AGEOD Forums

I have yet to test these events, but am posting them here in case anyone spots anything wrong with them... They are AI events designed to change generals according to historic plausibility and action.

Union...



Here are the CSA events...

You left out Pope. Other than that, they look great!

Jabberwock wrote:You left out Pope. Other than that, they look great!

Pope commanded the Army of Virginia, the goal of the current events was to show the chain of command for the Army of the Potomac (and revise Patterson's dismissal). Other armies (out west and possibly the Army of Virginia out east) will be developed after I make sure that the eastern chain (used as a template) works 100%.

I plan for the AI to have army commands for the Army of the Mississippi/Army of Tennessee (based on national morale, timeframes, and random chance) for the CSA. I am also going to work on the Army of the Tennessee and Army of Ohio/Cumberland as well (along the same basis).

The McDowell-McClellan and Patterson events are experienced by both the player and AI (to simulate the change of McClellan and demotion of McDowell, the existing event really did nothing but promote McClellan, but kept McDowell in command). The rest are meant solely for the AI, so they will 'transfer command' from general to general.

I am thinking of a few other 'possible' command shifts, such as the AI being very successful, or very unsuccessful (such as getting Lee out early, if the CSA AI is getting thrashed in 1861), as well as some randomness (see the 1st Hooker and Burnside event, they both have a 50-50 chance of replacing McClellan, meaning each game should be random to a point as to who gets command when for the AI).

Is there a chance that this might harm the AI by suddenly yanking a general away from a stack that e.g. is about to go into battle?

Rafiki wrote:Is there a chance that this might harm the AI by suddenly yanking a general away from a stack that e.g. is about to go into battle?

I am trying to coincide most of the events to gameplay 'downtimes'. Most general switches occur during the late fall, or early winter timeframe (as did most of the 'calm transfers of command').

However, since the event triggers at the end of a turn (after all movement), and the appearance of the HQs (out East this is easy) is very close to the front line (Washington and Richmond), there should be no problem in re-establishing Corps command during the next turn, without any severe mixup in battleplan.

The major negativity could be the general removed from the corps to be used as the new army commander. For example, Johnston leaving his Corps to command the Army of Northern Virginia from Beauregard. Yet, I am hoping that the AI will have time to react and regroup (akin to the chance of a general dying in battle they would be in the same situation).

Turn runs, event fires, AI reorganizes forces and issues orders, next turn runs.

I believe weighting the returns is what is important here. 1 turn of possible disorganization in a unit per 'era' vs. experiencing different commanders. I don't think it will be an easy task to train the AI to switch command as players do (to drag an appropriate general to the correct stack, then determine they are the one who is to be the new commander, then switch them). Personally, I would like to see different commanders out East (the major problem, when the AI keeps McDowell and Beauregard for the rest of the war even though Lee and Meade are eager and able to take command).

Again, this is a mod idea, nothing intended to be pressed into the basic game, so it won't be forced down people's throats. If by chance it is accepted, I am sure that there could be a tag that could disable this option.

McNaughton wrote:Again, this is a mod idea, nothing intended to be pressed into the basic game, so it won't be forced down people's throats.

Never took it to be anything else

However, knowing how there are quite a few players who strive for the correct "historical feel" and who'd like to see historical commanders to a larger degree than what is usual in vanilla games, I find this to be a very interesting aspect that you are exploring, and once again have to commend you on the work you're doing

Have you been able to test this any yet? To which degree will the AI "reverse" what the events do, i.e. redistribute commanders according to it's own wishes?

Thanks for the encouragement. While this mod won't really affect the player that much (they don't get much out of it), the AI will appear to react to their situation (specific conditions will trigger a change of command, such as low National Morale). So, it will appear as if the AI is gradually getting better and better commanders in the field (as what truely happened), so hopefully will be more of a challenge as the game goes on.

The events (the ones I have tested) seem to work fine, with the generals appearing in their proper regions with the HQ attached, ready for action. The AI does not appear to undo the actions of these events, the armies tend to stay where they are deployed (i.e., deployed out East they stay out East).

A small NM boost might be appropriate for the first or second command changes.

These are really great. I think you are underestimating their appeal.

The Eastern events are 'done', I am just thinking about the Western events. I don't know how reliably they can be done, although I can try and implement the command where a unit will apper in the largest friendly stack in the region (is done with some of the events when new generals arrive).

Indeed, I am planning to implement the Eastern Army events to have the HQ appear directly in the largest friendly stack in the region (instead of in Washington and Richmond). I am also planning on creating the Army of Virginia in events (where you get a 50-50 chance it is Fremot or Pope, to add some variation as both were potentially up for command) for the AI.

In concert with this I have developed some AI specific events when generals arrive. Some generals (the Western generals, A.S. Johnston, Pope, Buell, Grant, Curtis, Van Dorn) appear attached to army commands. It will only happen for the AI, as humans still have to build their own army HQs.

Also, I am toying with some 'alternate history' events for both the Human and AI.

Take the West Virginia theatre in 1861. It made and postponed carreers. It made McClellan's career, and delayed Lee's (since Lee was officially in charge he experienced the fallout). Basically everything revolves around Grafton (it controls the rail hub, is a central supply of the state, etc.). If the Union wins, they get McCellan as Army Commander (which also spurns some other interesting events, such as getting extra corps commanders) replacing McDowell. If the Confederates win, it is a victory for Lee, which may trigger an early release for him (maybe a 30% chance of being released in 1861). You have to weigh where to put your resources, as maybe stopping McDowell is more important than securing Grafton (which is on the other side of a mountain range).

In my last game, I actually had the CSA AI beat the Union Force that was protecting Grafton and secured the city. So, in game terms, the CSA would have won and released Lee early (even though histoircally it didn't).

It is all a part of my plan to make events based more on player action, than simply just waiting for an event to trigger. By being 'aggressive' things will pay off (I plan a major revision of the Kentucky and Missouri events).

A NM Boost makes sense, since most events will trigger if NM is below 100 (basically, the situation is grim, replace the ineffective commander, then with a new leader spirits have raised).