HOI4 Dev Diary - Synthetic Dawn

[him_15]

This is the change that I have been waiting for so long, very glad that it is finally getting implemented! Though the potential bug risk and game balancing issue remain my biggest concern.

 

[C-Breeze]

Interesting diary... At first blush it seems to have the potential to add some much needed depth to the economic system, which in my humble opinion, could be a very good thing.

 

[PinkBear]

I am beyond impressed with this developer team. Podcat is seriously a genius.

Guys, these are amazing ideas! Can't wait to play 1.5

 

[Sourlol]

@Axe99

This may be a bit gamey, but why not make nickel a late game resource? Medium tank 3, tier 3 naval, tier3 prop plane and beyond all require nickel.

Nickel deposits are added to the map through nf, date gated event, or just not used on any pre 1940 tech equipment.

 

[stjern]

Nice change. Does it go with a change making the AI take more notice of resources? A key issue I find is that the AI CICs only ever build factories - more and more of them, regardless of the resource restrictions it is under. Adding more of a balance to the AI to balance resource availability (including convoy capability for imports) with ICs and other features (Infra, and maybe even some radar!!!) would be good. Really nice, too, would be some sort of diminishing returns on ICs - like having them consume manpower, reducing the pool from which recruitment is taken. Maybe then critical policies like the use of women in the workforce could be represented in some way?

In the current version of the game the AI construction logic is sort of disabled and follows "build CIC before year X and MIC after". X can be set in defines files as you may knw, its called "freer construction" someting. However I explained to them that mods have added dynamic AI construction logic so the AI knows exacaly what to build based on game state and infra level. I showed them the code so hopefully this is what they are implementing now.

Can we get a confirmation of this? Or is the construction AI going to reworked (reinvented)?

 

[Secret Master]

there's this huge continent and there's absolutely no reason to actually invade it.

We trade it back and forth between the Axis and Allies in just about every MP game. Our Axis players really love going after it.

And after some serious losses in Egypt and Morocco in recent games, it makes sense. The Allies can't let Italy run her NFs for oil in Libya without at least trying to get it. That's just leaving oil on the table. The Axis can't let the Allies hold Tunis without a fight if only because of the air fields there. The Allies in many cases want Casablanca and Oran in friendly hands; in that situation, even if Gibraltar falls, the Allies can ship troops faster by sailing them to Casablanca, moving them by rail to Algeria, then shipping them to wherever they need to go, rather than rounding the horn. Egypt and the Suez needs to be secured, and letting the Italians walk from Tobruk to Cairo is a losing strategy. And if Spain willingly joins the Axis, the Allies need to be positioned right there in Morocco to seize the African side of the strait. You don't want the Italians to leave the Mediterranean. (That's how Case Green ended up being executed last game; it was a Regia Marina/Kreigsmarine co-production.)

Hell, I've seen (and participated in) an Axis Torch after losing North Africa to the Allies. Went from Casablanca to almost Tunis before the Allies realized we had landed and occupied it all. That operation, coupled with Guderian's attack from Syria to the east bank of the Suez, resulted in the Allies evacuating Italy and freaking Rome after both Gibraltar and the Suez became contested. No supply convoys could get through to Allied forces in the Italian boot.

I'm also sure the Allies don't want South Africa capitulated by Axis forces walking from occupied Egypt to Pretoria.

 

[nauticalweasel]

The size of a state might be estimated.

The defines profess the following:

View attachment 304294

Calculate the number of pixels per state.
Multiply the number of pixels by 7.114 x 7.114 = 50.06 km**2

For simplicity's sake, each pixel is 50 square km.

To make it even more "simple", don't even convert to km's...base the size on the number of pixels .

(but, counting the number of pixels per state might not be possible be tricky. But I know that the Paradox team could do it!).

I meant that the game can't tell how big or small a state is, to it two tiny provinces and two massive ones are still just two provinces.

 

[Sir Garnet]

Resources & Infrastructure
This is something we have been wanting to do for a long time. Each level of infrastructure now adds +10% resources in those states. This means that we can have resource amounts that actually grow later in the game. This should give you more reasons to upgrade low infrastructure areas to take advantage of the resources there, and will also allow bombing to impact normal resource gathering and not just refineries. A low infrastructure area with resources is now a great opportunity for expansion.

Together with this change we have improved the construction interface.

Is there a point where the incremental infra construction yields diminishing returns? You make it sound linear, which among other things encourages piling up infra in high resource areas rather than spreading development with cheap early infra on second tier resources offering more benefit than trying to squeeze more from a richer resource. providing better returns .

