Production Cost/Maintenance

Production Cost lowers both the buyout cost and the number of turns to complete.

The cost of an item is expressed in manufacturing. Every 1 point of manufacturing generated by a planet is 1 point it can put toward building a building or ship etc.

So say we have a planet that's churning out 50 manu per turn. It wants to buy a 300-manu ship. It takes 6 turns to do so.  If it wants to buy the ship, then it has to spend 10 times the manu value in cash - so it can buy it out for 3000 money. 

Now, say you've picked up a tech that reduces manufacturing cost by 10%. The ship now costs 270 manu. You still take 6 turns to build it, but you do get to carry over 30 manufacturing for the next thing in the build queue. Or you can buy the ship out for 2700 cash.

The real problem is, it's not very difficult to create a planet which can manufacture anything in 1 turn by itself with no help from any other planets. So manu reduction is kinda pointless atm.

Maintenance has a similar problem, in that cash is obscenely easy to generate. You're never really under any pressure financially. So maintenance reduction is also largely worthless right now.

This means that logistics and mass alteration are the only two choices worth taking.

You can break them down into the 'positive' choices which increases bonuses (increased mass and increased logistics) and 'negative' choices which reduce penalties (reduced cost and reduced maintenance) and the concept appears to line up with you needing to take one positive and one negative. I suspect it's meant to be a choices between logistics mixed with maintenance versus mass alteration mixed with production cost reduction (so large expensive ships vs hundreds of cheap ones), but currently we're able to use both positive ones because the existing penalties are insignificant.

You're thinking the same thing I am, then.

Without addressing the manufacturing cost of each part one by one, the economy really needs a kickstarter and some deflation for the penalty reductions to matter.

Mixing and matching is nice, but without more obvious tradeoffs, even the mixing and matching is pointless. I don't even have to boost research as things stand, just create an industrial base, set up a few trade routes and the game plays itself if I make a big enough and mean enough ship design where manufacturing cost is no object.

I'm wondering if we can't get it so that we can more reliably fit a single engine on all ship designs, where stacking engines creates rapidly diminishing marginal returns and then up the manufacturing costs of subsequent engines or adjust them for other parts so most people are stuck between 2 and 5 tile ship designs without making massive tradeoffs, AI included.

Making those changes is going to completely screw with the metagame, however, and may be better off modded in than officially patched. The problem is definitely not one of too large or small tech trees, but what costs are associated with each, and needs someone with stronger math skills to tweak things.

I wonder if I should dig up some modding tutorials and then play with the values since none of it involves creation or new assets...

You can implement a stacking penalty on engines pretty easily tbh, though I'd instead suggest just giving each hull class an extra -20% speed, so you need more engines to get the same speed from a big hull.

The way the game is built that you can only produce/buy 1 thing per turn makes production costs/maintenance irrelevant, and large overproduction a waste. hence the need to micro sliders depending on what you're building.

Even if the those bonuses were x2 or x3 their listed amounts I would still pick miniaturization.

Reason? As it was said, production points are cheap, the true resource is hulls/game turn. And credits are very easy to maximize on your empire.

In fact without crew or leaders xp on ships or fleets, fleet upgrades are a waste or money.

You also have to consider location amortization time costs: Ships don't appear on the front lines. So with better hull capacity you get more components per turn towards the enemy and this is also the reason why you should be building always bigger ships (large ships = more space = more components per turn towards the enemy).

Hull capacity always wins.

I don't know that hull capacity by itself independent of any other factors can be fixed (since fleet logistics will just take over where hull capacity doesn't win)

But in conjunction with reworking maintenance and production costs to be more advantageous for smaller empires while rapidly expanding empires need the logistics and hull capacity more, should the global hull-based capacity bonuses get lopped off and the extra hull capacity added into a new miniaturization tech tree made up entirely of exclusive techs?

So every time you go up the tech tree you can only miniaturize one tech at a time, rather than having the miniaturization bonuses within the tech itself. Certain techs, particularly around missiles and defenses, should mostly be untenable to use past the entry-level point without miniaturization research and other than racial bonuses to architecture, ship hull capacity should be pretty consistent with only minor variation across races.

If you can make miniaturization tech tree research durations equal or exceed the tech they are trying to miniaturize, you should be able to get huge payoffs on miniaturization for heavy investments, or you will be forced to make do until you have no choice but to look into either miniaturization or logistics as a larger army, whereas minimal production or maintenance costs should let smaller empires outproduce larger ones defensively but be unlikely to have the logistics or hull capacity to stick on the engines and life support necessary to efficiently travel larger chunks of empty space...

If you do this right between these four axes, then the presence or absence of tech trading and tech brokering becomes more critical and diplomacy bonuses become more imperative to not have to hemorrhage credits assuming the 'AI buys everything forever from each other and then laughs at the player heartily until the player's specialized strategy trumps the generalist AIs' problem in the early-mid game. HP bonuses also become that much more critical because players can't just make tank ships with extra speed and threat and stacked defenses, and may need the ships themselves to be more durable for some weapons loadouts.

Life support systems only have value for scouts, survey ships and freighters. For everything else you can build cheap constructors and lay out a route of starbases toward your goal. Once you conquer or settle a new planet in the region, you can decide whether to keep the starbases or scuttle them. And if it comes to war the AI will never attack a starbase - I've never seen it even try.

I used to always pick the miniaturization bonuses; now for weapons I usually pick a range extension and then some power increases. Miniaturization doesn't accomplish anything that a power increase doesn't - at least not that I can see. For engines, I always pick miniaturization. Adding +1 to movement is less valuable than being able to add a second (or third) engine. Get a logistics increase or two and you can put 50-100% more ships in a fleet than the AI can, and since combat is weighted in favor of numbers over quality, a large number of inferior ships will win (not that your ships are likely to be inferior). (By that I mean that 8 ships doing 5 points of damage apiece will fare better than 5 ships doing 8 largely because as a defense fails it absorbs all of the last shot - 5 in one case and 8 in the other - and because 8 small ships have more hull points than 5 small ships and so lose less firepower per hull destroyed). Not that you are likely to have inferior ships...

Seagoing warships once needed high tactical speed; nowadays they need to be able to steam at relatively high speeds for long distances (say, Norfolk, VA to the Persian Gulf). The relevant techs in GalCiv would be thrusters (which I never ever use unless I have 6 points I can't use any other way) and engines. If I can build ships that can get to the front in ten turns and the AI builds a ship of comparable power that moves half as fast, then assuming I can build as many ships I will probably win. The AI NEVER builds ships of comparable power to mine (it builds a lot of basic-level tiny-hull garbage) but my point is that high movement speed across the map allows you to get first to planets, resources, artifacts and to battle. And if you are fighting in the enemy's own space, superior speed allows you to take out multiple fleets, starbases and planetary defenses in one turn - and do it again the turn after, and then again until you invade and conquer.

The single 'constraint' on warship design in GC3 is having to put all three defenses on every combat hull (or deciding which one of the three you could live without). Getting a lead in research over the AI is tougher early on when you are covering the map with colony ships and constructors, but once you get settled and start researching techs to improve your research the AI can't keep up. Plus, it usually takes the 'wrong' choice, like reducing manufacturing cost. I have found that high weapons strength is less valuable than a range advantage, rapid-fire and stout defenses.

So far I haven't played on 'godlike' but at all of the lower levels I can out-settle, out-research and out-build the AI. I've had campaigns where my fleets of small and medium ships wipe out enemy armadas and never take a point of hull damage. That's gratifying - but not really fun. But to recap, I find that researching logistics techs, and having 2-3 engines and at least 1 of each defense on a warship, is the recipe for annihilating the AI.

Actually, you don't have to do that. This trick can be accomplished by combining capital and escort role ships into a fleet:

- Design an attack ship that has no defenses. Leaving out defenses completely leaves you lots and lots of room for weapons, weapon augments, and other stuff you want your attack specialist to have. Set the role of this ship to "capital". Change the computer offered role if it's something else. You want all your attack ships to be "capital" even if they are tiny hull ships.

- Design a defense ship that is all defense and hitpoints and augments that make it even tougher. Ignore all weapons. All room is for defenses and whatever drives you need. This ship needs no weapons. It's all defense. Set the role of this ship to "escort".

- Build a fleet where 2-3 ships are defense ships and the rest are attackers.

In battle the enemy will always target the defense ships before targeting the attackers. This is because the role "escort" has a higher target priority than "capital" in every ship role's target prioritization list. There are no ship roles that would target capitals over escorts. This means that the enemy must destroy your ultra-tough defenders before it can even get to try to hit your, very soft, attack ships. Those defenders can take a lot and that takes time and in all that time your attack ships keep pouring fire on the enemy. Essentially, your attack ships use shared defenses leaving them lots of extra room to pack more weapons.

It's actually quite a beautiful sight when a big fleet of ships crammed full of miniaturized missile tubes flush their birds...

Usually this tactic leaves your ships completely unharmed as the defenders can get really tough when they have nothing but defenses. Sometimes you can lose a defending ship if the enemy is very strong. Of course, if they do get past the defenders your attackers are toast after that but adjust the number of defenders to your level of paranoia. The less defenders you need the more powerful your attack wing.

That's an interesting strategy and I will have to try it. The US Navy has kicked around an 'arsenal ship' concept for decades, that being a cruiser-or-larger hull stuffed with missiles (and exotics like a rail gun) but having no defensive equipment. It is efficient but leaves a fleet prone to lose a large amount of its firepower if equipment should fail or be damaged. By temperament I prefer to build more of a 'general-purpose' ship - makes building fleets less finicky - but I see your point.

Just had an interesting experience, in the sense of 'May you live in interesting times.' I declared war on a minor race. My fleet of destroyers (small) took out his ships in space and his starbase defended by three tiny hulls. Then I sent my frigate (medium) fleet against his homeworld... and got smashed. The loss ratio was about 1-to-1 as I traded frigates for tiny hulls (ouch!). That was half of all my frigates at the time...

I immediately put the one planet in the area on full military production building destroyers (1 per turn) and put all of my core planets on full military building frigates (1 per turn at 5 different shipyards). While the new fleets are in transit I've been able to use my destroyers to keep him from putting up a shipyard. And 20 more frigates are on the way...

So this highlights the superiority of numbers (30 tiny hulls) over quality (5 frigates), at least in a situation where the AI can mass its ships. It also shows what happens when you get overconfident.

A single ship can target only one enemy ship at a time. That puts a bigger hull at a disadvantage against a swarm of smaller hulls as it renders most of its larger firepower useless. It must kill the smaller ships singly while they can attack en masse. This can be seen especially clearly with carriers where the swarm effect is made even more powerful by the extra logistics points you get from using the fighter modules.

I wish, too, the AI were more intelligent in that. In planetary defenses it gets this automatically right (if we ignore its poor ship designs...) as the ships are combined in a fleet by default but elsewhere I have encountered many times it just stacking individual ships in a single hex without making any fleet at all thus allowing you to shoot them down one-by-one with your fleet. Makes no sense. Hope SD fixes that one.

The attack-defense specialization strategy is by no means mine. The specific galciv implementation of it is described in the wiki. The tactic itself is, of course, much older than that. Basically, it's the space version of the medieval pikes+archers battle formation.