Game too easy

[Mannstien]

11 days to find out if I need to nerf myself

 

[KevinG]

Not allways, I followed some Hoi3 MP groups, as they posted their campaigns on this forum and some guys cheased the game and the other players alot. Thats not fun for me. Without house rules you have allways guys who wants to outwit you.

This is why you need to do some research before playing MP. If you take a look at the current MP groups in the HOI3 multiplayer forum, each group has a rules list 2-3 posts long. It may seem excessive, but once everyone is on the same page on "exploits", the game becomes much more enjoyable. I know that all I need to do to beat the UK as Italy in SP is build a 5 carrier stack, let the AI suicide its ships at me in the med, and then do a sealion with a few corps once the RN is dead. The AI will fall for it every time and it can't effectively coordinate its fleets or airforce. Its boring.

In my current MP game as Italy, the battle for Malta lasted an entire session. Literally over a hundred aerial battles were fought over Malta to keep the skies clear for my bombers to bring the port and infrastructure down to 0. Then once the garrison was out of supplies, the German player launched a massive paratrooper assault. The UK player threw the entire RN at Malta and I sent 10 carriers + the entire Luftwaffe to intercept them. Despite being outclassed at sea by the UK's 10 CVs + 5 CVLs and 15 BBs, Axis air superiority over the med allowed us to weaken the RN's carriers before the Italian carriers even got into range which resulted in a decisive Axis victory. The UK player then sacrificed most of his battleship fleets to keep my carriers busy while his carriers escaped.

There's no way the AI will ever have the ability to plan such a battle out. There's an almost infinite amount of variables for it to take into account that makes it not feasible with our current way of scripting gaming AI. If you want the full HOI experience that makes you feel like you're playing a strategy game and not solitaire then you have to give MP a try.

 

[blue_yonder]

It's a shame there has never been a DD on how they intend to approach difficulty levels this time - it's the one major component we know very little about. The last stream did indeed look like my idea of a 'Normal' setting; i.e. a competent player gets an entertaining game with no real headaches. Personally I hope that harder levels aren't designed solely around combat debuffs, but are made from a mesh of all the game's elements: longer foci and research, fewer PP, and so on. By spreading out the debuffs, no single one of them will seem artificially unfair and onerous. I wasn't thrilled to see Switzerland gobbled up in two minutes, but on reflection I didn't fully appreciate the speed Daniel's pc was running at, and I also think/hope/speculate that on a harder level, this suicidal undermining of the Maginot you've been building since the Great War would promptly get what it deserves.

 

[Salatmander]

(to lazy to read through all 7 pages if this was mentioned before, sry there :3 )
But... to me a difficulty level shouldn t regulate mali given to the player but influence how well the AI does or how good it is in exploiting your strategies...
That every Grand Strategy game only seems to regulate the "difficulty" in how much "income" or "happynes" you get but lets the AI be as stupid/exploitable as before is one of the few things i really hate in this genre.

(Anyway i know that its really hard to code a different/"better" AI again, but wanted to point that out)

 

[blue_yonder]

a difficulty level shouldn t regulate mali given to the player but influence how well the AI does or how good it is in exploiting your strategies...

Your great grandchildren will play HOI25 in just this way.

 

[Seen207]

As others have pointed out: There are systems in the game designed to create strategic decisions for the player, such as the weather system. Danial *completely ignores* the weather system as if it doesn't matter. Is he playing sub-optimally? Why would a player who has thousands of hours in the game be playing sub-optimally? The concern is that the weather system doesn't have enough impact to represent historical events, and so a player can just ignore it. In many other games I have played on this scale, attacking in the mud/snow is almost useless and always dangerous.

He also doesn't seem to care about terrain. The only concern they've ever shown for terrain is "let's put mountain troops in the mountains." That's great and all, but there are a whole slew of terrain effects for different types of units. My immediate thought when the fight with spain broke out was "he should be moving his armor along the coast to avoid the worst of the mountains" but he just let the AI blast through the mountains with tanks, presumably taking much more attrition than should have been required.

Should terrain matter? I think it should, but since their best player just ignores it, that seems to imply the game isn't properly penalizing you for not playing optimally.

If Daniel seems to make it look easy (crushed china in 2 years when historical Japan couldn't do it in the entire war; held poland for 4x longer than historical; Took over all balkins with no real resistance as Hungary; Fought a 3 front war with France with no issues), but I'm spotting all these mistake's he's making, the game will likely be too easy for me. MP might be my only hope until the AI can provide a challenge. But I am optimistic that PDX (and modders) will continue to improve the AI to the point where it will be able to provide a challenge.

