Print

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

[DownTheDrain]

Right now you are not expected to say anything. Provide the video with the given Brawler variant commanded by the officer with the given skill set against the Mule variant represented in your video on the previous page. That's all.

I don't really have a dog in this fight, at least not anymore, but why would he need to pick that specific variant with that specific officer?
What would that prove even if it worked? That there's one perfect set of circumstances that gets the job done?

Because certain individual purposefully tinkering with the evidence to prove its agenda.

Well, so are you if you're limiting it to a specific hullmod and officer.
Not that I really care, my point was that the Brawler should just shrug it off and armor-tank regardless of the loadout, but Alex already explained why that doesn't happen.

To prove that Brawler can do it you need only single case. However it is not that only this given variant is usable.

Proving the Brawler can do it under a certain set of circumstances wouldn't accomplish anything.
Nobody actually playing the game is bothered about one very specific example or how that could be tweaked to do one particular thing. The original post wasn't about Brawlers but about the AI in general not advancing enough and being too risk-averse. Alex already explained why that happens. I'm not thrilled about it because I prefer to not always lead the charge, but I understand the reasons behind it.

What you two are doing right now seems rather pedantic and not really helpful.

[Lucky33]

Right now you are not expected to say anything. Provide the video with the given Brawler variant commanded by the officer with the given skill set against the Mule variant represented in your video on the previous page. That's all.

I don't really have a dog in this fight, at least not anymore, but why would he need to pick that specific variant with that specific officer?
What would that prove even if it worked? That there's one perfect set of circumstances that gets the job done?

Because certain individual purposefully tinkering with the evidence to prove its agenda.

Well, so are you if you're limiting it to a specific hullmod and officer.
Not that I really care, my point was that the Brawler should just shrug it off and armor-tank regardless of the loadout, but Alex already explained why that doesn't happen.

To prove that Brawler can do it you need only single case. However it is not that only this given variant is usable.

Proving the Brawler can do it under a certain set of circumstances wouldn't accomplish anything.
Nobody actually playing the game is bothered about one very specific example or how that could be tweaked to do one particular thing. The original post wasn't about Brawlers but about the AI in general not advancing enough and being too risk-averse. Alex already explained why that happens. I'm not thrilled about it because I prefer to not always lead the charge, but I understand the reasons behind it.

What you two are doing right now seems rather pedantic and not really helpful.

It proves that it proves. The fact that it is working. This makes any attempts at generalizing null and void. Leaving us with certain bad variant as a specific case.

"AI bad in general" turns into "AI bad while using certain variants".

[Lucky33]

Take this as an statement:

I got a regular Brawler so I could run a regular, balanced combat ship capable of dishing out each type of damage. Gave it a heavy MG, a frag MG & 60 breachers. Jumped into try mode, ran it against a mule, murdered it without taking hull damage, handed it over to my reckless officer and, well... it wont even engage a freighter running an autocannon and nothing else. So I reinstalled OBS, reset the game, stripped out all my mods (again) except for the mod that adds showcase missions (since theres no mission with a brawler, apparently) & the speedup mod (you'll see soon why thats needed). The one thing is that the only equivalent to the heavy vulcan is double flak, but it made no real difference.

And this as an illustration:

[Ishman]

Spoiler

oh yeah, much better

[close]

Wow you got eviscerated by this.

Spoiler

And this as an illustration:

[close]

But also yeah, your initial fleet's fits are actually pretty weak, there's not enough of a critical mass of beamspam for it to be effective (and it's wasting a fury's offensive potential to utilize it like that). Since beams are only dealing soft flux damage, you'll be wanting to bring about 2000 flux/s actual value (this overcomes the majority of non redacted dissipation values).

So if we're facing .7 shields (nerfed apogee tier) with hardened shields, officer cr bonuses and skills like reduced energy damage, (and possibly solar shielding), we can lump it all together for ~.45 efficiency against energy. Which means you'll need ~4400 worth of supporting tactical beams (reduced by any HIL/graviton combos you field). That comes out to around 58 tactical lasers having the range and line of sight to shoot at the same thing.

Which is why monofleets of Beholders can beat almost anything in a fleet composition tournament, but beams are otherwise completely ignored (except to punish dropping shields with the HIL/tachyon).

[Alex]

I have a brief recollection of entering a sim of a ship from the refit screen then hitting autopilot ignores the personality of that ship (treating it as if it were a player commanded ship because its the ship the player is currently flying). Is that still the case?

