Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[MesoTroniK]

Eh, missile AI is part of a missile's balance and intended feel; it doesn't necessarily need to be "good" or "bad". And I'll note that vanilla missile AI is mainly intentionally the way it is, e.g. target tracking could obviously be more effective but then missiles would be too hard to dodge. I don't particularly think target picking is something most vanilla missiles need to worry about. Locusts do, but then they do target-picking decently. But primarily I think target-picking is the player's job (or the ship AI's). Otherwise you get a game that plays itself, or at least a move in that direction...

As far as ships vs fighters, I think it's mostly - like I said earlier, and like Morrokain said - an issue with not deciding to commit. Which is more or less unavoidable. If you've got a good example of poor vs-fighter targeting making a big impact, though, I'd love to see it! (Preferably simulator with vanilla loadouts etc...)

The sorts of things I was talking about doesn't override player (or AI) targeting decisions most of the time except when it is the only move that truly makes sense. Like let me give some examples:
- A PD missile won't target ships *at all* unless there are no missiles or fighters within range.
- A big slow derp missile, will never target fighters unless there are no ships within range.
  - A big slow derp missile if the original target is destroyed or retreats or whatever, will retarget the nearest non fighter ship within range basically as per the above. This is a big one and one of the big issues with vanilla missile AI. Oh I fired this Harpoon volley at an overloaded ship, but something else killed it as they were enroute. Well, they retarget... Fighters nearby, rather than a destroyer just a bit farther away. This is terrible! And conceptually generally the same issue as big huge alpha slow cannons being generally wasted on fighters in most situations and all of them in the case of some really enormous mod ones.

As far as examples to show regarding guns, will see what can be done. I have seen it hundreds and perhaps thousands of times but will take a little bit of digging to list a reproducible scenario.

[Alex]

Spoiler

I also wonder, is there design space for anti-massed fighter/missile spam weapons?  I mean, I only recently discovered the proximity charge launcher as being effective, and there's also flak, but it just doesn't stack the same way as fighters do.  You can't currently combine all the PD of the fleet in one place to handle a spike in fighters, but fighters can stack to focus all their fire power on one ship.  And due to flux mechanics, spike damage is much more powerful than sustained damage.

Remember, the player controlled spam fighter fleet is far more dangerous than the AI spam fighter fleet, since a player can call coordinated bombing runs which any individual ship can't survive.  They can pull back to a way point, gather strength, and then go again and take out another ship.

But what if the defending fleet could gather effectively an entire fleet's worth of PD in one spot?  I do know you've modified the Paladin to have an AoE component, which will be interesting to see, given it can shoot over allied ships to hit fighters on the other side.

If you do get enough AoE damage in one place that can last long enough, it is possible to bring down a swarm of sparks.  I remember doing something like that against a 12 Drover spark swarm in the simulator using an Onslaught XIV with hand picked Officer (Advanced Countermeasures, Armor Skills, Damage skills, Flux skill) + armor/flak focus on the ship itself.  With the magnified armor, it survived the first pass swarm and took out enough fighters along the way with its purely flak loadout, it eventually became immune to the uncoordinated return flights, and just slowly burn drived down each Drover.

So is there a place for a missile or maybe high tech energy weapon that chains between enemy fighters/missiles within a certain distance from each other like lightning, hitting each target only once, but then jumping to all targets within range of that target, and so on until all possible targets are exhausted?  A couple of these and sufficient density of fighters suddenly makes the damage spread to the entire swarm.  Below a certain density, it only hits a few and then runs out of range.  And against single wings, well, it only hits a single wing.  And against a ship, assuming its not bumping another ship, you get a single hit.  So now, you've got an anti-fighter weapon which fired from different directions on the swarm, hits all of them.  Potentially concentrating fire from multiple ships.

If there were some dedicated PD variant files that used such a weapon, and they showed up enough, they would give over concentrations of fighters some pause, at least in some engagements.

Not sure how easy or hard to script such a weapon, although it'd be a bit like ion damage jumping, but between fighters/ships instead of just on the ship.

