HOI4 - Development Diary - 12th of August 2016 **read OP edit**

[podcat]

Friday again and time for another update on the 1.2 patch (Sunflower).

Custom Difficulty
First up, despite some sickness we have managed to add a neat feature that I hope is going to help bring more challenge to the game (in combination with the AI improvements). There is now a game setup screen with custom difficulty settings for AI.

In the above screen you can buff certain nations and make them stronger, so for example as Germany if you want a tough soviet opponent you can pull up the "Strengthen Soviet" slider. The actual bonuses are a bit work in progress at the moment and you might see some changes after we have had time to playtest more.

By the way the system is easily moddable for adding more settings. Settings can work with any regular modifier bonuses and you can specify which nations you want to apply them to, so making a "Strengthen european minors" slider is possible, or a slider just to control american industry etc.

AI
Time to hand the mic over to @SteelVolt to give us an update on AI:

Some of you have voiced concerns over certain aspects of the AI, and we have heard you. In the upcoming patch you can look forward to things like an AI not sending 75% of its army to fight in Africa as well as an AI that is able to efficiently swap out old templates for new ones, both for units in training and units in the field. For that last one it turned out that it was not only incapable of changing light armor to medium, but it only considered a single division every 14 days for upgrade. Needless to say, this was a massive handicap, particularly for larger countries with big armies. At this very moment I am working on making this progression fully scriptable so we can have much better control over how the AI evolves its division templates.

Apart from this there are of course minor tweaks and fixes here and there being done also (like managing straits and choke points better).

Next week I should be able to give you a rough date for release of Sunflower. See you then!

Edit

btw after wading through a few pages of hostility, it stops now. if you cannot post in a constructive manner and on topic, don't post.

 

[podcat]

These custom difficulty settings, are they achievement-friendly?

Some achievements where you say have buffed your allies will be disabled, but making your enemirs harder is ok

A new custom difficulty setting and some AI improvements... I mean, that's all good, and AI stuff is very important (crucial, even) but it doesn't seem very ambitious for the next full patch. When will we start seeing player feedback really being implemented?

Well, the top 2 items of feedback i see on the forum is improvement of AI and higher difficulty, so i think we are focusing pretty well on what the community wants so far. Do keep talking about what you want though, we are listening!

 

[Darkrenown]

You mean you give us a cheat slider to buff the AI with wild boni since you cant get the AI to work as it is supposed to be? That does not sound too good.

Reading between the lines...

I'd say you haven't thought about the difficulty settings fully. With a blanket difficulty slider you have problems like: You play as the UK, you want a challenge and so set the game to Very Hard. You now have some penalties and the AI has some bonuses....all the AIs have bonuses, including your likely allies and sorta allies, the USA and SOV which effectively and counter-intuitively make the game easier in some ways. With custom difficulty sliders you can solely boost your likely enemies when you want a challenge.

 

[podcat]

Thank you for your hard work and I hope everyone feels better soon. Good excuse for hot toddies. I recommend either Wild Turkey 101 or Blanton's, you will either feel better or won't care.

While i am sure these issues are on your to do list an idea of when they might be tackled would be appreciated:

Invulnerable and overpowerful Port Strikes
Invulnerable supply and oversea colony resource convoys
Naval production
Naval combat balance
AI partisan suppression

These are issues that are difficult if not impossible for modders to tackle and directly affect play.

May I also suggest you look at the Enhanced Peace Conference mod? It really helps prevent peace conference silliness.

We are working on several naval issues but i also do not want to delay the patch too long either, so some stuff may need to wait for 1.2.1 to make sure we keep the fixes flowing

 

[podcat]

@podcat. Looks great. A big question - will the MP system be tweaked ? (i.e. taking allocation to non-combat units into acocunt and avoid monior division spamming) ?

Cheers,
A

Not short term, but i see the amount of divisions you can maximally field as an issue. I think the solution lies in upkeep cost rather than manpower

 

[podcat]

I'd like to suggest a button which can collapse the difficulty slider window and keep it closed/open for every new game's country selection depending on what its last state was. I'd recommend a arrow which points either up or down depending on whether the window is open or not.

you can toggle its opening with a button thats below the tooltip in the screenshot

 

[Darkrenown]

Wow. I seem to remember somebody at Paradox once having a dig at the Creative Assembly guys about the Total War AI being a joke and creating difficulty with buffs and bonuses for the AI...

