FTM Invasions...too many, too early!!

[Oof]

I wouldn't like to see hardcoded limits, either, besides the very basic ones. If e.g. France is completely empty and Germany is occupied elsewhere, then I don't see the reason why the AI shouldn't attack. However, if there is no second front, then attacking is just stupid, even if there are several empty ports. How long will they last? A week? Two weeks? A month? Inevitably, they will be destroyed.

If i understand Secret Master's post this can already be done. Apparently the AI doenst invade when there are two units in a port. So withdrawing one would prompt an invasion...

Dont get me wrong i rahter see another solution. I think most invasions are too easy and too unrealistic anyways. They just dont take into account the huge amount of planning, logisitcs, etc, involved and the overall strategic situation.
Even the 2 unit port garrisons are not a real solution to the problem which is: If the AI does an invasion there must be a realistic chance of succes without increasing the risk of other theatres being overrun.

For example an invasion of France by the UK in late 1940 would always be unsuccesfull because the English forces would have to face the entire Wehrmacht on their own. But the result would also be a neglect of the African and Far East theatres. Such an invasion would only have a chance of succes if Germany is at war with Russia and the USA and at least a stalemate situation in N-Africa and the Far East occurs.

 

[Oof]

It's not a cheat. It's an exploit, perhaps, to know that the AI or human players are disallowed from invasions until certain dates and conditions are met. That's the problem with something like this. If the conditions are hard-coded, then the player simply plays around them and we are back to not having to garrison the west while attacking the Soviets.

I agree but arent we already doing this? We know Germany will attack Poland late august/early september 1939. Germany will not attack France before april 1940, They will not attack the USSR before april 1941, Japan will not attack the USA until november 1941 and the Allies will not invade France properly before 1942...

But i do get your point and i really dont like hard coding either but until a new expansion or version of HOI something needs to be done. Because until the AI more or less starts to understand if an invasion is smart an viable considering the overall strategic situation the game can be easily unbalanced.

 

[Darlor]

I agree but arent we already doing this? We know Germany will attack Poland late august/early september 1939. Germany will not attack France before april 1940, They will not attack the USSR before april 1941, Japan will not attack the USA until november 1941 and the Allies will not invade France properly before 1942...

Not entirely correct. I have seen Germany invade France in Oct of 1939, and quite successfully. FTM 3.05 seems to have more suprises for the player than you would suggest.

 

[Secret Master]

I agree but arent we already doing this? We know Germany will attack Poland late august/early september 1939. Germany will not attack France before april 1940, They will not attack the USSR before april 1941, Japan will not attack the USA until november 1941 and the Allies will not invade France properly before 1942...

But i do get your point and i really dont like hard coding either but until a new expansion or version of HOI something needs to be done. Because until the AI more or less starts to understand if an invasion is smart an viable considering the overall strategic situation the game can be easily unbalanced.

I did some experiments last night regarding this whole mess. Garrisoning my ports with stacks of MIL warded off Allied invasions as I stated before. The AI did, however, decide to strike at the "soft underbelly" as Churchill would have wanted. It landed around Nice and Genoa and made some progress until I managed to SR three corps down there to help the Italians and the French.

The AI saw an under-defended part of the continent and attacked. It didn't realize that within an hour of the invasion, I would be able to rebase 10 TACs, 10 INTs, and 6 CAS to begin bombing them to death and that in few days the garrison along the Atlantic would be there to form a defensive line.

In one sense, it was a smart offensive move to attack the Italians. Again, it was doomed to failure because I had a strategic reserve strong enough to wipe it out easily. But what surprised me is that the AI bothered to attack continental Europe while the battle in North Africa was still undecided.

Maybe the UK needs to weight North Africa more so that it does not spread its forces too thin. All those divisions in Nice could have helped obliterate the Italians and French in North Africa, securing a nice chunk of land for the UK and freeing up resources for an invasion of Europe. (I wasn't wasting my time helping out the Italians in North Africa. I had better things to do. )

 

[Oof]

Maybe the UK needs to weight North Africa more so that it does not spread its forces too thin. All those divisions in Nice could have helped obliterate the Italians and French in North Africa, securing a nice chunk of land for the UK and freeing up resources for an invasion of Europe. (I wasn't wasting my time helping out the Italians in North Africa. I had better things to do. )

It does. I reckon the UK, both ingame and historically, is the cornerstone of Allied resistance and victory in WWII. The decisions made by the UK if AI controlled, determine pretty much the outcome of the war. If it loses N-Africa, Italy will have card blanche in Africa and the Med and the Germans can use all their forces against the USSR. But it also gives Japan a card blanche in India.

