Long, rambling post about ship hulls (v0.4.5 b2015-09-01)

Describe your experience with the latest version of FreeOrion to help us improve it.

Moderators: Oberlus, Oberlus

Forum rules
Always mention the exact version of FreeOrion you are testing.

When reporting an issue regarding the AI, if possible provide the relevant AI log file and a save game file that demonstrates the issue.
Message
Author
Atarlost
Space Floater
Posts: 24
Joined: Sat Oct 24, 2015 1:58 am

Long, rambling post about ship hulls (v0.4.5 b2015-09-01)

#1 Post by Atarlost » Thu Oct 29, 2015 9:22 am

The hulls in the game don't appear balanced. I get the impression that they're largely placeholders.

Historically I think the issue is internal slots. The first Free Orion version I played shields were ablative rather than reducing the damage of every shot individually, but stacked. Internal and external slots could both be used for stackable durability. After this changed the ship hulls were not modified. They still haven't been modified except for the addition of core slots or replacing what I think were single internal slots with core slots. If the hulls were balanced before the change must have wrecked that balance and if they placeholders then they're placeholders till. While there are some internal devices for scouting and raiding the only internal device that contributes to combat is still the shield and while drives can be thrown on anything there are diminishing returns for combat designs: a fleet moves at the speed of the slowest ship, only moves one jump a turn into guarded enemy territory, and the highest reasonable fleet speed (150 or 160 excluding the Interstellar Logistics and Interstellar Lighthouse boosts for most capital ships other than the sentient and solar) isn't really enough to not keep forces on the border for whom speed is irrelevant.

This is most apparent on the organic hull tree. The organic hull has 3 external 1 internal, same as the robotic hull or the basic large hull. The symbiotic hull loses an external slot, cutting its offense in half or its structure by around a quarter (but its initial structure by around two fifths for zortrium). It can mount a stealth device without giving up shields, but the first stealth tech costs about 2.7 times as much to research as the hull itself. The protoplasmic hull adds another internal slot, which it has little use for. It regenerates fuel between combats and doesn't really hit hard enough to make deep raiding terribly attractive with just two external slots. It's fast enough that a drive upgrade is unnecessary. On the other side of the tree the static multicellular hull gains an internal slot at the cost of the original organic hull's regeneration. It isn't suited for scouting or raiding and so far as I can tell is better than the earlier hull only as a troop transport. The endomorphic hull is the first to actually have more external slots, but as a dead hull it has little advantage over inorganic hulls. Total cost to research the endomorphic hull is 460. Total cost to research the asteroid hull with the same slots is 230. But that doesn't tell the whole story: many of the prerequisites will be researched anyways, but 126 points of the cost of the endomorphic are just hulls while only the last 50 points of the asteroid hull are hull tech. The endomorphic is a ot faster, but the rock is a little cheaper and has more structure. Being a weaker combatant than something half as far into the tech tree isn't good even if it is faster. The endosymbiotic hull is potentially a good raider since it is a living hull with four external slots, but it appears to be the only bright spot. Farther up the tree the Bio-Adaptive is an inferior combat hull to the endosymbiotic while requiring just 4 RP less than the endosymbiotic hull and all non-shared prerequisites that don't lead to lifecycle manipulation. Enough people swear by lifecycle manipulation that I think it's fair to not count it against the endosymbiote in this. I am counting trans-organic sentience against the bio-adaptive since it's a theoretical prerequisite for techs that aren't automatic early game choices. It regenerates completely, but it's less likely to live to regenerate. It would be worth it for raiding, but stealth gives no combat advantage to make raiding viable. The ravenous hull is another corpse. It has a reasonable slot arrangement for combat, but not as good as self-gravitating hull or heavy asteroid hull. The self-gravitating hull is 745 RP in and the heavy asteroid just 590. The ravenous hull is 1036 RP in. The ravenous is faster and cheaper, but the self-gravitating has an extra external slot and a core slot and a lot more structure and the heavy asteroid is far earlier and still has an extra external slot and an extra internal it can use to partially make up the speed deficit.

The living organic hulls tend to be lousy at combat and the dead organic hulls tend to be both later than comparable inorganic hulls and less combat capable. Most of them also have more internal slots than they actually need. I find it hard to really justify filling the second internal slot on a combat design. The organics already have an excess internal slot worth of speed over the robotic hull line.

The robotic hull line seems much stronger. The robotic hull is 48 RP in rather than 42 for the organic and is more expensive, but it's also the prerequisite for the fleet repair techs that everyone should want given that most of the organic combat hulls are zombies. On the other hand the robotic hull requires a building you can produce before researching the tech and that your capital starts with. There's a gap in the "big ship" upgrade path where the organic has the endomorphic, but it comes out ahead with the self-gravitating hull, gets its own rapid regenerator in the nano-robotic hull, and for actual combat the nano-robotic hull is as good as the sentient hull where the organic line stops while the robotic hull line still has the titanic hull and the logistics facilitator coming. I feel the titanic hull is probably somewhat overpowered. The trans-spatial hull is completely useless, but the drive is good. Actually, let me complain about the hull a little. It's not really a strike against the tech since the drive is worth the price, but all you can do is slap the drive and a scanner on it. It's too expensive, slow building, and flimsy for combat and with no internal slots it's useless as a scout as soon as 40 stealth stops being good. Or you could put an absorption field on a camoflaged asteroid hull or anything organic more than a thousand RP sooner. I get that scouting isn't supposed to be a strength of the robotic hull line and that's fine, but this is still not a hull with much purpose.

