Issues with AI Germany is why Germany loses the war in 1941.

[Gwydion5]

The shuffling problem was much worse in 1.0.0. It has been gradually getting better

Agreed.

and has certainly gotten better instead of worse in 1.3.3.

I'm not certain about that. With AI Germany now garrisoning conquered provinces, it has added another AI branch of theater/army/division realignments to current force disposition. As evidenced here, where AI Germany while at war with Russia reduces the Eastern Front army down to 8 divisions, while increasing the Europe Theater Garrison army to 209 divisions. In 1.3.2 it is unlikely it would've done that because AI Germany didn't really do garrisons.

Germany simply does not manage to produce enough units by the Spring of '41 to accomplish its required tasks. It is short handed. That means that shuffling is more likely to be called for.

To add on to that, it also doesn't manage its theater/army make up very well either. Like using MARM's for garrison duty, and not producing cavalry or divisions specific for garrison duty, or as evidenced above reducing an active combat front beyond any sensible proportion, or putting 60 divisions in a single province, or shifting 5/6ths of a 200+ division front to the poorest infrastructure areas on the front, etc....

 

[Meglok]

Both of those seem to imply that Germany's current plight was in some way premeditated. I can assure you that it was not. Here is the history of 1.3.3 in a nutshell. They introduced a few new features and fixed some long standing bugs. We then had it in public beta for most of a week and that involved a few hotfixes to other bugs (some of them new).

Each fix alters the balance. Some of the bugs, like hundreds of US and UK divisions getting stuck at sea for lack of transport, heavily favored Germany. Fixing those egregious bugs brings with it the cost of significant balance shift.

Some of the changes in 1.3.3 were very fundamental, things like the way infrastructure now affects construction and the way resource shortages are handled. Pre-release, such fundamental economic changes would have required a few weeks of testing just to tweak the balance. For an update that was a free patch only, I can understand not being able to invest that much time in testing.

The entire rebalance situation will be radically different again in 1.4 most likely.

Not quite sure how you discern that from my comment. Podcat said that since most people prefer to play Germany they tend to focus on how Germany does. My comment was they should consider why most people play Germany in HOI4, is it preference or that playing many other majors is a bad experience,

That being said, I don't think PDS made a conscious decision to nerf Germany and the Axis. My thought is they did not play test the production and infrastructure changes enough in single play mode to comprehend the adverse effects on the Axis nations and minors when they were coupled with the scripted templates in 1.3.3. And they either didn't catch or understand the return of the convoy stockpile would mean the return of mini invasions, suicidal troop transports, and the Axis sending troops all over the map again instead of focusing on needs closer to home. These were seen by some players during the inadequate open beta time frame but by then I guess it was too late to debug and fix. It is either that or from a cynical point of view they did see these issues but released anyway due to marketing and budget deadlines figuring they will just fix things later in 1.4.

 

[Gwydion5]

It is either that or from a cynical point of view they did see these issues but released anyway due to marketing and budget deadlines figuring they will just fix things later in 1.4.

That is one thing that I truly don't understand. How did PDS ok the release of 1.3.3 knowing that the devs would be indisposed for 7-10 days or whatever. What harm was there in letting 1.3.3. sit in open beta for another week and then addressing any issues that the community reported, or at least being aware of those issues and making a more informed decision when deciding to release 1.3.3.

This type of action is the absolute opposite of my experience, and that of my friends who also work in IT. No way would I allow or be allowed to patch/upgrade a system or service and then have the core team be on a week long hiatus immediately after. Now granted this is just a game, so impact and severity are a bit reduced compared to the the systems and services I support, but it is just one more point of concerning behavior that gives me doubt and concern.

 

[Secret Master]

Actually that problem is not new with 1.3.x. Germany has been plagued with that kind of nutty behavior on her eastern front ever since the game first launched.

:cough: HOI3 :cough:

Seriously, though, unit shuffling isn't new.

 

[Dalwin]

I'm not certain about that. With AI Germany now garrisoning conquered provinces, it has added another AI branch of theater/army/division realignments to current force disposition. As evidenced here, where AI Germany while at war with Russia reduces the Eastern Front army down to 8 divisions, while increasing the Europe Theater Garrison army to 209 divisions. In 1.3.2 it is unlikely it would've done that because AI Germany didn't really do garrisons.

