Dockyard output doesn't scale enough

[Alexander 'The Grape']

Compared to military factory output, dockyard output doesn't scale well at all.

Dockyards have a lower base output than mills. But because dockyards effectively always have 100% production efficiency, their output is roughly similar.

However, while Military Factory out scales greatly into the late game due to improvements to production efficiency cap/retention, Dockyards lag behind. Dockyards only gain a max +50% to Dockyard Output from tech, which is inferior to the additional +25% output from Concentrated or the additional Retention from Dispersed.

This makes it harder to build a navy, making naval power levels too dependent on starting fleets.

It also makes it extremely difficult to build later model ships in time to impact a war.

 

[Gran Strategist]

I am a bit torn on this one.

Tovstart with ieel we need to separate large and small ships or capital vs screens and subs.

For destroyers and subs you can generally build 10+ a year per production line and for light cruisers 3-4 a year which seems fine.

I always feel a bit put off building capital ships but then in reality if you commit to them they don't take too long to build, the question is are they really worth it?

For the most part unless you are US or Japan and building aircraft carriers (and capital ships to screen them) they probably aren't which is true to history.

Looking at later models, these agai. seem wasted as by the 1944 models in particular you are likely near the end of the war and it is obvious you aren't going to build them in time but these are more for if you plan to go into alternate history side and carrying on after ww2 has ended and like modern tanks or jet planes these represent post war models.

When it comes to building a surface navy or expanding one, unless i am Japan or US, i just spam out light cruisers with as much light attack as possible and torpedo armed destroyers. Heavy ships find small ships very hard to hit, your light cruisers will take out enemy screen ships leaving their capitals at the mercy of your torpedo armed destroyers.

This may sound like gamey cheese but it is again more true to real life than you expect.

 

[SeekTruthFromFx]

I think the *status quo* is correct because there's just more you can do to increase efficiency with smaller vehicles than you can with ships. Although the US in particular achieved amazing things by prefabricated parts that were assembled in the shipyard (median time before launch for Liberty ships was 39 days!), complex warships were not built on a production line. You had to finish one ship before the workers could start the next, with all the inefficiencies that implies. And if a warship takes a year to build anyway, long-lead items are ordered 2 or 3 years before launch, so it's hard to speed things up as you go along.

 

[goodcigar]

Naval production for the majors is well below historical numbers. But increasing production would just make the AI create massive fleets of Screen ships because of how they have the naval AI setup. They basically gave up on programming a realistic/enjoyable naval AI (literally won't build any Capital ships after the ones in production at game start) and just set it to build goofy meta gameplay fleets. It just builds a static predefined ratio of ship types for the most part (mainly Screens). There's not much dynamic response (few dynamic responses for a few majors). There's no real strategic AI thinking.

And the AI will start and cancel ships at random whenever it changes its mind/suffers losses. Regardless of how expensive/long they've been building a ship. The AI immediately cancels all heavy ships under 25% production or so at game start. The AI doesn't assign enough Dockyards to expensive ships to actually get them built. It's a real mess. Needs major work.

The mine response mechanic is hilarious. Will cancel all ships in production to build a few minesweepers lol.

What is needed (beyond common sense AI improvements) is an actual strategic naval AI for each major. A strategic AI based on the nation, the Focus chosen, the war situation, etc. Germany built 90% Submarines during the war for a reason.

 

[stormsand9]

They basically gave up on programming a realistic/enjoyable naval AI (literally won't build any Capital ships after the ones in production at game start) and just

I agree with alot of your points but this is straight up wrong. The Ai does build new capital ships.

Post BBA Japan game

 

[HugsAndSnuggles]

This makes it harder to build a navy, making naval power levels too dependent on starting fleets.

And what's wrong with that? It still largely depends on industrial capacity: even expert AI US can get over 80 dockyards by '40 - if they had no starting fleet, they could build a new one by the time it's needed.

It also makes it extremely difficult to build later model ships in time to impact a war.

I'd argue that this has little to do with production output: modern tanks and jets are in the same "boat" as '45 ship hulls - you won't produce any in time to impact war (unless you horribly mismanage the whole thing); while '40 models can be produced in relevant numbers to make an impact - if you rush them via research bonus (same as tanks or planes)... that is if you don't mean "navies entirely composed of '40 hulls" by "impact". Also, sanity of decision to rush navy tech (often in place of air or land stuff) is more of a question of overall navy relevance than production output.

The Ai does build new capital ships.