[close]

I think something like that could work, yeah. Basically AoE that's specifically more effective vs fighters  - most other AoE doesn't quite get the job done due to the explosion radius not being enough. I mean, a hypothetical flak with 500 explosion range - a couple of those would chew up just about any number of fighters. The Doom is quite good at this sort of thing already, btw, but that's the closest there is currently to a hard counter for fighters.

I think massed fighters will already be weaker in the next release, though - less buffs through skills, and the buffs scale down sharply with the number of fighter bays in the fleet. So it's discouraging a fighter mono-fleet from another angle. Plus, anti-fighter officer skills are also more effective.

For example, you can get... I think +50% fighter replacement rate, fleetwide - but only at 6 fighter bays in the fleet - not deployed, but fleetwide total. (Numbers could be tweaked, of course; 6 feels like it might be a bit high.) So you could have one or two very effective carriers, with some support - or you could have a ton of more lackluster ones. Etc.

The sorts of things I was talking about doesn't override player (or AI) targeting decisions most of the time except when it is the only move that truly makes sense. Like let me give some examples:
- A PD missile won't target ships *at all* unless there are no missiles or fighters within range.
- A big slow derp missile, will never target fighters unless there are no ships within range.
  - A big slow derp missile if the original target is destroyed or retreats or whatever, will retarget the nearest non fighter ship within range basically as per the above. This is a big one and one of the big issues with vanilla missile AI. Oh I fired this Harpoon volley at an overloaded ship, but something else killed it as they were enroute. Well, they retarget... Fighters nearby, rather than a destroyer just a bit farther away. This is terrible! And conceptually generally the same issue as big huge alpha slow cannons being generally wasted on fighters in most situations and all of them in the case of some really enormous mod ones.

Gotcha. My point really is more that this isn't "good" or "bad" but a design choice. E.G. the Harpoon being basically useless in that scenario is ... just how it works. I think it'd be a mistake to think of an addition of more intelligent re-targeting in this case as a strict improvement - it'd be a change that makes the weapon more effective, but it's just a change, with different gameplay consequences.

Like... is it making the missile AI better at doing damage? Yes. Is it making the game better because it now does that? That depends, so: design choice.

As far as examples to show regarding guns, will see what can be done. I have seen it hundreds and perhaps thousands of times but will take a little bit of digging to list a reproducible scenario.

Thank you! Honestly, I probably wouldn't spend too much time on it, but if you have something fairly easily reproducible, then yeah, I'll check it out. A bunch of various changes already on my end, so it'll be interesting to see if it still happens with the dev build, and to what degree. (I'm more leaning towards including the "don't fire at fighters" hint, btw. The idea of calling it out explicitly is really selling me on it; could see maybe using it for a few vanilla weapons, even.)

[FooF]

Re: Weapons being billed as "will not shoot at fighters:

I'm sure you'll come up with more nuanced version but for me, if one rule was followed, I'd be very happy with such a tag.

*Any weapon that costs >300 flux/shot won't shoot at fighters.*

To be explicit: Mjolnir, Guass Cannon, Heavy Blaster, Plasma Cannon, Hypervelocity Driver, Hellbore, AM Blaster, Mining Blaster. (Heavy Mauler is about there, too, along with the Mk. IX. I'm not sure how you would look at high-flux beams like Ion Beam, HI Laser or, Tachyon Lance, but the Phase Lance is fine.) Outside of the Mjolnir, every other weapon is a high-damage, relatively slow-firing projectile that has little probability of hitting a particular fighter and would almost always be a Pyhrric victory even if it did.

My only concern is the number of weapons I listed: 1 Small, 2-4 Medium, 4-7 Large. Large Weapons have the easiest excuse not to target small/fast-moving targets but that does leave a lot of Large Weapons not firing at fighters, even if it does make gameplay sense.

[Hiruma Kai]

Re: Weapons being billed as "will not shoot at fighters:

I'm sure you'll come up with more nuanced version but for me, if one rule was followed, I'd be very happy with such a tag.