The asteroid hull line starts a bit late, but I expect most people would start with the robotic hull to get the fleet repair techs and it still gets a good combat hull sooner than the organic line. The asteroid hull is the first four external combat hull. The mini-asteroid hull is a workable fighter. I don't like it as much as the spatial flux hull because shields aren't as cost efficient as armor, but it works okay. The camoflage asteroid hull may be useless. It can't mount sensors, which seriously impacts its utility as a stealthy scout. The small camoflage asteroid hull does have an external slot, which I suspect is an oversight. It is more useful for it, though. The mini-asteroid swarm is hands down the best light gunship hull in the game with two external slots, more structure than the non-asteroid hulls of similar size, and free shields equivalent to the second tech. I suspect this was overlooked when shields were nerfed. The heavy asteroid hull at six external three internal is an okay heavy combat hull and the research cost is tiny. The only expensive tech in the whole tree is monomolecular latticing, which is still a mere 10 points more expensive than xentronium which it lets you skip. On the end of the tree is the scattered asteroid hull, which has +3 shields, provides that bonus to everything else in the system, and even without that is only a little behind the excellent titanic hull in combat power while being 242 RP earlier and including 18 and 30 point armors as prerequisites that the titanic hull user has to research elsewhere for 1800 RP. I feel this hull is also somewhat overpowered.

The last ship line starts at 654 RP, but the first hull is all but useless. First, you need a white or blue star. Second, there's only one external slot, which means no armor or shield to give any semblance of durability if you want a weapon. The stealth is inadequate with no way to mount a stealth component. At 1494 RP in you get one of two actual useful hulls that can only be built at a blue star. Fractal can't mount shields but can mount a lot of armor and guns or makes the best troop ship with quite acceptable speed and tons of slots for troops and enough base structure to not die to mines before deploying them if it doesn't spend one on armor. Zero internal slots is a noticeable drawback, especially with force field harmonics as a prerequisite. The other choice is the quantum energy hull. At seven external slots it's a very good combat hull even if not in the same class as the titanic and having an internal slot for shields is a plus over the fractal. The other two internal slots aren't so useful. 160 isn't that much better than 120 and a combat ships tend to never move out of supply. The solar is ridiculously expensive to research and is the strongest hull in the game. And has five useless internal slots because with more than 2 drives it will outrun any other ship except the sentient hull and it's kind of expensive to field a fleet of nothing but solar hulls. This line is really slow to tech and can't be where you start. It doesn't make sense to tech this after robotics because apart from the fractal hull making an outstanding troop transport it's inferior to the titanic hull until the very late solar hull. It doesn't really make sense to tech it after asteroids because the same can be said of the scattered asteroid hull. You could tech it after organic hulls, but it's a bit late to help. Once you're teching to ZPE anyways for eg. neutronium it's less expensive but it still seems like an afterthought.

The asteroid hull tree seems obviously superior to other hull groups. The organic hull line has the best stealth raiders, though while the small asteroid hull retains an external slot it's probably better at scouting. The problem with stealth, though, is that there's no selective aggression. Aggressive stealth forces will attack combat fleets and die horribly to similarly teched opponents. Non-aggressive stealth forces don't do anything except extend vision. When the self-gravitating hull appears after 768 RP raiding starts to become questionable and that's 108 points RP less than the bio-adaptive hull and if the game isn't over by the time titanic hulls appear all organic ships are obsolete. The other problem with raiding is that stealth and detection are in the basement: the big stack of basically linear tech chains at the bottom of the tree display. There are no barriers to teching detection to catch up to stealth and anyone who doesn't keep detection up is asking for trouble when planetary stealth passes detection and they can't even counterattack a planet lost on the last turn because they can't find a colony they put their themselves on a planet describing a predictable newtonian orbit around the nearest star. An empire pursuing a heavy combat strategy can ignore stealth techs entirely, but the detection range increases are important to everyone.

The AI seems to tend to choose the asteroid, robotic, self-gravitating, and titanic hulls for combat, though it appears to research everything. It may be underestimating the self-repair of the nano-robotic hull, but for the most part whatever algorithm it's now using to choose hulls seems to mostly agree that the organic line is only useful for noncombat roles and that the titanic hull is the gold standard for combat hulls.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#2 Post by MatGB » Thu Oct 29, 2015 2:18 pm

I'm at the end of paragraph two, so I'm going to start replying now before I go into paragraph three (for future "long rambles" can I introduce you to my best friend, the bullet point button? :wink: )

Rebalancing the hull lines is an ongoing task that is, irrevocably, linked to internal slot parts. For starters, we have plans for damage control parties of various abilities plus science teams and manufactories, ships can already in the code act as resource centres but there isn't support in the scripts for it yet. Consequently determining internal slots balance wise is impossible (and for their production cost they're largely ignored, I based the current costings on external slots, speed and structure). At the same time I also, massively, reduced the costs of the engine parts and improved them in order that internals had at least some use.

Re core slots, the first hull with one, the Self Gravitating, lost two external slots to gain them, all others had them added, with the exception of, IIRC, the Solar Hull, which simply didn't have enough space for another slot, so it did lose an internal. We've added more core slot parts since then, and more are planned.
The organic hull has 3 external 1 internal, same as the robotic hull
Robotic has 4 external slots, the organic hull is a faster, cheaper, 'cannon fodder'/zerg rush ship that mostly excels at non combat stuff, I recommend the Static Monocellular for combat.
The symbiotic hull loses an external slot, cutting its offense in half or its structure by around a quarter (but its initial structure by around two fifths for zortrium). It can mount a stealth device without giving up shields, but the first stealth tech costs about 2.7 times as much to research as the hull itself.
The Symbitotic Hull, on its own, is invisible until an opponent has Active Radar and with a stealth part they need Neutron Scanners, it's not meant to be, nor is it best used as, a combat vessel, it is exceptionally good at other tasks, I find it the best cost/benefit scouting hull in the game even without stealth parts, the vision boost is an excellent advantage.
The protoplasmic hull adds another internal slot, which it has little use for
Ha!