I believe that the only time HOI3 was any difficult was with mods especially black ice; and anyway in HOI3 vanilla you can ignore the weather and terrain mostly except for mountains and marsh so I don't understand why we are holding HOI3 as the Gold standard for difficulty. Sure it had a steep learning curve but once you knew the game it was incredibly easy to exploit. It seams that Daniel is using this same mentality in HOI4 and while this is concerning, from what I have seen HOI4 at this moment is a solid base game with definite flaws but that I look forward to being built upon with new and interesting mechanics like EU4 and CK2 before it and I am excited to sink my teeth into a new grand strategy game.

 

[Greenboot]

I don't mind a casual fun game. And would enjoy harder modes to have a challenge. We could also see some mods that make additional challenges?

 

[angelosnau]

I'm quite optimistic that the game will be well balanced.Every other paradox game I've played was well balanced (usually played on hard-ironman).

 

[Murmeldjuret]

I thought the AI looked promising.
Some points:

• The AI actually used encirclements, cutting off armies, and thrusts, and has so for the whole WWW series. It already looks better than most games I've played.
• Nations are not supposed to be ready for the Fascist nations sudden and aggressive diplomacy. That is the whole reason there is a WW2 and HoI games. Doing so increases world tension and will enable the allies, especially US go into war-economy, and as an Axis player that is going to be a problem. It is weird France can just invade Switzerland, that should definitely require more work to perform. I mean they follow the armed neutrality policy, though that has nothing to do with AI or easy, but with diplomacy being to flexible.
• It was France that started the war, attacking a numerically inferior spanish with superior equipment. The Spanish held the mountains well, but as soon as those fell it was all over. Again, if you have superior numbers and equipment you should win. Though it is strange that all the Pyrenees were part of the attack, as I would send most troops through the coast, but the Spanish were vastly outnumbered so it might be why it worked so well regardless.
• I have never played a game except for simpler numerically solvable games such as chess where the AI is close to challenging without penalties to the player. Even those games where they hail the AI as hard, it never is without penalties. Though it usually is better policy to buff the AI rather than penalize the player, as you have to use different build orders for different difficulties because the difficulty penalties do not translate uniformly. As an AI researcher, my estimate is that actually AI difficulty is at least a decade away.
• Infantry spam sounds pretty much like what actually happened. The bulk of all the war material was still WWI tech during the early years of the war. I mean, there was AT, turreted tanks, airplanes, rockets, and so, but a vast majority of all men were still riflemen. The main german weapon, Karbin98K is a modified version of a rifle from 1898. The Russian used modified Mosin-Nagant, which is from 1891. It is only in the later half of the war that you actually start to get things other than cheap infantry spam. It feels represented in game.

 

[Starisc]

What is the purpose of the WWW stream?
My guess are the would be something along the lines of this:

1) Show off as many features as possible in a limited time
2) Demonstrate effective UI and information flow so hardcore user are not frustrated with lack of control or influence
3) Stress that the game is "fun" to play (difficult word in game design, but compared to HOI3, which leans more closely to "work" and still deeply satisfies player even without a classical definition of easy "fun)

In all cases demonstrating how difficult the game and AI could be would be detrimental to these goals. @Da9L would have to revert even more to exploit and "cheese" tactics to win, focus on optimal builds which reduce the width of shown features (goal 1), stress Daniel with loads of intensive micro (goal 3) and give the commentators less time to explain UI prompts, information in dialogs etc (goal 2).

Hence, @cKnoor is right to pit the best player in the world against a normal AI and have a good time for commentators, Daniel and ultimately the viewers.

If we clamor loud enough we can still convince them to do a ultra hard playthrough stream after launch.

So far I will lean back, enjoy the show and try to restrain my anticipation until Deadline Day.

@podcat , keep at it and good luck in the last couple of super busy days with the bughunt. I hope the crunch pizza tastes well and the unforeseen last minute catastrophes are resolved by your dedicated team.

 

[Doctoxic]

What is the purpose of the WWW stream?
My guess are the would be something along the lines of this:

1) Show off as many features as possible in a limited time
2) Demonstrate effective UI and information flow so hardcore user are not frustrated with lack of control or influence
3) Stress that the game is "fun" to play (difficult word in game design, but compared to HOI3, which leans more closely to "work" and still deeply satisfies player even without a classical definition of easy "fun)

to play devils advocate here it could be argued to some extent that it failed in all 3

1) it failed to show that other countries had an awareness of the political situation and some things were just dumb like GER attacking the Maginot Line

2) there seems to be frustration with the AI in that manual intervention is often needed

3) its no fun to me watching "seemingly" dumb and unrealistic things (see previous threads) happening just so they can paint the map red

OK, so these comments are deliberately overstated (and ignore the huge amount of positives that this game has) just saying that whilst i have enjoyed the WWWs they all show things that, to me, don't look "right", particularly bearing in mind how long the game has been in development

Having said all that really i am really looking forward to the release - even more than Stellaris

 

[MateusMat]

This is exactly the problem people are having. You're judging the game based upon watching someone who created the game playing it. Someone who has more experience playing it than the majority of us will have in the first 6 months. Which is probably the worst possible way to judge a video game. Don't believe me? Go watch practically any skill based game on twitch and tell me the professionals don't make the game look easy.