Hmm, I don't actually remember this. Just double-checked and it seems fine.

There *is* an issue - fixed in-dev, but not in -RC15 - where if you run the game in dev mode, go into Edit Variants, check the "player max skills" box, and click on the portrait to select a personality other than steady, then that choice would be ignored. But it was limited to exactly these circumstances. (To be clear, the order in which you check the max skills box and select the personality doesn't matter.)

"A 1v1 with an enemy that can't really hurt it" - that's practically impossible for it to determine, right?

... Can it not? I've never confirmed this before; a lasher pursuing a fleeing buffalo, can it not tell that it's not chasing, say, another lasher or something? I didn't know that.

I think I phrased this rather poorly. It's aware of these things, but it's hard to make this kind of "hard call" that something is in fact harmless. Skills, current state of armor, specific of weapons (and any scripted effects they might have), and importantly "what the current risk tolerance should be based on the larger context"...

RE: the rest of it; that makes a lot of sense. It genuinely helps to just show footage of what its doing so it can confirmed that the problem is not the game but my expectations

*thumbs up*

Proving the Brawler can do it under a certain set of circumstances wouldn't accomplish anything.
Nobody actually playing the game is bothered about one very specific example or how that could be tweaked to do one particular thing. The original post wasn't about Brawlers but about the AI in general not advancing enough and being too risk-averse. Alex already explained why that happens. I'm not thrilled about it because I prefer to not always lead the charge, but I understand the reasons behind it.

(First, I really appreciate your understanding!)

To clear up a point - I think you can fit ships, and give them reckless officers, so they can lead the charge successfully. It's just that for some builds that won't work, but for some builds it *will*.

(If I could safely improve its ability to armor-tank, though, I'd love to. Again, maybe tying that to "reckless" might be a way to go, since anything general-purpose runs too great a risk of getting ships blown up.)

[intrinsic_parity]

The best 1v1 brawler weapon combo is probably 2x HVD or HVD + mauler, but I almost never use that ship so I can't say for sure. That's what I would use if I had to put one in my fleet though.

I mean, at that point were just saying that any gun with less than 1k range isn't viable then, arent we?

No, I'm saying long range guns work better for an AI brawler in a 1v1... Why would you jump to such extreme generalizations instantly?

To make short range loadouts work, you need:
- aggressive personality (either by officer or fleet doctrine) because this lowers the threshold for when the AI backs off
- sufficient speed/maneuverability to close distance quickly (brawler is ok with maneuvering jets but has mediocre top speed)
- enough effective shield HP so that the AI doesn't feel like it's at a disadvantage and backs off (you can increase shield HP by increasing capacity, or by improving shield efficiency).

If you do these things, it works better and if you don't it works worse, as clearly highlighted by the clips shared in this thread...

[Linnis]

After watching all those brawler videoes. Makes you really appreciate how much better omni shields are on frigates.

[Deshara]

Wow you got eviscerated by this.

not really. all it proves is that the brawler will use the HMG if you grind out exactly those 5 skills, get exactly that 1 hullmod & nothing else. It means that the brawler is literally unplayable for me if I happen to be running a fleet with a burn level of 10 because the 4 OP I put on bumping its burn speed up by 1 ruins the vessel's AI.
I mean, if I sold you an electric car that you then say isn't useable & I sent u a video of it doing donuts in my driveway bc I charged it at a battery station in Austin Texas where I live, I haven't really owned you if you don't live in Austin Texas -- except, I guess, in the sense that I've robbed you blind.
I mean, one of the core mechanics of the game is the ability to customize ships, if you can't customize a ship's hullmods when you put these two guns together bc the only gun out of its 2 that the AI will use is the only that can never break a shield, on an enemy with shields, then the game is breaking one of its central promises.
We weren't saying the dude was wrong that it might work with those exact skills (well I was to antagonize him for telling me what youtube videos to make & upload so he'd do it instead), we were saying it shouldn't

[intrinsic_parity]

all it proves is that the brawler will use the HMG if you grind out exactly those 5 skills, get exactly that 1 hullmod & nothing else. It means that the brawler is literally unplayable for me if I happen to be running a fleet with a burn level of 10 because the 4 OP I put on bumping its burn speed up by 1 ruins the vessel's AI.