It makes damage and such harder to show, and its a core design philosophy for HOI that people like to make stuff better rather than less bad. its mostly a vague difference but green numbers going higher is a lot nicer than red numbers becoming smaller
ai will care more about areas like that yea

Important question for game play - any fix for reversed undersupply attrition (higher attrition for smaller units being bizarre)

 

[Axe99]

@Axe99

This may be a bit gamey, but why not make nickel a late game resource? Medium tank 3, tier 3 naval, tier3 prop plane and beyond all require nickel.

Nickel deposits are added to the map through nf, date gated event, or just not used on any pre 1940 tech equipment.

This is getting way outside anything I've got much of a clue about now - this is more a question for Bauxite @billcorr . That said, there were deposits of other resources that existed at the start of the game, but weren't used at that time because there were uneconomical, that were brought back into service during the war because the demand for them skyrocketed. At the moment, these are only represented in as much as their distribution matches those resources actually in use (or those provided by NF's).

That's not to say I that don't think the system we're getting is a good idea - it is . It's a reasonable enough abstraction that will be a big step forward in managing the increase in resource demand - it's just a bit limited in terms of reflecting the where the new resources will pop up (as it's locked to the distribution of in-use resources at game start, or NFs) relative to what I understand (and again, Bauxite Bill's a better person to comment on this than me) happened in terms of resource extraction during 1936-48. As long as the resources are 'close enough' to where they were historically in a strategic sense, it should still provide the same strategic challenges it did for the players and AI to manage.

 

[myzael]

(but, counting the number of pixels per state might not be possible be tricky. But I know that the Paradox team could do it!).

It's trivial, all it takes is to iterate over each pixel in the map once.

 

[Serensius]

That would require it moved to the provincial level, which I am for.

Or it could be based on the average modifier of all provinces in the state. Let's say (for the purpose of this example) that plains is a 0% construction speed penalty, forest is 50% and mountain is 100%. A state with 1 plains, 2 forest and 3 mountain would therefore have a penalty of (0 + 50 + 50 + 100 + 100 + 100) / 6 = 66%.

Separate province-level rail for strategic redeployment, supply and/or resource convoys is interesting to consider for the tactical possibilities (like, for instance, cutting off forces in central africa by capturing rail), but I'm wondering if it's not just a lot of extra micromanagement for a relatively small change in how the game works.

It also seems like one of those problems that the player is much better at solving than the AI, since it involves a lot of planning ahead and surveying the map for optimal locations.

 

[katanatan]

@Developers @podcat

I like where this is going. But it seems to me with all those starting numbers infrastructure is (in terms of ressources) still negligible (not counting Japs and italy, but they should trade IMO).



But.

Please think about redistributing infrastructure levels and maybe adjusting infrastructure focuses and costs. I think only germany (as an federalistic high developed state) should have high infrastructure, britain and france as ultra-centralistic colonial empires should have BAD infrastructure.

(Example as of (centralistic-obsessed) french infrastructure: In ww2 and the Franco-Prussian war they had significant mobilization issues because of their rail system (rail was always during the war the main transportation asset for the troops). If you wanted to get from Bordeaux (SW-France) to Marseille (SE-France) you had to take a train through paris. Every intercity railroad had to go through paris. So make paris and maybe some big cities infrastructure 8-10 but the rest 2-3)

(For the US, they had bad infrastructure because the US used a lot of shipping and was focused (industrial and population-wise) at the coasts.
If you adjust the levels there maybe the US in HOI 4 had finally a reason to do a real new deal?)

 

[Porkman]

Is there a point where the incremental infra construction yields diminishing returns? You make it sound linear, which among other things encourages piling up infra in high resource areas rather than spreading development with cheap early infra on second tier resources offering more benefit than trying to squeeze more from a richer resource. providing better returns .

You actually want expanding returns.

AOD had infra tied to resource production and what was called Concentration effects. @Balesir can correct me on this but the way it worked was that Industry would get a bonus just from being next to other industry. So 10 factories in a province would behave like 11. This incentized you to build on what you already had.

Infrastructure also increased factory yield and resources gain. Building infrastructure in a place was faster (I don't think cheaper) if there was already some infrastructure there and it was modulated by terrain type.

What this encouraged was piling a lot of stuff into a few provinces... which was historical. But this made them more vulnerable to bombing and capture. (For China this is a huge deal as almost all of the industrialized areas with any infra are near the coast and thus at risk).

It was historically cheaper to build more industry where there was already industry and roads where there was already roads and mines where there were already mines.

It should actually cost more to spread out and diversify which is why most countries didn't do it unless directly threatened.

 

[Porkman]

@Developers

I like where this is going. But it seems to me with all those starting numbers infrastructure is (in terms of ressources) still negligible (not counting Japs and italy, but they should trade IMO).



But.

Please think about redistributing infrastructure levels and maybe adjusting infrastructure focuses and costs. I think only germany (as an federalistic high developed state) should have high infrastructure, britain and france as ultra-centralistic colonial empires should have BAD infrastructure.

