Brainstorm: Discussing fleet role overlaps

[Wingflier]

This is a problem that's been around for quite awhile, which I think we first started discussing back in the ~6.0 era, but which, in light of other, possibly more important issues, has been put on the backburner numerous times. I think now, at the start of 8.0, is a good time to bring it up again, and see where the community stands on this issue.

My current beef with the modern system of balancing is that it leaves a lot of overlap between roles. In fact the roles themselves have often been muddied to the point that they don't really matter that much anymore. I typically don't find that it matters what is in my fleet composition as long as I'm keeping it alive and moving it around together. When choosing what ship best fits my composition, there is almost always a clear and easy choice, because with so many overlapping functions, one always comes out on top.

Many of the roles have become so superfluous that some of them are barely used, and are clearly (mathematically) the inferior option. To give an example, when it comes to the issue of crowd control, Beam Frigates, Grenade Launchers, Electric Shuttles, Lightning Starships, Lightning Torpedo Frigates, MLRSes, and Spire Railclusters all do the same thing in slightly different ways. However, if your goal is crowd control (killing many small ships), in the most cost efficient way possible, there are clear choices much better than the rest. This is because in the end, the verdict comes down to pure DPS and survivability anyway. The armor and ship hull mechanics make almost no difference, and this is applicable in almost every other situation as well with our current balance.

The problem is exacerbated by the fact that small, high cap ships are clearly inferior in unquantifiable ways to their less numerous, tougher counterparts. In this case, the Spire Railclusters and the Lightning Torpedo Frigates are both immune to most forms of instagibbing, yet pack nearly the same punch as their weaker, more plentiful cousins. Maybe the cap DPS is higher for the higher cap ships, but why does it matter when half of them get obliterated in a single wave of a needler guard post or laser guardian? Especially in attacks on the AI Homeworld, high cap units just DISINTEGRATE within seconds. They are also vulnerable to Ion Cannon fire, swallow and the like, and to top matters off, their higher quantity makes attacking into Eyes an even bigger threat.

There is almost no advantage to having higher cap numbers in this game. I can't think of a single advantage. Lower caps are usually more cost efficient (you can repair them instead of rebuilding), have numerous immunities, don't trigger Eyes as badly, are less vulnerable to aoe attacks, and can be microed more easily to ensure individual survival.

To conclude, I want to point out some of the major flaws with the current balance:

1. Hull type and armor mechanics nearly irrelevant.
2. Low caps significantly better than high caps in almost every way.
3. Major ship overlaps in almost every role, yet the winners of said overlap are usually the low cap/high dps option.
4. Many unique mechanics are thus rendered obsolete, such as polarization and armor rotting either because the mechanics these work on are useless or because the best ships are immune.
5. So many immunities makes some of the most exciting mechanics irrelevant (parasitism used to be amazing, now half the cast is immune).
6. Players are punished further for using the worst kinds of ships (AI Eyes).
7. Starships a more viable option in most cases than fleet ships because of their low cap/high immunity (the fact that they cost more is irrelevant, players aren't going to do major attacks without full caps anyway).
8. Cost mechanics as a balancing method makes almost no difference, it just increases the time a player has to wait between attacks. More expensive, durable units are always worth it.

I want to point out that all the balance flaws I've pointed out in no way makes AI War a bad game, it is still fun as hell to play even though (in my opinion) 25% of the bonus ships are vastly superior to the others. However, I think that if we came up with a viable solution to these problems, and gave each unit more unique roles and balancing techniques, it could be even better than it is now.

So let's discuss.

[Faulty Logic]

1. Hull type and armor mechanics nearly irrelevant.

I'm pretty happy with not having to dig through stats very often, while still sometimes needing to look at all the bonuses/armor, etc. So I don't think this is an issue. Also, veteran players just know what ships to use against what, which is bonuses/armor, even if one doesn't necessarily think directly about it.

2. Low caps significantly better than high caps in almost every way.

YES. There are two main problems. The first is damage decay, as high-cap ships lose dps at a much faster rate. I think a starting solution would be to simply multiply the high-cap ships ship cap by 1.5, leaving individual ship stats unaffected.

The second problem is, as Wingflier mentioned, the immunities. Low-cap ships have most or all of the immunities to their most intuitive counters, like insta-kill, reclamation, and tractor beams. These should simply be removed (not from starships, though).