Now, if only someone would tell AI that carriers are supposed to carry planes to be effective... =)

 

[Alexander 'The Grape']

And what's wrong with that? It still largely depends on industrial capacity: even expert AI US can get over 80 dockyards by '40 - if they had no starting fleet, they could build a new one by the time it's needed.

This is technically true, though the navy produced would be smaller, cost wise, than the starting US fleet.

Assuming the 80dy USA builds them with giant wakes around late 37, then they will average around 60 docks over the 5 year period before a 1941 war. These 60 docks will have an output in the late game of around 4 IC per day (base 2.5, +60% from stab, tech and free trade). 4ic times 60 docks times 5 years equals ~440,000 ic - significantly less than the starting US or UK fleets.

And this doesn't account for the need to build convoys!

I'd argue that this has little to do with production output: modern tanks and jets are in the same "boat" as '45 ship hulls - you won't produce any in time to impact war (unless you horribly mismanage the whole thing); while '40 models can be produced in relevant numbers to make an impact - if you rush them via research bonus (same as tanks or planes)... that is if you don't mean "navies entirely composed of '40 hulls" by "impact". Also, sanity of decision to rush navy tech (often in place of air or land stuff) is more of a question of overall navy relevance than production output.

The point is twofold - firstly that for expensive hulls, naval tech is essentially delayed by 1-2 years to actually get them combat ready, and that it is very difficult to get them in significant numbers compared to the influence of raw starting fleet power.

Due to the recent Nav nerfs, navy is extremely relevant in high level MP. There are Japan strategies that revolve around building up to 100 dockyards to raid allied shipping and deny them rubber. However, even those builds don't bother with researching advanced hulls for expensive ships because once you get the tech, you simply can't build them in time for the war!

 

[Alexander 'The Grape']

I think the *status quo* is correct because there's just more you can do to increase efficiency with smaller vehicles than you can with ships. Although the US in particular achieved amazing things by prefabricated parts that were assembled in the shipyard (median time before launch for Liberty ships was 39 days!), complex warships were not built on a production line. You had to finish one ship before the workers could start the next, with all the inefficiencies that implies. And if a warship takes a year to build anyway, long-lead items are ordered 2 or 3 years before launch, so it's hard to speed things up as you go along.

> Innovation was fundamental to the success of Bay Area shipbuilding during World War II. Shipbuilding prior to the war tended to be bound by long-standing traditions and methods. The transition from wooden ships to iron and then steel was slow. During World War I, steel shipbuilding followed tradition, calling for riveted hulls with each vessel custom built on site, a labor intensive, relatively slow process. In 1917, for example, a typical steel vessel took 12 to 14 months from keel-laying to delivery. At the peak of production in World War II, the work could be accomplished in four to six days.

World War II Shipbuilding in the San Francisco Bay Area (U.S. National Park Service)

www.nps.gov

Arc welding in particular is called out here - which is the icon for a tech that gives no bonus to naval output!

 

[HugsAndSnuggles]

Due to the recent Nav nerfs, navy is extremely relevant in high level MP. There are Japan strategies that revolve around building up to 100 dockyards to raid allied shipping and deny them rubber. However, even those builds don't bother with researching advanced hulls for expensive ships because once you get the tech, you simply can't build them in time for the war!

I still think it's in no small part due to more advanced hulls being about as cost-effective as for tanks. Besides, Japan only has a single realistic option for rushing: cruisers (with 300% bonus). They also get two 100% bonuses: for battleships (if you forgo sortie efficiency branch) and destroyers - hardly ideal to rush tech with (although with DDs it can be done if there's actual need to have them ready in time for war).

 

[marcelo r. r.]

What is needed (beyond common sense AI improvements) is an actual strategic naval AI for each major. A strategic AI based on the nation, the Focus chosen, the war situation, etc. Germany built 90% Submarines during the war for a reason.

Allied AI need 2 "stances", the normal one, and another when UK, USA stripped of colonies/allieds. The "second stance" would be a bunker mode: limited exports/closed economy, because actually they keep running "free market', and end in lacking of steel, crippling naval production.

 

[davidc929]

I think the *status quo* is correct because there's just more you can do to increase efficiency with smaller vehicles than you can with ships. Although the US in particular achieved amazing things by prefabricated parts that were assembled in the shipyard (median time before launch for Liberty ships was 39 days!), complex warships were not built on a production line. You had to finish one ship before the workers could start the next, with all the inefficiencies that implies. And if a warship takes a year to build anyway, long-lead items are ordered 2 or 3 years before launch, so it's hard to speed things up as you go along.

