The ship classes discussion...

[Foraven]

Well, i've been toying with the idea of adding more ship classes to my mod and i'm wondering what i should truly add. The vanilla classes already cover everything we could possibly want, at least when it comes to matching ship sections and weapons. Any new ship classes isn't likely to add anything truly new, only more of what is already there. From what i seen on the net, what most modders have done is add more powerful version of what already is in the game. More ship sections, bigger ship size, a ship class for each new spaceport tech... But does it change anything to the gameplay or how we actually play?

I can easily add more ship classes, but i'm wondering what would really make a difference rather than just more of the same.

 

[Comm Cody]

Suggestions that I've heard of,

A fleet support ship, cruiser size, can carry three auras, able to mount only small and medium weapons. Weak in combat, but can support a fleet with its Auras. Potentiua

A flagship like 3-4 time larger than a battleship, gives influence, but is weaker than conventional battleships, and gives you a happiness and influence malus if you lose them. In effect, a giant political ship to keep the masses in line but not an end all be all weapon. No limits on it except for fleet capacity and money.


Recon/spy ship, massive sensors. weak. ability to stealthy infiltrate other empires, but if caught can cause a major war.

 

[Foraven]

Suggestions that I've heard of,

A fleet support ship, cruiser size, can carry three auras, able to mount only small and medium weapons. Weak in combat, but can support a fleet with its Auras. Potentiua

A flagship like 3-4 time larger than a battleship, gives influence, but is weaker than conventional battleships, and gives you a happiness and influence malus if you lose them. In effect, a giant political ship to keep the masses in line but not an end all be all weapon. No limits on it except for fleet capacity and money.


Recon/spy ship, massive sensors. weak. ability to stealthy infiltrate other empires, but if caught can cause a major war.

Hum... Support ship is easy to make, but not sure about 3 auras (and AI don't seem to know how to use them). Flagship is meh, big fat target that take lots of fleet space and provide little combat wise (not a fan of such ships). Recon ships is easy to make, but stealth and spying isn't something we can currently mod in. I'm also thinking how the AI will deal with the ships, both empire AI and fleet AI.

What seem to me the most practical and useful is specialized ship classes, ships that have fewer options but great bonuses at certain things. Like bonus to sensors, speed, ftl ; recon ship. Or bonuses to armor/shields ; heavy cruisers. All medium/large guns : Battlecruisers. I see quite a few possibilities on that front. If limited by specialized tech, that could greatly change how fleet are composed since you would not have access to all the possible classes, and that make it easier for the AI to manage.

 

[Naselus]

In my opinion, adding ship classes is only really necessary if you're extending the length of the game. Given the modular construction of ships, I think specialization by class is unnecessary - we can make role combinations based on section and module picks. I reckon the most fun option is for class to simply represent hull size and let the player specialize however he wants from there.

 

[Foraven]

In my opinion, adding ship classes is only really necessary if you're extending the length of the game. Given the modular construction of ships, I think specialization by class is unnecessary - we can make role combinations based on section and module picks. I reckon the most fun option is for class to simply represent hull size and let the player specialize however he wants from there.

What i have in mind is a bit different than just swapping modules and ship parts; the new ship classes have special modifiers, different number of utility slots, different cost and build time. They are especially good for the AI since i can set fleet ratio for them and different weighs for parts and modules so they end up being different designs than the base ships (the AI currently has only one design per ship class). Unlike the base classes of ships, those specialized classes may be linked to special tech or ethics so not everyone gets them. You also have the choice of not researching them at all if they are not useful to you.

 

[Naselus]

What i have in mind is a bit different than just swapping modules and ship parts; the new ship classes have special modifiers, different number of utility slots, different cost and build time. They are especially good for the AI since i can set fleet ratio for them and different weighs for parts and modules so they end up being different designs than the base ships (the AI currently has only one design per ship class). Unlike the base classes of ships, those specialized classes may be linked to special tech or ethics so not everyone gets them. You also have the choice of not researching them at all if they are not useful to you.

Yeah, I understand your premise. I just think that that ultimately railroads the player into using certain classes for certain roles (because by definition, it does), which undermines the whole point of having a ship designer and ultimately leads to one 'best' way to kit out any given ship class - which is precisely why it's easier to write AI for it (which is really the only sound argument in favour of the idea, imo).