7. Starships a more viable option in most cases than fleet ships because of their low cap/high immunity (the fact that they cost more is irrelevant, players aren't going to do major attacks without full caps anyway).
8. Cost mechanics as a balancing method makes almost no difference, it just increases the time a player has to wait between attacks. More expensive, durable units are always worth it.

I strongly disagree here about cost. It is always relevant, because you're always limited by it in some way, even if it's just the speed at which you can crank out replacements.

Starship-heavy can work, but I don't think it's strictly superior to fleetship-centered by any means.

[Wingflier]

I'm pretty happy with not having to dig through stats very often, while still sometimes needing to look at all the bonuses/armor, etc. So I don't think this is an issue. Also, veteran players just know what ships to use against what, which is bonuses/armor, even if one doesn't necessarily think directly about it.

My point was that it seems rather irrelevant anyway because you're just fleetballing everything around. Even the Fighter/Bomber/Frigate triangle no longer seems to make much of a difference, at least to me.

I strongly disagree here about cost. It is always relevant, because you're always limited by it in some way, even if it's just the speed at which you can crank out replacements.

I haven't seen the disadvantage of sitting in my base, putting the game speed on +10, and waiting for the more expensive, more efficient units to rebuild.

Starship-heavy can work, but I don't think it's strictly superior to fleetship-centered by any means.

The idea was that even in fleetship-centered builds, the fleetships that are the best are practically starships anyway. I find myself just ignoring the fleetships most the time except as base defense. What high cap fleetships do you often use in your army?

[Vacuity]

The poor balance of high-cap fleetships has surfaced and re-surfaced time and again.  I don't think you'd need to look very far back to find a thread either about it or where it gets mentioned.

Starships themselves are pretty tough, but have a very poor damage output for the resources you put in.  Fleetships provide a much better damage/resources return, that's why I always build all my fleetships before I build my starships at the beginning of the game, although over the course of a game I use my starships more than quite a lot of my fleetships.

As for ship roles; while the fleetball is simple and functional, it's also very inefficient and is probably exacerbating the losses you're taking with your ships.  While I don't know offhand all the multipliers and armour types, I make sure I get familiar enough with the synergies between the AIs unlocks and my ships in each game to try and make sure I'm using the right ship for any given job (and sometimes the fleetball *is* the right choice), or just get certain ship types out the way.  This generally reduces your losses greatly and you'll find that the ship types actually *do* matter.  I'll go for Anti-Armor ships as a good example here: stick them in your fleetball and you're going to spend vast quantities of resources rebuilding them over and over again, and they won't seem to do anything much (because they're always dead).  Use them in the right place at the right time and they'll take few casualties themselves and save any of your other ships from taking any damage.

If you think resources and time don't matter, try turning on Automatic AIP.  Add in stronger Exos or other AI Plots.  Try a Quadratic AI type.  All these provide strong reasons to get the job done quickly and efficiently because you really don't want to spend time rebuilding things, and that in turn means you need to focus on using the right tool in the right place at the right time.

For genuine high-cap ships, most of them are pretty weak, but this thread has a pretty good guide to which ships are useful for what, and some of them are not low-cap ships.

[Hearteater]

Fleet balls will always be superior while we lack the ability to issue more complex orders. I'd love to have my units englobe a warp point but I am not manually positioning them for that. I had a wave of Zenith Beam Frigates last night on 10/10...spreading out would be great, but it just takes too much micro. I can't order hit-and-run for units (attack target, back out of range until ready to fire again, repeat). Micro'ing while a Space Dock is churning out replacements messing up your control groups is also hard. Starships require a lot of babysitting to keep alive (Riot, y u so close! 23k range bro!), especially Heavy Bombers. At least they are tough but I like to use them as assassins and bouncing in-and-out is tough to set up especially if you want them to wait on being repaired. You also can't tell your ships to stay out from under force fields, that's all manual. We are limited to a single preferred target for ships, and we can only set that when that ship is in the same system between runs.

So yeah, I can totally see fleet balls being ideal. But to me it isn't because the ships are stepping on each other's toes too much. We don't have the controls needed to make the most out of them. I'm working on some new controls in my UI proposal which I hope to have done sometime this weekend.

I can agree that high-cap ships are often too week. I think part of that is low cap getting insta-kill immunity too often. But it is also true there is some overlap among bonus ships and a few (Armor Rotter, I don't choose you) have less than ideal abilities.