Very few battleships were started once the war began. For the UK only four were started after they entered the war and three of those had been ordered before the war. For the Americans only one of the battleship's that entered service during the war had been started after the start if the war. Two more were started and never completed.

As far as I can see no other country completed any battleships during the war that they hadn't begun beforehand.

Aircraft carriers are different. But when it comes to battleships it is pretty accurate of most countries depend on those they already had.

 

[Dimmie_Dumm]

While it's true that MIC output quadruples from industry tech alone and NIC receives merely a 50% bonus, there might be valid historical reasons for that and I myself gonna leave it at that for now.

However, what I find disturbing is that wartime shipbuilding is no different to peacetime one, and all we have is the same slot allocation gradually covering that full cost. No cutting corners, no usage of existing left-overs or prebuilt stuff. Hell, we can't even switch designs midway to accomodate new modules once they unlock even if it's just a radar. Really?..

The game itself bashfully admits there is a problem by having a load of Focus Tree Magic where ships simply enter the queue being already half-constructed. I'm not actually buying an argument that this is allowed because allegedly the work had already started for quite some time of which we were simply unaware - for a simple reason the same logic could have been applied to 'normal' construction process too, to give us a speed-up button in some form.

Historically, a lot of ships as I recall were delayed by awaiting for new guns and turrets to become avaliable, by sorting out reliability issues of the innovative engines etc. Sticking to already proven solutions then might/should provide a construction speed bonus. As usual, I dare summon @Axe99 for proving me wrong

My personal solution though would be allowing +5 slots allocated for each ship once at war, but the efficiency of each added one would drop by say 10-15%, i.e. the last one would contribute only half to quarter of its normal IC output. This way we could have an option of switching from efficient peacetime construction in parallel to wartime rushed churning, using both as we see fit.

Whether the AI could be taught to handle these nuances is a different issue. Because what [naval] things the AI actually does handle right?..

 

[Captain Palmtree]

Personally I think naval needs a full rework.

However, what I find disturbing is that wartime shipbuilding is no different to peacetime one, and all we have is the same slot allocation gradually covering that full cost. No cutting corners, no usage of existing left-overs or prebuilt stuff. Hell, we can't even switch designs midway to accomodate new modules once they unlock even if it's just a radar. Really?..

I think you have a good point here, maybe the construction tab could be changed to something more realistic where the base hull is constructed with armour, engines and turret sizes and numbers set.

Then it could be that up until its finished for the specific turrets, radar, fire control systems and other more minor components like aircraft you can change them without effecting the overall construction. This would make it more worth while to go for upgraded guns, torps, radar and FCS earlier on.

This way you can pump out a 1936 battleship at game start and then upgrade the modules just before its construction ends with the latest tech.

I feel this would also help with refits of older ships too as they could be thrown in for a quick refit to upgrade them and make them useful.

This all would need dockyards to be far more flexible than they are right now as well as more scaling in that flexibility.

Check out my suggestion for naval engines too: https://forum.paradoxplaza.com/forum/threads/naval-engine-rework.1566477/

 

[Gran Strategist]

Personally I think naval needs a full rework.


I think you have a good point here, maybe the construction tab could be changed to something more realistic where the base hull is constructed with armour, engines and turret sizes and numbers set.

Then it could be that up until its finished for the specific turrets, radar, fire control systems and other more minor components like aircraft you can change them without effecting the overall construction. This would make it more worth while to go for upgraded guns, torps, radar and FCS earlier on.

This way you can pump out a 1936 battleship at game start and then upgrade the modules just before its construction ends with the latest tech.

I feel this would also help with refits of older ships too as they could be thrown in for a quick refit to upgrade them and make them useful.

This all would need dockyards to be far more flexible than they are right now as well as more scaling in that flexibility.

Check out my suggestion for naval engines too: https://forum.paradoxplaza.com/forum/threads/naval-engine-rework.1566477/

This is similar to thoughts i have posted previously, specifically in regard to upgrade costs of ships which can often cost more than building from scratch.

For tanks and planes it is quite cheap and easy to upgrade to new models with it costing a fraction of the cost of a new plane or tank.

When it comes to upgrading ships it is generally not worth it for anything over light cruisers. I think made worse by the fact that by the time you complete one the tech has advanced so much you can be jumping multiple techs ahead which seems to amplify the cost even more as, from my understanding, upgrade cost being based on the difference between the original and new design.