*Any weapon that costs >300 flux/shot won't shoot at fighters.*

To be explicit: Mjolnir, Guass Cannon, Heavy Blaster, Plasma Cannon, Hypervelocity Driver, Hellbore, AM Blaster, Mining Blaster. (Heavy Mauler is about there, too, along with the Mk. IX. I'm not sure how you would look at high-flux beams like Ion Beam, HI Laser or, Tachyon Lance, but the Phase Lance is fine.) Outside of the Mjolnir, every other weapon is a high-damage, relatively slow-firing projectile that has little probability of hitting a particular fighter and would almost always be a Pyhrric victory even if it did.

My only concern is the number of weapons I listed: 1 Small, 2-4 Medium, 4-7 Large. Large Weapons have the easiest excuse not to target small/fast-moving targets but that does leave a lot of Large Weapons not firing at fighters, even if it does make gameplay sense.

I don't think heavy blasters projectiles are that slow.  I actually rely on auto-fire heavy blasters turrets on my SO Aurora builds to help swat fighters quickly.  I don't aim them, but I do back up at with plasma burn on, and fighters tend to fly straight at the Aurora, making them easy targets for the auto-fire turrets.

Similarly, if the fighters are as thick as flies, I want the AI Plasma Cannons firing since they have pass through and will hit multiple targets.   Hypervelocity drivers shots are fast and long range enough that I see them picking off bombers on attack runs or returning all the time.  Hellbore is probably slow enough you wouldn't want to risk a high deflection shot, but against a bombing run coming straight in, its fine.  Also, wasn't pass-through recently added to it as well?

And if a ship has sufficient flux dissipation to cover firing the weapon, why is it a bad idea to fire it?  An SO Aurora can be flux neutral firing 3 heavy blasters.  Why wouldn't you want to fire them in that case?  It doesn't harm your ship in any way when your shields are up - since that dissipation is going to waste anyways. I don't want my ship sitting there using only 1/10th of its soft flux dissipation firing PD weapons only while surrounded by fighters.

If the fighters are moving in a straight line towards or away, like bombers on an attack run against a capital ship, firing big guns can be a good idea with decent odds of hitting.

Certainly on most vanilla variants, flux dissipation doesn't equal or exceed flux generation, but on player designed ships, it can.  At which point such rules start to be detrimental rather than advantageous.

To be honest, if there was a section to weapons setup in addition to groupings that let a player customize target priority and willingness to fire in something like a matrix (i.e. 1,2,3,4 and never, split along missiles, fighters,ships for each weapon, like how we setup weapon groups) that would be really cool.  Although perhaps that is too much fiddling for vanilla.

In theory, is such a thing possible for a mod?  Not sure how well they can hook into the UI like that.  Given we already have a hull mod that makes PD weapons prioritize missiles over fighters, as well as make all small weapons prioritize missiles, I'm guessing it'd be possible under the hood, but making an easy to use interface is the hard part.

Then players who want to have plasma cannons fire at missiles with max priority, could.  And the players who never want plasma cannons to shoot at missiles or fighters could as well.  I mean, as a player, I'm using Plasma cannons all the time to shoot down reapers or atropos on my Odyssey.

[Nafensoriel]

Not sure I like the idea of "x flux = cant shoot at fighters". We have tons of references(honorverse off the top of my head as the main one) that reflect "capital" ships utilizing mainline guns to strengthen PD. It's squarely in "design" and "balance" choice wise though so it's up to Alex.

If anything allowing all projectiles to pass through fighters to continue doing damage to the wing would probably have more impact than a special PD weapon.
If plasma batteries were tweaked to fire very quickly they are extremely effective fighter sweeps.


-----------
As to some guns having more "sweep" power... just give them conical flak with long lifetimes and low individual damage after the round expires or hits a fighter. PD saturation would then increase as the AI retreats from fighter swarms and on the initial approach.

[TaLaR]

*Any weapon that costs >300 flux/shot won't shoot at fighters.*

That's a horrible idea.  Cleaning up the approaching fighter wave with all available firepower is very much viable strategy. You don't even need to be accurate to hit large fighter cloud, just shooting in their general direction is good enough - this could be different if fighters dodged like bullet hell protagonists, but they just fly in patterns.
It's not like I always have just one weapon group that could be controlled manually (like 2 Plasma Odyssey).