I don't think it's necessarily bad - many games have used this approach, it's logically consistent and if handled well it works. It's just not the philosophical approach that I'd take to the entire question of ship design. I'm more inclined to stick to the idea of 'you get a new class which is X% bigger every Y years' and then letting the player kit them out however they wish, while attempting to achieve a balance between all the different elements. So while I wouldn't object to the idea of role-based bonuses on specific sections or modules, I'd rather the player had the option to make his cruisers all support ships or his battleships all concentrate on point defense rather than having 'hull type X gets +100% bonus to PD' which preordains that it will now always be used for point defense duties.

 

[Foraven]

Yeah, I understand your premise. I just think that that ultimately railroads the player into using certain classes for certain roles (because by definition, it does), which undermines the whole point of having a ship designer and ultimately leads to one 'best' way to kit out any given ship class - which is precisely why it's easier to write AI for it (which is really the only sound argument in favour of the idea, imo).

It also work better with the way the game is set up. The game can generate all kind of ship designs but there is no way to define what it's best used for; players may know but the AI won't. I don't see why you think it's bad to "railroad" players into doing things, it's far easier to balance a game when you know how players are likely to play it. And adding new ship classes don't remove the stock ones (at least i don't plan to), the freedom to design ships how you want them remain. New ship classes just add options you don't have otherwise (i don't want to make base ship classes obsolete, that's why i restrict freedom on specialized ships).

I don't think it's necessarily bad - many games have used this approach, it's logically consistent and if handled well it works. It's just not the philosophical approach that I'd take to the entire question of ship design. I'm more inclined to stick to the idea of 'you get a new class which is X% bigger every Y years' and then letting the player kit them out however they wish, while attempting to achieve a balance between all the different elements. So while I wouldn't object to the idea of role-based bonuses on specific sections or modules, I'd rather the player had the option to make his cruisers all support ships or his battleships all concentrate on point defense rather than having 'hull type X gets +100% bonus to PD' which preordains that it will now always be used for point defense duties.

Restricting what can or cannot be used on a certain ship class doesn't remove all design choices (unless only one possibility remain). If i give a certain ship class bonuses for PDs or fleet defense in general, yes i do restrict that ship to a certain role but that's the whole point. If you use that ship class as a front line ship despite the bonuses for defending the fleet, either the player doesn't understand what it's meant for or there is some unforseen flaw in that shipclass parameters that allow it to be used effectively at something else (possibly making everything else less useful or obsolete in the process).

 

[Naselus]

It also work better with the way the game is set up. The game can generate all kind of ship designs but there is no way to define what it's best used for; players may know but the AI won't. I don't see why you think it's bad to "railroad" players into doing things, it's far easier to balance a game when you know how players are likely to play it. And adding new ship classes don't remove the stock ones (at least i don't plan to), the freedom to design ships how you want them remain. New ship classes just add options you don't have otherwise (i don't want to make base ship classes obsolete, that's why i restrict freedom on specialized ships).

Restricting what can or cannot be used on a certain ship class doesn't remove all design choices (unless only one possibility remain). If i give a certain ship class bonuses for PDs or fleet defense in general, yes i do restrict that ship to a certain role but that's the whole point. If you use that ship class as a front line ship despite the bonuses for defending the fleet, either the player doesn't understand what it's meant for or there is some unforseen flaw in that shipclass parameters that allow it to be used effectively at something else (possibly making everything else less useful or obsolete in the process).

See, I don't think it really does add new options. It takes away other options instead.

If we return to the hypothetical +100% PD ship, for example, then not only is that ship only really good for that role, but more importantly once you have that hull there's no good reason to ever design any other ship for that role either. You've gone from having a bunch of options to having 1 sensible option and a lot of bad options. Sure, I can still use a regular destroyer hull (or whatever) for it... but it'll be rubbish compared to the +100% PD one. It creates a 'best' option, which is inherently bad in a strategy game - if you make one strategy or design much better than all others, then it's a false choice and so the player shouldn't even be presented with the option, it should just be the default. This is pretty much why it's so much easier to balance, too - you are actively reducing the number of valid strategies, so of course it's easier. It's also restricting player choice, which I personally really dislike doing.

