, and took Manassas on turn 3! BTW, I didn't even manage to destroy the depots in those cities!
2. Butler. Butler attacks in each game from Ft Monroe but it was the FIRST TIME his attack made sense. He was clearly making for Fredericksburg to join with Hunter there! He HAD a chance to establish a corridor from Hampton Roads to Fredericksburg which would supply Union force by the i.e Hampton Roads harbor.
3. Lyon. He came out of nowhere and sieged Columbus, KY with his division. I HAVE NEVER seen such a move from Union AI so far!
4. What did I do? I moved to attack Fredericksburg just to FIND Hunter has slipped out and now is moving towards Charlotsville! Oh, and Hooker with his 1000 strength stack (FULLY cohesive) went back to Washington through Alexandria in ASSAULT posture! The big battle that ensued was a disaster for me and I had to REPLAY it! He mowed my Jackson and Beauregard stacks, losing his 6500 men and killing my 9500 men in the process. Not to mention 14of my units he destroyed (he had LOTS of artillery)! Jackson' stack was down from 350 strength to 10. Basically Hooker assualted so fast that Beauregard couldn't reinforce Jackson!
AND IT IS ONLY TURN 6 !
Now, I DON'T KNOW what happened but this low aggresiveness/small FOW bonus/competenet Union generals in charge (Hooker, Lyon,even Butler) produced the most aggresive and cunning opening I HAVE EVER SEEN!
I've tried this in BoA
Jacek wrote:Now, I DON'T KNOW what happened but this low aggresiveness/small FOW bonus/competenet Union generals in charge (Hooker, Lyon,even Butler) produced the most aggresive and cunning opening I HAVE EVER SEEN!
Jacek,
I've tried the same settings with BoA and found that indeed the AI seemed to play a little smarter, ie less aggressive all over but more concentrated on taking important objectives. Can't wait to find AACW in a store near me.
Omnius
Omnius wrote:Jacek,
I've tried the same settings with BoA and found that indeed the AI seemed to play a little smarter, ie less aggressive all over but more concentrated on taking important objectives. Can't wait to find AACW in a store near me.
Omnius
Agreed. I have just fought off the third attack towards Mannassas. It is Oct 62 and McDowel came at me again with a huge stack, over 100k. It was a very close thing. But we held. Barely.
At the same time. I am sieging St. Louis. The AI is sallying forth every turn. Various generals commanding. Dang the AI likes to stack alot of federals in St. Lo. I may have to lift the siege and back up to regroup. Johnston is taking a beating. McCullough's cav div. was wiped out. Healing at nearby Rollo.
This is all on the lowest settiongs. And with a small leader randomization. McDowel is a 3-3-3. He has been on the attack consistently, every few months. ( You'l notice his elevated strat score)
Nial
-
McNaughton
- Posts: 2766
- Joined: Wed Mar 21, 2007 8:47 pm
- Location: Toronto, Canada
There are more priorities for the AI based on geography than victory points. I believe that Harper's Ferry and St. Louis are not only VP cities, but also supply and ammunition depots. The AI most likely prioritieses these supply centres as places to take and defend (which was why, when Harper's Ferry and Winchester had VP they were such a massive target by the AI).
I believe that the AI can be 'trained' to fight more logically based on tweaking priorities for targets (something from AGEOD's POV), or even modifying VP locations based on things like transportation hubs, passes through mountains, etc., to have the AI move along logical and historic routes to greater objectives (something that Modders can do).
Also, you could test to see what the AI does if units are given more VP when destroyed. Each unit has a VP allowance given to the opposing side if the unit is destroyed. Some units have higher VP than others.
This could do three things, make the AI more aggressive to kill more enemy units. Make the AI more passive in order to protect their own units. Or do absolutely nothing!
I believe that the AI can be 'trained' to fight more logically based on tweaking priorities for targets (something from AGEOD's POV), or even modifying VP locations based on things like transportation hubs, passes through mountains, etc., to have the AI move along logical and historic routes to greater objectives (something that Modders can do).
Also, you could test to see what the AI does if units are given more VP when destroyed. Each unit has a VP allowance given to the opposing side if the unit is destroyed. Some units have higher VP than others.
This could do three things, make the AI more aggressive to kill more enemy units. Make the AI more passive in order to protect their own units. Or do absolutely nothing!