[Tartiflette]

A decision like "don't fire at the fighter because a better target is not in range, but will be in range soon" is too difficult to make well. And it's more decision-making that I think autofire should do - it should be predictable rather than "smart" since it's more of a control for the ship's main AI than a separate brain.

I could see a use case for something not firing at fighters *at all* though. I just don't think it would make sense for this to apply to the vast majority of weapons, e.g. nothing in vanilla.

I mean, I think I understand where you're coming from! Just, to me it seems like "having a set of tags that make target selection better" sounds a lot better than it would actually be, for practical reasons.

Right, a IGNORE_FIGHTER is limited, but it will already help immensely with cruiser/cap-mounted spinal weapons, weapons with long chargeup, strike weapons with clips etc.

Now another thing that makes a lot of weapons vastly underperform against fighters is the turret rotation dampening while firing and cooling down. It was designed when the game was much slower overall: There were a lot less mobility systems, ships were slightly slower and fighters were a LOT slower. Maybe it should be toned down or removed entirely given the current balance of the game? Or at least there could be another weapon hint like NO_ROTATION_DAMPENING?

[Goumindong]

I could see a use case for something not firing at fighters *at all* though. I just don't think it would make sense for this to apply to the vast majority of weapons, e.g. nothing in vanilla.

Antimatter Blasters
Ion Pulsar*
Autopulse Laser

Are three that come to mind. The AB's delay makes it really hard to hit fighters. the Pulsar and the APL just dump charges. Though an accuracy clutch might fix these as well.

Though the main AI tag i would like would be if some weapons refused to fire at your selected target and instead maintained PD/fighter priority. Particularly for things like burst PD and Flak cannons.

*the ion pulsar could probably use a strike tag too. And it would be interesting to see what would happen to the game if HE/Frag weapons in general were given an "anti-armor" tag that caused them to fire "if the enemies shield was down or if their flux was high"

[intrinsic_parity]

I really don't think firing at fighters with high flux weapons that are unlikely to hit is more understandable than not firing. Firing a weapon and missing is actively worse than not firing so weapons with slow projectile speed/ low rate of fire/wind ups /slow turn rate are really not suited too shooting fighters down the majority of the time. If they're likely to miss, then you're just wasting flux. I think turn rate and projective speed are a bigger factor. Maybe there could be some logic to not shoot at things moving too fast to hit/track?
I think most weapons are fine to shoot at bombers because they move slow and are easy to hit (and its very valuable to shoot them down before they fire), but shooting at fighters/interceptors with big guns can be really detrimental.

I'm also of the opinion that spending lots of flux capacity to clear fighters with inefficient weapons is bad as well. You're basically doing the fighters job for them by running up your own flux. It depends on your loadout and the situation of course, but it's definitely bad to fire some weapons at fighters, even if they hit.

Also, backpedaling to deal with fighters is very frequently the right choice, even for the player. You usually can't just push through fighter swarms to go for the carriers. Fighters are really strong, I don't think the AI is necessarily to blame for having a tough time dealing with them. I don't mind that the AI is not prone to making desperado charges at carriers through fighters, and the player is left to make decisions about aggression. I do wish the player had a bit more control since it is their responsibility to make those decisions though. The AI ignoring my orders is pretty annoying and the control point system can hurt too. I wish there was something in between eliminate and engage. Eliminate is borderline suicidal and engage is more of a suggestion.


Another thought: what if the player could set weapon groups to have behaviors like 'strike', 'PD', 'anti fighters' etc.? That might let people have the control they want, and as long as there was a 'general purpose' setting that was default, I don't think it would be too much of a complication for beginners.

[Yunru]

Don't know if doing this specifically is the best plan, but: If the hypothetical weapons that didn't autofire at fighters clearly said so on their stats card, that would make the behavior clearly-not-a-bug.

That's... a very good point. Yeah! I like this a lot, coming from sort of an in-fiction reason why it doesn't/can't do that, rather than being a purely an attempt to adjust the AI. So it stops being an AI thing and becomes an explicitly billed feature of the weapon.

