A Detailed Suggestion for an "Army Goals" system that avoids micro

  • We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Yaitz331

Second Lieutenant
32 Badges
May 3, 2018
170
408
  • Crusader Kings II: Way of Life
  • Victoria 3 Sign Up
  • Crusader Kings III
  • Imperator: Rome Sign Up
  • Crusader Kings II: Holy Fury
  • Crusader Kings II: Jade Dragon
  • Europa Universalis IV: Third Rome
  • Crusader Kings II: Monks and Mystics
  • Crusader Kings II: Reapers Due
  • Hearts of Iron IV: Colonel
  • Hearts of Iron IV: Cadet
  • Crusader Kings II: Conclave
  • Europa Universalis IV: Cossacks
  • Crusader Kings II: Horse Lords
  • Europa Universalis IV: Common Sense
  • Europa Universalis IV: El Dorado
  • Europa Universalis IV
  • Crusader Kings II: Charlemagne
  • Crusader Kings II: Legacy of Rome
  • Crusader Kings II: The Old Gods
  • Crusader Kings II: Rajas of India
  • Crusader Kings II: The Republic
  • Crusader Kings II: Sons of Abraham
  • Crusader Kings II: Sunset Invasion
  • Crusader Kings II: Sword of Islam
  • Europa Universalis IV: Art of War
  • Europa Universalis IV: Mare Nostrum
  • Europa Universalis IV: Conquest of Paradise
  • Europa Universalis IV: Wealth of Nations
  • Europa Universalis IV: Call to arms event
  • Europa Universalis IV: Res Publica
  • Crusader Kings II
So, the idea of giving an army a goal to work towards.

Dev response on the Diary: “What we're considering adding is a method of prioritizing the various targets in the war, and setting custom targets, on a national (not Front- or General-) level. What we need to be careful with here is to not add methods of control that make the player technically able to control with precision how Generals act in every moment by microing their priorities.”

I think many people are unsatisfied with this, and want a way to give an army a specific goal. I agree wholeheartedly with this. Let’s look at a historical example; the February 1862-July 1863 campaign for control of the Mississippi River.

The Mississippi River was vital to both the Union and Confederacy, for primarily economic reasons. For the Confederacy, loss of the Mississippi meant the loss of the vast supplies of the Trans-Mississippi, with the loss of the vast farmland of Texas felt particularly hard. For the Union, as long as the Mississippi was in Confederate hands, the entire Midwestern economy ground to a halt; without the river route to the sea, goods produced in the Midwest could only be exported across land, which was far less efficient. I think both of these are things Vic3 would have to represent, beyond economic and logistic in nature.

Now, let’s look at the system as it is. Ulysses S. Grant’s army is put on “Advance” orders, and starts pushing into Confederate territory. With no way to tell him to take the Mississippi, he just advances blindly; the Midwest suffers, and the Confederates have their supply lines. This is not only untenable for gameplay, it’s also unrealistic; everyone knew the Mississippi was the Union’s goal because its value was obvious. Given that I doubt the AI will be robust enough to “know” the strategic value of the Mississippi, the player needs a way to push the AI to do that.

So, we need some way to give goals to individual armies. But how can we avoid that turning into somewhat-hand-off microing?

Here’s my suggestion.

Scattered across the map, there are “Strategic Targets”. These can be of two types; a province, like a national capital or major city, or a line, like a large river or important railroad. On both Advance and Defend orders, the player has the option to select one of these Strategic Targets to either focus the attack on or focus the defense on. That army will then work towards achieving that goal. If the player does not select a particular goal, the AI will instead generally work towards those Strategic Targets. Strategic Targets give a slight boost to warscore, but their impact is primarily on affecting the enemy’s ability to continue fighting rather then directly on warscore.

So how do we avoid manipulating the system to allow micro? My suggestion comes in two parts.

1. The player has no control over how the AI goes about achieving this goal. An Aggressive general might launch an all-out offensive directly at his target, a Maneuvering general might attempt to get around his enemy, a Cautious general might just very slowly push towards the target. Maybe the general will even ignore orders; if the general’s home region is occupied by the enemy, he might forgo his goal in favor of trying to liberate his home. The player can only choose a goal, and that only from a Strategic Target; no “Alright, move to this town and then to that one” or “Take this city by advancing this way”, only “Take this major city” or “Secure this railroad”.