You're acting like it just randomly happens to work with specifics skills because the AI is weird, and doesn't with any other combination. But it's not random, it works because the skills specifically improve and enable the ship and AI. I just laid out the factors that play into the AI's effectiveness at using short range weapons. It's not about specific skills or hullmods working with the AI, it's about making the ship better at tanking damage and closing distance quickly. These factors make the ship better at using the weapons independently of the AI. If you piloted the ship personally, having those skills and hullmods would make the fight much easier too.

It's easy to see that most of the changes between your loadout and his very clearly improve the ship in either shield HP or speed which are the key factors in using short ranged weapons effectively, and the aggressive personality increases the chances the ship commits to a fight and takes risks. If you removed some of those factors, the ship would probably work less well, but it wouldn't instantly become useless if you spent 4 OP in a different way... In fact, dropping 4 caps  would probably not decrease effectiveness noticeably at all.

At the end of the day, if a loadout doesn't have the speed to close quickly into weapon range and the shield HP to tank damage while closing, it will not work well, regardless of how good the AI is. You're getting angry at the AI for not winning an uphill battle you constructed yourself. If you just make the battle easier, then the AI does fine.

Also, if you could just throw any combination of weapons and hullmods together and still get a good ship, the entire loadout portion of the game would be pointless. There's no reason to try and improve your loadouts when a random loadout works just as well. The whole point of that aspect of the game is that certain combinations work (a lot) better than others. That's what makes it fun. I think your expectation to be able to put random weapons and hullmods together effectively is unreasonable.

[Lucky33]

Wow you got eviscerated by this.

all it proves is that the brawler will use the HMG if you grind out exactly those 5 skills, get exactly that 1 hullmod & nothing else

That's another lie.

[torbes]

one last thought before I put this thread behind me; I do find it interesting that now that the game has plenty of options to mitigate the loss of low-value ships, from base ship recovery to industrial skills to recovery-guaranteeing hullmods, that the game now suffers from a severe lack of a way to have your ships behave as if their loss is expected and/or acceptable. As-is I would never run a hulk fleet campaign just bc my ships are already piloting themselves so cautiously. I think, if there was a hullmod that guaranteed the ship was recoverable but instead of making it more survivable made it less survivable by making it much more willing to face-tank to turn its remaining armor into free shield to apply pressure, I'd use it to do hulk playthrus


OH it just clicked in my head!! I'm playing a kite slinging extreme anti-armor loadout, waiting for my fleet to Do Work so I can swoop in & assist, while my fleet is waiting for the player to Do Work so it can swoop in & assist. The game expects me to be piloting the strongest ship in my fleet with a generalized enough loadout that I can always lead the flow of battle, & I'm piloting the tiniest ship in my fleet with such a specialized loadout that I am so incapable of leading the flow of battle in a way that it forces me to sit outside of the battle, and the AI just isn't aggressive enough to carry battles without me like that (shy of just fighting enemies too weak to fight back). Makes sense; the game just isn't built to use a kite as my flagship

The first paragraph is a neat idea.

The second is the salient point. More accurately that, as intrinsic alludes to, that if you want your AI to dictate the flow of battle they need to be 1)fast and 2)have enough shield ehp to feel safe doing so. I think it's reasonable to have discussions about how much of 1 and 2 ships need in the current version and if there are ways to enable similar outcomes without redo'ing the AI or causing shocks to the system by making it too reckless.

[Retry]

Can we get some pictures of the ships from the saved linked here?  I'm away from my desktop that has Star Sector so I can't check the builds personally (and I'd probably not have the patience to download the specific modset used anyways)

Based off Hiruma Kai's incomplete description of the fleet in question, the builds used spark very little confidence.

[BigBeans]

Can we get some pictures of the ships from the saved linked here?  I'm away from my desktop that has Star Sector so I can't check the builds personally (and I'd probably not have the patience to download the specific modset used anyways)

Based off Hiruma Kai's incomplete description of the fleet in question, the builds used spark very little confidence.

I've managed to work out that the bulk of his complaint is due to trash ship builds.

AI defo has issues and is very cautious and has issues during station combat but its not that bad. That video he posted with the Brawler was entirely due to that deformed ship armament he made. Even if he had the heavy vulcan from mods that would've been a trash brawler.

[Deshara]

its only a trash ship if the AI doesnt know to get into weapons range, which they used to

[Hiruma Kai]

Can we get some pictures of the ships from the saved linked here?  I'm away from my desktop that has Star Sector so I can't check the builds personally (and I'd probably not have the patience to download the specific modset used anyways)

Based off Hiruma Kai's incomplete description of the fleet in question, the builds used spark very little confidence.