This is ultimately just a design philosophy point, though. As I say, it's not necessarily objectively 'better' or 'worse'. It's not the decision I'd make, but I don't think it's automatically a stupid choice to go for. The advantages are very valid; it's just the disadvantages are, too, and it's a trade-off - you can't try and have both systems co-existing.

What I would say is, you can't really take half-measures - if you want to specialize the hulls, then go the whole way with it and do make the base-game ships obsolete. Split each basic class into 2-3 separate ones (attack, def, buff) and just get rid of the normal ones, so you can maximize the advantages of the system. Having both systems side-by-side is likely to mean you have lots of hull types which aren't used.

 

[Foraven]

See, I don't think it really does add new options. It takes away other options instead.

If we return to the hypothetical +100% PD ship, for example, then not only is that ship only really good for that role, but more importantly once you have that hull there's no good reason to ever design any other ship for that role either. You've gone from having a bunch of options to having 1 sensible option and a lot of bad options. Sure, I can still use a regular destroyer hull (or whatever) for it... but it'll be rubbish compared to the +100% PD one. It creates a 'best' option, which is inherently bad in a strategy game - if you make one strategy or design much better than all others, then it's a false choice and so the player shouldn't even be presented with the option, it should just be the default. This is pretty much why it's so much easier to balance, too - you are actively reducing the number of valid strategies, so of course it's easier. It's also restricting player choice, which I personally really dislike doing.

But that's how it is in the real world; a ship specialized in shooting down missiles and fighters will always be better at that job than a jack of all trade ship that wasn't meant for that role. The flip side is a pd ship, while great at it's job is worst at everything else (or why pay more for a feature you don't use). There is still room for ships having PDs when shooting down missiles or fighters isn't your main concern and you want those ships to still provide real firepower against the enemy fleet. A jack of all trade has no real strength but no real weakness either.

This is ultimately just a design philosophy point, though. As I say, it's not necessarily objectively 'better' or 'worse'. It's not the decision I'd make, but I don't think it's automatically a stupid choice to go for. The advantages are very valid; it's just the disadvantages are, too, and it's a trade-off - you can't try and have both systems co-existing.

What I would say is, you can't really take half-measures - if you want to specialize the hulls, then go the whole way with it and do make the base-game ships obsolete. Split each basic class into 2-3 separate ones (attack, def, buff) and just get rid of the normal ones, so you can maximize the advantages of the system. Having both systems side-by-side is likely to mean you have lots of hull types which aren't used.

I disagree there. I want the option to specialize, but i don't want it the be THE only option. I'm perfectly fine with being able to build jack of all trades ships or having the option to swap around parts and weapons. It's a question of efficiency vs versatility, both are good answers.

 

[Naselus]

But that's how it is in the real world; a ship specialized in shooting down missiles and fighters will always be better at that job than a jack of all trade ship that wasn't meant for that role. The flip side is a pd ship, while great at it's job is worst at everything else (or why pay more for a feature you don't use). There is still room for ships having PDs when shooting down missiles or fighters isn't your main concern and you want those ships to still provide real firepower against the enemy fleet. A jack of all trade has no real strength but no real weakness either.

Yes, but there's different, more logical places that the specialization can be inserted. Such as on the computers, the sections themselves, the weapons etc. This maintains the player's freedom to shift between different roles for different ship sizes over the course of the game, rather than some roles becoming 'full' when a given hull is researched.

I disagree there. I want the option to specialize, but i don't want it the be THE only option. I'm perfectly fine with being able to build jack of all trades ships or having the option to swap around parts and weapons. It's a question of efficiency vs versatility, both are good answers.

It will be the only option, though. The bonuses that the specialized ships get will be cumulatively better than the jack of all trades ships can manage, making combined fleets of specialists a better option than balanced. This is pretty much the problem that GC3 had - specialized planets were an order of magnitude better than balanced ones, and it was impossible to correct it. Why would I ever need versatile ships when I can just make my fleet out of a combination of specialists? The fleet is just as versatile as one made up of jack-of-all-trades ships, but is also pound-for-pound better at every task.