OK, yeah, a wonder of the game is the variety of ways to play, I'm always torn with my protos as to which part to give them, speed, stealth and fuel are always useful for scouts.

Re the research cost of stealth, and the production cost, I'm going to look at that further in my next balancing pass, we're talking of overhauling stealth and I really want to make it more strategically possible, I have varying ideas. But generally a statement like "X costs too much to research ergo Y is useless" is less helpful, because we can always change the research cost of X, it will in fact be changed sooner rather than later as I'm going to split up planetary and part based stealth very soon thus reducing research costs. I haven't done it yet as I've been busy with other things and life got in the way.
I am counting trans-organic sentience against the bio-adaptive since it's a theoretical prerequisite for techs that aren't automatic early game choices
The BiAd isn't an early game hull, it shares prerequisits with Pure Energy Metabolism, which is currently in my opinion the second most powerful (overpowered) tech and I always rush to get it even when not playing Organic line, when I am it simply makes it easier.
It regenerates completely, but it's less likely to live to regenerate. It would be worth it for raiding, but stealth gives no combat advantage to make raiding viable.
Completely untrue, a stealthed ship is invulnerable in round one of any combat, meaning that a) it gets shot at twice while shooting three times and b) anything it destroys in the first round can't shoot back.

I, regularly, use Absorption Fielded BiAds as raiders/first attack ships, they are lethal in numbers and their low cost means you can field them in number easily (don't bother with shields, they don't need them if you're doing it right). The way stealth parts interact and become obsolete within the research tree is something that needs further work on but I'm fairly sure that BiAds will be slightly overpowered once that work is done, as they are anyway.
The ravenous hull is another corpse. It has a reasonable slot arrangement for combat, but not as good as self-gravitating hull or heavy asteroid hull. The self-gravitating hull is 745 RP in and the heavy asteroid just 590. The ravenous hull is 1036 RP in. The ravenous is faster and cheaper, but the self-gravitating has an extra external slot and a core slot and a lot more structure and the heavy asteroid is far earlier and still has an extra external slot and an extra internal it can use to partially make up the speed deficit.
For what it's worth, I am considering completely dumping the 'zombie' hull approach and have all organic line ships have regenerative abilities, but it's not something I've decided on yet and it would mean finding another advantage for the EndoSym other than stealth (which it is good at in the early/mid game versus the AI).

However, I do think you're trying to compare apples and oranges. It can be argued that I reduced the cost to build of the Self Gravitating too much, and also that the Organic line lacks a 'big beast' hull to build battlecruiser/battleship size vessels, both are probably true. Each hull line has strengths and weaknesses, I personally find the slow speed of the asteroid line debilitating, I can win a conquest victory far far quicker with organics, their biggest drawback is fleet upkeep, but we've, again, talked about that.

We also need to do more about fuel to make it useful, currently supply lines are too easy to extend and I'm not sure I've done enough to nerf the fuel capacities of the asteroids. Then there are building production/location costs, etc. A Self gravitating hull requires an incredibly expensive building to be present, whereas the organic lines require cheap but longer to build buildings, I'm not actually sure that's balanced, I'm currently testing allowing the two expansion organic shipyards to be built alongside the incubators, that seems to work better.
The robotic hull line seems much stronger. The robotic hull is 48 RP in rather than 42 for the organic and is more expensive, but it's also the prerequisite for the fleet repair techs that everyone should want given that most of the organic combat hulls are zombies.
We've already agreed to change that prerequisite, the other damage control techs require construction prerequisites, keying Basic off the Robotic Hull was a stopgap a few years ago that will be changing.

The Robotic Hull is meant to be the 'best' hull for a given definition of best, it's a simple, straightforward blunt instrument combat machine. But the line lacks decent/cheap troop carriers and is reliant, on its own, on the basic standard hull for outposts and colonies, etc unless you want to pay through the nose for robotic hulled outposts, not good in the early game.
I feel the titanic hull is probably somewhat overpowered.
Agreed, I'm waiting to do another cost pass on it until we've more internal parts but I think I did make them a bit too good with the last pass.
The trans-spatial hull is completely useless, but the drive is good. Actually, let me complain about the hull a little. It's not really a strike against the tech since the drive is worth the price, but all you can do is slap the drive and a scanner on it. It's too expensive, slow building, and flimsy for combat and with no internal slots it's useless as a scout as soon as 40 stealth stops being good.
The ship plus drive is invisible until an opponent has Sensors, which for the AI is a very late game pick, not normally showing up until turn 250+. It can get past virtually every monster in the game so can scoot around places most other ships can't get and at speed 120 it can complete virtually every starlane jump in the game in one turn meaning you never lose visibility.

If the AI ever starts prioritising detection in a way that it currently does not, you'd have a point, but given that Sentinels and Wardens can't, currently, see it, it's one of the best scouting hulls, its drawback is the NanoRobotic prerequisite being too expensive to research, but, um, that's only a current drawback.