You are misinformed, we have had difficulty sliders in every game at least as far back as EU2 where I started playing these games, and in every case they involved giving the AI bonuses and the player a penalty for the harder levels. At some point EU4 split them into a difficulty slider and a handicap player/ai/none setting, which is the only exception I can think of.

 

[Darkrenown]

Do you also believe Coke adds life, and Bud Lite will make you better at volley ball?

Thats ridiculous. If the AI was competent, your allies are not going to be super awesome while the enemy is only competent. All the AI is the same. If you increase the AIs ability, it can handle the AI, and is more of a challenge to the player.

Ugh.

This HOI team is starting to piss me off.

Stellaris and CK and whatnot is making this team look like it has no idea what its doing and wont be able to get it together to fix the problems we have. Not just not in the short term, but at all.

Sliders?

Really?

Sliders? This is your solution to the woefully inadequate AI? Mess with some .laus and add sliders?

You need to get into the .EXE paradox, and fix the bloody AI.

The Allies tend to have more of everything than the Axis, if you increase the all by the same percent then the Allies gain more absolute numbers. There's no trick or marketing claim, just maths. FWIW Faction or nation specific bonus settings have been something Podcat and I talked about back when we did the HoI3 expansions, and as other have mentioned it's apparently something the BlackICE mod does already.

I'm also not sure if you just replied after reading the first half of the DD, but you should be able to see Steelvolt talking about the other AI stuff he is working on so there's no need to act like the sliders are the only change happening.

 

[Darkrenown]

Where were the difficulty sliders in HOI3?

And uh:


http://europauniversalis.wikia.com/wiki/Difficulty_level_(Europa_Universalis_II)

No there arent.

Please stop lying to people. There are no sliders.

HOI2 had no sliders.

The only sliders n HOI2 were for government. Like Hawk vs Dove. It was the same Easy, Normal, Hard, Very hard from every other game. There have never been sliders for difficulty.

Uh, what? Like, are you claiming there is some difference between having difficulty on a slider with X settings and having a easy/normal/hard/very hard toggle? They are just the same thing with a slightly different UI. HoI3 had a difficult setting on the country selection screen, that EU2 article you linked clearly talks about the difficulty levels and the AI bonuses/player penalties they set.

 

[Darkrenown]

And you should know that from community feedback that the lucky nations modifier that exists in EU3/EU4 is not very well liked. Abstract penalties/AI benefits is just lazy implementation

There's an option to disable them, just as you are free to not use these sliders though?

Sliders.

Toggle.

Sliders.

Toggle.

Sliders.

Toggle.

Read what you wrote again. The same thing but different? Is that a real statement? Im not trying to be a jerk here, but come on, calm down and try and remember were on the same side here.

Either way, the point is still, that no one likes toggles, or sliders, that come with bonuses and maluses to cover up bad AI routines.

I have no idea what you think is the difference between a slider and a toggle is for difficulty levels, but practically speaking it's just a slightly different UI to select a setting. You could also have a drop down menu or a list with radio buttons. Would it help to do back to my post and imagine I said difficulty setting instead of difficulty slider? The post I replied to didn't mention sliders or toggles, just the concept of buffing the AI for difficulty.

"The same but different" isnt the same. Thats non-sequitur.

Lucky I didn't say that though, I said they were the same thing with a slightly different UI: Which is exactly what they are.

 

[podcat]

This and buff signal companies as well, which will provide help to Grand Battle Plan doctrine which is clearly the 2nd weakest of the doctrines.

I'v buffed signal companies for this patch yea (cheaper and better bonuses)

Are you aware of the issue where AI nations will task divisions across the lines to enemy ports, heedless of success or failure?

yea, its on our list of important issues

We just want you guys to recognize that the fundamental AI is not adequate. Ill wait until hell freezes over if you guys tell me thats the goal.

Did you read this dev diary, or the previous one? both state that improving ai is our most important task.

@podcat and co
Are some of you at Gamescom next week? Would like to buy you a beer as thx for hours over hours of fun i had with this game

Thanks, but the team is staying in stockholm to work on the game this gamescom. As good as beer is, delaying a patch wouldnt be worth it.

Good to hear. I like the idea of the custom difficulty.