2. There is a difference between a Target chosen to defend and a Target chosen to attack. A Target chosen to defend can be changed at will based on where the enemy might be threatening; however, to change an Offensive Target is an admission of defeat. Once an Offensive Target is chosen for an army, to change it is to admit “We have failed to achieve our goal.” Naturally, this will severely influence public opinion to the negative; war exhaustion/displeasure with the government will rise. If a small army is redirected, maybe few will notice, but if a large army changes its Offensive Target, that will incur a very severe penalty; admitting your large force failed is a heavy blow to public morale. Even if an army with an Offensive Target is changed to a Defensive stance, if switched back to Advance, it will keep the same Target. Only once an Offensive Target is achieved will choosing a new one be capable.

Let’s look at two historical examples to see how this system might function.

Mississippi Campaign: An army under Ulysses S. Grant is given orders to secure the Mississippi River. This entails making sure no province bordering the river is under Confederate control. Grant prefers Maneuver to outright Attack; facing Confederate strongholds at first Columbus, then Memphis, and finally Vicksburg, Grant moves to cut off those strongholds, forcing the Confederates to either retreat or be besieged. Meanwhile, Confederate forces in the area have orders to defend the Mississippi. At Columbus and Memphis, the Confederates choose to evacuate rather than be besieged; at Vicksburg, they choose to hold. Grant wins the subsequent siege and take Vicksburg. With a second army in Louisiana pushing north from New Orleans with the same goal taking Fort Hudson, the Offensive Target has been achieved; Grant can now be redirected to join Rosecrans for the Battle of Chattanooga.

Eastern Theater: At the start of the war, the main Union army selects as its Offensive Target the Confederate capital of Richmond. As the war goes on, various attempts to take Richmond have failed, though Confederate counterattacks have also been halted. However, despite the immense difficulty of taking Richmond, changing the Offensive Target of the largest army in the field would devastate national morale, so the Union army continues to try for “On to Richmond!” Even as bloodshed mounts, the Union is unwilling to take the penalty for admitting failure, and so the bloody push to Richmond continues to the bitter end.

Now what would happen if someone tried to use this system to micro? They decide they want their army to move northeast and then double back southeast, so they choose an Offensive Target to the northeast. Once their army has reached the point where they want it to move southeast, they change the Strategic Target. This then incurs the heavy penalty for changing an Offensive Target, and public frustration rises to the point where the war effort is severely hampered. The attempt to abuse the system to micro has failed utterly.

I think that this system would achieve both of the critical things any system of Army Goals need to; it gives the player the ability to strategically direct a war, but still prevents microing. With a list of Strategic Targets automatically chosen by the game (say, select rivers, major railroads, national capitals, and cities past some threshold of size), this also keeps player control firmly at the strategic level, while giving them plenty of control at said level. It also adds only a tiny amount of complexity, keeping it easy for the player; if you have a large number of small armies, you can just not give them specific Goals.

So, what do you think? Please do comment with any thoughts.
 
Last edited:
  • 91Like
  • 16
  • 11Love
  • 4
  • 3
Reactions:
You've clearly put a lot of thought into this, I really wish the devs will see this and consider your suggestions. That being said, I think that instead of setting targets at the province level, it would be more suitable to set targets at the state level. Such as "Push towards Alabama" or Defend West Virginia" . Precisely which provinces are attacked or occupied depends on the General's traits. Instead of suffering a debuff to war exhaustion and morale every time you change a target, it would make more sense to simply cost time. Let's say that it takes 3 months to change one target to another, and once the target has been set, you can't change it again within the next three months. Of course this could be modified according to general traits and army composition.
 
  • 22
  • 6Like
  • 3
Reactions:
I really like this idea. I've been all for the crackpot theory this whole time, but I was hoping for at least a little bit more control than what was showcased. I think this looks like a great compromise.
 
  • 8Like
  • 2
Reactions:
Very good suggestions, I very much like the idea that giving up a target has negative consequences and therefore discourages exessive micro.

I'd like to add that while generals will and should obviously go at their own pace advancing towards your targets, it might be interesting to have the option to order your generals to act more quickly (because you think timining is important, i.e. one participant of the other side hasn't mobilized yet). The downside of that would obviously be that the officers and troops aren't prepared, so they'll do much worse than if you let them take their time.