I'm actually genuinely surprised the current weapons do get used on fighters.
Between point-defense weapons and Integrated Point Defense AI, I'd assumed that all weapons by default didn't target missiles and fighters.

[Igncom1]

They don't missiles, but fighters are almost treated like tiny frigates by the AI.

[Grievous69]

Well it would appear I'm disagreeing with the majority here since I'd rather have ships fire all of their guns in the general direction of fighters rather than wait for them to come into almost melee range. Yes, you're running up your own flux, but that's soft flux, if fighters come too close, then you're stuck with hard flux and they're even harder to get since they start circling around ships instead of coming in a straight line. A weapon not being accurate matters only if you're targeting a single wing. When there's a huge swarm coming, and your ships start firing everything, most of the shots actually connect. And that then makes the whole fight easier since I won't probably see a ship of mine get instantly nuked by fighters.

The only weapon in vanilla I'd say is a horrible choice to fire at fighters in every scenario is the Antimatter Blaster. Everything else depends on the situation. I get that it would be super useful for mods tho, I'm just talking about vanilla weapons here.

But one thing that pisses me off the most with AI is indecisiveness, constant target switching. This is easily observed with beam weapons. AI rarely focuses on one ship, instead it fires at something a few times, and then when it's high on flux or even overloaded, it switches to some random frigate/destroyer waaay back or to the side instead of finishing the ship directly in front of it. I guess something even more annoying would be if it started blasting at fighters in the middle of a duel, which actually started this whole discussion because of what happened in the tournament.

Re: Missile picking thing
I don't get how the game would play itself if the AI actually started firing HE missiles at right targets instead of small speedy ships it can never hope to hit. This is why I stopped using Hurricane MIRVs on AI ships, it's absolutely ridiculous seeing it fired at a Wolf or a Hound. And since capital spam is trying to get solved by making enemy fleets more balanced, it's gonna be even more useless. Locusts will be a better choice 99% of the time. Not to mention the horrible spread without ECCM on a HE missile...

[DatonKallandor]

Having big guns shooting fighters would be more understandable if depleting a carriers fighters supply was more of an option. But as it stands, replacements are so frequent it's a losing maneuver. And many fighters don't even get destroyed by big gun hits, just disabled and they limp back to the carrier or take a second shot to kill. And if a fighter causes a big gun to shoot twice, the fighter has already done massive damage.

[Grievous69]

I blame Expanded deck crew for that, it's a no brainer on every dedicated carrier.

[FooF]

I really don't think firing at fighters with high flux weapons that are unlikely to hit is more understandable than not firing. Firing a weapon and missing is actively worse than not firing so weapons with slow projectile speed/ low rate of fire/wind ups /slow turn rate are really not suited too shooting fighters down the majority of the time. If they're likely to miss, then you're just wasting flux. I think turn rate and projective speed are a bigger factor. Maybe there could be some logic to not shoot at things moving too fast to hit/track?
I think most weapons are fine to shoot at bombers because they move slow and are easy to hit (and its very valuable to shoot them down before they fire), but shooting at fighters/interceptors with big guns can be really detrimental.

I'm also of the opinion that spending lots of flux capacity to clear fighters with inefficient weapons is bad as well. You're basically doing the fighters job for them by running up your own flux. It depends on your loadout and the situation of course, but it's definitely bad to fire some weapons at fighters, even if they hit.

Also, backpedaling to deal with fighters is very frequently the right choice, even for the player. You usually can't just push through fighter swarms to go for the carriers. Fighters are really strong, I don't think the AI is necessarily to blame for having a tough time dealing with them. I don't mind that the AI is not prone to making desperado charges at carriers through fighters, and the player is left to make decisions about aggression. I do wish the player had a bit more control since it is their responsibility to make those decisions though. The AI ignoring my orders is pretty annoying and the control point system can hurt too. I wish there was something in between eliminate and engage. Eliminate is borderline suicidal and engage is more of a suggestion.