To be honest i think the UK needs a lot of work and so does the AI. The key to the problem is making the AI understand the overall strategic situation. In the end the invasions in this thread are only a symptom of the real issue. But making the AI understand might prove difficult, cause certain decisions need human intelligence...

I am currently working on some ideas on how to improve the UK. At the moment the text is to long to post though (3+ pages) so i have to find a way to downtone it a bit so it can be posted for comments and suggestions.

 

[Oof]

Not entirely correct. I have seen Germany invade France in Oct of 1939, and quite successfully. FTM 3.05 seems to have more suprises for the player than you would suggest.

Havent seen it yet but i reckon when France is player controlled and you keep the Northern borders properly guarded it will attack in 1940. Anyway the exact date doenst matter, my point is that there a quite a few things ingame that are very predictable including the aprox moment. And we all abuse that knowledge

 

[Darlor]

Havent seen it yet but i reckon when France is player controlled and you keep the Northern borders properly guarded it will attack in 1940. Anyway the exact date doenst matter, my point is that there a quite a few things ingame that are very predictable including the aprox moment. And we all abuse that knowledge

Keep things predictable and we take advantage of the predictability and then complain that the game is too easy to beat. Allow it to vary too much from the historical and we complain that it is too sandboxy and we can not anticipate and thus take advantage. Seems like we speak out of both sides of our mouths. But I still think the AI is getting better at taking advantage of our small mistakes. Look how many were surprised by all the invasions. Granted, not very good invasions but still it forces us to take less advantage of our knowledge of history by causing us to garrison places where many did not before. A small step forward in making us play more realistically. Now if we can only teach the AI to think on a bit larger scale and to effectivly utilize its forces at the strategic level, it might just get real interesting.

 

[No idea]

Definitely something has to be made to tame the kamikaze UK bull dog. The UK AI invades crazily and often until it runs out of units, which leaves France with no help at all. I think the UK AI should invest its units on the defence of France. Apart from that, it should be taught that units can be loaded onto transports, so that the germans don´t capture all its units. I have seen the UK AI make several good landings with quite a good number of units and taking ports and several provinces, but the AI should be taught that in 1939 or 1940 the landings are doomed, because the UK simply doesn´t have a big enough army to be able to conquer Germany, not even maintaining a beachhead indefinitely. So, if UK units reinforced France instead of launching kamikaze invasions against Germany, I think we´d get a far better game.

 

[Oof]

Keep things predictable and we take advantage of the predictability and then complain that the game is too easy to beat. Allow it to vary too much from the historical and we complain that it is too sandboxy and we can not anticipate and thus take advantage. Seems like we speak out of both sides of our mouths. But I still think the AI is getting better at taking advantage of our small mistakes. Look how many were surprised by all the invasions. Granted, not very good invasions but still it forces us to take less advantage of our knowledge of history by causing us to garrison places where many did not before. A small step forward in making us play more realistically. Now if we can only teach the AI to think on a bit larger scale and to effectivly utilize its forces at the strategic level, it might just get real interesting.

The aI is getting better and better. And i reckon that as long as we come up with new ideas and suggestions how to improve the game it will get better with every new expansion and version. Cant wait for HOI15

 

[Oof]

So, if UK units reinforced France instead of launching kamikaze invasions against Germany, I think we´d get a far better game.

Thats exactly why the thread was started by PlasticPanzers.

 

[Cybvep]

Keep things predictable and we take advantage of the predictability and then complain that the game is too easy to beat. Allow it to vary too much from the historical and we complain that it is too sandboxy and we can not anticipate and thus take advantage

This sounds rather illogical, especially since there are difficulty levels. Few players play below Normal and I've rarely heard complaints about the game being too hard because of its unpredictability. Usually, I hear the opposite.

People don't like sand-boxes when they frequently result in completely implausible scenarios. Remember the HOI3 in its early days? It was not unusual to see Japanese invasions of Finland and Brazilian invasions of France. Half of the world was in the Allied camp in 1938, including the USA and Switzerland. Few want to play games like that.