How do you envision defensive targets to work? Influencing the troop concentration along the front or something like that?
 
  • 4Like
Reactions:
How do you envision defensive targets to work? Influencing the troop concentration along the front or something like that?
Not OP, but I'd say that this could tie in to the general traits. Taking the example above, the CSA player assigns General Pemberton and his army to defend the Mississippi River. The AI then takes Pemberton and his battalions and "deploys" them to the theater.

If Pem has a cautious trait, the battalions would probably stay concentrated, keep near to defensible terrain, and be more willing to retreat.

But if he has an aggressive trait, the AI may be more willing to engage in pitched battles, and may even launch counter-invasions, even though he would be on a "defend" order.
 
  • 6Like
  • 5
Reactions:
You've clearly put a lot of thought into this, I really wish the devs will see this and consider your suggestions. That being said, I think that instead of setting targets at the province level, it would be more suitable to set targets at the state level. Such as "Push towards Alabama" or Defend West Virginia" . Precisely which provinces are attacked or occupied depends on the General's traits. Instead of suffering a debuff to war exhaustion and morale every time you change a target, it would make more sense to simply cost time. Let's say that it takes 3 months to change one target to another, and once the target has been set, you can't change it again within the next three months. Of course this could be modified according to general traits and army composition.
There are two problems with state-level targets; firstly, it still fails to model things like pushes down rivers along along railroads, and secondly, it would break the "You can't change targets without penalties until your target has been achieved" system, because you would need to occupy the entire state in order to change your target. I also disagree with it only taking time to change targets, because that still allows for manipulation of the system to allow finer control then you should be allowed.

Very good suggestions, I very much like the idea that giving up a target has negative consequences and therefore discourages exessive micro.

I'd like to add that while generals will and should obviously go at their own pace advancing towards your targets, it might be interesting to have the option to order your generals to act more quickly (because you think timining is important, i.e. one participant of the other side hasn't mobilized yet). The downside of that would obviously be that the officers and troops aren't prepared, so they'll do much worse than if you let them take their time.
Would that not essentially be an "Act more aggressively" order? If you want your general to rush as fast as he can, appoint an extremely Aggressive general.

How do you envision defensive targets to work? Influencing the troop concentration along the front or something like that?
Not OP, but I'd say that this could tie in to the general traits. Taking the example above, the CSA player assigns General Pemberton and his army to defend the Mississippi River. The AI then takes Pemberton and his battalions and "deploys" them to the theater.

If Pem has a cautious trait, the battalions would probably stay concentrated, keep near to defensible terrain, and be more willing to retreat.

But if he has an aggressive trait, the AI may be more willing to engage in pitched battles, and may even launch counter-invasions, even though he would be on a "defend" order.
Basically, yes. If the general is Cautious, the Defensive Target might only influence troop concentration along the front. If the general is moderately Aggressive, maybe he would attack an enemy army threatening his Defensive Target (despite being on Defensive), but wouldn't attack an army threatening elsewhere along the front - he needs to keep the enemy away from his DT.

I think this is an excellent idea that really needs to be implemented in the game before release.
Excellent Idea!
I really like this idea. I've been all for the crackpot theory this whole time, but I was hoping for at least a little bit more control than what was showcased. I think this looks like a great compromise.
I really like this idea, and hope that they end up implementing something like it.
Thank you very much! Here's to hoping the devs will notice this idea!
 
  • 5Like
  • 1
Reactions:
I have posted in a few places that I think the minimum military system would have the following:
1) The ability to position generals on the front (as opposed to assigning a general to the entire front).
2) The ability to give more complex orders than advance/defend (which generals can then ignore - like McClellan repeatedly ignored Lincoln).
3) The ability to set strategic targets.

I am not picky about how these are implemented, so your system could do for #3. However, I still think that we need #1 and #2. Otherwise, you assign Grant to the thousand plus mile front, then tell Grant to take the Mississippi. Grant appears in Maryland and starts his march across the Confederacy to take Mississippi. Without the ability to position generals, strategic targets are insufficient for a workable system.
 
  • 16
  • 4Like
Reactions:
I have posted in a few places that I think the minimum military system would have the following:
1) The ability to position generals on the front (as opposed to assigning a general to the entire front).
2) The ability to give more complex orders than advance/defend (which generals can then ignore - like McClellan repeatedly ignored Lincoln).
3) The ability to set strategic targets.