 

[Foraven]

Yes, but there's different, more logical places that the specialization can be inserted. Such as on the computers, the sections themselves, the weapons etc. This maintains the player's freedom to shift between different roles for different ship sizes over the course of the game, rather than some roles becoming 'full' when a given hull is researched.

It would indeed be better that way, but limitations in the code limit the possibilities. For example there is no way to influence build time with ship components (or at least none of my tests worked). There is also no way to make a ship section unlock with a certain tech, that's why we get strike crafts on cruisers and battleships even when we never researched them. And last, the AI cannot use combined arms the way it is coded, it won't make different designs for different tasks like players can.

It will be the only option, though. The bonuses that the specialized ships get will be cumulatively better than the jack of all trades ships can manage, making combined fleets of specialists a better option than balanced. This is pretty much the problem that GC3 had - specialized planets were an order of magnitude better than balanced ones, and it was impossible to correct it. Why would I ever need versatile ships when I can just make my fleet out of a combination of specialists? The fleet is just as versatile as one made up of jack-of-all-trades ships, but is also pound-for-pound better at every task.

My personal workaround is add both bonuses and maluses. A ship with better firepower and speed? Less hull hitpoints, no point-defenses. Great armor, shields and hull; ship is slower and expensive. Ship is great with pds? Cannot use large weapons, shorter range for it's weapons. To avoid specialists from owning generalists, i give major flaws to the specialists so there is a weakness to exploit to beat them. Combined arms is great but only up to a point; it's a management hell when you need to count how much of everything you want in and make sure every specialists are at the right place at the right time. A specialist ship you don't need (or want) at a particular time is a dead weight you have to pay for and may be liability. A generalist fleet don't have such issues.

 

[FullMetalFox]

Honestly both systems can coexist, but as with most coexistance they have to both adapt to each other.
For specializations, the question is how those specialisations are done, and how much risk/disadvantage they have, even high specializations can make an ship not be the only option.

An example, that theoretical +100% PD ship, okay lets say that thing is littered with small slots and gets some good bonuses, but it doesnt have any large, or maybe even medium weapons.
I wouldnt use that. Why? Because even if say 10 of those in a fleet of 100 ships could do the same job as if 100 ships were equiped with some PD, that means every single one biting the dust reduces the whole fleets PD by 10%. Then there is also the question of coverage and how much those ships would tend to YOLO into the front lines.

Now comes the "but for those PD you save on the other 90 ships you could use some other weapons instead, so the rest of the fleet would get stronger" argument, well small slots, with Stellaris's section designs i will always have some small weapons unles i go with only L-BS, and after midgame those small weapons on big ships will rarely see more than the last 10-20% of a fight, the rest is dominated by long range high accuracy weapons, which are mostly L exclusive unles you also count M size torps.

 

[Qilue]

There is one ship class that needs to be added, an armed science cruiser perhaps with a limitation where you can only add up to medium weapons. The mod that adds ship classes has one, but it's little more than an upsized science ship due to an engine limitation where civilian ships can't be armed.

 

[Foraven]

Honestly both systems can coexist, but as with most coexistance they have to both adapt to each other.
For specializations, the question is how those specialisations are done, and how much risk/disadvantage they have, even high specializations can make an ship not be the only option.

An example, that theoretical +100% PD ship, okay lets say that thing is littered with small slots and gets some good bonuses, but it doesnt have any large, or maybe even medium weapons.
I wouldnt use that. Why? Because even if say 10 of those in a fleet of 100 ships could do the same job as if 100 ships were equiped with some PD, that means every single one biting the dust reduces the whole fleets PD by 10%. Then there is also the question of coverage and how much those ships would tend to YOLO into the front lines.

Now comes the "but for those PD you save on the other 90 ships you could use some other weapons instead, so the rest of the fleet would get stronger" argument, well small slots, with Stellaris's section designs i will always have some small weapons unles i go with only L-BS, and after midgame those small weapons on big ships will rarely see more than the last 10-20% of a fight, the rest is dominated by long range high accuracy weapons, which are mostly L exclusive unles you also count M size torps.