-
Zoetermeer
- Sergeant
- Posts: 71
- Joined: Sun Jun 03, 2007 4:08 pm
- Location: Memphis, TN
Here is what I tried recently in 1.05a game as the CSA, July 1861 Campaign
Normal difficulty
Low aggressiveness
Small FOW bonus
It was a good challenege till the end of November when immobilised Union stacks started suffering from bad weather. Twice I witnessed McDowell sitting with his big army in mountains north of Harper's Ferry, enduring blizzard for THREE turns. Do I need to mention that all local railroads were cut by my cavalry? That was no fun - I took Washington by Early January 1862. So I spiced things a liitle bit and changed the settings a bit.
Normal difficulty
Low agrressiveness
Small FOW bonus
No Activation checked
Now Union stacks MOVE!
1. East: Union formed corps under Banks and Butler, army stack under McDowell. Butler winds up in Alexandria going through Fredericksburg for rendez-vous with McDowell. Mansfield moves to Ft Monroe! Strong divisions (400 points) are formed under Hooker and Hunter. Burnisde gets his stack too. For the first time in many games Johnston's corps leaves Harper's Ferry without fight because Bank's Corps is considerably stronger. Yet McDowell still sucks at army buliding and moves his behemoth stack with penalties of 20-35% so when he is trapped and surrounded in Stafford,VA the battles there generate some bizarre loses. His corps though are much better at manouvering or retreating.
2.West: McClellan moved to and held at Evansville, though Union lost Vincennes. Fremont formed a division which sits at Paducah and threatens Hardee's stack south of it. East Kentucky is no brainer: Union invades and takes Lexington and Clarksburg; Prestonburg holds for a few more turns then falls. Strangely, in all of my games Union fails to reinforce Bowling Greene and never tries to take possesion of the railroad from Louisville to Bowling Greene.
3.Cavalry raids. Union raided Kentucky, Tennessee, South Carolina.
My general impression is that these settings free MANY Union units which would otherwise sit in cities with their inactive leaders (McClellan, McClernand, Banks, McDowell etc.). Many units are now in Kentucky, considerable numbers of them in Missouri. East has become a manouver war to some extent each side having 1 army stack + 2 corps stacks.
Yet there is a downside of this setting: McDowell. Every time he will take his big army to Fredericksburg and get surrounded in the process in Stafford,VA.
Since it is hostile territory he suffers additional combat penalties. Loopsided battles follow and voila - 25,000 Union casualties!
Probably this can be offseted by yet another change in the settings:
Normal difficulty > Up difficulty one level
Low agrressiveness
Small FOW bonus
No Activation checked
This may provide for some interesting gameplay...
Normal difficulty
Low aggressiveness
Small FOW bonus
It was a good challenege till the end of November when immobilised Union stacks started suffering from bad weather. Twice I witnessed McDowell sitting with his big army in mountains north of Harper's Ferry, enduring blizzard for THREE turns. Do I need to mention that all local railroads were cut by my cavalry? That was no fun - I took Washington by Early January 1862. So I spiced things a liitle bit and changed the settings a bit.
Normal difficulty
Low agrressiveness
Small FOW bonus
No Activation checked
Now Union stacks MOVE!
1. East: Union formed corps under Banks and Butler, army stack under McDowell. Butler winds up in Alexandria going through Fredericksburg for rendez-vous with McDowell. Mansfield moves to Ft Monroe! Strong divisions (400 points) are formed under Hooker and Hunter. Burnisde gets his stack too. For the first time in many games Johnston's corps leaves Harper's Ferry without fight because Bank's Corps is considerably stronger. Yet McDowell still sucks at army buliding and moves his behemoth stack with penalties of 20-35% so when he is trapped and surrounded in Stafford,VA the battles there generate some bizarre loses. His corps though are much better at manouvering or retreating.
2.West: McClellan moved to and held at Evansville, though Union lost Vincennes. Fremont formed a division which sits at Paducah and threatens Hardee's stack south of it. East Kentucky is no brainer: Union invades and takes Lexington and Clarksburg; Prestonburg holds for a few more turns then falls. Strangely, in all of my games Union fails to reinforce Bowling Greene and never tries to take possesion of the railroad from Louisville to Bowling Greene.
3.Cavalry raids. Union raided Kentucky, Tennessee, South Carolina.
My general impression is that these settings free MANY Union units which would otherwise sit in cities with their inactive leaders (McClellan, McClernand, Banks, McDowell etc.). Many units are now in Kentucky, considerable numbers of them in Missouri. East has become a manouver war to some extent each side having 1 army stack + 2 corps stacks.
Yet there is a downside of this setting: McDowell. Every time he will take his big army to Fredericksburg and get surrounded in the process in Stafford,VA.
Since it is hostile territory he suffers additional combat penalties. Loopsided battles follow and voila - 25,000 Union casualties!