I am not picky about how these are implemented, so your system could do for #3. However, I still think that we need #1 and #2. Otherwise, you assign Grant to the thousand plus mile front, then tell Grant to take the Mississippi. Grant appears in Maryland and starts his march across the Confederacy to take Mississippi. Without the ability to position generals, strategic targets are insufficient for a workable system.
Agree with this and with OP. If we're going to have essentially fixed fronts, then I want at least SOME way to dictate how my forces are spread over that front. And it goes without saying that I should be able to give armies broad strategic targets such as the capture of a port, railroad, or major city.

The concern that these sorts of features would allow "micro by the backdoor" would seem to be able to be easily solved just put putting a sufficient cost or cooldown on such actions. I'm a huge proponent of abstracted warfare, but the system shown in the most recent DD really does seem to make the outcome of warfare entirely down to your economy and military. If the player is going to be kept at such great distance from how wars are prosecuted, then it might as well be entirely automated - but that would seem to violate the objective of warfare having a strategic element which the player is involved in.
 
  • 9
  • 3Like
Reactions:
I have posted in a few places that I think the minimum military system would have the following:
1) The ability to position generals on the front (as opposed to assigning a general to the entire front).
2) The ability to give more complex orders than advance/defend (which generals can then ignore - like McClellan repeatedly ignored Lincoln).
3) The ability to set strategic targets.

I am not picky about how these are implemented, so your system could do for #3. However, I still think that we need #1 and #2. Otherwise, you assign Grant to the thousand plus mile front, then tell Grant to take the Mississippi. Grant appears in Maryland and starts his march across the Confederacy to take Mississippi. Without the ability to position generals, strategic targets are insufficient for a workable system.
I am assuming that #1 is true; otherwise, I have no idea what assigning multiple generals to a single front (which is explicitly mentioned in the DD) would even MEAN.

I don't actually think #2 is so critical. The orders McClellan ignored were just "Attack. You have all of these forces; do something with them." While I agree there should be a bit more nuance (like, say, if your army is on Advance then you can choose to have it "Devastate", like Sherman's March to the Sea or Britain's ultimate strategy in the Boer War), I think the Advance/Defend dichotomy combined with Strategic Targets is enough for a basic system (with maybe more options being for a DLC). I don't demand too much, just a system at release that allows to me have some basic control over what my armies are actually trying to do.

And #3 is, as you said, exactly what my suggested system is.
 
  • 1Like
Reactions:
I am assuming that #1 is true; otherwise, I have no idea what assigning multiple generals to a single front (which is explicitly mentioned in the DD) would even MEAN.

Well, assuming I understand the system, #1 is currently not true. You assign a general to a front, then the game decides where to put them.

I don't actually think #2 is so critical. The orders McClellan ignored were just "Attack. You have all of these forces; do something with them." While I agree there should be a bit more nuance (like, say, if your army is on Advance then you can choose to have it "Devastate", like Sherman's March to the Sea or Britain's ultimate strategy in the Boer War), I think the Advance/Defend dichotomy combined with Strategic Targets is enough for a basic system (with maybe more options being for a DLC). I don't demand too much, just a system at release that allows to me have some basic control over what my armies are actually trying to do.

Critical? Is it critical? That is a difficult word. Every time that I start typing that "okay, you are right, it is not *critical*", I can't quite get it out. I do think it is necessary. That is the word I would use. McClellan was a bad example. I was just rattling off a general. I agree that his instructions were just attack. So, the example that I used is one that the system as designed does meet. Well put.

I was actually thinking of Stone Wall Jackson's campaign in the Shenandoah when I first put it in the list. With a force 1/3 the size, Jackson managed to fight a bunch of skirmishes with three Union armies, which kept those armies from reinforcing the campaign around Richmond. Not every interaction between armies is to the death. Jackson's goal was somewhere between attack and defend. I would describe it as "attack, but do not fight a pitched battle." I think that option needs to be in there. Could you have a system where Jackson lines his men up and the Union generals line their men up and there is a fight to the death where one army is left standing and the other is either fleeing or dispersed? Sure, you could, but it wouldn't be a very good system. In that case, Jackson's forces are routed and the entire point of the campaign would be lost.