Also? Glass cannons. I have messed around with giving it other core slot parts. It's very very silly. But fun. As more core slot parts are created, the hull can have more uses, when I split the hull and the part up I envisaged it to be a 'testbed' vehicle, not every ship needs a combat role, but it doesn't need to have a use in and of itself in every game.
if the game isn't over by the time titanic hulls appear all organic ships are obsolete.
Nonsense. Complete nonsense. Titanic Hulls and Omnni Scanner, yes, absolutely, but I'll put my turn 150 fleet of Sentients and Biads up against your Titans anyday ;-)
The other problem with raiding is that stealth and detection are in the basement: the big stack of basically linear tech chains at the bottom of the tree display. There are no barriers to teching detection to catch up to stealth and anyone who doesn't keep detection up is asking for trouble when planetary stealth passes detection and they can't even counterattack a planet lost on the last turn because they can't find a colony they put their themselves on a planet describing a predictable newtonian orbit around the nearest star. An empire pursuing a heavy combat strategy can ignore stealth techs entirely, but the detection range increases are important to everyone.
I agree with this, but most games (and all my test games) are against the AI, which doesn't priorise stealth or detection. I plan to split them up, make them more interesting and make stealth-as-a-strategy viable in a variety of ways. Having Detection Strength go up to a possible 200 but only really using the first 70ish for most things is a bit of a downer, this will change.

Basically, the problem in the currently isn't the relative balance of the shiplines, asteroids are clearly 'better' for defence, robotic easiest for pure aggression. energy really rather cool but in need of some work (clearly still the most powerful line and you can field Fractals on turn 100 if you want, the Energy Frigate will help them (I need to finish that ASAP) and Organics subtle and in need of more strategic support.

The problem is the lack of internal slot variety and the blunt edge way stealth works. Both are being worked on.

You're right on some stuff, and I am still thinking Heavy Asteroid is overpowered for what it is (but I've nerfed it so many times now I'm running low on ideas, it may need to lose an external slot).

The tech tree, and research costs, needs a balance pass, I've barely touched that and need to, there are definitely some anomalies, but the entire tech tree is unbalanced currently.

Also, apologies if my reply started out a bit too defensive, I've put a lot of work into getting things as balanced as they are but I do agree with and appreciate many of your points, but do think you're undervaluing stealth and the organics as they are, let alone what they can/will be.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Atarlost
Space Floater
Posts: 24
Joined: Sat Oct 24, 2015 1:58 am

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#3 Post by Atarlost » Mon Nov 02, 2015 4:30 am

MatGB wrote:Also, apologies if my reply started out a bit too defensive, I've put a lot of work into getting things as balanced as they are but I do agree with and appreciate many of your points, but do think you're undervaluing stealth and the organics as they are, let alone what they can/will be.
I can't value what things can/will be. I'm not even looking at the dev branch, much less stuff that hasn't been done yet. That's why the post says v0.4.5. What I can do is try to analyze the version I played.

Most of the knocking on the organic hull line comes from trying to use them. With stealth parts. I got a lot of scouting, but couldn't use the intelligence because my ships were inferior and divided. I suppose I could have concentrated them if I'd abused metaknowledge I don't have about when the AI techs various sensor techs, but even if they couldn't see my planets they could have isolated all my systems if I didn't keep a border force they didn't want to attack.
MatGB wrote:Rebalancing the hull lines is an ongoing task that is, irrevocably, linked to internal slot parts. For starters, we have plans for damage control parties of various abilities plus science teams and manufactories, ships can already in the code act as resource centres but there isn't support in the scripts for it yet. Consequently determining internal slots balance wise is impossible (and for their production cost they're largely ignored, I based the current costings on external slots, speed and structure). At the same time I also, massively, reduced the costs of the engine parts and improved them in order that internals had at least some use.
So, yes, they're unbalanced placeholders. The engine parts don't really work as an always applicable internal slot use because only fleet speed usually matters. Slow hulls can use all their internal slots, but fast hulls don't get anything while your fleet composition includes slow hulls. That's my problem with the Protoplasmic. The symbiotic hull can mount a shield and a stealth device and two external slots. The protoplasmic hull can mount a shield and a stealth device and two external slots. Both regenerate fuel, making fuel pods not critical even for deep scouting or raiding. If you're raiding you probably want to lump your new hulls with your old so drives are useless and so is additional fuel since the symbiotic hulls would hold the fleet back either way. I think they even both have the same stealth rating. The same issue crops up with the solar hull and sentient hull.
MatGB wrote:Re core slots, the first hull with one, the Self Gravitating, lost two external slots to gain them, all others had them added, with the exception of, IIRC, the Solar Hull, which simply didn't have enough space for another slot, so it did lose an internal. We've added more core slot parts since then, and more are planned.
The organic hull has 3 external 1 internal, same as the robotic hull
Robotic has 4 external slots, the organic hull is a faster, cheaper, 'cannon fodder'/zerg rush ship that mostly excels at non combat stuff, I recommend the Static Monocellular for combat.
Organic hulls can't be zerg rush ships because they start with tiny base HP and grow to max very slowly. A few are good for noncombat roles, but not all that good. The base and asteroid hulls are both good for noncombat purposes other than scouting. The static monocellular isn't a good combat hull either. It doesn't suffer from organic growth even though all the other zombie hulls do, but it still doesn't compare to the inorganic options.
MatGB wrote:The Symbitotic Hull, on its own, is invisible until an opponent has Active Radar and with a stealth part they need Neutron Scanners, it's not meant to be, nor is it best used as, a combat vessel, it is exceptionally good at other tasks, I find it the best cost/benefit scouting hull in the game even without stealth parts, the vision boost is an excellent advantage.
If the AI doesn't research Active Radar early I'd consider that an AI bug. Without active radar you can't see all native populations or space monsters and that's bad. There are better techs, but I'd want it before any of the hull techs. The symbiotic hull isn't useless, but "other tasks" is just scouting. You could put a colony pod on it, but there's no real advantage to using it rather than its prerequisite. And as a scout there's no reason to tech any further. It has 35 stealth and 100 speed and can mount a shield, a stealth device, an armor plate, and a scanner and it regenerates fuel.
MatGB wrote:
The protoplasmic hull adds another internal slot, which it has little use for
Ha!