This may have been answered already, I'm too lazy to go through the thread, but will the custom difficulty option work in conjunction with the standard difficulty or as a replacement? For example could I set it to Recruit but have one of the enemies buffed or would I need to choose one or the other?

It works in conjunction with standard difficulty. The standard one only affects player, these affect other ai nations.

It is not about sliders in general. Yes, I despise of them for various reasons. But developers seem to like them. It is about sliders now. No one called for them. Show me a single thread

I saw a lot of posts from people saying they had a good time playing with the difficulty improvement mods around (the most popular trio here on the forum seemed to be making africa impassible, harder difficulty and ai tweak mods). The interfaces for those mods and implementation are not as good as they could be though because they are only mods and has to be tacked on in unnatural ways (like giving player a bunch of events on startup with choices).
Thats whats behind the decision to include this.

If you dont like difficulty sliders just keep playing on "Regular" difficulty and dont set any of the custom ones, nobody is going to force you. I think there are a lot of people who will like them and play with them though.

 

[Havebeard]

A new custom difficulty setting and some AI improvements... I mean, that's all good, and AI stuff is very important (crucial, even) but it doesn't seem very ambitious for the next full patch. When will we start seeing player feedback really being implemented?

You mean you give us a cheat slider to buff the AI with wild boni since you cant get the AI to work as it is supposed to be? That does not sound too good.

Personally I hate balancing the game around handicaps and not around properly responding to the in-game situation AI. It seems to me like choosing an easy way around. Don't get me wrong, I understood that they are also going to put effort into improving AI but I'd me more happy seeing all the production capabilities being put into the AI and broken game mechanics rather than short term fix of the problem which are custom handicaps. Just my opinion tho.

Shouldn't you first have a balanced AI first before you start trying to customize it? The current AI is incompetent. Paradox released the game as such, do we really want their focus to be testing adverse effects to custom sliders when their focus should be fixing the AI?

Actually, even though you obviously want both, customization without competence is still an improvement and it is one that can be gained quickly. AI competence is ultimately more important but it is also much much more difficult to achieve and will take much more time. This latter item will happen gradually over the course of the next 2-3 years. I'd rather not wait two years to get the customization.

Not replying directly to you @Dalwin. Just addressing the issue.
Custom strengths for nations never took time away from an AI programmer. We have people on the team with different skill sets.

Reading between the lines...

Please don't. You're likely to get it wrong, and only hurt the discussion.

I really don't mean to sound disrespectful here, but you guys need to do better. I know I am not the only one who has been frustrated with the lack of communication since the last hot fix, and Paradox really needs to take a hard look at how they are communicating to he HOI community.

Main reason is that people have been away. When people get back there is usually a lot of work to do, so time is spend on the game instead of the forum.
A contributing factor can be found in the tone of many posts. Since reading and posting on the forum, is something we usually do besides our normal work tasks, it would be great if it was a nice break from the usual work. Something we could look forward to.
Instead we get:

Cool. Thanks guys.

*sigh*

Sliders....ffs.

So....

Git gud?

Stellaris and CK and whatnot is making this team look like it has no idea what its doing and wont be able to get it together to fix the problems we have. Not just not in the short term, but at all.

"This is paradox fixing bugs"

And this is why the slider had this hostile reception. Any no AI improvement told in this DD could have been anyething most people expected.

And now I'm pretty sure someone will respond with: "If you had made a better game to begin with, we would all be happy and congratulating you!"
If that's your view and reason for speaking then so be it. But it's not something that encourages discussion.

========

Paradox, none of us want damage/defense bonuses as the go- to way of making a campaign harder! It is not fun having russian peasents beating my SS HArm, and it break immersion. When you do that the game will revolve around numbers and trying to game the system, instead of trying to outsmart the AI opponent.

No one? I'm going to reply with quotes.

The ability to strengthen other nations is nice. It will add some variety to playthroughs. I used that option with some mods in HOI3 at times, and liked it.

Well considering I made two separate threads asking for precisely this feature (the ability to buff specific countries) and a few people agreed, I think this is a helluva good start, Bravo PDX!

========

Im not trying to be a jerk here

Are you sure?

Wait - let me ask this explicitly, can difficulty be changed mid game, or only at game start???

Game start only.

Unless you think UI guys should code AI? Hey, maybe that's why the AI is in such a bad shape Paradox! Stop putting your UI team on AI script coding

This guy has a point. And humor.

