Grand Strategy Victoria 3

[Fedora Master]

And now there isn't even much possibility of modding that shit out. Great job.

 

[Average Manatee]

I don't think so. DH has tons of completely ahistorical mods, and the AI is just as competent in any setting. It's not flawless, of course but "AI leaves too small a garrison to defend England against invasion" is an orders of magnitude smaller problem than "AI sits by and watches as I blitz through its lines and conquer the heartlands."

It's just incompetence.

I think a lot of it comes down to # of provinces. DH generally has fewer, and figuring out where to send 30 divisions when the front is 5 provinces is pretty easy to get close to optimal. When there's 20 provinces and you still only have 30 divisions its a lot harder, and if you've played DH mods with a lot more provinces you'll know that the AI tends to degrade similarly to HoI3/4. DH also restricts naval invasions to a few specific provinces which similarly helps the AI a lot.

At the same time I think the constant player demands in paradox games for more provinces is partially because "omg watch me do this epic encircle" maneuvers are what the playerbase likes and its poor AI reactions to wide fronts that often allows for this.

I mean they literally could reuse frontline system from HoI but make units invisible/AI only and teleport for naval invasions and it would work out better...

Yeah, I'm not sure why they didn't do this. Like a lot of things to do with AI, making a convincing fake system behind the curtains is almost as difficult as making a real system and ends up with so many weird bugs because your fake system isn't based on anything in reality.

 

[thesecret1]

I think a lot of it comes down to # of provinces. DH generally has fewer, and figuring out where to send 30 divisions when the front is 5 provinces is pretty easy to get close to optimal. When there's 20 provinces and you still only have 30 divisions its a lot harder, and if you've played DH mods with a lot more provinces you'll know that the AI tends to degrade similarly to HoI3/4. DH also restricts naval invasions to a few specific provinces which similarly helps the AI a lot.

The problem with this theory is that this behavior persists in HoI 4 even across small fronts. For example: In the spanish civil war, I saw the AI leave an unprotected gap in the front as it illogically shuffled troops back and forth. So I sent some tanks there. The AI made no effort to deal with the breach, so I just drove onwards while infantry held the front. Before long, everything except the front was mine, the AI only holding one long noodle full of its troops. All done with virtually no combat or opposition. This would never ever happen in DH, even on large fronts like during Barbarossa.

At the same time I think the constant player demands in paradox games for more provinces is partially because "omg watch me do this epic encircle" maneuvers are what the playerbase likes and its poor AI reactions to wide fronts that often allows for this.

If they make shit AI on purpose to provide people with ebin screencaps for Reddit, then they're retarded and the whole franchise deserves to burn.

 

[Average Manatee]

The entire eastern front in DH is like 10-15 provinces depending on how much you spread out though, while the SCW starts at more like 30. And the amount of frontline provinces changes much more frequently in HoI4 with how much more mobile the combat is. It's also the number of divisions to province ratio that matters more than just number of provinces.

The AI basically works as such:

Less than 1 division per frontline province: instant collapse.
1-2 divisions per frontline: tries to cover everything but easily overwhelmed anywhere along the line and once a breakthrough is made it goes haywire trying to scramble divisions around.
2-4 divisions per frontline: much more stable, breakthroughs generally take longer and the AI can react in a reasonable timeframe and doesn't quite go haywire.
5+ divisions per frontline: AI can fairly competently hold this unless you have a super strong set of breakthrough divisions (which is another problem, those don't really exist in HoIs before 4's busted unit designer and the AI can't cope with them).

 

[thesecret1]

The entire eastern front in DH is like 10-15 provinces depending on how much you spread out though, while the SCW starts at more like 30. And the amount of frontline provinces changes much more frequently in HoI4 with how much more mobile the combat is. It's also the number of divisions to province ratio that matters more than just number of provinces.

If you make a wedge, you can easily get a 30 province front even in DH. Yet the HoI4 behavior never emerges. Similarly, the AI manages to handle itself competently enough even if it doesn't have enough units to man the front. Furthermore, HoI4 AI does not get better if you reduce it to a smaller front. That's not to mention stuff like GB having its units virtually everywhere and managing several theatres at once.

Again, it's not like DH AI is flawless or anything, but you'd never see such glaring bullshit like just ignoring a breached front.

 

[Space Satan]

They banned me but left a backdoor for me to mock devs even further.
Diplomacy feedback