Yup, you sum it better than i managed to. Ships "yoloing" might be a big issue when we have no control on what the ships are doing...

 

[Macaurelius]

I was thinking about this very question as I was looking through your mod, Foraven. My major concern is that it creates an optimum fleet composition.

An air-defence warship is still expected to be able to engage other surface combatants, and submarines, and land marines, and so on. It just does it slightly worse than the dedicated hull. In the grand scheme of things, the difference between the various types of destroyer in use by the Royal Navy is probably more down to module choice than fundamental hull design. A British Type-23 anti-submarine frigate is about 4,900 tonnes, whilst a Type-42 guided-missile destroyer was 4,500 tonnes. These are, in Stellaris terms, both "Destroyer"-class hulls. The USN Ticonderoga-class guided-missile cruiser is almost double the size at 9,600 tonnes (According to the USN, and Stellaris, I'd call this a Cruiser). It does the same thing as the Type-42, but is of a different hull class.

This example is instructive to me. It suggests to me that hull-type and role are independent of each other. Just as you can have super-carriers, carriers, and aircraft-cruisers, you can have any hull-class doing any role.

The way I am approaching this issue in my ship-components mod is as follows (I'm still stuck on working my way through Master of Orion-inspired components and crew types):

1) Make variable-size PD mounts (meaning that you can have something that fits into M and L slots too)
2) Make more modules that provide more slots
3) Make components that provide a bonus to PD-fire-rate / accuracy.
4) Make more intermediate PD-type weapons rather than the 3 vanilla weapon-types

Thus you provide the option for players to make anything from PD-zilla to a jack-of-all-trades, as their whim and budget allows.

EDIT: What convinces me that a role should be independent of a hull-size is the idea of the PD-destroyer. If I want to make a PD-battleship, you're telling me that the ship is WORSE even though it has more guns?

 

[ShadowDragon868]

There is one ship class that needs to be added, an armed science cruiser perhaps with a limitation where you can only add up to medium weapons. The mod that adds ship classes has one, but it's little more than an upsized science ship due to an engine limitation where civilian ships can't be armed.

Actually, the science cruiser now has a devastatingly effective weapon suite, courtesy of a destructo-aura which scales very nicely the more enemies who are present, yet remains able to wipe out any hostile wildlife. It's actually kind of hilarious; If you sail a few of those into a gigantic fleet engagement, they'll wreck the enemy fleet while the enemy are busy shooting at your nominal warships.

 

[Foraven]

EDIT: What convinces me that a role should be independent of a hull-size is the idea of the PD-destroyer. If I want to make a PD-battleship, you're telling me that the ship is WORSE even though it has more guns?

Depend on what you consider "better". A battleship will be able to kill more fighters due to it's amount of guns and be able to take whatever thoses fighters throw at it... But is it a practical design? A battleship is expensive, it takes a long time to build and is slow as heck while PD weapons are rather short ranged; do you think it's a good choice to make such a ship? Nothing prevent you from making one but the game will teach you that there are better things you could do with that battleship hull than put pds on it.

Yeah, i created new ship class while i could have made modules instead; but not all the effects i wanted could be done that way and some modules would have been hard to balance. Easier to make a whole new hull, and the plus side it's easier to get the AI to use the new features that way.

 

[Macaurelius]

The choice, in 'real-world' navies is determined by doctrine and finance. They choose a hull size that fits the role in their navy. In the Soviet Navy, this was a battle-cruiser (Kirov-class battlecruiser at 28,000 tonnes), in the French and Italian navies this is a large destroyer (Horizon-class, at ~ 8,000 tonnes), in the US-navy a small cruiser (Zumwalt-class at 15,000 tonnes), and so on. (For reference, the HMS Vanguard (WW2 British Battleship) and US Iowa-class battleship weighed in at 45,000 tonnes, whereas the German pocket battleships of the same period were ~30,000 tonnes).

I don't necessarily agree with the idea that the "the game will teach you", because we are discussing ways to change the rules. A single PD-battleship might be such an efficient PD-platform that it is worth the time and cost to build because it makes your fleet immune to missiles and fighters; it is a question of whether or not the resulting mod makes them efficient. If the AI never uses fighters or missiles, then they're a waste of money. Or if the AI builds super-efficient PD-battleships all the time, then their utility will necessarily change.