To add on to that, it also doesn't manage its theater/army make up very well either. Like using MARM's for garrison duty, and not producing cavalry or divisions specific for garrison duty, or as evidenced above reducing an active combat front beyond any sensible proportion, or putting 60 divisions in a single province, or shifting 5/6ths of a 200+ division front to the poorest infrastructure areas on the front, etc....

That is not the shuffling to which I refer. I am speaking of the side to side nonsense that happens at the front. This primarily concerns the behavior of the battle planner. The problems to which you are referring are more commonly labeled as front abandonment or force allocation. Those things are still problems, but I think they are not much closer to solving them.

 

[cellinis]

That is not the shuffling to which I refer. I am speaking of the side to side nonsense that happens at the front. This primarily concerns the behavior of the battle planner. The problems to which you are referring are more commonly labeled as front abandonment or force allocation. Those tings are still problems, but I think they are not much closer to solving them.

Well, I do hope they solve them because the force allocation is, in my opinion, a way bigger problem for the AI than side to side shuffling. In my current game, when the German Panzers attack, I haven't a chance in hell at stopping them. The template is (I think) 8 MARM (with a very decent variant), 4 Mot, plus MSPG. Each time those Panzers attack, the front is pushed back by half a dozen provinces. Then, they disappear from the front. Sometimes they're sent to Norway (btw, via French Ports) or to Garrison duty in Western France. Though I don't have the stats to back it up; I'd say that the front logic is the reason for at least thrice as many losses than what I've managed to inflict upon the German Army.

 

[Chondrite]

Well, I do hope they solve them because the force allocation is, in my opinion, a way bigger problem for the AI than side to side shuffling. In my current game, when the German Panzers attack, I haven't a chance in hell at stopping them. The template is (I think) 8 MARM (with a very decent variant), 4 Mot, plus MSPG. Each time those Panzers attack, the front is pushed back by half a dozen provinces. Then, they disappear from the front. Sometimes they're sent to Norway (btw, via French Ports) or to Garrison duty in Western France. Though I don't have the stats to back it up; I'd say that the front logic is the reason for at least thrice as many losses than what I've managed to inflict upon the German Army.

In the current state of the game, you should reload the game every month or so during wars with large fronts. Reloading the game usually causes the AI to delete and recreate its front which in turn cancels all the current unit orders that make them to behave like lemmings and redistributes the units properly along the entire front. Running a hands-off game with and without mid-game reloads can lead to vastly different results on the east front.

Of course this doesn't fix the innate unit shuffling problem with the battleplan AI, the occasional front abandonment nor the 1.3.3 unit shuffling problem with the AI's occupation armies which leads to hundreds of Panzers resting on the bottom of the Atlantic because it constantly sends Panzer divisions to Norway through the British fleets.

 

[Gwydion5]

In the current state of the game, you should reload the game every month or so during wars with large fronts. Reloading the game usually causes the AI to delete and recreate its front which in turn cancels all the current unit orders that make them to behave like lemmings and redistributes the units properly along the entire front. Running a hands-off game with and without mid-game reloads can lead to vastly different results on the east front.

In my experience loading from a save game breaks the AI. As evidenced here against Max Boosted AI Germany. You can compare that to this where there was no reload.

 

[Chondrite]

In my experience loading from a save game breaks the AI. As evidenced here against Max Boosted AI Germany. You can compare that to this where there was no reload.

It doesn't break it but certainly it can swing the war to either direction depending on which side messed up and how badly. I've spent literally hundreds of hours running test games and looking at what the AI does (but I only test my mod so the situation is not exactly comparable to your vanilla tests).

What I call the "lemmings issue" is fixed by reloading the game. The problem is that on very large fronts, units for whatever reason start shifting toward either end of the front and leave the center undefended. Eventually they become huge stacks of units that are ordered to one end of a front and upon reaching it, they get redeployed to the opposite end of the front (assuming they dont get stuck in combat and likely encircled). When you reload the game, the error log shows "[13:45:46][theatre.cpp:3958]: Theaters or fronts has been fixed, as some spots was incorrect as in the savegame.", it deletes the front, cancels the unit orders and after a few ticks recreates it. That is my interpretation anyways.

 

[Gwydion5]

It doesn't break it but certainly it can swing the war to either direction depending on which side messed up and how badly.

I suppose I should've worded my response better by saying "can" versus "will". I might have to run a few simulations from that save and see what the ratio is though in terms of Russia advancing their front towards Berlin.

 

[Razor]

I believe the Soviet Union became a little too strong after the latest patch.