It was only for a month or so, but the Confederacy also did not immediately fight a pitched battle during the peninsula campaign in 1862. After McClellan's troops arrived at Yorktown, a far outnumbered General Johnston did not march his forces out onto the battlefield to be slaughtered. Instead, they fought more of a rearguard action while defenses were built around Richmond and troops were redirected from other parts of the Confederacy. Again, "defend, but do not fight a pitched battle".

I mean, it is such low hanging fruit. How hard is it to put some options in other than just attack and defend?
 
  • 7
  • 1Like
Reactions:
A solid base idea. Strategic Targets and the number you have at once could probably be limited by some combination of laws, military size and conditions at the time and war goals, and take a least a few weeks to start applying as analysts figure out HOW to do it and distribute the plans. Changing wastes that time and might result in complications and disorder when not everyone gets the new info at the same time, and risk the planning coordinators/order givers seeming fickle or incompetent.

This could also serve to limit the scope of conflicts; If the UK and Netherlands get into it over Borneo, setting a goal to invade Holland right away would cause outrage when the people of Britain can't see how that's appropriate for such a small goal. But setting a goal to invade Sumatra would be more tolerable, and could lead to steady escalation as more fronts open up.
 
  • 2Like
  • 1Love
Reactions:
I was actually thinking of Stone Wall Jackson's campaign in the Shenandoah when I first put it in the list. With a force 1/3 the size, Jackson managed to fight a bunch of skirmishes with three Union armies, which kept those armies from reinforcing the campaign around Richmond. Not every interaction between armies is to the death. Jackson's goal was somewhere between attack and defend. I would describe it as "attack, but do not fight a pitched battle." I think that option needs to be in there. Could you have a system where Jackson lines his men up and the Union generals line their men up and there is a fight to the death where one army is left standing and the other is either fleeing or dispersed? Sure, you could, but it wouldn't be a very good system. In that case, Jackson's forces are routed and the entire point of the campaign would be lost.

It was only for a month or so, but the Confederacy also did not immediately fight a pitched battle during the peninsula campaign in 1862. After McClellan's troops arrived at Yorktown, a far outnumbered General Johnston did not march his forces out onto the battlefield to be slaughtered. Instead, they fought more of a rearguard action while defenses were built around Richmond and troops were redirected from other parts of the Confederacy. Again, "defend, but do not fight a pitched battle".

I mean, it is such low hanging fruit. How hard is it to put some options in other than just attack and defend?
We haven't seen how battles would work yet, so more tactical actions like delaying or refusing to fight a pitched battle may be simulated inside battles. Or maybe it won't. If battles are sufficiently detailed, I think that just having 2 options, attack and defend will be enough. And based on the traits of the general, they will decide whether or not they want to be involved in a pitched battle. Then again, it's also possible that all battles in Vic3 will be a fight to the death, in which case I will be rather disappointed.
 
Well, assuming I understand the system, #1 is currently not true. You assign a general to a front, then the game decides where to put them.



Critical? Is it critical? That is a difficult word. Every time that I start typing that "okay, you are right, it is not *critical*", I can't quite get it out. I do think it is necessary. That is the word I would use. McClellan was a bad example. I was just rattling off a general. I agree that his instructions were just attack. So, the example that I used is one that the system as designed does meet. Well put.

I was actually thinking of Stone Wall Jackson's campaign in the Shenandoah when I first put it in the list. With a force 1/3 the size, Jackson managed to fight a bunch of skirmishes with three Union armies, which kept those armies from reinforcing the campaign around Richmond. Not every interaction between armies is to the death. Jackson's goal was somewhere between attack and defend. I would describe it as "attack, but do not fight a pitched battle." I think that option needs to be in there. Could you have a system where Jackson lines his men up and the Union generals line their men up and there is a fight to the death where one army is left standing and the other is either fleeing or dispersed? Sure, you could, but it wouldn't be a very good system. In that case, Jackson's forces are routed and the entire point of the campaign would be lost.

It was only for a month or so, but the Confederacy also did not immediately fight a pitched battle during the peninsula campaign in 1862. After McClellan's troops arrived at Yorktown, a far outnumbered General Johnston did not march his forces out onto the battlefield to be slaughtered. Instead, they fought more of a rearguard action while defenses were built around Richmond and troops were redirected from other parts of the Confederacy. Again, "defend, but do not fight a pitched battle".