OK, yeah, a wonder of the game is the variety of ways to play, I'm always torn with my protos as to which part to give them, speed, stealth and fuel are always useful for scouts.
Stealth is always useful for scouts, but are you sure speed and fuel are real benefits? There doesn't seem to be any advantage to being the first to see a system you can't get a colony to so it seems to just be a matter of collecting intelligence. Sitting a few scouts around stationary does that.
MatGB wrote:Re the research cost of stealth, and the production cost, I'm going to look at that further in my next balancing pass, we're talking of overhauling stealth and I really want to make it more strategically possible, I have varying ideas.
I'd suggest making it not all or nothing. Make a detection roll for each fleet or planet based on stuff like how big the fleet is or how developed the planet is and how close it is to the limit of the observer's detection envelope as well as the lowest stealth rating in it. That way the final detection tech doesn't have to completely render all stealth obsolete.
MatGB wrote:The BiAd isn't an early game hull, it shares prerequisits with Pure Energy Metabolism, which is currently in my opinion the second most powerful (overpowered) tech and I always rush to get it even when not playing Organic line, when I am it simply makes it easier.
No, it's not particularly early, which is why I'm inclined to not count early game must have techs against the total research cost of its competitor. The non-shared prerequisites in the PEM line aren't early. Cyborgs enables hostile environment colonization, but hostile environment colonization isn't necessary for multi-species empires, making that whole tech line skippable.
MatGB wrote:Completely untrue, a stealthed ship is invulnerable in round one of any combat, meaning that a) it gets shot at twice while shooting three times and b) anything it destroys in the first round can't shoot back.
Well then, we've found a documentation problem.
MatGB wrote:I, regularly, use Absorption Fielded BiAds as raiders/first attack ships, they are lethal in numbers and their low cost means you can field them in number easily (don't bother with shields, they don't need them if you're doing it right). The way stealth parts interact and become obsolete within the research tree is something that needs further work on but I'm fairly sure that BiAds will be slightly overpowered once that work is done, as they are anyway.
I wasn't using nothing but gone designs, but the AI completely trashed my BiAds once it started building titanic hulls.
MatGB wrote:However, I do think you're trying to compare apples and oranges. It can be argued that I reduced the cost to build of the Self Gravitating too much, and also that the Organic line lacks a 'big beast' hull to build battlecruiser/battleship size vessels, both are probably true. Each hull line has strengths and weaknesses, I personally find the slow speed of the asteroid line debilitating, I can win a conquest victory far far quicker with organics, their biggest drawback is fleet upkeep, but we've, again, talked about that.
What role does the ravenous hull fill that the others don't? It's not good at stealth, it doesn't regenerate, it's too flimsy to be a useful troop ship since organic growth makes it die to residual system defenses. It's a straight forward combat hull and not good at any other role. The asteroid is the only ship of the three with a noncombat purpose since it's cheap and does have enough base structure to work as a troop ship against mine and network regeneration defended systems.
MatGB wrote:We also need to do more about fuel to make it useful, currently supply lines are too easy to extend and I'm not sure I've done enough to nerf the fuel capacities of the asteroids. Then there are building production/location costs, etc. A Self gravitating hull requires an incredibly expensive building to be present, whereas the organic lines require cheap but longer to build buildings, I'm not actually sure that's balanced, I'm currently testing allowing the two expansion organic shipyards to be built alongside the incubators, that seems to work better.
The buildings don't and probably can't balance the ships built at them. They're one time costs and if they were expensive enough to effect a well industrialized empire they'd be so expensive they'd completely bar the option for a weaker empire. The existence of different map size, empire count, and special frequency settings pretty much removes any hope of finding a cost point where the building makes a difference between hull paths.
MatGB wrote:The Robotic Hull is meant to be the 'best' hull for a given definition of best, it's a simple, straightforward blunt instrument combat machine. But the line lacks decent/cheap troop carriers and is reliant, on its own, on the basic standard hull for outposts and colonies, etc unless you want to pay through the nose for robotic hulled outposts, not good in the early game.
Relying on the basic hulls for noncombat roles isn't a large disadvantage. The lack of good scouting hulls might be, but I didn't find that scouting gave me any actionable intelligence. A slight dip into asteroids gets an adequate colony/outpost hull in the small asteroid and an adequate troop hull in the standard asteroid. Going deeper into asteroids is cheaper than researching the normal armor techs once you already have orbital industry.

MatGB wrote:The ship plus drive is invisible until an opponent has Sensors, which for the AI is a very late game pick, not normally showing up until turn 250+. It can get past virtually every monster in the game so can scoot around places most other ships can't get and at speed 120 it can complete virtually every starlane jump in the game in one turn meaning you never lose visibility.

If the AI ever starts prioritising detection in a way that it currently does not, you'd have a point, but given that Sentinels and Wardens can't, currently, see it, it's one of the best scouting hulls, its drawback is the NanoRobotic prerequisite being too expensive to research, but, um, that's only a current drawback.
There are two problems with this. First, intelligence about stuff blocked by monsters you can't beat kill isn't actionable. Second, if it weren't so expensive to research the robotic line would have a good scout and that would defeat your prior point about the robotic line only being good at straight forward combat.
MatGB wrote:Also? Glass cannons. I have messed around with giving it other core slot parts. It's very very silly. But fun. As more core slot parts are created, the hull can have more uses, when I split the hull and the part up I envisaged it to be a 'testbed' vehicle, not every ship needs a combat role, but it doesn't need to have a use in and of itself in every game.
In 0.4.5 the only core slot parts are the ZPE "fuel" part and two drive parts. I don't believe a testbed vehicle is a viable role unless some sort of prototyping requirement is instituted before parts can be used and I don't think doing that is a good idea.