[TechSY730]

Re: Hull types and other "flavor" not mattering
Um, hull types and bonuses actually matter a lot in the early to mid game, and in the end-game where the AI core and AI homeworlds have enough defenses to threaten even late game fleets. In those cases, if you want to get anything done without excessive losses, you need to be able to pay attention to bonuses, ranges, hull types, and on occasion, even armor. In these stages of the game (early especially), minimizing losses is very important. This is especially true on higher difficulties.

I will grant you though that one of the most "memorable" phases of the game (mid game to approach to and preparing to attack AI HWs) is one of those places where it doesn't matter nearly as much, and is one of the most "blobby" stages of the game, which is unfortunate.

Re: The game's tendency to promote "fleet blob" play
Based on past discussions, the "death ball"/"fleet blob" nature of the game of the game has two big causes. One is the lack of UI options and order options. Two is the relatively simple combat model. The facing direction of units doesn't have any say in where units can fire (or more accurately, units can change facing direction instantaneously), there is no "cone of where I can fire from my current facing direction" aspect really. Almost all ships can fire while moving. There is very little of "terrain" to speak of (forcefields, gravity, and tachyon range being pretty much it, and even then, that is controlled by units, not inherent to the map).

All of these together "conspire" to produce a setup where micro is not only hard to do, but isn't all that rewarding. So, why not fleet ball once your economy is strong enough to be able to handle somewhat more than technically necessary losses? Pretty much the only "army management" is making sure your truly expensive stuff stays alive and avoiding a total fleet wipe once you hit that stage.

More order options and UI options for control will help some, but the fundamental "simplistic" combat and environment model will still be a factor.

I guess one of the reasons why it is so "simplistic" is the scale of this game. A fancy, Starcraft (I or II) or Command and Conquer style combat model might work well for a few hundred units, but what about a few thousand? Thousands of units vs thousands of units with a "complex combat model" is enough to tax even modern processors; heck it is even taxing some processors under the current simplistic model. This is compounded by the single threaded nature of the simulation. So it isn't as simple as "putting more depth into the combat model", simply doing that will tank performance.

Re: Low cap vs high cap ships
Agree with low cap fleet ships getting too many immunities by default. And the pros and cons of high cap, mid cap, and low cap (under the current combat model; yes even under the current combat model there are cons with low caps and pros of high caps, even if they aren't strong enough to truly bring things into balance), and the necessary balance adjustments needed to counteract destabilizing pros and cons has been discussed many times. Someone else can crunch the numbers (for example, average quantitative effect of DPS "decay" due to ship losses vs DPS "wasted" due to overkill), which is what we really need to give concrete balance suggestions for this particular aspect.

Re: Ship role overlap
I would agree that some bonus ships may be too niche or too similar to other types of ships available. I would also say that some roles are disproportionately useful (like the bomber role), though whether that is a good or bad thing is up to debate (your goal is to kill a structure after all, the AI HW. That primary objective alone gives extra value to bomber roles). I would argue though that the problem isn't nearly as bad as you seem to be making it out to be.

Re: Starships being more "useful"
I get the current starship balance (expensive, low cap, high cap HP compared to fleet ships, low DPS compared to fleet ships) and I do think this is a nice role distinction from fleet ships. But I too am beginning to feel that maybe the current balance pushed them too far in that direction...

The contrast in how you manage and use fleet ships and starships does seem a little...extreme right now.

[Draco18s]

Fleet balls will always be superior while we lack the ability to issue more complex orders.

And...that's why I don't use etherjet tractors.  There is NO WAY to "select all etherjets that are not currently tractoring something" and tell them to re-engage.

[Hearteater]

I just want to throw this out there on Starships: The Neinzul Enclave Starship is so powerful it completely negates a player's choice of bonus ship (the Enclave SS puts out more damage while begin 100% safe from all damage) as well as hull bonus considerations in the early game thanks to free drones that conveniently deal with nearly all hull types as if not more effectively than your regular fleet ships. I'm currently holding off 10/10 waves using just Mark I & II Enclaves + Spider Turrets. Just toss in some Tractors and Gravity Turrets for cannon fodder and to slow down the wave until all their engines are destroyed and you're done. There is no reason to waste my regular fleet ships.

[TechSY730]

The Enclave starship has been such a balance issue for so long I am wondering if it should just be disabled and removed until we can figure out how to balance it or how to redesign it into something more manageable balance wise.

Or if that is too much, nerf it into the ground (and say so in the description) until we can.