To be fair, sometimes you just can't find the weapons you want.  Especially in a heavily modded game where weapon pools can be very diluted.  Then there's building fleets to do different campaign jobs.  The fleet in the save is clearly not focused on combat, but rather campaign speed, stealth and sensors.  To be honest, it looks like a reasonable fleet for smuggling or stealth missions. Combined with Navigation and Sensors, the ship has a ridiculously small profile, strong sensors, and a base burn speed of 11.  Other player skills are Helmsmanship, Impact Mitigation, Shield Modulation, Missile Specialization, Wolfpack tactics, Energy Weapon mastery, and Field Repairs.  The skills themselves are also modded.

Anyway, I present a break down of the ships as requested for those who don't want to download the mods/save.

The Fury has:
0 Caps, 3 Vents
s-modded Dedicated Targeting Core, s-modded Expanded Missile Racks
Insulated Engine Assembly (lowering profile)
Fuel Additive (15 OP for +1 burn, 20% less fuel consumption, but makes engine more unstable, 25% more maintence)
ECCM Package
Pirate Handiworks (0 OP, 20% less profile)
Converted Hangar (Broadsword wing)
1x Ion Beam, 1x Graviton Beam, 1x Kneecapper MRM (dual regenerating energy/emp missile,base 6 ammo), 2 tactical lasers, 2 mining lasers, 2 PD lasers

In the saved file, the Fury has a cautious officer with Helmsmanship and Strike Commander - but as noted by Deshara, officers were shifted around.  I think the ship does better with an aggressive officer instead of cautious personally.  Reckless would also likely be an improvement.  There's a level 5 Reckless officer, level 3 aggressive, and level 1 aggressive as well, plus the player.

The Moray (phase skimmer destroyer) has:
0 Caps, 0 Vents
s-modded Expanded Missile racks, s-moded Fuel additive (see above)
Insulated Engine Assembly, Pirate Handiworks, Expanded Magazines, Converted Hangar (High Tech Talon Interceptor, 7 base OP - 11 in this hangar, 4 fighters, light machine gun, bright missile each), High Resolution Sensors
2 Heavy Burst Lasers (forward mounted), 1 burst laser (rear mounted), 1 Kneecapper MRM (forward mount)

Brawler (TT) had:
0 Caps, 9 Vents
Comes with insulated engine assembly built in
s-modded fuel additive, s-modded high resolution sensors
Pirate handiworks, Stabilized Shields, Resistant Flux Conduits, Hardened Shields, ECCM Package, Armored Weapon Mounts
2 Salamander MRM, 1 Ion Beam, 1 Graviton Beam

Hedone class yacht (Time splitter frigate) had:
0 Caps, 10 vents
Civilian Grade Hull built in
s-modded fuel additive, s-modded militarized subsystems (note burn speed 10 + 2 = 12 with this setup)
Insulated Engine Assembly, Pirate Handiworks, Reinforced Bulkheads, Escort Package, High resolution sensors
2x Light dual machine guns (forward mounted), 1 IR pulse laser (360 degree turret)

Kite (A) had:
1 cap, 1 vent
Civilian Grade Hull, Militarized Subsystems built in
Pirate Handiworks, Expanded Missile Racks, Insulated Engine Assembly, Escort package, ECCM package, Reinforced bulkheads
2x Hammer torpedoes (so with Missile skill, 12 total torpedoes), 1x Prickler PD Turret

Given the 4 additional civilian ships (Buffalo, Valkyrie, Phaeton, Salvage Rig) are also configured for minimum signature, maximum speed (burn 10), and sensors (high resolution sensors), the insulated engine assemblies on the frigates unfortunately aren't doing much, given the largest 5 ships are used to determine sensor profile.

As I noted before, letting the fleet fight, it looks like it is engaging the enemy to me according to officer AI.  But given it's a speed/stealth build instead of a more combat focused build, it doesn't fight as well as it might, which can certainly cause problems.  The other way to look at it, is the purely AI enemy side (for example the Merchant fleet) has no issue engaging and destroying ships in the fleet.

While I think it's perfectly fine to make suggestions about the tactics the AI might want to consider using (run in with shields down and armor tank), I'm not seeing this unwillingness to engage.  Especially if I hand edit all the officers to all reckless.

Certainly in this case, the suggested behavior for the mission brawler would be non-ideal for the beam Brawler, Hedone or Kite, for example.