On other idea, what if shields was a "fleet wide defense" only. Aka the shield value of all ships was added together into 1 big shield, basically a "fleet HP" value?

Right now, defenses are local, the only way to ensure a ship gets to benefit from a defense is either:

1) For that ship to have it
2) To use combat roles to ensure that ship is never targetted.

But of course that last one only works so well. But if shields were a fleet resource, you can always ensure every ship benefits from teh protection. Now in this model shields would likely be weaker than armor in terms of bang for your buck....but again with armor the protection is totally lost if the wrong ship is targetted, but with shields they always have value....that's the trade off you pay for.

Strategically then, if you can optimize your role types against an enemy fleet, than high armor on the right ship would provide you the greatest benefit. Otherwise, a high shield fleet ensures a certain standard level of protection that is good against a variety of fleets, never as optimized as an armor fleet can be, but never as vulnerable either.

Hard counters being special mods sounds good to me.  I wonder if there's a way to get the math to work out in such a way that defenses do counter certain weapons better.

Just going down this rabbit hole you might be able to get things to work out in this order. First I would balance out the comparative damage per phase by increasing the damage of both missiles and beams.  CDP shouldn't be exact and kinetics should still have the highest CDP as this calculation doesn't factor in range.  This leaves missiles with the highest damage output and beams in a middle ground.  Right now beams and missiles get one volley before being in range of kinetics where they are decimated.

Comparative Damage per Phase (CDP) = Accuracy*Damage/(Mass*Reload)

Using your explanations of defenses above we can now see how they will stack against different weapon types.

Evade vs :

Kinetic:  Because kinetic in low damage, evade is far more likely to trigger against this option.

Beam:  Evade would be second best against beams due to beams higher damage output

Missile: The high damage output of missiles rarely can be evaded so evade tends not to work against missiles

Deflect works pretty well against all forms of damage acting as kind of a middle ground just absorbing incoming damage until they reach zero.  While not specialized it also tends to be the weakest form of damage mitigation because at some point it will fail and the ship will take all the damage.

For mitigating damage I would create a rule set saying:

If (incoming damage) > (armor) then roll a number between 0-(armor) reduce incoming damage by that roll.

If (incoming damage) < (armor) then roll a number between 1-(incoming damage/X) reduce incoming damage to that roll. X is probably a number between one and two.

Basically some amount of damage will always get through with this rule set.  This means you could set it up such that:

Mitigate vs :

Kinetic:  Because kinetic has the highest fire rate it will do consistent damage to the enemy fleet and be the strongest against armor overall.

Beam:  Mitigate would be second best against beams due to the decent fire rate on beam weapons

Missile: Armor negates most of missiles power if the armor has a higher rating than the missiles because if reduces their damage so much.

I can't tell you how excited I am to hear this!  I am a huge fan of military history having followed it for a long time.  My only suggestion is make sure their are multiple viable endgame strategies so that carriers and cruiser spam don't become the only options available. 

When it comes the the A.I, I would suggest the ability to save player loadout templates that the A.I. can then access and use.  In my games when I've left ship designs available to an A.I. they almost always choose to build my designs over their own.  For example I built a ship called the Sea Lord when I played the Torian's awhile back.  During one of my recent games before the last update, the A.I. built a bunch of Sea Lords rather than the ships they designed.  They had a few of them, but the A.I. seemed to think that ship was better overall and it did perform better in combat from what I could tell.

Sea Lord Loadout:

My suggestion is in a future DLC add a option that will save the loadout as well storing it in a place the A.I. can access and use it for their ships.  

Better yet have this occur whenever the player designs a ship.  The more ships loadouts the player puts together the more the A.I. has access to.  It is my assumption here the the A.I. would have an easier time choosing a loadout that fits what it needs over trying to design its own.  I could be mistaken though.

In discord we discussed how currently kinetic weapons dominate ship battles.  I've found missiles and beams to have little use outside their barrage functions.   I think this is in large part due to the fact missiles and beams have relatively low damage output compared to kinetic weapons as shown below.

An interesting question in the context of this game is how to make weapons relevant?

In an FPS a shotgun is good in close quarters but not very useful in an open field.  A sniper is the opposite of that.  If I was developing an FPS it is easy to know which weapons I'm designing around based on how I layout the map.