I don't believe a testbed vehicle is a viable role. Nonsense. Complete nonsense. Titanic Hulls and Omnni Scanner, yes, absolutely, but I'll put my turn 150 fleet of Sentients and Biads up against your Titans anyday ;-)[/quote]

I pit my Biads against the AI's titans and lost horribly.
MatGB wrote:Basically, the problem in the currently isn't the relative balance of the shiplines, asteroids are clearly 'better' for defence, robotic easiest for pure aggression. energy really rather cool but in need of some work (clearly still the most powerful line and you can field Fractals on turn 100 if you want, the Energy Frigate will help them (I need to finish that ASAP) and Organics subtle and in need of more strategic support.
The problem is the lack of internal slot variety and the blunt edge way stealth works. Both are being worked on.

You're right on some stuff, and I am still thinking Heavy Asteroid is overpowered for what it is (but I've nerfed it so many times now I'm running low on ideas, it may need to lose an external slot).

The tech tree, and research costs, needs a balance pass, I've barely touched that and need to, there are definitely some anomalies, but the entire tech tree is unbalanced currently.[/quote]

That sounds good, but I feel like you're not coming from the same place as me. No doubt you're a better player since you've played more and know the game scripts, but I can't believe I'm undervaluing speed by that much. I'd bet you don't primarily play at the default aggression, map size, and empire count. That's going to change things. I've seen it said that aggression level isn't supposed to be a difficulty setting. To me that implies that the non-rushing game also needs to be balanced and I worry that the non-rushing game isn't getting as much high level testing.

User avatar
Vezzra
Release Manager, Design
Posts: 5265
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#4 Post by Vezzra » Mon Nov 02, 2015 10:13 am

Atarlost wrote:
MatGB wrote:Re the research cost of stealth, and the production cost, I'm going to look at that further in my next balancing pass, we're talking of overhauling stealth and I really want to make it more strategically possible, I have varying ideas.
I'd suggest making it not all or nothing. Make a detection roll for each fleet or planet based on stuff like how big the fleet is or how developed the planet is and how close it is to the limit of the observer's detection envelope as well as the lowest stealth rating in it. That way the final detection tech doesn't have to completely render all stealth obsolete.
We deliberately avoid try to design our game mechanics without any kind of "rolls", so that's not an option, at least not this way. But rolls aren't necessary anyway, you could still determine visibility by factoring in distance, detection range and stealth level. Originally stealth and detection actually worked that way, but there are quite serious issues with that approach. There have been extensive discussions on that topic, we've yet to come up with a good detection&stealth mechanic that addresses all issues and concerns in a satisfactory way.

You can read up on the discussion concering this here.
Well then, we've found a documentation problem.
Definitely. But we have a lot of these, with the game being WIP and in alpha stage and all that. ;)

That doesn't mean that we don't try to address it, but it's very difficult to keep documentation current, even only to a certain degree. We are a very small team, and most of us have only very limited time we can spend on making this game. Of course, that shouldn't keep you from pointing out such things. We need those reports/feedbacks, even if we won't be able to address all of them.

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#5 Post by Kassiopeija » Mon Nov 02, 2015 12:06 pm

Just a few personal opinions from me on some stuff I read:

- for my personal playstyle Asteroids are too weak, so I usually ignore that line completely. Too slow. I play usually without Stargates during the first 200 turns so I need fast ships.
- the Asteroid-Swarms shield need a nerf I guess they were forgotten when shields were generally nerfed.
- the Scattered Asteroid is the only one that always has a positiv contributing function because of the shield bonus.
- I will always use Symbiotic Hulls for Colony/Outpost ships and Troopers because it has more hitpoints + detection range, better stealth & is faster than the base Organic hull - and this sometimes saves them from trouble, eg when a new Dyson Forest spawned etc pp. It's also cheaper to build. And the extra fuel is sometimes needed. Even with 3 fuel there is sometimes trouble when I have to evade a Vaccuum Dragon outside my supply etc.
- The Protoplasmic is even better but you'll need another shipyard so sometimes I don't bother. Usually only when taken AI worlds already have them. The AI builds organic shipyards fairly well so I can easily produce troopships directly at the front, quickly use them therefore decreasing the shipcount-penalty.
- Later the BiA make the very best Trooper because they regen away all damage from planets or mines and they can also fit in the role of damage soaking support ships for my main fleets, if needed.
- I don't encounter trouble loosing any of these ships to planetary defenses or mines formostly because I completely outtech/produce the AI. For example I've never seen him build Titanic Hulls, games are over long before. I currently play a 800 system map vs 20 AI, around turn 175 I have 7000 prod 1000 res, according to the graphs, this is around 50% output than all other 17 remaining AI combined. Ultimately that means I can ignore a large part of the tree because it'll never come to use, eg. garrisons, planetary defenses + shields + mines, Asteroids, Energy Hulls, stealth, hitpoint mods... so the other techs will come much faster onto my table.
- I try to separate different designs in order to get good speed results. Typically fast troopers following the attack fleets. Early attack fleets are Robocruiser. Later ones are Self-Gravitating, these are faster because I will always use engines on them (they got enough fuel). This is even good because sometimes, when planets are set to defense-focus, it is wanted that the Self-Gravitating arrive more early because they can take the hit where Robocruiser would be oneshoted.
- Fractals don't make good troopers IMO. Too expensive, cannot be build everywhere, needs extra shipyard that the AI doesn't deliver, and with the advanced troop pod you'll loose just too much production in overkill. You also need a boatload of extra research for something that otjer lines do better, and are more early accessable. That said I see no reason for Solars too because the game is over long before.
- Spatial Flux is an excellent early non-stealth far-away scout because of the 6 fuel. It might be destroyed here and there but it's so cheap to build that doesn't hurt. In a strong research game I always get them early.
- The reason why Asteroids are so weak is the AIs inability to move ships away from danger if you can reach them in a single turn. He doesn't react to that. With Asteroid fleets you oftentimes have to chase him back & forth, or divide your fleets in order to push him back, faster hulls are so much better if you just hit him with all you've got. Max starlane length is 120 mostly, so that speed can reasonably reached with Self-Gravi hulls.
- The fuel regen of organics (+0.1 per turn more) isn't sufficient to ignore extra fuel tanks on them. Do you really want to wait 5 turns every time you want to make a jump when deep scouting? Spatials can go invest, then return, then go the other lane... or go really deep and be stationairy acting as a sensor buoy.
- The basic hulls are all trash. There's no reason to not invest a very tiny amount of research to get, at least, Symbiotics, for Colony/Outposters/Scouts