Right now, feedback seems to be that it is OP in the unfun way (aka, trivializes the game way), which, IMO, is justification for more drastic short term action.

EDIT: Just a thought, the enclave starship has been a big enough issue that it probably deserves its own thread.

[Draco18s]

Enclave Star is amazing.  I played a game once utilizing them + Lightning Torpedo Frigates.

"What Eye?  Oh that thing.  Well it's a [reclamation, translocation, etc.] and the drones are (effectively) immune."

[Hearteater]

EDIT: Just a thought, the enclave starship has been a big enough issue that it probably deserves its own thread.

Done.

[Wingflier]

Alright, here are my list of minor suggestions to fix some of the current issues with the game:

1. Remove most immunities. I see no reason why a low cap ship can't be reclaimed, paralyzed, or swallowed. If low cap ships, for example, could be reclaimed, that would just make the game 100x more awesome.

2. AI Eyes no longer work based on individual number, but on total firepower (this is a no brainer).

3. Rethink the roles of each bonus ship so that they no longer overlap in such obvious ways and attempt to fill a certain niche the others currently don't.

However, I will say that I do think this solution is just a band-aid to the actual problem, which will keep cropping up in the current system. However, if we're only interested in fixing the smaller issues for the moment, this is what I would recommend.

[TechSY730]

A "real" fix to this issue would probably require a reduction to the "scale" of the game.

This includes, but is not limited to:
- Number of ship types (especially bonus ship types)
- Reducing economic potential
- Reducing number of ships (this includes caps (bring it down to 100s vs 100s sized fleets, how much one can afford to build/replace per-phase of the game, how much the AI gets, etc)
- Reducing average lifetime of units (not to SC2 levels, that is too short, but right now, it is almost to the point where counters are nearly overshadowed by how long things tend to live, even for fleetships)
- Average map sizes (number of planets mostly, planet "size" (the "grav well") seems about right)
- Average time to win/lose (this would naturally follow from the above)

This would give enough "wiggle room" to afford adding some new ways for combat to be "deeper" (firing angles and turn rate, shots interacting with obstacles, more ships that can't fire when moving, harder collision detection making optimal positioning more important). Due to engine limitations I gave above, such depth cannot be added simply as the game is now.
This would also make it easier to manage "role overlap" as there would be less things to give roles to, and less extremes (like the difference between low cap and high cap ships, the difference between early game and late game armies, etc, wouldn't be quite as disparate) to balance for. EDIT: The number of bonus ship types right now is so huge it is near unsustainable.

EDIT2: Also, player limitations, reasoning, and reward structure (How can you reward microing "away" from "fleet ball" when there are 1000s of ships, most of which are disposable? How can a human reasonably do anything but fleet ball when there are 1000s of ships?) are a big contributor to how the current scale limits combat depth.

However, this would be a HUGE change. Bigger than the major change of ship bonuses that happened from 3.0 to 4.0. So big that I would not call the end result AI War anymore. It would be AI War 2 (a new product really).

As much as Chris is against it, IMHO, we are very close to the point where the game is too "bloated", and they will have to start stripping out (possibly some core) features or design goals to keep the game from imploding into a giant, unweildy, not "fun to play" mess, which no one wants for a great like AI War.

Even though such radical stripping seem to be against the spirit of AI War major version revisions, the current style isn't sustainable.
I hate to say it guys, but I think the need for an AI War 2 like project is soon going to be needed if we want more innovation for AI War...

Thankfully, we are not quite there yet, but we are very close to that point.


In the short term, I think "band-aids" to fix the current state will work. Maybe once we have community and commercial* consensus, basically "freezing" AI War (no new ship types, AI types, plots, minor factions, or expansions. Only UI tweaks, bug fixes, and minor balance changes accepted, and maybe map types (those are pretty low impact balance changes)) may be what we will have to go for soon, so we will no longer need a "band-aid bonanza" set of patches after every major shift (like is becoming apparent is needed now), as there would be no more major shifts.


*aka, sales starting to sharply increase the rate of decreasing in rate in ways attributable to the sheer scale of the game being too overwhelming/paralyzing to newcomers OR the time where continuing to do new major features and content to AI War stops becoming profitable because of the aforementioned reason

[Cinth]

A "real" fix to this issue would probably require a reduction to the "scale" of the game.