Probably this can be offseted by yet another change in the settings:
Normal difficulty > Up difficulty one level
Low agrressiveness
Small FOW bonus
No Activation checked
This may provide for some interesting gameplay...
To me, it sounds like the AI armies get surrounded and decimated pretty easily. I think this needs some addressing or the AI loses validity in a large measure. Not meant to sound nasty there, but it is a concern.
My top concerns for the AI, in no particular order:
My top concerns for the AI, in no particular order:
- Encirclement, leading to disaster - no maneuvering in response to enemy
- Inactivity (Union especially)
- Activity during harsh winter
-
McNaughton
- Posts: 2766
- Joined: Wed Mar 21, 2007 8:47 pm
- Location: Toronto, Canada
I think that general's inactivity has something to do with the US difficulty in attacking. Maybe there could be an option about the activation rule to have it active or inactive for a particular side? I personally like to play with it on, but if it hurts the AI, then I don't want it on for them (it probably affects the CSA less than the USA, because the average CSA general has higher strategic ratings).
In regards to getting surrounded, possibly tweaking the VP rates, making a 'logical' path of advancement (i.e., VPs along critical junctions, supply routes, etc.) to encourage the AI to move on a broad front, rather than at specific objectives (which gets them surrounded).
Have Manassas and Fredricksburg both as VP cities (Clovis' mod does this), but possibly due to Manassas' supreme importance as a transportation hub for both North and South, maybe have it as a 2 VP city? I think that this may help the CS and US AI, as Manassas is the key to the East. If you are moving against Washington or Richmond, you need Manassas' supply route. Also, even some empty rail lines might benefit by getting VPs, symbolizing their importance as supply routes, as well as getting the AI to secure them before moving too far forward.
Suggestions.
#1. Have separate toggles for USA and CSA Activization in the setup screen (so the AI can recieve no activation, yet players can still use this feature for their side).
#2. See what happens if there is a revamp in VP (to expand on Clovis' tests before asking for it to be included in an official patch). Transportation hubs and important rail lines should have VP values to encourage the AI to take and secure them, as well as to show their importance to the military cause (even some non-urbanized centres could use some VPs).
In regards to getting surrounded, possibly tweaking the VP rates, making a 'logical' path of advancement (i.e., VPs along critical junctions, supply routes, etc.) to encourage the AI to move on a broad front, rather than at specific objectives (which gets them surrounded).
Have Manassas and Fredricksburg both as VP cities (Clovis' mod does this), but possibly due to Manassas' supreme importance as a transportation hub for both North and South, maybe have it as a 2 VP city? I think that this may help the CS and US AI, as Manassas is the key to the East. If you are moving against Washington or Richmond, you need Manassas' supply route. Also, even some empty rail lines might benefit by getting VPs, symbolizing their importance as supply routes, as well as getting the AI to secure them before moving too far forward.
Suggestions.
#1. Have separate toggles for USA and CSA Activization in the setup screen (so the AI can recieve no activation, yet players can still use this feature for their side).
#2. See what happens if there is a revamp in VP (to expand on Clovis' tests before asking for it to be included in an official patch). Transportation hubs and important rail lines should have VP values to encourage the AI to take and secure them, as well as to show their importance to the military cause (even some non-urbanized centres could use some VPs).
McNaughton wrote:Suggestions
#1. Have separate toggles for USA and CSA Activization in the setup screen (so the AI can recieve no activation, yet players can still use this feature for their side).
I'd like to make an enhancement here, as I do think Options toggle is the way to address this:
Give the AI an additional toggle of [color="Blue"][SIZE="3"]BONUS[/size][/color], which gives the AI a -1 modifier to all activation die rolls. This way a player can keep some of the activation concept rolling for both sides. The AI has a better chance to activate, but it isn't automatic and units parading around willy nilly from the get-go. It keeps the historic angle in play a bit longer than just free activation.
-
McNaughton
- Posts: 2766
- Joined: Wed Mar 21, 2007 8:47 pm
- Location: Toronto, Canada
PBBoeye wrote:I'd like to make an enhancement here, as I do think Options toggle is the way to address this:
Give the AI an additional toggle of [color="Blue"][SIZE="3"]BONUS[/size][/color], which gives the AI a -1 modifier to all activation die rolls. This way a player can keep some of the activation concept rolling for both sides. The AI has a better chance to activate, but it isn't automatic and units parading around willy nilly from the get-go. It keeps the historic angle in play a bit longer than just free activation.
Maybe it could be a toggle that ranges (separate for the CSA and USA) the level of activation.
Full Activation, -1 Activation, No Activation
Just so players can fully customize.
Return to “AGEod's American Civil War”
Who is online
Users browsing this forum: No registered users and 10 guests