As a cheap and dirty work around for upgrading specifically maybe we could be allowed to use 10 dockyards for upgrading as auxiliary equipment can be constructed off site and transported in or the ship isn't stuck in dock as the core of the ship is untouched unlike the ethos for actually building a ship where it is done on site as your building from the keel up.

 

[Captain Palmtree]

As a cheap and dirty work around for upgrading specifically maybe we could be allowed to use 10 dockyards for upgrading as auxiliary equipment can be constructed off site and transported in or the ship isn't stuck in dock as the core of the ship is untouched unlike the ethos for actually building a ship where it is done on site as your building from the keel up.

Exactly my point it seems weird to me that it takes so long to upgrade turrets and radar on ships when you can just pull the turret or radar system off and slap a new one on in a matter of days most of the time, its called a class mod for a reason.

 

[HugsAndSnuggles]

As a cheap and dirty work around for upgrading specifically maybe we could be allowed to use 10 dockyards for upgrading as auxiliary equipment can be constructed off site and transported in or the ship isn't stuck in dock as the core of the ship is untouched unlike the ethos for actually building a ship where it is done on site as your building from the keel up.

And at which point would you differentiate "refit" from "building ship in stages"? Latter already benefits from increased speed via Refit Yards (making it somewhat on-par IC-wise with just building a ship) and saves metal (because magic); buffing it further by doubling amount of dockyards you can assign to refitting would turn it into the only way to build ships competitively.

 

[Gran Strategist]

And at which point would you differentiate "refit" from "building ship in stages"? Latter already benefits from increased speed via Refit Yards (making it somewhat on-par IC-wise with just building a ship) and saves metal (because magic); buffing it further by doubling amount of dockyards you can assign to refitting would turn it into the only way to build ships competitively.

Not saying my workaround is perfect but it seems like the simplest option to make refitting larger ships worthwhile. At least then you can refit ships in timescales that are quicker than building a new ship.

When it takes longer to refit an ok ship than it takes to build a good one, what is the point in refitting to get one good ship when you can build a good ship quicker and still use the ok ship while you build the new one. And then once the good one is finished you have both an ok ship and a good ship?

 

[HMCS Warrior]

When it takes longer to refit an ok ship than it takes to build a good one, what is the point in refitting to get one good ship when you can build a good ship quicker and still use the ok ship while you build the new one. And then once the good one is finished you have both an ok ship and a good ship?

Maybe you could have some way of accounting for the need for periodic refits. IRL ships need periods in dock every so many years/miles for overhauls of key systems. This would give you a reason to undergo major work anyway.

 

[PorkchopSandwiches247]

Ive always thought it was odd dockyards didnt have the production efficiency element.

Unfortunately, the game already kind of has an issue with later ship models and modules not really being well balanced, and adding a production efficiency element would make them even less worthwhile (since going from say 1936 hull to 1940+ would tank your efficiency and thus production). For example, I think the ‘meta’ is still to use heavy weapons 1 and never upgrade them, since they reduce visibility so much in future versions. Which is really pretty strange to me.

So youd probably end up with a quantity over quality thing in game, even though (as I understand it) naval research advances in real life were particularly sensitive to quality based improvements. Ie a ship based on 5+ year old tech would have not only lost to but wouldve been completely outclassed by a more advanced version of that same class of ship.

 

[davidc929]

This is similar to thoughts i have posted previously, specifically in regard to upgrade costs of ships which can often cost more than building from scratch.

For tanks and planes it is quite cheap and easy to upgrade to new models with it costing a fraction of the cost of a new plane or tank.

When it comes to upgrading ships it is generally not worth it for anything over light cruisers. I think made worse by the fact that by the time you complete one the tech has advanced so much you can be jumping multiple techs ahead which seems to amplify the cost even more as, from my understanding, upgrade cost being based on the difference between the original and new design.

As a cheap and dirty work around for upgrading specifically maybe we could be allowed to use 10 dockyards for upgrading as auxiliary equipment can be constructed off site and transported in or the ship isn't stuck in dock as the core of the ship is untouched unlike the ethos for actually building a ship where it is done on site as your building from the keel up.

How about you use dockyards to build the main part of the ship, so the hull and engines. It military factories are used to build the guns, radar, torpedoes, etc. Then ships become a bit like divisions. You select you want to build five battleships. It then pulls together the equipment from the stockpiles, say five completed hulls, the correct number of guns and so on. And it takes a period for these to be fitted out. If you are short of specific equipment it can't be fitted out until the shortfall is made up.