If you've gone for a PD-specific variant of one hull-class, why not create a PD-variant of each? For example, a PD-corvette, a PD-destroyer, a PD-cruiser, and a PD-battleship? For that matter, why not create a missile-variant, a beam-variant, and a projectile-variant of each hull-class also?

I hadn't appreciated the difficulty in getting the AI to use modules. I had taken the module-approach because I didn't want to design lots and lots of new ships. I think I may just have to wait for a few patches then

 

[Foraven]

The choice, in 'real-world' navies is determined by doctrine and finance. They choose a hull size that fits the role in their navy. In the Soviet Navy, this was a battle-cruiser (Kirov-class battlecruiser at 28,000 tonnes), in the French and Italian navies this is a large destroyer (Horizon-class, at ~ 8,000 tonnes), in the US-navy a small cruiser (Zumwalt-class at 15,000 tonnes), and so on. (For reference, the HMS Vanguard (WW2 British Battleship) and US Iowa-class battleship weighed in at 45,000 tonnes, whereas the German pocket battleships of the same period were ~30,000 tonnes).

Doctrines and finances does influence what their navy will be like, but so is technology. One reason battleships have been phased out after being the end-all of military tech for almost a century is the development of guided missiles technology; you don't need big armored ships anymore.

I don't necessarily agree with the idea that the "the game will teach you", because we are discussing ways to change the rules. A single PD-battleship might be such an efficient PD-platform that it is worth the time and cost to build because it makes your fleet immune to missiles and fighters; it is a question of whether or not the resulting mod makes them efficient. If the AI never uses fighters or missiles, then they're a waste of money. Or if the AI builds super-efficient PD-battleships all the time, then their utility will necessarily change.

It is a sci-fi game and we do have some leeway on the rules the game will follow, but i try to remain consistent in the way i mod things. If a ship is big and slow it better have some incredible weaponry and range to make for it (in FRB they do, unlike Vanilla). Giving such ship short range weapons is both a waste and a liability, the battle may be over before it gets there to do anything. Yeah, i could make a special PD weapon for mattleship that cover the whole fleet but then it would be incredibly OP and wreak the balance of the game. Battleship should be powerful to make up for their cost and build time, but if they are too strong it becomes a race of who can stack the most BBs and not that great gameplay. Realism wise, one thing they learned before battleships became obsolete is that maximizing their firepower with the all big guns approach (Dreadnaughts) made battleship way more effective than mixing all kind of weapons so they ship can defend itself against everything.

If you've gone for a PD-specific variant of one hull-class, why not create a PD-variant of each? For example, a PD-corvette, a PD-destroyer, a PD-cruiser, and a PD-battleship? For that matter, why not create a missile-variant, a beam-variant, and a projectile-variant of each hull-class also?

Because i don't believe i should. Besides, my escort was meant to have bonuses to PDs but that did not work (could not make a bonus to pd weapons like we can for lasers or missiles). Instead they are just ships that are a bit faster, thougher and accurate than regular ships at the expense of firepower and range. Good enough to defend the fleet against missiles, strike crafts and swarm of corvettes but little help against anything bigger.

I hadn't appreciated the difficulty in getting the AI to use modules. I had taken the module-approach because I didn't want to design lots and lots of new ships. I think I may just have to wait for a few patches then

The AI use weights when making it's designs, you don't need to design ships for them unless you want them to use pre-made designs (like i did for my pirates since they have no AI to design ships for them). Properly made that system can make rather good designs but that doesn't tell the AI how to use those ships. Also the AI has only one design per ship classes, makes them really vulnerable to counter designs...

 

[Foraven]

There is another great reason to use modified hulls rather than modules; retrofit. Retrofit allow you to swap around modules as you wish, but if you want to disallow a module to be removed after the ship is built, there is no means to do that currently. For example, if i want to make a light corvette that is cheaper but weaker than a normal corvette, i don't want players to just retrofit that ship afterward to be a normal or heavier corvette. Why? Because they get corvettes faster with no real drawbacks for doing so, defeating the whole concept.