HOI4 Dev Diary - License production and map changes

  • 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.
Why is the AI still producing level 1 ships in 1944?

The AI tends to prioritise other research over ships, so it tends to get to ships 'last-ish' (which, in the context of the game mechanics and starting balance of forces, is often a sensible decision for many nations). Most nations benefit more from better tanks, infantry equipment, industry tech or aircraft more than they do from better ships, on average. Not saying I think this is the way it should be, just the way it seems to be at the moment.
 
The AI tends to prioritise other research over ships, so it tends to get to ships 'last-ish' (which, in the context of the game mechanics and starting balance of forces, is often a sensible decision for many nations). Most nations benefit more from better tanks, infantry equipment, industry tech or aircraft more than they do from better ships, on average. Not saying I think this is the way it should be, just the way it seems to be at the moment.
As an "example human", I tend to focus on ships (as far as I want to develop them) hard early on, then essentially abandon them later. This is, on reflection, a function of the fact that the game has a fixed "end date" (and an "expected time scale). Ships have a long lead-time, and researching something that's not going to be available in sufficient numbers to really make a difference before the show's over is a waste of a precious resource. The real actors in the real war, of course, did not have the luxury of seeing the timescale in advance...
 
Why is the AI still producing level 1 ships in 1944?

Is that all kinds of ships? I have noticed USA making a bit too many tier 1 BC, but they also get tier 3 CV which is good.

The AI tends to prioritise other research over ships, so it tends to get to ships 'last-ish' (which, in the context of the game mechanics and starting balance of forces, is often a sensible decision for many nations). Most nations benefit more from better tanks, infantry equipment, industry tech or aircraft more than they do from better ships, on average. Not saying I think this is the way it should be, just the way it seems to be at the moment.

Pretty sure the question is in regard to ai naval production lines set on infinity production of level I starting tech that never switch out. Because the production is set on infinity the ai will not switch out when new tech is learned. @SteelVolt acknowledged it is an issue on his list. What he suggested as a temporary fix was to change this common/defines code below:

PRODUCTION_MAX_PROGRESS_TTO_SWITCH_NAVAL = 0.3

change the value from 0.3 to 1.0.

I can confirm this does work. It immediately switches to the new tech, no matter how close the current production is to finishing. This of course sucks if the ai is a few days from launching a new ship but on average the switch is beneficial. And it does block ironman, if that is an issue for anyone. The other option would be to go into the history files and change all infinity production to numeric quantities for major powers. Not sure if the ai would then start infinity production for new lines, have not thoroughly tested this option.

Otherwise the ai will just continue to pump old ships out for anything originally set at infinity in the history file.

The generic ai does regard ship tech as a lesser priority. Some majors like Japan, USA, and the UK have higher priority settings. Hopefully the whole naval research and production script code is something that will get a through review and upgrade in 1.5.

Add Aluminum in Asia
 
PRODUCTION_MAX_PROGRESS_TTO_SWITCH_NAVAL = 0.3

change the value from 0.3 to 1.0.
Aha *click* - because this switch only triggers once on techup, of course. I wonder whether it's possible to make an event, triggered by the tech completion, that switches any extant ship lines for the replaced level(s) from "infinity" to "1". This could be made AI only, or even give an option to the player to keep it - I could be tempted!
 
Aha *click* - because this switch only triggers once on techup, of course. I wonder whether it's possible to make an event, triggered by the tech completion, that switches any extant ship lines for the replaced level(s) from "infinity" to "1". This could be made AI only, or even give an option to the player to keep it - I could be tempted!

I have been dipping into the history/unit files and looking at the starting production a little this morning. I am not seeing anything that starts with infinite production quantities for Japan, the USA, or the UK. But i know from tagging after the game has run a while these nations will be running infinite production lines. This is telling me this problem is a production script issue. If I have time later I will delve into them. There is probably a script line saying to go to infinite somewhere that can be changed to a small finite quantity. This should force the ai to recalculate needs and shift to higher tech. If I can find it.

@Axe99 you know anything regarding naval scripts, or maybe @Elouda has dealt with this?
 
I have been dipping into the history/unit files and looking at the starting production a little this morning. I am not seeing anything that starts with infinite production quantities for Japan, the USA, or the UK. But i know from tagging after the game has run a while these nations will be running infinite production lines. This is telling me this problem is a production script issue. If I have time later I will delve into them. There is probably a script line saying to go to infinite somewhere that can be changed to a small finite quantity. This should force the ai to recalculate needs and shift to higher tech. If I can find it.

@Axe99 you know anything regarding naval scripts, or maybe @Elouda has dealt with this?

I'm afraid I've been deliberately waiting until 1.4 to get my head back into these (and other naval defines/scripts) issues. Outside of what's in the starting OOBs (which are, as far as I can remember, always for a specific number of ships - I'm pretty sure there aren't any 'infinity' production runs set up for the start of the vanilla game) the production decisions for ships, as far as I can tell, are governed entirely by AI strategies at the broad ship type level (ie, capital ships, screens, carriers, transports). These, however, are by no means the only factors taken into consideration, as even with screens, for example, clearly more heavily weighted than capital ships, the AI for some reason plunks most of its NIC into capitals.