This includes, but is not limited to:
- Number of ship types (especially bonus ship types)
- Reducing economic potential
- Reducing number of ships (this includes caps (bring it down to 100s vs 100s sized fleets, how much one can afford to build/replace per-phase of the game, how much the AI gets, etc)
- Reducing average lifetime of units (not to SC2 levels, that is too short, but right now, it is almost to the point where counters are nearly overshadowed by how long things tend to live, even for fleetships)
- Average map sizes (number of planets mostly, planet "size" (the "grav well") seems about right)
- Average time to win/lose (this would naturally follow from the above)

This would give enough "wiggle room" to afford adding some new ways for combat to be "deeper" (firing angles and turn rate, shots interacting with obstacles, more ships that can't fire when moving, harder collision detection making optimal positioning more important). Due to engine limitations I gave above, such depth cannot be added simply as the game is now.
This would also make it easier to manage "role overlap" as there would be less things to give roles to, and less extremes (like the difference between low cap and high cap ships, the difference between early game and late game armies, etc, wouldn't be quite as disparate) to balance for. EDIT: The number of bonus ship types right now is so huge it is near unsustainable.

However, this would be a HUGE change. Bigger than the major change of ship bonuses that happened from 3.0 to 4.0. So big that I would not call the end result AI War anymore. It would be AI War 2 (a new product really).

The only thing I can agree with here is that it would cease to exist as AIW.  And it would kill what the game is to me as a player (and I've been a player since not long after it launched).

I love the sheer scale of this game.  I love that I don't have to micro.  I love that I can spend a week playing a single match.

[TechSY730]

A "real" fix to this issue would probably require a reduction to the "scale" of the game.

This includes, but is not limited to:
- Number of ship types (especially bonus ship types)
- Reducing economic potential
- Reducing number of ships (this includes caps (bring it down to 100s vs 100s sized fleets, how much one can afford to build/replace per-phase of the game, how much the AI gets, etc)
- Reducing average lifetime of units (not to SC2 levels, that is too short, but right now, it is almost to the point where counters are nearly overshadowed by how long things tend to live, even for fleetships)
- Average map sizes (number of planets mostly, planet "size" (the "grav well") seems about right)
- Average time to win/lose (this would naturally follow from the above)

This would give enough "wiggle room" to afford adding some new ways for combat to be "deeper" (firing angles and turn rate, shots interacting with obstacles, more ships that can't fire when moving, harder collision detection making optimal positioning more important). Due to engine limitations I gave above, such depth cannot be added simply as the game is now.
This would also make it easier to manage "role overlap" as there would be less things to give roles to, and less extremes (like the difference between low cap and high cap ships, the difference between early game and late game armies, etc, wouldn't be quite as disparate) to balance for. EDIT: The number of bonus ship types right now is so huge it is near unsustainable.

However, this would be a HUGE change. Bigger than the major change of ship bonuses that happened from 3.0 to 4.0. So big that I would not call the end result AI War anymore. It would be AI War 2 (a new product really).

The only thing I can agree with here is that it would cease to exist as AIW.  And it would kill what the game is to me as a player (and I've been a player since not long after it launched).

I love the sheer scale of this game.  I love that I don't have to micro.  I love that I can spend a week playing a single match.

Agree, I don't want to lose all of what I said; even though my wording didn't make it clear that all of those kinds of "scale" wouldn't need to be removed (some scale would still remain).
Also, if there was an AI War 2 that went this route, the scale of these could still be much larger than the traditional RTS, but maybe by only .75-1.5 orders magnitudes as much, maybe not the 1.5-2 orders of magnitude like many of those are right now.

I think pretty much the biggest thing to address the OP's primary concern (fleet role overlaps) would be bonus ship type reduction, which is entirely feasible to do and still be "AI War". However, it might mean removing ship types already advertised as features in expansions' feature list (which people have paid for), so that might be a no go.

However, even all the bonus ships we have now, something can be done about this. Note that I said the sheer number of bonus ship types right now is near unsustainable, not quite there yet.

Finally, I was NOT try to advocate a start on AI War 2 (as much as it seemed that way, now that I look back on that post). Rather, it was to point out:
1. Some (but nowhere near all) of the things in the OP are basically fundamental design downsides of the nature of this game (this very same nature brings its own set of upsides; those upsides, for me at least, are one of the big things that drew me to AI War in the first place)
2. The current game is near "saturated" in terms of how much can reasonably fit into one product.


EDIT: Grammar and clarity