Judging a games' difficulty based upon videos alone just doesn't work, even more so when those videos are by the creators of said game. I can understand why the concern arises, but use some logic and realize that these aren't your average gamers playing.

A few days ago I saw my friend playing Dark Souls 3... I had never played, or even watched, Dark Souls before. So I was quite surprise that, that was the game everyone talked about being the most difficult game ever. It looked so easy... just a normal RPGesque adventure.

Boy I was wrong... I felt like a 5 year old with a controller.


People here are projecting the skills of someone who played for more than 1000 hours... onto themselves just because they watched him play for only 20. I wanna see the people saying the game is easy to do a twitch stream on D-Day playing as 1939 Poland. I wanna see how easy they think the game is them.

 

[Bridger15]

My only question is why is Hungary capable of invading and completely annexing Austria in 1936 without any international (let alone) German reprisal? How does anything we've seen actually attempt to faithfully recreate the political climate of the mid-20th century?

Because what you describe requires the coders to anticipate such things and script in AI reactions. Obviously they either hadn't had time to add in that reaction or hadn't thought to do so until after that stream. It's very difficult to consider every wild ambition of every player.

 

[Bridger15]

to play devils advocate here it could be argued to some extent that it failed in all 3

1) it failed to show that other countries had an awareness of the political situation and some things were just dumb like GER attacking the Maginot Line

It's worth noting that the GER AI does *not* have omnipotence. It actually doesn't know how well the maginot line is defended until it actually attacks it (unless it has far superior decryption, which is very unlikely at that point in the game). The AI made a calculated risk similar to what a player would do. "Hey look, france is fighting on two other fronts, I wonder if they left the Maginot line weak?" The only way to find out is to attack. It could be argued that it continued to attack for too long, but attacking in the first place is the only way to be sure. It did stop attacking eventually, and immediately went about justifying war on the low countries (via national focus) in order to open up the front and make use of it's superior numbers.

Of course, it also left the line hilariously weak, which is probably something that needs tweaking.

 

[amalric de g.]

This is why you need to do some research before playing MP. If you take a look at the current MP groups in the HOI3 multiplayer forum, each group has a rules list 2-3 posts long. It may seem excessive, but once everyone is on the same page on "exploits", the game becomes much more enjoyable. I know that all I need to do to beat the UK as Italy in SP is build a 5 carrier stack, let the AI suicide its ships at me in the med, and then do a sealion with a few corps once the RN is dead. The AI will fall for it every time and it can't effectively coordinate its fleets or airforce. Its boring.

In my current MP game as Italy, the battle for Malta lasted an entire session. Literally over a hundred aerial battles were fought over Malta to keep the skies clear for my bombers to bring the port and infrastructure down to 0. Then once the garrison was out of supplies, the German player launched a massive paratrooper assault. The UK player threw the entire RN at Malta and I sent 10 carriers + the entire Luftwaffe to intercept them. Despite being outclassed at sea by the UK's 10 CVs + 5 CVLs and 15 BBs, Axis air superiority over the med allowed us to weaken the RN's carriers before the Italian carriers even got into range which resulted in a decisive Axis victory. The UK player then sacrificed most of his battleship fleets to keep my carriers busy while his carriers escaped.

There's no way the AI will ever have the ability to plan such a battle out. There's an almost infinite amount of variables for it to take into account that makes it not feasible with our current way of scripting gaming AI. If you want the full HOI experience that makes you feel like you're playing a strategy game and not solitaire then you have to give MP a try.

Exactly this. What you describe, in your example, is a normal and fun mp game. The AI in Hoi 3 is a push over. Thats the reason, when i played Germany, i allways quited the game after i conquered the SU, only one game i invaded the US and even that was a cake walk, my battle hardened Panzergrenadier and Tank Divs mowed the US Divs like grass.

I talked about really cheasing other players, example one guy played Canada, he tech rushed fighters to 1942 tech, even before the war started and build only fighters and put them all into southern UK, the german players Luftwaffe got destroyed by his Airforce alone and the game was over in 1941 as the Allies stomped Italy and Germany. That was for sure a real fun game for the US and SU players.

 

[FieldMedic]

I guess we'll all see for ourselves just how hard or easy the game is in about a week!

