AI launching larger offensives - Just a suggestion

[Chromos]

I hope supply will play a bigger role in HoI4.
E.g. as you need more supply when fighting as when just guarding the frontlines watching what is going on..

So an offensive would need huuge amount of supply. And that way it was also historically. Big operations got suprlus on supply to "get it going"..
In HoI2 we had a button that gave a bonus but also cost much more supply.

I would like to see:
1. Units need more supply when fighting.
2. A button that can priorize my units for supply(like in HoI3) as also one that is amassing supply for that units as it is about to move in and fight ist way to Berlin(similar to HoI2)..

AI is of course a key factor here. It needs to be able to prepare such operations. Battleplaner would be very helpful for us humans though.

 

[Jazumir]

Haha, sorry, that was a fair play. That said, I think there's a bit more potential for this than the theatre/AI system thing. By all means laugh at me post-launch if it doesn't work out, but I've got a good feeling about this . While I respected PDS' efforts with HoI3 (and still enjoyed the game immensely, particularly from Semper Fi onwards), I didn't have as good a feeling about it prior to launch.

Okay, i´ll spell it out: I was refering to the chain of command feature of HoI3 - which could as well have been served as a crutch for the AI to get itself organized - if only it had ever learned to utilize it in the slightest. Instead, it was completely unable to take any bearing from it, got screwed over by it, and the feature gets axed completely (for all we know) in 4. I don´t think it is unreasonable to fear something similiar happening to the battleplan feature, just one iteration later - though of course i sincerely hope i will be utterly wrong.

Please forgive the bitterness i display here - it is just that to me, one (the battleplans) makes little sense without the other (CoC) and both are entirely pointless without a proper AI, not only being able to utilize both features, but also getting much better because of them.

 

[apfelstrudel]

Wow, didn't even think of that before. I only thought about the battleplans from the human player's perspective.
It was indeed always a major a.i. problem that it would simply constantly attack throughout the entire front line.

Historically armies would plan and amass troops for large scale offensives that would focus on a few sectors.
The human player does exactly the same. You think of a Barbarossa plan, deploy your troops accordingly and make sure that the timetable works out.
Then once the mud dries you launch your offensive without the a.i. having a clue where your main drives are going to be. That's how warfare works.

I hope the the HoI4 a.i. will do likewise:
- long-planned large scale offensives INSTEAD OF continous attritional attacks
- few selected breakthrough points with several axis of drives towards several targets INSTEAD OF single attacks on a broad front line

 

[MaximillanVoss]

I'm glad I managed to start this broad discussion. I hope Devs will at least consider this.

 

[Axe99]

Okay, i´ll spell it out: I was refering to the chain of command feature of HoI3 - which could as well have been served as a crutch for the AI to get itself organized - if only it had ever learned to utilize it in the slightest. Instead, it was completely unable to take any bearing from it, got screwed over by it, and the feature gets axed completely (for all we know) in 4. I don´t think it is unreasonable to fear something similiar happening to the battleplan feature, just one iteration later - though of course i sincerely hope i will be utterly wrong.

Please forgive the bitterness i display here - it is just that to me, one (the battleplans) makes little sense without the other (CoC) and both are entirely pointless without a proper AI, not only being able to utilize both features, but also getting much better because of them.

Aye, I know where you're coming from, and understand the bitterness. From my experience with HoI3 (ie, from my limited perspective) the AI didn't really use it's chain-of-command much at all - rather than a tool to help it do things (beyond allocating units to theatres, which it did seem to use, and half-competently) it seemed to be something that was primarily designed with the human player in mind, to allow them to toggle different parts of the war effort to AI/human control. However, when working out how to attack, the AI just seemed to move the whole front forward, rather than play for breakthroughs per se (you'd still get salients, but more because of different parts of the defensive line collapsing at different rates, than because of a concentration of forces and an offensive in the sense we understand them).

The Battleplan system, though, is a framework that sets objectives and axes of advance, then allocates divisions to those objectives. Potentially (as far as I can see it), it's a powerful tool around which an AI offensives-based combat system could be developed, far more so than the CoC system.

Anyways, I'm sure we both hope it works out . I agree that the removal of the CoC (beyond the single layer of 'forces assigned to a general for a plan' - although I still have my fingers crossed for some kind of optional unit management system for the player, if not the AI) is a shame, but in many ways the CoC system could have held back the AI from developing offensives, particularly with a battleplan system alongside it (with a CoC, the AI has to balance its various HQs with the different objectives, and has a lot more work to do to work out which units to go where - which gets even worse if it considers terrain and the best units for the job - without a CoC, the AI just sends the units that do the job best to the appropriate objective).