I mean, it is such low hanging fruit. How hard is it to put some options in other than just attack and defend?
As far as I understand it, the operational level will not be modeled within the game. While we know that control of provinces is modelled, it was confirmed over here that province-based unit movement is not, meaning that operational details will be abstracted. Strategically, Jackson's orders in the valley were to defend it against Fremont's and Banks' armies; his aggressive approach to that defense was operational in nature. Similarly with the Peninsula Campaign; Johnston's strategic objective was to defend Richmond. The method he used to defend Richmond was to retreat, gather his forces, and counterattack (Seven Pines).

I do not know how this will be modeled; the Battle DD will presumably cover it. I presume it will be handled abstractly somehow; maybe a "Delaying" battle of some sort. However, what's directly modeled is only the strategic level, and those more precise actions are on the operational level. I'm also not sure if Jackson had orders to fight an aggressive defense in the Shenandoah, and Johnston in fact had orders not to retreat as he did, so those examples would still not be orders you, as a player, would give. I definitely agree that modeling such an aggressive/delaying defense should be in the game, but it would be at the abstracted operational level rather then the direct strategic level. Maybe, if a "Battle Stance" system exists, you could be able to give orders to an army for a specific stance to take (that could, of course, be disregarded), but until the Battle system is revealed, we don't know enough to suggest things like this.
 
  • 1
Reactions:
As far as I understand it, the operational level will not be modeled within the game. While we know that control of provinces is modelled, it was confirmed over here that province-based unit movement is not, meaning that operational details will be abstracted. Strategically, Jackson's orders in the valley were to defend it against Fremont's and Banks' armies; his aggressive approach to that defense was operational in nature. Similarly with the Peninsula Campaign; Johnston's strategic objective was to defend Richmond. The method he used to defend Richmond was to retreat, gather his forces, and counterattack (Seven Pines).

I do not know how this will be modeled; the Battle DD will presumably cover it. I presume it will be handled abstractly somehow; maybe a "Delaying" battle of some sort. However, what's directly modeled is only the strategic level, and those more precise actions are on the operational level. I'm also not sure if Jackson had orders to fight an aggressive defense in the Shenandoah, and Johnston in fact had orders not to retreat as he did, so those examples would still not be orders you, as a player, would give. I definitely agree that modeling such an aggressive/delaying defense should be in the game, but it would be at the abstracted operational level rather then the direct strategic level. Maybe, if a "Battle Stance" system exists, you could be able to give orders to an army for a specific stance to take (that could, of course, be disregarded), but until the Battle system is revealed, we don't know enough to suggest things like this.

According to a quick Google search, Jackson's orders were to create a diversion. Neither attack nor defend. That was strategic level. The way in which Jackson created the diversion was the operation level. I couldn't figure out whether Johnston specifically had orders to fight a rearguard action until reaching Richmond, but my impression is that those were his orders and I am certain that there have been generals that were given that order at the strategic level. I think we can just agree to disagree on this. I think that generals received more complicated orders than attack/defend, which is why I want the player to have the same capability, but if you don't believe that generals received more complicated orders than attack/defend, then it makes perfect sense why you would oppose the player/AI having the option.
 
  • 3
Reactions:
According to a quick Google search, Jackson's orders were to create a diversion. Neither attack nor defend. That was strategic level. The way in which Jackson created the diversion was the operation level. I couldn't figure out whether Johnston specifically had orders to fight a rearguard action until reaching Richmond, but my impression is that those were his orders and I am certain that there have been generals that were given that order at the strategic level. I think we can just agree to disagree on this. I think that generals received more complicated orders than attack/defend, which is why I want the player to have the same capability, but if you don't believe that generals received more complicated orders than attack/defend, then it makes perfect sense why you would oppose the player/AI having the option.
I'll concede on Jackson - I'm not sure what his precise orders were - but Johnston's decision to retreat to Richmond was his own. He wrote about his initial arrival on the Peninsula “Before nightfall, I was convinced that we could do no more on the Peninsula than delay.” Davis had given him orders to hold, which was why he held at Yorktown for a week before falling back. Lee later wrote “Johnston should never have fallen back to Richmond, but should have delivered a general battle on the narrow isthmus at Williamsburg where McClellan’s superior numbers would have been neutralized by the character of the ground.” Not only did Johnston choose on his own to retreat, once he did, he didn't bother to inform his superiors of his further intentions until his attack at Seven Pines.