 

[anam]

It's worth noting that the GER AI does *not* have omnipotence. It actually doesn't know how well the maginot line is defended until it actually attacks it (unless it has far superior decryption, which is very unlikely at that point in the game). The AI made a calculated risk similar to what a player would do. "Hey look, france is fighting on two other fronts, I wonder if they left the Maginot line weak?" The only way to find out is to attack. It could be argued that it continued to attack for too long, but attacking in the first place is the only way to be sure. It did stop attacking eventually, and immediately went about justifying war on the low countries (via national focus) in order to open up the front and make use of it's superior numbers.

Of course, it also left the line hilariously weak, which is probably something that needs tweaking.

And i'll add a bit more to this The Ai doesn't even know it's the maginot line. Probably all it knows is that it's a fortified province - if it even knows that - and that there is an uncertain number of opposing forces there....so it's should try to attack until it goes above it's "acceptable losses limit for a single province" weigth.

Now why is that. Well if you hardcode it to not attack the maginot line it might end up never doing it regardless of actual forces. Thats bad.
And it doesn't deal with any other fortresses it can come accross either. Hardcoding them all is just silly because new might be buildt.
So a better solution is to let it actually attack until it reaches a certain limit....and that also let it deal with any other ones with the same lines of code.
Problem ofc is that in this instance it's a line of defences 10 provinces(or so, don't recall exact nr atm) wide and it have to test them all. So naturally there is quite a lot of losses before it decides to call it of....
And i don't see how to avoid this unless you let the AI cheat and always se how how many forts are in a province togheter with number of divisions...


NOw the second part there is ofc that it should have gone through Switzerland. But that that's mountains area and it probably has some weight against that too.....so there might be an unbalence between fort attack and opposed mountain attack. But as i said it might not know that there are forts(or how many anyway) and then it makes sence for it to test the "plains but forts and forces" attack before the "mountains with forces" attack

And i don't think it decided to go around , so it picked the Focus it did.....i'm pretty sure it picked that focus as soon as it was at war with france(after finishing the current focus it was doing). As the focuses it's picks are scripted.

 

[KevinG]

Exactly this. What you describe, in your example, is a normal and fun mp game. The AI in Hoi 3 is a push over. Thats the reason, when i played Germany, i allways quited the game after i conquered the SU, only one game i invaded the US and even that was a cake walk, my battle hardened Panzergrenadier and Tank Divs mowed the US Divs like grass.

I talked about really cheasing other players, example one guy played Canada, he tech rushed fighters to 1942 tech, even before the war started and build only fighters and put them all into southern UK, the german players Luftwaffe got destroyed by his Airforce alone and the game was over in 1941 as the Allies stomped Italy and Germany. That was for sure a real fun game for the US and SU players.

That's why MP groups also have a GM that makes balanced teams. Canada tech rushing fighters shouldn't be a problem because an experienced Germany should also always be tech rushing fighters. There's a whole MP competitive metagame to HOI3 that is fairly balanced once everyone in the group learns it.

Unfortunately the only solution to letting the US player have some gametime is making Japan DOW USA when they DOW the allies as part of the rules. Judging from what I've seen so far I think it'll have to be the same in HOI4.

 

[amalric de g.]

That's why MP groups also have a GM that makes balanced teams. Canada tech rushing fighters shouldn't be a problem because an experienced Germany should also always be tech rushing fighters. There's a whole MP competitive metagame to HOI3 that is fairly balanced once everyone in the group learns it.

Unfortunately the only solution to letting the US player have some gametime is making Japan DOW USA when they DOW the allies as part of the rules. Judging from what I've seen so far I think it'll have to be the same in HOI4.

Sure, but Germany or Italy can´t compete tech wise, if the players of Canada, France and UK gang up, i only talked about the Canadian player, the French one tech rushed ATs and the Brit one tanks. They deployed british Tank Divisions and French Inf Divs on all river provinces and thats game over for Italy and Germany.
Also Italy has no chance against this in Africa or on the French Border.

 

[ArcandSpark]

It's a shame there has never been a DD on how they intend to approach difficulty levels this time - it's the one major component we know very little about. The last stream did indeed look like my idea of a 'Normal' setting; i.e. a competent player gets an entertaining game with no real headaches. Personally I hope that harder levels aren't designed solely around combat debuffs, but are made from a mesh of all the game's elements: longer foci and research, fewer PP, and so on. By spreading out the debuffs, no single one of them will seem artificially unfair and onerous. I wasn't thrilled to see Switzerland gobbled up in two minutes, but on reflection I didn't fully appreciate the speed Daniel's pc was running at, and I also think/hope/speculate that on a harder level, this suicidal undermining of the Maginot you've been building since the Great War would promptly get what it deserves.

That would be a really good way of making the game harder at higher difficulties so that you can't just stick with the same strategy or general idea through every difficulty. The main points being longer foci times, less PP/MP/the other points that I can't remember right now, and longer research times.