Just posting to show some appreciation for your efforts and communication.
Please don't give up seeing all this rant.

Thank you.

These things do not eat up that much time and are immensely helpful and cathartic

Wrong. Reading through posts and answering takes immense amounts of time. Making a road map suited for public view takes a lot of work.

What concerns me the most is the lack of attention and immediate concern spent on this game right now compared to Stellaris but I made the mistake in thinking Hearts of Iron IV would be the flagship title. Just goes to show that sales numbers dictate where resources and dedication should go. I just feel bad for those few left to work on Hearts of Iron IV now. Just seems pointless to me when all that caution in not making the same mistakes with Hearts of Iron III's release and then it ultimately ends up in development hell anyway.

Why are you presuming to know about internal Paradox matters, like team size?
How many do you think are working on HOI now? How many before release?

It makes me so sad seeing how toxic part of this community has become so quickly.

I think toxic is a dangerous word. People are frustrated, and are having a hard time expressing it in a way that leaves room for an informed discussion. Maybe when the dust settles, we will be able to work together on improving the game further.

This is problem that's well know, and has been reported many times. Germany has trouble taking Copenhagen. Where in WW2 they surrendered after 6 hours.

Agreed. We're working on it.

But no one is saying this. No one is saying this. No one is saying Paradox is evil, etc... etc... This get thrown out there every time, and it's not true.

It sure sounds like it.

======

Do you Paradox play your own game SP?

Yes

How can it be that each title at the series is inferior than its predecessor?

I disagree, but if you think so, fair enough. Sad to hear it.

Is there ANY love in what you re doing?

Yes! Is there any love for us in your post? ;'(

Can't speak about the money matters you asked about. It's not my department.
======

I personally have enjoyed Hearts Of Iron 4 immensely and I have gotten 160 hours of entertainment from the game, but I feel like the game needs more depth. Not the point of Hoi3 depth, but enough to the point where you can always feel like you're learning things and finding more optimal ways to do things. After 160 hours, I feel like I've found the most optimal paths to go.
I don't think Hoi4 was as good as it was hyped to be, but it's still an enjoyable experience that I've had a lot of fun to play with.
Can't wait for the next few patches that can help the game realize it's potential.

Glad to hear it you have gotten so many hours of enjoyment from it.


So that's it from me. We're listening. We're working. Sorry if you feel we're not communicating enough. Answering takes a lot of time away from development.
Please be considerate in your posts. Pointing out problems is fine, but think about how you can make it constructive for us.

//Havebeard

 

[podcat]

Some might say that after 3 plus years in development the AI should be in better shape than it currently is. That might be true if the rest of the game had been finished three years or even 12 months ago. The fact is that many of the major game systems were not even in place until 6 months before release and even then they had bugs and needed balancing. It is perhaps unfortunate that AI is not only the hardest and most time consuming aspect of design but that it must by its very nature wait until almost everything else is finished.

Its also about expectations. I think the AI is close to (but not better) than HOI3 with all expansions and patches in level. Since games are completely changed you start from scratch with ai every time you cant really build on previous work. In fact I'm pretty sure the hoi4 ai is likely the best ai we have done on release for any of our titles. Stuff that people call "broken ai" like Germany not stomping everything is mostly due to changes in the game allowing minors to be more fun and playable, and also, and this is important, actually having AI's that are tryign to play the game, not just sit there and wait to get rolled over.

Getting german ai to perform more consistently and stronger is of course incredibly important (soviet players will want a challenge and not be forced to declare war on allies to get it every time for example).

However, that being said, I am tremendously disappointed with the naval AI, this has always been equally important as land

yeah, its not good at all. We are working on improvements for it. I hope they will be in 1.2, but might have to arrive in a 1.2.1 hotfix. we'll see.

but I think it was a reasonable expectation to expect something substantial this past week. When that didn't happen, people started to vent their frustrations. It is a matter of opinion on whether you think the content in this past week's DD was substantial or not, but I personally do not believe it was.

I think showing a feature and talking about fixing the #1 most complained about ai issues on the forum is pretty substantial. This is where your expectations should be sitting. What would you have wanted to see to call it substantial? Remember, we arent in the process of writing expansion diaries so I'll mostly talk about whats currently going on in the team.

for next DD I would like to know about what is being done in diplomacy, battle plans and combat (land, naval and air)

Thats not happening. Diaries will cover things we have done, or are in the process of doing. Not promises of the future. if our naval combat changes are done on friday I'll talk about them, otherwise that will be talked about later.