(Example as of (centralistic-obsessed) french infrastructure: In ww2 and the Franco-Prussian war they had significant mobilization issues because of their rail system (rail was always during the war the main transportation asset for the troops). If you wanted to get from Bordeaux (SW-France) to Marseille (SE-France) you had to take a train through paris. Every intercity railroad had to go through paris. So make paris and maybe some big cities infrastructure 8-10 but the rest 2-3)

(For the US, they had bad infrastructure because the US used a lot of shipping and was focused (industrial and population-wise) at the coasts.
If you adjust the levels there maybe the US in HOI 4 had finally a reason to do a real new deal?)

What you're talking about is the need to separate rail infrastructure from road infrastructure.... Which the devs will never do... even though it absolutely needs to be done.

 

[aiakira]

Sorry
my English is very bad because of i will write in my own language,Turkish
then i will write English
If someone who speaks Turkish translates this writing into English.

I
It's a very nice change but I will have some questions

1st: You wanted to bomb Japan with strategic bombardment.
2.Sorum: But there is an order. Strategic bombing You have to bombard all the targets, facts, factories, anti-aircrafts, ports, fortresses and airports.
Result
Now, with so much effort, you will not be able to survive, even if you are going to dry up resources for at least 1000 with at least 10,000 strategic bombings.
If they make this change, they will not do anything if they add the strategic bomber command without changing it.
how changes should be
Attacks on roads, factories, anti-aircrafts, ports, fortresses and airports must be separated as in the first HOIs. Separate command is required for each target and must be a point bomber option

There is only one thing to say if they bring that patch before they do these changes
YOU ARE KIDDING US

 

[fellaz007]

@podcat does the 10% infrastructure bump also affect the synthetic factories' output or is it just natural resources that improve?

nope, only "natural resources"

Suggestion: Maybe there can be a minimum infrastructure level to built synthetic factories, as these are quite high tech and would require according infrastucture.

 

[King-Salomon]

nope, only "natural resources"

Quick question: will ressources gain/created from NF count as "natural resources" or will they be not affected?

If they are affected, will they get more ressources if the infrastructure was build before getting them by NF or is the "infrastructure effect" wasted when getting them after building the infrastructure?

 

[Balesir]

You actually want expanding returns.

AOD had infra tied to resource production and what was called Concentration effects. @Balesir can correct me on this but the way it worked was that Industry would get a bonus just from being next to other industry. So 10 factories in a province would behave like 11. This incentized you to build on what you already had.

The concentration effect in AoD doesn't (or, at least, didn't - I'm not fully au fait with the latest versions) affect resource production, which @Sir Garnet specifically referred to, here, and I don't think it should. In retrospect, I think resource production should have a tail-off, or more precisely perhaps an 'S-curve' - accelerating gains early on, but diminishing returns at the top end. This would encourage fast development of resources discovered, but going to the upper extreme only if demand was high with no other recourse. I agree with the poster who said that 0% Infra should mean zero resources; having resources in zero infra states not shown on the map would allow for resource "discoveries", too.

Industry is another beast. Here the AoD 'concentration effect' applied, as well as infra effects, making concentrating your ICs into nice, vulnerable blobs attractive. I don't find this entirely persuasive from an economic standpoint, but as a game mechanism it works really well

ICs also reduce the overall manpower pool in AoD. I think this is important, as it makes the IC creation a balance, rather than "more is always better".

Infrastructure also increased factory yield and resources gain. Building infrastructure in a place was faster (I don't think cheaper) if there was already some infrastructure there and it was modulated by terrain type.

Infrastructure in AoD is also extremely important from the point of view of the supply network. A bottleneck somewhere can really degrade supply efficiency for the whole area beyond it. The supply network characterisation in HoI4 could really do with a fairly radical overhaul, but I have sent some ideas to @podcat on that already, by pm ('cos it's long and the latter bits are rather technical/tedious to non-designers).

 

[Balesir]

The size of a state might be estimated.

The defines profess the following:

View attachment 304294

Calculate the number of pixels per state.
Multiply the number of pixels by 7.114 x 7.114 = 50.06 km**2

For simplicity's sake, each pixel is 50 square km.

To make it even more "simple", don't even convert to km's...base the size on the number of pixels .

(but, counting the number of pixels per state might not be possible be tricky. But I know that the Paradox team could do it!).

OK, but that is only an average value, though, because the map is not an interrupted projection. Pixels at the north and south of the map represent a much smaller area than pixels in the middle of the map. The way to address this would actually be to give each province a longitude and latitude and use grand circle route calculations (as used by airlines) to derive the distances between provinces.

 

[sprites]

so what's that yellow thing in province wiew, above the ressources?