Semper Fi Preview-AAR

[unmerged(18239)]

Before getting too worried, keep in mind that the success of Fall Gelb was highly dependant upon luck, mainly the French being stupid and leaving the hinge of their entire defensive plan unguarded and committing their entire operational reserve prior to identifying the main German thrust.

This AAR seems to be a much fairer representation of what would have happened if both sides were making at least some of the right calls. It looks like the German AI is fighting smart, looking for weak spots and attempting to exploit them. If the AI were actually to let the historical course happen, there would be some that complain the Allied AI is all jacked up (kinda like real life )

Exactly my thoughts. The odds in this battle are near on even.

 

[unmerged(18239)]

Actually this AAR reminds me very strongly of a mirror image of the 44-45 campaigns in this area.

Yes it does indeed. I just read When the Odds were even (a bad book) on the '44 Vosges campaign.

 

[Alexander Seil]

Is it just me,but (German) AI is performing catastrophicaly in offensive?

Its allmost One MONTH from atack and Holland, Belgium still stands on their legs?????

I find the panic these SF AARs generate to be rather hilarious. So, you think that a hallmark of a great AI would be superior forces in great defensive terrain collapsing at the drop of a hat? How would this very scenario play out in multiplayer? Much the same, I suspect...

 

[unmerged(65128)]

Is it just me,but (German) AI is performing catastrophicaly in offensive?

Its allmost One MONTH from atack and Holland, Belgium still stands on their legs?????


Or is it that my dark fears fulfilled,that enhanced defence by AI in Semper-Fi could make offensive much less efective ruining the gameplay? ...

Singleton Moresby and others please your opinion, urgently!

Are you saying you're actually sad that the AI is doing good? :rofl:

The french and Beneleux countries are putting up one h*ll of a fight and the germans are trying to outmaneuver the defenders. To me it's just awesome seeing this. Looks great

 

[nessin]

The invasion of France is the absolute worse possible example to use when demonstrating the AI, because there is a fundamental problem with the scenario that the AI can't handle without specific scripting. Namely, the Maginot Line. The problem that has existed in every HoI game (that modders will often fix via scripting or other factors) is that the AI must assess the threat to the border with France and garrison it accordingly. The problem is the AI can't handle the decision of what is just enough forces to hold the line long enough to swing around the north. Instead the AI devotes an excessive number of divisions along that border (thus severely weakening the thrust north) in order to compensate for the fact that most of the French army is there as well. After war is declared the AI will slowly shift up some troops and it will balance out eventually, but it takes longer because the AI still leaves a lot of troops bordering the forts.

The problem is, from a resolution perspective, is that without specific scripting, the AI can't be programmed to make that decision like a human could. Thus you have this situation which will always be a problem, and judging the AI for something that is beyond the scope of a modern game AI is ridiculous.

 

[unmerged(18239)]

Chapter eight: A moment of glory for the BEF (and Monty)​

As the II corps near Den Bosch fights of another heavy armoured attack Monty devises a daring plan. Strike east, with an armoured column through Eindhoven, capture the bridge over the Meuse at Venlo, and the Roer a few miles beyond. Then move on and breach the Rhine striking for the heart of the Reich. While putting down his plan before high command he also asks for para-support. As there is none, he can have none, too bad for him!

Against all odds Monty's IV corps succeeds in driving the German forces from Eindhoven and enters the city, celebrating its liberation.

To the north, the situation deteriorates and even the RAF can't hold the Germans from pounding II corps into submission

By June 9th Den Bosch has fallen to the Germans, swiftly followed by the fall of Eindhoven two days later. Serious problems ensue as the BEF is pushed out of Dutch Brabant and embattled in Belgian Brabant as well. The III corps reserves are moved in to defend Liege, and even attempt a counterattack, but have to return to the defensive quickly.

An overview of the entire front learns us the Germans have a lot of troops in reserve areas, way back from the front. Next to that the Dutch Waddeneilanden have a massive stack of divisions doing nothing. The gap near Belfort could have been easily exploited as troops are available in abundance, but are not. I have no idea what the AI is doing here but it isn't as effective as it should or could be.

 

[cougar46]

Disappointing to see the German AI being a bit passive, I see a couple of spots ripe for exploitation. This is the 1940 scenario though correct? Hopefully giving the German AI those 4 years to build their forces will make a difference in their tenacity.

 

[Jorath13]

If this is the new SF scenario, perhaps it needs some more tweaking? Perhaps the German AI can be given some scripted events or such to help push it along in the direction it needs to go. I am rather happy though that the defensive AI is as good as it is. My initial fear was that the game would become WW1 redux...

 

[S0ny B1ack]

If this is the new SF scenario, perhaps it needs some more tweaking? Perhaps the German AI can be given some scripted events or such to help push it along in the direction it needs to go. I am rather happy though that the defensive AI is as good as it is. My initial fear was that the game would become WW1 redux...

Check out Darkrenown's handoff AAR, Germany was successfull on all fronts. So I don't think that the German AI needs scripted events...

 

[unmerged(18239)]

Disappointing to see the German AI being a bit passive, I see a couple of spots ripe for exploitation. This is the 1940 scenario though correct? Hopefully giving the German AI those 4 years to build their forces will make a difference in their tenacity.

I guess it will. As it is now they don't have enough troops and don't know how to circumvent the Maginot fortifications. At the moment I am slowly being beaten back to the coast though. Not historical perhaps but quite enjoyable!

 

[unmerged(196349)]

I guess it will. As it is now they don't have enough troops and don't know how to circumvent the Maginot fortifications. At the moment I am slowly being beaten back to the coast though. Not historical perhaps but quite enjoyable!

You do of course always pass over the main advantage the human player has that most people seem to forget- foresight.

 

[cougar46]

You do of course always pass over the main advantage the human player has that most people seem to forget- foresight.

Well the problem is the German's don't seem to be taking advantage of a few soft spots and have quite a few reserves hidden in the back which means they're playing defensively, more then likely because they feel they don't have the means to successfully crush the Allies. The fact Mosby is playing UK I don't think has much influence here.

I think the problem is Paradox went with an historical OOB which in reality gives the Allies an advantage on paper. They may have to cheat a bit and slash some of the allied units in order to achieve anything even remotely historical. The German Invasion of the West unfortunately was successful for reasons that can't really be implemented in the game.

 

[unmerged(148761)]

I find the panic these SF AARs generate to be rather hilarious. So, you think that a hallmark of a great AI would be superior forces in great defensive terrain collapsing at the drop of a hat? How would this very scenario play out in multiplayer? Much the same, I suspect...

Don't you remember the 1.0 release? The early AARs showed completely bizzare game events, but everyone dismissed them as being just one game, etc. Then the game came out and we realized the Russians really would just let Finland walk all the way to Moscow, China would occupy Korea in every game, etc.

As far as I'm concerned, though, this just shows that the scenario set-up, rather than the AI, is bad.

 

[unmerged(105989)]

Aside from the Germans not being as aggressive as they could be, I haven't seen too much bizarre here. Rather, this appears to be a fairly plausible layout of how things would have gone down without the sickelschnitt, or if it had failed.

 

[XHR]

Don't you remember the 1.0 release? The early AARs showed completely bizzare game events, but everyone dismissed them as being just one game, etc. Then the game came out and we realized the Russians really would just let Finland walk all the way to Moscow, China would occupy Korea in every game, etc.

As far as I'm concerned, though, this just shows that the scenario set-up, rather than the AI, is bad.

In my opinion this AAR shows how effective a human-supported defensive AI can be. I will try a hands-off 1940 campaign just to see the results without human intervention.
Could be interesting

btw. this also goes for the full hands-off AAR by DarkRenown where the SOV got beaten by the Germans after over 3 years war. Image a capable human player defending along with the AI - this could really make the difference. Isn't this what we all want to see? Us, the armchair general, making all the difference in WWII?

 

[unmerged(162341)]

Are you saying you're actually sad that the AI is doing good? :rofl:

The French and Beneleux countries are putting up one h*ll of a fight and the germans are trying to outmaneuver the defenders. To me it's just awesome seeing this. Looks great

As a French I must say it will be good if in real WW2 allies in 1940 performed so well.

But historicaly Germans were so fast that only partialy French troops and BEF managed to establish combat contact and form line, so fast they were that even forming of clasical frontline was very difficult.There is still allive my relatives which storries could confirm this.
They were send to some forward line where Germans were to be expected according to inteligence to form defence, but Germans were not anymore there-but here-NOW and atackin french troops in preparation,actualy usualy some german tanks were usualy allready flanking them!!!
A speed not seen ever before in history.
Germans in this AAR are not performing anything of that- and exactly THAT was only way that could enable them victory..
Its of course truth that if Germans were not so fast,and if Allies were not so clumsy and abitious in fast engaging them in belgium war will go differently.
But in THIS game Germans cant be so fast,they can not perform large scale deep breaktroughs with concentration of armor,while defencive AI is set to form defence much better than in original HOI III,plus shortening line etcetera..

If I will command Germany I gues I will breach them and smash .

But I dont want to see game in which I am say Finland, and ww2 cant go on since German AI is allways (sometimes is realistic of course)stuck in the positional warfare with French for one year ,even more.

WW 2 was dinamic war,Shifiting the balance on the side of defenders could result in stalemates,and make offensive poor choice both for AI and player.

I rised question fiew months ago that pollishig Defencive AI in semper-Fi,and not in the sme time enabling Offensive AI to make true fast pincer moves and concentrations of armour-could result in to much stalemates.

I repeat I will behapy if I am wrong.

 

[unmerged(18239)]

I repeat I will behapy if I am wrong.

Darkrenown's handoff AAR showed the AI is fully capable of defeating the French.

 

[Zeddicus]

Next to that the Dutch Waddeneilanden have a massive stack of divisions doing nothing.

Maybe the passage from Waddeneilanden to the mainland is blocked by Allied naval forces? It used to be a pain to conquer those islands as Germany due to that.

 

[unmerged(162341)]

Darkrenown's handoff AAR showed the AI is fully capable of defeating the French.

Oh I see,yes its true,I checked there.

Than its I gues OK.

I gues in your AAR,your BEF made the difference!

Cant wait to see myself.

 

[Gankenstein]

I rised posed question fiew months ago that polishig Defencive AI in semper-Fi,and not in the sme time enabling Offensive AI to make true fast pincer moves and concentrations of armour-could result in to much stalemates.

I repeat I will behapy if I am wrong.

It's an ongoing effort to improve the AI, innit? In this iteration, the zerg-proofing is upgraded (and, it is promised, naval + amphib ops). Very good for those of us who like to attack and prefer decent resistance (though the AI still lets the Germans most unhistorically take Moscow ).