Further, the AI has a substantial preference for low NIC cost, even if the proportionate increase in capability is greater than the increase in NIC (and it's worth noting, many larger ships were more expensive relative to their increased capability and still worth building) which may also be part of the problem (as the ships in the lower part of the tech tree are cheaper in terms of NIC). In the naval mod I'm working on with historical (as close as I can get) ship stats, the AI has a strong preference of cheaper vs more capable ships, and the stats beyond value it does appreciate are not clear from my general observations.

Reason I haven't looked into it is that doing a bunch of testing now that might already be obsolete isn't a great use of time/capacity. Once 1.4 hits, I'll give it another look, but I'm hoping that at some point we get more control over shipbuilding priorities, so it's possible, for example, to have Britain prefer cheaper destroyers while the US and Japan prefer more expensive destroyers.
 
I'm afraid I've been deliberately waiting until 1.4 to get my head back into these (and other naval defines/scripts) issues. Outside of what's in the starting OOBs (which are, as far as I can remember, always for a specific number of ships - I'm pretty sure there aren't any 'infinity' production runs set up for the start of the vanilla game) the production decisions for ships, as far as I can tell, are governed entirely by AI strategies at the broad ship type level (ie, capital ships, screens, carriers, transports). These, however, are by no means the only factors taken into consideration, as even with screens, for example, clearly more heavily weighted than capital ships, the AI for some reason plunks most of its NIC into capitals.
I suspect it might go even deeper than the AI scripts; if the player starts a ship costruction line, the default setting is a serial run of infinity. All the script has to do is omit to change that, and you will get infinite runs of every ship started after the game start set are built.
 
Paradox, have you considered changing the resource displacement in the world? Possibly adding more rubber to the Belgium Congo perhaps?

Rubber in Africa was already corrected in the patch 1.3.3 Torch last February, based on hard figures I provided in an earlier thread (may 2016). Thanks to @podcat and the team !

- Added rubber to states in Africa. Equatorial Africa - 4 rubber | French West Africa - 2 rubber | Gabon - 1 rubber | Central Congo - 5 rubber | Angola - 1 rubber | Portuguese Guinea - 1 rubber | Liberia - 7 rubber | Uganda - 4 rubber | Nigeria - 3 rubber | Tanganyka - 2 rubber.

You will find the figures and sources in this thread: https://forum.paradoxplaza.com/forum/index.php?threads/rubber-an-asian-affair.937515/page-3
 
Last edited:

Impassable areas


A common problem that seems to pop up no matter how many times we think we have solved it is AI over-committing to Africa. There has been a lot of community feedback on this and we were hesitant to just make a hole in Africa as some mods do, but we still felt something needed to be done so we decided to go with a slightly different approach in Oak and have implemented a new kind of impassable areas. It's fairly realistic that large troop movements can not be done through Sahara in HOI4's time frame, so we have decide to block them off for troops and avoid any case where AI may be tempted to commit more troops there than what makes sense. It seems to help the axis AI out a lot and creates more interesting small-scale fighting for players as well. Right now there is no way to un-impassable these areas, but I feel like its a logical step in the future where people may be playing past the WW2 era to be allowed to do some expensive project and build a highway across.


We have also set up several borders as impassable to block passage across some areas of the Himalayas and others mountain ranges in that area (Afghanistan now is even tougher to take with more border choke-points like the Khyber Pass). While you can't enter impassable areas controlling surrounding areas will flip impassable states so if you control Africa it will look that way on the map also without say a hole in the middle.

How would one go about modding the map to add impassable areas? Is this even possible? I have some ideas for impassable areas that I would like to make into a mod
 
Also fascist Spain wanted to license build the Italian Litorio-class BB and did build 2 british CA (I think it should be added to the production queue)


Which CA did Spain build? I am looking in Janes but can't see anything

K
 
Which CA did Spain build? I am looking in Janes but can't see anything

K

The Canarias class - basically modified Counties, but iirc the modifications were still done by a British designer.
 
The Canarias class - basically modified Counties, but iirc the modifications were still done by a British designer.


I suppose if I try to blot out the superstructure the hull form is a County :)

K
 
I really, REALLY don't like the impassable concept. Make it passable by specialised troops and some mechanized units only, at insane costs, speed and attriction but don't block them completely, it's not realistic and an immersion-killer to me. :(
 
I really, REALLY don't like the impassable concept. Make it passable by specialised troops and some mechanized units only, at insane costs, speed and attriction but don't block them completely, it's not realistic and an immersion-killer to me. :(
That is in a way what they tried, but the AI couldn't handle it.
 
will be impasable terrain overcomed if you build big enough infra here ?