 

[Darlor]

There appears to be a wide difference of opinion as to what is implausable. Go back to the first post where an invasion of Denmark by a unit from Denmark just after Denmark falls is called implausable. My point is that the "impossible", as in your examples, is different from highly unlikely to have happened.

An invasion should either look like a well thought out multi division activity with a reasonable chance of success based on what knowledge is available to the AI or it should look like a brigade sized raid with provisions for withdrawal of the small force after inflicting some amount of damage to a target: air base, naval base, manufacturing, etc. Right now air cover for the invasion, provisions for withdrawal of a force if things are not going well, adequate sized initial invasion force and follow up forces, and a more rational process for determining when and where to launch an invasion all appear to need some work.

 

[PirateJack]

Maybe limiting the Allies to invading only when they have troops above a certain level, so that there is still a risk of invasion from them, but it's unlikely to happen more than once before the USA joins in. Also, the AI needs to understand when and where to retreat, as well as when to reinforce the beachhead (and with what); but those are both difficult to code so perhaps the first option would be more appropriate for now.

 

[Oof]

Maybe limiting the Allies to invading only when they have troops above a certain level, so that there is still a risk of invasion from them, but it's unlikely to happen more than once before the USA joins in.

Thats not bad because thats pretty much the way it is right now. As a player you know that your forces in France can defeat any English invasion as long as you're not at war with the USA.

 

[Oof]

An invasion should either look like a well thought out multi division activity with a reasonable chance of success based on what knowledge is available to the AI or it should look like a brigade sized raid with provisions for withdrawal of the small force after inflicting some amount of damage to a target: air base, naval base, manufacturing, etc. Right now air cover for the invasion, provisions for withdrawal of a force if things are not going well, adequate sized initial invasion force and follow up forces, and a more rational process for determining when and where to launch an invasion all appear to need some work.

Thats one of my earlier points and why i came up with the planning option. Problem is that its probably hard to implement because there are too many variables. A player can decide where and when he wants an invasion, but it would be pretty difficult for the AI unless coding is used. But this would result in predictable invasions...

On the other hand if you take a look at some of the ideas by the community in this thread how to improve invasions with techs etc, i reckon its worth a try for PI. It would greatly improve the game and all nations would benefit. I reckon that it can also improve western allied cooperation and the Japanese conquest in the Pacific.

 

[Secret Master]

Well, I saw the AI UK invade the soft under-belly again. This time, they invaded puppet Norway, and again only when they saw an ungarrisoned port (my MIL hadn't SR'ed there yet).

Yeah, I think the AI UK needs to be consulting ENIGMA a bit more to determine invasion feasibility. The strength of coastal garrisons does not equate to strength of reserve forces. The AI needs to be told the general brigade and wing strength of enemy forces in an area of operation (Theater?). I mean, it launches some nice invasions, but what can 200,000 Brits do to the entire German Army if I'm not at war with the Soviets and they don't have air superiority?

 

[Darlor]

Another thread on defending against a German invasion of Britian has proposed a more aggressive building of radar in Britian and research of radar techs to help with the naval defense against invasions because it seems to help the AI with deploying its units. If that were implemented, it could also provide the information for determining whether an invasion of the continent should go forward or not. First not allow anything beyond very small raids until a certain amount of occupied territory around a potential invasion point could be seen with the radar. Then require local superiority of forces in that area to allow the invasion to move forward. I don't know what information the AI has available to it now but at least this could put it on a par with what the human knows. This would decouple the decision to invade from the need for USA involvement and Barbarossa thus keeping the Germans more on their toes. Does the AI know what forces are in an area? Does it remember unit sightings from air units scouting? Does it have access to and use the same information that a player does? So much I don't know.

 

[Oof]

Not a bad idea. This could be combined with the intel partisan bases provide.

 

[Lord Duppy]

This thread needs a bump. I'm tired of seeing the UK invade Holland in November 1940, get wiped out, and then try the same thing again at least once every year. Why is no one else talking about this? Has anyone found a fix?

 

[Oof]

This thread needs a bump. I'm tired of seeing the UK invade Holland in November 1940, get wiped out, and then try the same thing again at least once every year. Why is no one else talking about this? Has anyone found a fix?

There is fix in the sense that when Germany is players controlled. Just deploy 2 MIL Div in each port and the AI will not attack. It will cost some MP though.