User avatar
Vezzra
Release Manager, Design
Posts: 5265
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#6 Post by Vezzra » Mon Nov 02, 2015 12:33 pm

Kassiopeija wrote:the Asteroid-Swarms shield need a nerf I guess they were forgotten when shields were generally nerfed.
Actually, they have been nerfed (a few months ago IIRC). The Scattered Asteroid Hull used to give a shield bonus of 5, which has been reduced to 3. For a high end flagship hull that seems about right IMO.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#7 Post by MatGB » Mon Nov 02, 2015 3:47 pm

Two points.

First, I didn't say the robotic line was meant to be pure combat (although that's mostly what it is), I said the robotic hull is, it's a simple, straightforward, expensive 'cruiser' hull that isn't particularly good at any non-combat task, it's too expensive to be a mainstay troop carrier, etc, whereas other hulls in the line can be (I've used gravs, titans and nanos as troop ships, but rarely if ever the base robotic hull, cost/benefit isn't there).

Second, we have a differing strategic approach
I got a lot of scouting, but couldn't use the intelligence because my ships were inferior and divided. I suppose I could have concentrated them if I'd abused metaknowledge I don't have about when the AI techs various sensor techs, but even if they couldn't see my planets they could have isolated all my systems if I didn't keep a border force they didn't want to attack.
I work on the assumption that nearby AIs are always just about to research the next detection level and watch what they can see (we do need to improve detection rating awareness within the UI somehow both for your detection and opposing but I have zero clue how).

I always want to have eyes on all my neighbours, a if they're massing ships near me instead of a different neighbour I need to know.

I don't keep much of a border force, pickets at best, I mass my fleets to attack and concentrate entirely on this, with intelligence I get enough advanced warning of an actual threat that I can send a newly built fleet to deal with an aggressor on an undefended front.

I strongly agree with you that the AI doesn't research detection strength early enough, but the AI team disagrees and I don't push it too heavily, it will become massively more obvious when the revamp of the mechanics comes through though, but that's a big task with lots of different tweaks needed that, as you rightly say, is outside the remit of this discussion.

A mass fleet of organic ships can, and does, work well in a zerg rush approach (actually, not exactly, zerg rush in so incredibly early but this takes a bit longer), each individual ship is dirt cheap and fast.

For starters, no shields. Until recently I rarely used shields at all, now they're cheaper it's less of a comparative advantage but it can be effective. Scout hulls with stealth simply don't need an expensive shield, if they get shot at they're dead anyway, avoid fights and run away. Combat vessels don't always need them, BiAds certainly don't, a fleet of ten BiAds with stealth can take out AI titans as long as you don't let them concentrate, if you get stuck in a knock down fleet battle you're in trouble, use speed and stealth to avoid them.

Organic line is, hands down, my favourite line to play with, it's a challenge and they're very, very adaptable.

However, I'm very tempted to ramp up the power level of the Ravenous Hull almost immediately, they need a bigger ship than they've got, sentients are about right at 6 externals as they're more than just a combat vessel and the bonuses are cool, maybe jump Ravenous up from 5 externals to 7, give them more of a chance?

What say others? Also, would anyone object/think it unbalanced if we gave all Organic hulls the repair/refuel ability and dumped the 'zombie'/dead hull idea? They don't all need Growth, but it is confusing that they don't all have repair/refuel (maybe make refuel dependent on star type, they're sorta phototrophes?)
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
Dilvish
AI Lead and Programmer Emeritus
Posts: 4768
Joined: Sat Sep 22, 2012 6:25 pm

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#8 Post by Dilvish » Mon Nov 02, 2015 6:07 pm

MatGB wrote:I strongly agree with you that the AI doesn't research detection strength early enough, but the AI team disagrees and I don't push it too heavily,
Please feel free to experiment. The AI priority Active Radar ("SPY_DETECT_2") is currently fixed entirely by its position in the various lists in TechListsAI.py -- it appears in the tech_group_2_blah lists; there are 4 of those, in each of them you could move them to (or near) the top of that list, or perhaps even move it to the end of the group_1 lists.

If you find a spot that seems to work better than where it is now, let me know, I'll be happy to check it out (or just plain adopt it if I'm too busy at the time to really check it).
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#9 Post by Kassiopeija » Mon Nov 02, 2015 10:26 pm

Vezzra wrote:
Kassiopeija wrote:the Asteroid-Swarms shield need a nerf I guess they were forgotten when shields were generally nerfed.
Actually, they have been nerfed (a few months ago IIRC). The Scattered Asteroid Hull used to give a shield bonus of 5, which has been reduced to 3. For a high end flagship hull that seems about right IMO.
which ship are you referring to?