Anyway, this whole thing is just me debating the examples you gave, not my main point of argument. My main point of argument wasn't that orders from on high were never nuanced beyond Attack/Defend, my point was that almost all of that nuance would be handled at the operational and not strategic level. There are exceptions - armies having orders to devastate land they pass through is an example - but things like "Act to distract the enemy" or "Conduct a fighting retreat" are operational orders rather then strategic orders. At the moment, while we have a good idea of Paradox's planned strategic system, we know nothing about how the abstract operational system will work. Presumably it will be in the Battles DD, but until then, we can't discuss that operational system. I do agree that there should be nuance rather then just having the two armies throw themselves at each other, but we don't know whether that already exists, nor do we know how it could be implemented if it doesn't already exist. The discussion around operational nuance is not one that can be had meaningfully until the details of the abstract operational system are revealed.

When I originally responded to your second point, I believed you meant more complexity of strategic orders. I do not believe that effective strategic orders of this period were more nuanced then Attack/Defend. The Schlieffen Plan, the classic example of a "battle plan" in the period of Vic3, derailed precisely because Von Moltke Jr. didn't grasp the full complexity of the plan. The original plan involved swinging around Paris and attacking the rear of the main French army trying to retake Alsace-Lorraine. Von Moltke Jr. instead was drawn into attacking Paris once it seemed within reach, opening himself to Entente counterattack and defeat at the Battle of the Marne. The plan fell apart precisely because of the inability of upper authorities to effectively develop and execute complex plans. On the strategic level, it makes sense for orders to be an Attack/Defend duality, so long as there are objectives to attack or defend (and preferably an option to set the army to Devastate). And until we know how the operational level is modelled, the debate can only be on that strategic level.

If you disagree on the strategic level, and do believe strategic orders should be more complex, then yeah, we'll agree to disagree.
 
So, the idea of giving an army a goal to work towards.

Dev response on the Diary: “What we're considering adding is a method of prioritizing the various targets in the war, and setting custom targets, on a national (not Front- or General-) level. What we need to be careful with here is to not add methods of control that make the player technically able to control with precision how Generals act in every moment by microing their priorities.”

I think many people are unsatisfied with this, and want a way to give an army a specific goal. I agree wholeheartedly with this. Let’s look at a historical example; the February 1862-July 1863 campaign for control of the Mississippi River.

The Mississippi River was vital to both the Union and Confederacy, for primarily economic reasons. For the Confederacy, loss of the Mississippi meant the loss of the vast supplies of the Trans-Mississippi, with the loss of the vast farmland of Texas felt particularly hard. For the Union, as long as the Mississippi was in Confederate hands, the entire Midwestern economy ground to a halt; without the river route to the sea, goods produced in the Midwest could only be exported across land, which was far less efficient. I think both of these are things Vic3 would have to represent, beyond economic and logistic in nature.

Now, let’s look at the system as it is. Ulysses S. Grant’s army is put on “Advance” orders, and starts pushing into Confederate territory. With no way to tell him to take the Mississippi, he just advances blindly; the Midwest suffers, and the Confederates have their supply lines. This is not only untenable for gameplay, it’s also unrealistic; everyone knew the Mississippi was the Union’s goal because its value was obvious. Given that I doubt the AI will be robust enough to “know” the strategic value of the Mississippi, the player needs a way to push the AI to do that.

So, we need some way to give goals to individual armies. But how can we avoid that turning into somewhat-hand-off microing?

Here’s my suggestion.

Scattered across the map, there are “Strategic Targets”. These can be of two types; a province, like a national capital or major city, or a line, like a large river or important railroad. On both Advance and Defend orders, the player has the option to select one of these Strategic Targets to either focus the attack on or focus the defense on. That army will then work towards achieving that goal. If the player does not select a particular goal, the AI will instead generally work towards those Strategic Targets. Strategic Targets give a slight boost to warscore, but their impact is primarily on affecting the enemy’s ability to continue fighting rather then directly on warscore.

So how do we avoid manipulating the system to allow micro? My suggestion comes in two parts.