GalCiv combat and space in general is different.  There isn't really a map layout unless you're in nebulae or asteroid fields.   It really creates a difficult problem to make each weapon feel useful and not allow one to dominate the encounters.  I've thought about how I would do it and here's two ideas I came up with:

1.  Go For Balance:  In this case you would attempt to get the weapons closer to a balanced state by increasing the CDP of missiles and beams significantly.  In this case I would still keep their CDP below kinetics, because they get the first volley's and have the barrage function, but this would make sure they're still functional in combat. Kinetics should still be the top weapon choice for ship to ship combat.

Then focus on making the ship classes feel different and placing the emphasis of combat in fleet composition and construction rather than what weapons they're fielding.  This would be my preferred option overall.   

2.  Right Weapon for the Right Job:  In this case the player is attempting to pick weapons for the job they can do and each new Tier of the weapon improves or adds to that weapons capability to do a job.

Current CDP of each weapon type.  (I think plasma pods is supposed to have a higher CDP than disruptor banks, but currently it doesn't.)  

I'm only accounting for weapons that don't use special materials in this graph.

This is how I think I would go about making weapons useful for the right jobs.  Either give them special perks that just tell the player how they should function (easy) or make the math work out such that the weapons are better in certain cases (harder)

An example of this is missiles.  They already shoot first out of any weapon so we can lean into that initial volley and make it count by increasing the amount of damage they do in each higher tier.

Tier One:  These missiles are really only useful for they're barrage function and not much else.

Tier Two:  These missiles are somewhat functional in combat, able to destroy most unshielded fighters with two missiles

Tier Three:  These missiles are functional in combat, able to destroy most unshielded targets with their initial damage output, but slow reload will leave them sitting ducks if most of the targets survived by using shielding or evasion.  You might even further slow their reload just to make sure they get one volley before having to wait.

Tier Four:  Same as Tier 3 except with more power.

The CDP graph would look something more like this:

The large jump from 1 to 2 gives it the ability to destroy unshielded fighters, and the large jump from tier 2 to 3 allows it to destroy almost any unshielded target.  This could be done by either increasing the damage of each tier dramatically.  I used 2, 6, 20, 30 for each tier in my graph. Or you could add a perk that if enemy target is unshielded then this weapon gets +200% damage to that ship.

For beams I would lean into their high accuracy committing them to a fighter, bomber, frigate, and destroyer removal system.  They almost always land their shots, but it will take them a long time to whittle through the high health and defenses of cruisers, battleships, and dreadnoughts.  They would still need a boost to their damage output so they could do damage to those targets, but not so much that they can eat into the larger ships.

For kinetics they would struggle to hit anything smaller than a cruiser late game relegating them to more of a broadside cannon role in the end game.

Anyway I'd enjoy some more conversation on how other people think weapons should function.  This is the month they're planning to do a lot of that work based on the roadmap and discord comments.

I had one more idea I wanted to throw out here.

A ship ability called Weapon Scale

Basically this would function as a modifier to ship weapons and a way to build out and distinguish more ship classes.

The cannons carried by battleships were much larger than those carried by cruisers and frigates.  This modifier would help represent that.

You could do something like a small, medium, and large system.

Small:  Weapons stay about the same

Medium:  Increases the cooldown of weapons, but increased damage and range.  Small and tiny ships receive a slight evasion bonus against these weapons.

Large: Large increase to the cooldown of weapons, but a large increase to damage and range. Small and tiny ships receive a substantial evasion bonus against these weapons,  and medium ships receive a slight bonus.

This let's you do something like splitting tiny ships into 3 roles: Fighter, Bomber, and Muti-role each class using Small, Medium, and Large weapons respectively.

You could also limit Multi-role class to specific civilizations by making them unlock them in the culture screen 

A possible guiding principle for the combat system might be that a small weapon can't hurt a strong defense. As Asimov put it, "a shield that could take a Q-beam without working up radiation."

This suggests that a given defense can absorb most of the damage from each attacking weapon, if the weapon's damage is below a given level. Even many small attacks just wouldn't do much; each one would mostly just bounce off. It would be like a fleet of destroyers attacking a battleship with their guns; their shells just aren't strong enough to penetrate.

So the best weapons against heavily defended ships would be ones with high damage per attack. They would get that in trade-off of having a long cool-down – missiles, in other words. (Or large scale weapons, as Halicide suggests.)

But as you advance up the tech tree, and get more powerful weapons, even a kinetic weapon, which starts out with small damage per attack, could get strong enough per attack to get past an obsolete defense.