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#10 Post by MatGB » Mon Nov 02, 2015 11:29 pm

Scattered Asteroid got nerfed, giving a bonus of +3 not +5, which appears to be fine. I did look at the Mini Asteroid Swarm at the same time, but with only two external slots and low base structure I didn't consider the flat rate of 5 to be too overpowered so left it as it is.

I'd be very open to tweaking that if people do feel it's overpowered but I must've missed it if it's been explicitly called out as such in the past. Looking at it it's probably too cheap for what it is, given it's got a built in deflector shield equivalent, but I rarely use asteroids so haven't really tried to abuse it in numbers.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#11 Post by Kassiopeija » Tue Nov 03, 2015 3:17 am

I once used the Asteroid Line, just to check them out. Put nothing but 2*Crystallized Armour Plates on them, getting these stats:

- 71 HP (76 with Reinforced)
- 5 shields + 5 from Scattered = 10
- 26 production costs

that seemed dead cheap to me.

So for every Scattered Asteroid I produced 2 of these Swarms, which actually doubled the total HP from my fleets but it did cost only a mere +10% PP more than the Scattered Asteroids alone.
So I could completely neglect to put hitpoints on these Scattered Flagship and so could use up all external slots for weapons only.

One might argue that using this tactic could probably result in having problems with the fleet upkeep multiplier, but the advantages seem to override that. In effect, if you think about how the game randomly make a targetting decision, my Scattered Asteroids weren't taking much damage at all only 33% of before, the other 67% got soaked in by the swarms. They could fight endlessly because the amount ofd repair per turn from techs was already enough.
And if I ever lost a ship it was a Swarm that could be easily replaced In fact, I could separate damaged Swarms from the main attack fleet and send to a shipyard all the while the main attack fleet could still fighting, not loosing any of their firepower.
Swarms also can be produced in 3 turns, so if my main attack fleet needed additional HP before facing a strong enemy it could be fastly given to them.

All these things cannot be done by "normal" ships except the Instant-100%-Repair ships (who are, fair enough, more expensive to build and w/o shield), so I consider this borderline exploitative and never used it again.

Then shields got generally nerfed making the Mini-Swarm even stronger in relation....

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#12 Post by MatGB » Tue Nov 03, 2015 3:34 am

And the planned move for fleet upkeep to be per part instead of per hull would make that an even nastier combo.

Hmm, not something I had considered (although I do now vaguely recall you mentioning it in the past). They are dirt cheap, but also very late game tech, almost certainly a plan to reduce them a bit and/or do something, but I'd need to play a bit with them to see what I think first. Good catch though, I'd never really looked at them in that way, they just look too weak for a late game hull for me. But then, rock swarms are supposed to be lots of inconsequential rocks hiding a bigger threat.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
Vezzra
Release Manager, Design
Posts: 5265
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#13 Post by Vezzra » Tue Nov 03, 2015 8:41 am

Kassiopeija wrote:which ship are you referring to?
The Scattered Asteroid Hull, not the mini-swarm. Didn't realize that the latter didn't get its shield bonus reduced, guess that 5 is a bit much for a hull that cheap. Maybe reduce the bonus to 2 or even 1?

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#14 Post by Kassiopeija » Tue Nov 03, 2015 10:46 am

One thing to notice is that this thread (amongst others) is a fine example of several different players who raise a complete different opinion which hull/line-of-research is the best. Not saying now that me or those persons are ultimately/mathematically right/wrong, but if you look a bit behind the arguments raised it'll boild down to personal playstyle or general tactics to win the map in conjunction with a personalised research patters.

And isn't this one hell of a proof that the techtree is kinda balanced as is? Not saying that is perfect, no, but something that is openly overpowered is usually recognized by the majority of players as it, and soon becomes a mandatory no-brainer. Guess if I would have to name some of those, it would be a few in the production/research/population-enhancing regions.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Long, rambling post about ship hulls (v0.4.5 b2015-09-01

#15 Post by MatGB » Tue Nov 03, 2015 1:50 pm

Kassiopeija wrote:One thing to notice is that this thread (amongst others) is a fine example of several different players who raise a complete different opinion which hull/line-of-research is the best. Not saying now that me or those persons are ultimately/mathematically right/wrong, but if you look a bit behind the arguments raised it'll boild down to personal playstyle or general tactics to win the map in conjunction with a personalised research patters.

And isn't this one hell of a proof that the techtree is kinda balanced as is? Not saying that is perfect, no, but something that is openly overpowered is usually recognized by the majority of players as it, and soon becomes a mandatory no-brainer. Guess if I would have to name some of those, it would be a few in the production/research/population-enhancing regions.
That's my current take, there are issues, and some lines require a more subtle approach, but given one persons 'impossible' is anothers 'dead easy' I think we're close to balanced for the current state of the game. As we add new features and/or balance other aspects of the tech line then the hull line'll need more adjustment, and I do think we have a big gap in the ship size area between about 7 and about 14 external slots, there aren't many hulls at all in that size at all which means the big hulls really can/do dominate.

Plus, more internal parts'll mean they become more important and that will require more tweaking. For now, the big thing for me is to make stealth viable and fun, as currently the ships that rely on it are all/nothing in a Not Fun way, I have many many ideas but I don't know which will work.

Aside: Geoff introduced backend code for ship parts that can produce PP and/or RP so we can make factory ships, research vessels, etc. I haven't had time to even look at it, but it'd be cool to see, if anyone wants to give setting up such things a go they'd get help.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Post Reply