1. The player has no control over how the AI goes about achieving this goal. An Aggressive general might launch an all-out offensive directly at his target, a Maneuvering general might attempt to get around his enemy, a Cautious general might just very slowly push towards the target. Maybe the general will even ignore orders; if the general’s home region is occupied by the enemy, he might forgo his goal in favor of trying to liberate his home. The player can only choose a goal, and that only from a Strategic Target; no “Alright, move to this town and then to that one” or “Take this city by advancing this way”, only “Take this major city” or “Secure this railroad”.

2. There is a difference between a Target chosen to defend and a Target chosen to attack. A Target chosen to defend can be changed at will based on where the enemy might be threatening; however, to change an Offensive Target is an admission of defeat. Once an Offensive Target is chosen for an army, to change it is to admit “We have failed to achieve our goal.” Naturally, this will severely influence public opinion to the negative; war exhaustion/displeasure with the government will rise. If a small army is redirected, maybe few will notice, but if a large army changes its Offensive Target, that will incur a very severe penalty; admitting your large force failed is a heavy blow to public morale. Even if an army with an Offensive Target is changed to a Defensive stance, if switched back to Advance, it will keep the same Target. Only once an Offensive Target is achieved will choosing a new one be capable.

Let’s look at two historical examples to see how this system might function.

Mississippi Campaign: An army under Ulysses S. Grant is given orders to secure the Mississippi River. This entails making sure no province bordering the river is under Confederate control. Grant prefers Maneuver to outright Attack; facing Confederate strongholds at first Columbus, then Memphis, and finally Vicksburg, Grant moves to cut off those strongholds, forcing the Confederates to either retreat or be besieged. Meanwhile, Confederate forces in the area have orders to defend the Mississippi. At Columbus and Memphis, the Confederates choose to evacuate rather than be besieged; at Vicksburg, they choose to hold. Grant wins the subsequent siege and take Vicksburg. With a second army in Louisiana pushing north from New Orleans with the same goal taking Fort Hudson, the Offensive Target has been achieved; Grant can now be redirected to join Rosecrans for the Battle of Chattanooga.

Eastern Theater: At the start of the war, the main Union army selects as its Offensive Target the Confederate capital of Richmond. As the war goes on, various attempts to take Richmond have failed, though Confederate counterattacks have also been halted. However, despite the immense difficulty of taking Richmond, changing the Offensive Target of the largest army in the field would devastate national morale, so the Union army continues to try for “On to Richmond!” Even as bloodshed mounts, the Union is unwilling to take the penalty for admitting failure, and so the bloody push to Richmond continues to the bitter end.

Now what would happen if someone tried to use this system to micro? They decide they want their army to move northeast and then double back southeast, so they choose an Offensive Target to the northeast. Once their army has reached the point where they want it to move southeast, they change the Strategic Target. This then incurs the heavy penalty for changing an Offensive Target, and public frustration rises to the point where the war effort is severely hampered. The attempt to abuse the system to micro has failed utterly.

I think that this system would achieve both of the critical things any system of Army Goals need to; it gives the player the ability to strategically direct a war, but still prevents microing. With a list of Strategic Targets automatically chosen by the game (say, select rivers, major railroads, national capitals, and cities past some threshold of size), this also keeps player control firmly at the strategic level, while giving them plenty of control at said level. It also adds only a tiny amount of complexity, keeping it easy for the player; if you have a large number of small armies, you can just not give them specific Goals.

So, what do you think? Please do comment with any thoughts.
I think the penalty for changing an attack target should be less

in ww1 for an example the goal of many campaigns was basically never really met.

and the Public did care little

of course if you continuesly fail to take a target you loose manpower which likely affects the public opinions you and yadda yadda you don’t want that.

also changing the target if the general would have been capable of reaching or likely would upset the general who could
Have won prestige with this.

also you should be able to have multiple targets: This of course if taken to the extreme will severely restrict your ability to win the war by having too many pointwhere the general „focuses“

I would also like to have these points to be not stagnant but Variable

For an example in the start of the game in Germany the most important things are likely the agrarian fields and the towns but as you rapidly industrialize the Rheinland becomes the industrial heartland and there are many cities which’s are far more important then beforehand like Berlin becoming even more important then before due to the electric and chemical industry

So these points of interest if you will should be geographic locations like mountain ranges or rivers, infrastructure centers or lines, big cities and industrial centers most of these would be considered by an algorithm like a railroad that transports at least a certain amount of the enemy’s supply’s is one and your industry next to the enemy’s borders is one of your defendables
 
  • 1
Reactions: