• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

HoI4 Dev Diary - LaR UK AAR

Join me as I recount some of the highlights from my recent prerelease testing playthrough of LaR in this After Action Report (AAR) where played I as the UK.

Phase 1: The buildup

When playing the UK in singleplayer I don’t really like trying to hold France, as I feel doing so kinda ruins the pacing of a historical playthrough. So I spent all my time from game start only building civilian factories with about 85% of my mills making aircraft. I also began establishing my intel agency in late 1937. I focused on improving my intel generation and code-cracking ability first so I could give some force multiplication to my rather small army.

agency.png



Phase 2: Naval Dominance and Focused Defense

At the start of hostilities, I began cracking Germany’s encryption and used my airforce in the Mediterranean in conjunction with a large part of my navy. I figured it wasn’t worth sacrificing too much air strength against Germany in France at this point. I had just switched to building military factories from only building civilian factories, so I could not afford to use my more limited airforce recklessly. I also set up the majority of my operatives to set up intel networks in Germany.

I attempted to recruit mostly “seducer” trait operatives as they have a lower chance of being caught. In the process, I got some interesting seduction experts.

OperativeSeducerLAdy.png


Most of my rather small army was deployed in Egypt to hold the Suez. With air superiority and an intel advantage over the Italians, Holding Egypt was a great success. I was able to recruit the famous Nancy Wake and I decided to send her on a Roman holiday to help me get more intel on Italy since I would be fighting them in Africa for the foreseeable future.

Wake In Rome.png



Phase 3: Battle of Malta

After the Italian navy was largely defeated, I infiltrated the Italian airforce to help get a more clear idea of how close I was to breaking it. At this point, the Italian airforce started port striking my Mediterranean Fleet in Malta. After looking at their plane counts in the intel ledger, I built up some radar in Malta and deployed the airforce to intercept the Italians in the region. Baiting them to bomb my exposed, and no longer as useful, fleet worked as phase one of my plan to break the axis airforce.

Battle of Malta.png


Meanwhile, In Germany and occupied France, my intel networks had become rather strong and were providing good info on the state of the axis. I had at this point also infiltrated the German civilian govt’ and army to open up further options for operations and to get a more clear picture of their strengths.

By late 1940 I had broken both the German and Italian ciphers and had weakened both the German and Italian airforces by fighting in favorable conditions where I had a large radar advantage combined with my passive cracked crypto advantage. Having enemy ciphers broken increases interception efficiency as well as adds to air detection.

Phase 4: Battle of Greece


At the end of 1940, Greece was invaded by Italy and Germany. By this point, I had a significant intel advantage, was close to matching axis airpower, and had a large and equipped Free French volunteer force. I decided I would turtle southern Greece as long as I could and brutalize the axis in the air in the process. I scrambled a large part of my North African forces to Greece and deployed the majority of my airpower. At one point my defensive line was nearly broken. I was able to save it by activating my broken ciphers on Germany, giving myself a temporary 30day combat. Before the buff expired I was able to get some extra forces in and save Greece.

CrackTheCodes.png


By mid-1941 I had overtaken the axis in the air and southern Greece looked more and more secure. I decided it was time to start boosting resistance in France and laying the groundwork for eventual liberation. I also was well on my way to cracking the new Italian and German ciphers.

greece airwar won.png


Once the ciphers were cracked again and my tac bombers were no longer needed in Greece, I decided to start harassing the Germans with a strategic bombing campaign in their homeland. With my Intel levels, I was able to track how my bombing campaign was impacting Germany. I had also begun targetting resource-rich areas in France with targeted sabotage operations to further put stress on the German war machine.

StratBombingUnderway.png


Phase 5: Yugoslavian Uprising

Over the course of the next year, America and Vichy joined the war and a fight for North Africa broke out again. With Intel and Air advantage pushing Vichy France back was pretty easy. During the North Africa campaign, I noticed that Croatia was barely keeping occupied Yugoslavia under control. So I sent some of my Operatives to support the resistance there, pushing it over the edge and causing a full-scale uprising. Many of the Axis forces in northern Greece were then cut off and annihilated.

YugoRises.png


After a great victory in Yugoslavia, I dedicated my operatives to building a massive spy network across all of Germany. This resulted in several captured Operatives, as they are more likely to be discovered in large and powerful networks, but I decided it was worth it to keep my intel on Germany maxed and the mainland set up for my Arrival.

Phase 6: La Resistance and D-Day

By mid-1942 the French Resistance, due in no small part to my support, had become disruptive. It was not fully rising up in rebellion but was strong enough to disable strategic redeploy in northern France and was providing constant attrition to local Axis forces. This combined with local spy network buffs, general intel advantage, air superiority, and ongoing fighting on the eastern front made securing my beachhead in France very smooth.

ParisFallsAgain.png


After setting up a plan to drive the Germans out of France, I once again fully utilized my code-cracking for a 30day buff and battle planned the Germans back into their homeland. By late 42 The Axis was all but broken and crumbling on all fronts. The combined Allied air, land, and intelligence efforts proved to be too much and everyone was Home for Christmas of ‘42.

I hope you all enjoyed my war story! See you next time.
 
  • 1Like
Reactions:
Reason why Paradox doesn't update AI is that most players who play singleplayer are so bad that the AI is hard for them.

Obviously people on this forum would be good enough to shred the AI as they're interested, and so play more/understand more mechanics. But many people find Germany in vanilla hard lol. Just go on Steam community forums and you'll see.

So Paradox doesn't have a real incentive to update AI as those bad players will keep playing and not even notice the problems.

Given that Paradox has, on many occasions, updated the AI, I'm not sure the facts match your hyphothesis here. Indeed, Paradox's response to people finding the game difficult was to add in another easier difficulty mode (at the same time as making improvements to the AI).

As for incentive, I have absolutely no doubt that the team are highly motivated and want to make the best game they can, plus there's a decent amount of customer pressure for better AI in these parts - between them, that's a good explanation for the continuing iterative improvements in AI that are happening.
 
Given that Paradox has, on many occasions, updated the AI, I'm not sure the facts match your hyphothesis here. Indeed, Paradox's response to people finding the game difficult was to add in another easier difficulty mode (at the same time as making improvements to the AI).

As for incentive, I have absolutely no doubt that the team are highly motivated and want to make the best game they can, plus there's a decent amount of customer pressure for better AI in these parts - between them, that's a good explanation for the continuing iterative improvements in AI that are happening.
Seriusly? People know germany vanilla difficulty? What!? I play germany in very easy when i want relax my head! If i want a challenge i do mod!
 
Reason why Paradox doesn't update AI is that most players who play singleplayer are so bad that the AI is hard for them.

Obviously people on this forum would be good enough to shred the AI as they're interested, and so play more/understand more mechanics. But many people find Germany in vanilla hard lol. Just go on Steam community forums and you'll see.

So Paradox doesn't have a real incentive to update AI as those bad players will keep playing and not even notice the problems.

The small size of the HOI team is surprising to me and most likely contributes greatly to the AI problem. If the suits at Paradox see it like you say, then that size may never grow sufficiently.
 
As for incentive, I have absolutely no doubt that the team are highly motivated and want to make the best game they can, plus there's a decent amount of customer pressure for better AI in these parts - between them, that's a good explanation for the continuing iterative improvements in AI that are happening.

I do agree here. The size of the team is less than it should be, in my opinion. and the size of the customer base must stretch them in more directions than I care for.
 
As for incentive, I have absolutely no doubt that the team are highly motivated and want to make the best game they can, plus there's a decent amount of customer pressure for better AI in these parts - between them, that's a good explanation for the continuing iterative improvements in AI that are happening.
Issue with this theory is that Paradox's HOI4 dev team doesn't even have a dedicated AI designer. So the pace of AI improvements will be slow, and new mechanics will keep breaking it. If Paradox wanted to rapidly improve the AI, they could hire or transfer a AI developer to the HOI4 team. It's not the dev team's fault, but the company and its management. Where they value sales over game quality.
 
I follow the news about HOI4 semi-regularly in order to get an idea about the current state of the game. I've not played it yet and I know that HOI games require a large time investment, so I want to make sure that this game is in a really good shape before I pick it up. While the features in this expansion look quite cool, the AAR paints a rather sad picture of the overall game balance and the quality of the game's AI. In a British AAR focused around showcasing various new features, with several decisions that might seem rather questionable (like not defending France), ending the war with a decisive victory in 1942 is really not something that I would expect. The comments here are not encouraging, either.

So yeah, back to the wishlist with this one. Maybe one day...

Long time no see man :cool:

I had my fun with this game since 1.1 but the AI issues are the reason I´m dissapointed with the game. I´m playing other stuff like Mechwarrior 5 until the new group of patches.
 
Issue with this theory is that Paradox's HOI4 dev team doesn't even have a dedicated AI designer. So the pace of AI improvements will be slow, and new mechanics will keep breaking it. If Paradox wanted to rapidly improve the AI, they could hire or transfer a AI developer to the HOI4 team. It's not the dev team's fault, but the company and its management. Where they value sales over game quality.
I think it goes way beyond this. Whether you have a dedicated/expert/whatever AI designer, it takes a great deal of time and effort to create an AI that efficiently finds a good strategy for any complex game (such as HoI)...

AND...

...if someone succeeds in creating an AI that produces an "optimal strategy for (a) WW2 (game)", then that constitutes proof that it is a poor WW2 (strategy) game.

Quite simply, WW2 was a vast and endlessly complex struggle. We are still finding out new stuff about the details of what actually happened and why, never mind assessing a "perfect strategy" that should have been followed by any of the sides involved. So, the creation of a really good, absorbing WW2 game and the creation of a really "competent" AI are actually in conflict with each other. That's the scenario that I see the design team as having to overcome - and I don't expect it to be quick or easy.
 
I think it goes way beyond this. Whether you have a dedicated/expert/whatever AI designer, it takes a great deal of time and effort to create an AI that efficiently finds a good strategy for any complex game (such as HoI)...
I agree, but the AI now is pretty wonk. It doesn't understand game mechanics.
 
I agree, but the AI now is pretty wonk. It doesn't understand game mechanics.
Of course it doesn't 'understand' anything - it's a machine. It can only respond, or learn to respond, to specific scenarios, or try to project the benefit of specific priorities and possibilities that are assigned to it by a human creator. That's what today's "AIs" do :).
 
On that note, it's worth wondering why there has been no attempt (afaik, and I hope I would know) to test out basic evolutionary programming as an "AI trainer". I mean, I well understand why Paradox do not use it, but I'm surprised that the large and capable modding scene have not put out a test implementation.

Really, it could be fed relatively small fields as proof of concept, and the significant work would just be a breeder script much more straightforward than some of the things various Paradox mods have in their arsenals.