Hopefully they can do what Red Ball did and get a beta patch with the biggest changes out relatively soon which will allow them to fine tune the patch.

we will always do beta patches if possible, its a really effective way of finding issues.

I think I will have to go Multi Player and give that a go, as I am recently employed and can afford it, but my poor old Italy games of HOI3 of yore are simply gone forever, right?

yeah, the AI in HOI3 was on rails and would basically ignore non-historical expansion (like stuff italy did) so it was a huge exploit. If you want to do simialr things in hoi4, it might be best to keep your tension down initially so gemrany gets more, not join up with germany and make sure they remain the focus of allied anger.

 

[SteelVolt]

Are the AI naval invasions in the things to improve?

Yes! It may not be ready in time for sunflower, but it is being worked on by another coder as I write this.

 

[SteelVolt]

@SteelVolt Would it be inappropriate if I deeply tongue kissed you?!

I suppose it might be.

It's almost like you've been reading some of my posts regarding a deep desire for divisions to more aggressively swap to better templates!

Yours and others. There is a reason it was top of my priority list.

 

[SteelVolt]

@SteelVolt - glad to hear the AI is getting some love! One thing I haven't seen mentioned but I have noticed that would be a great improvement is the AI use of units with different "supply footprints". What I mean, I'll try to illustrate by example: if I give the AI battle manager in North Africa divisions including an armoured division (that requires lots of supply) and a couple of infantry brigades (that don't require much supply), it seems intent on putting the armour in the deep desert (where it will wither and die) and the infantry near the coast. If it could somehow consider placement along the front based on the supply "weight", that would both make the battle planner easier to use for me and help the AI nations who can't micromanage fronts, etc. Thanks!

Good feedback, and absolutely something I want to have a look at. Not going to be in this patch, though.

 

[SteelVolt]

@SteelVolt

Will there be any optimizations to how expeditionary forces work? Right now, one of two things sort of needs to happen with them IMO.

1) Either the AI recipient needs to do a better job of garrisoning the senders country from naval invasions and what not, or...

2) The AI needs to clamp down on the number of divisions it sends as expeditionary forces to other AI nations, but not to the player since the player can make better decisions about when and how to defend.

I think option 2 is probably better, because it will let say Italy or Romania or Nationalist Spain handle their own home defense, while still contributing a decent number of divisions to Germany. Right now, the AI is sending way too many of its forces away, to the extent that these countries are barely being garrisoned at all.

Yes, but not for the current patch.

 

[BjornB]

There is a difficulty with the development diary schedule for HoI4 where the DD's are posted on Fridays. If you look at the responses in this thread you actually received quite a lot of responses on Friday, Monday, Tuesday and Wednesday. For natural reasons there will be less activity on our part on Saturdays and Sundays.

Regarding the recent discussion that there should be Community Managers handling the forums to relieve the developers, it is true to some extent. I aim to do that (I've just returned from my vacation however), but what we consider to be a key ingredient to making good games is that the developers (and all of the organisation for that matter) is involved in the community. If we were to leave it to "communication specialists" I don't think Paradox would be the same at all. I do think you'll see a lot more of me in this forum from now on though.

 

[BjornB]

Yes they do. Just because you have not seen a reply or action does not mean it wasn't read.

A tip for suggestions, the title should be clear and concise regarding the subject. Generic titles that don't explain the post in one sentence probably means your post isn't clear and concise either. Long rambling posts are unlikely to be read, there is only so much time in the day.

I've written a small advice on how to best post your suggestions here.

 

[BjornB]

I kind of feel that the "strengthening" aspect was added as a bandage solution to making actual improvements on the AI, but I hope that I'm wrong here.

In the case as stated in the OP, these settings have nothing to do with the AI's shortcomings. It's merely to give players more choices of how they want to play (as stated by Had a Dad).

In the rare cases where we might use AI "cheats" as bandaids (it can happen), it would be only as temporary solutions. This could be implemented where an issue is really urgent, but we don't have a good proper fix for it in the near future. In those cases we might consider implementing some cheats to improve the current gaming experience. We are in full agreement that should not be a permanent solution however.

But again, what's mentioned in this DD is not a fix for the current AI issues people are reporting.