Another thought: what if the player could set weapon groups to have behaviors like 'strike', 'PD', 'anti fighters' etc.? That might let people have the control they want, and as long as there was a 'general purpose' setting that was default, I don't think it would be too much of a complication for beginners.

You elaborated where I didn't. Thank you.

That's what I meant by Pyhrric Victory: by firing these flux-intensive weapons at fighters, you lose even if you hit. If there is nothing else to fire at and you're at 0-flux, I get it, but the fighters are absolutely winning if the ship in question generates more flux trying to kill them than the fighters' weapons themselves.  At best, firing these big guns are grossly inefficient when they hit. At worst, you're dumping huge flux/shot and hitting nothing.

Tracking speed/shot speed is something I considered but that's a little more subjective. I do agree that it's probably a better standard than flux/shot but you might get weird edge-cases where the Heavy Blaster is still considered "ok" when it's probably the worst offender out there.

I do like the idea of designating certain weapons as "PD" or not, but unless you can designate a class of weapons or weapons group, that would become tedious. I would love the option, though, because it would make arbitrary rules like the one I suggested irrelevant.

Having big guns shooting fighters would be more understandable if depleting a carriers fighters supply was more of an option. But as it stands, replacements are so frequent it's a losing maneuver. And many fighters don't even get destroyed by big gun hits, just disabled and they limp back to the carrier or take a second shot to kill. And if a fighter causes a big gun to shoot twice, the fighter has already done massive damage.

Agreed. A Heavy Blaster shot on a shielded fighter just overloads it and doesn't actually impact the replacement rate. You're spending a lot of flux for literally no gain.

Well it would appear I'm disagreeing with the majority here since I'd rather have ships fire all of their guns in the general direction of fighters rather than wait for them to come into almost melee range. Yes, you're running up your own flux, but that's soft flux, if fighters come too close, then you're stuck with hard flux and they're even harder to get since they start circling around ships instead of coming in a straight line. A weapon not being accurate matters only if you're targeting a single wing. When there's a huge swarm coming, and your ships start firing everything, most of the shots actually connect. And that then makes the whole fight easier since I won't probably see a ship of mine get instantly nuked by fighters.

The only weapon in vanilla I'd say is a horrible choice to fire at fighters in every scenario is the Antimatter Blaster. Everything else depends on the situation. I get that it would be super useful for mods tho, I'm just talking about vanilla weapons here.

I understand this position, especially if the choice is "do nothing" vs. "do something, even it it's inefficient." Something like a Plasma Cannon probably can hit quite a few fighters that are bunched together. The question I have is whether or not it was worth 1,650 flux to kill 1-2 fighters. How long would it take those same fighters to generate that kind of flux against shields? Likewise, sustained fire from a Heavy Blaster is 720 flux/sec. Whole wings of fighters aren't putting out that kind of damage against shields, soft flux notwithstanding. You are right, though, everything is relative and situational. Firing a Heavy Blaster for a Paragon does not have the same relative "cost" attached to something like say, a Wolf. The former can afford to do so, the latter not so much. However, as a general rule, spending 720 flux/sec is non-trivial even to a Paragon and a poor use of flux to kill fighters.

In the case of having flux-parity between weapons and dissipation, as Hiruma Kai pointed out, that's a scenario where being able to fire all weapons, regardless of efficiency, is acceptable because there is no opportunity cost. However, I would argue that's the exception that proves the rule. Outside of finely-tuned ships, flux parity is extremely rare and my AI allies are the ones I have issue with using grossly inefficient weaponry to kill fighters. If the AI wouldn't use certain weapons as they backpedal from fighter swarms, they might actually be at flux parity and have a much greater chance of outlasting the swarm rather than punching themselves out and dropping shields. AI using big guns against fighters is a kind of "hyperventilating" where it's a short-term gain but long-term loss and if they had just "remained calm," they would have been fine. If a ship isn't at/near flux parity, which is more likely the case, those big guns are using flux that could otherwise be used for more appropriate/efficient guns from firing. Most of the big guns I listed are killing fighters incidentally or haphazardly anyway, so unless there is virtually no cost involved, the risk/reward is almost always too high.