Thargoid invasion - Next target systems?

[MVBanks]

Seems something weird is going on at Roggeveen Installation in Ebisu. I've dropped in at the planetary port a few times, but nothing seems to be happening there. It's on fire, but there's no CZ and no ships flying around it.

I also visited Roggeveen, it was bugged, no CZ. It was often before 14.02, with almost all CZs in open, but now, I don't expected same bugs.

 

[Kayvan]

I suspect that both the Control and the Invasion requirements were increased on Monday–Tuesday.

Any announcements of that? Why would they do that? We are still losing the war about 1:9 (systems saved vs lost), what reason would they have to make it even harder.
Personally I suspect that some players start feeling combat fatigue. Not yet enough to show a steep decline in participation, but mb noticeable in that the progress bars move slower. I mean, I notice it myself. We've played along over 2 months now, but ultimately it's always the same gameplay loop. The new modules (stabi and scanner) make the content more accessible to previously uninvolved players, but that machine may yet have to spin up. Veterans may experiment with the new toys, but this may actually result in reduced contribution for a while bc ships have to be built or refitted, and builds tested, naturally with varying degrees of success.

So in short, I think there are other explanations than "FD reverted the adjustments made last month".

--

Note about Pathamon: I fought there yesterday and tonight and was wondering why combat felt so hard. Now I realized that both planets are very high gravity, like around 1.7G both. Even a Chief's lateral thrusters struggle to balance that out. When you're close to the ground you have to take care not to be dragged into the dirt, and even at higher altitides the controls are incredibly wobbly, making it harder to aim (unless the reason for that is that my joystick needs to be recalibrated).
Oh yes AND - probably due to the patch changes - the ceptors fly much lower above ground, which aggravates the aforementioned problems.
All in all this session was relatively frustrating; tomorrow I'll find me a different system to fight in.

 

[Aleks Zuno]

The cycle is starting to take shape; a few more previous top targets return, joining the fight anew are HIPs 29596, 20065 and 21261, and a bit of Alert activity in Senocidi!

Top targets at 01:45 3rd February 3309:
Putas Invasion 22% — Hadad 23 Ly, 2 ports, 861 Ls outpost attack
Pathamon Invasion 18% — Leigong 22 Ly, 3 ports, 193 Ls planet + 193 Ls outpost attack, 104 Ls planet + 2695 Ls outpost damage
Ebisu Invasion 8% — Taranis 21 Ly, 5 ports, 105 Ls planet + 105 Ls outpost attack, 63 Ls outpost + 1797 Ls planet damage
Huile Invasion 8% — Oya 30 Ly, 1 port, 1543 Ls starport attack, 3853 Ls planet damage
HIP 26688 Invasion 6% — Taranis 21 Ly, 1 port, 230 Ls planet attack

Any announcements of that?

Of course not—but I hope by now one should know that the presence or absence of an announcement has very little bearing on reality. If I may say so, it amazes me a little to hear that question following the same update which without announcement just changed Combat bond amounts, and despite announcement does not appear to have included Conflict zones in Control systems (or if they exist, I would quite like to know why Tougeir and HIP 24329 do not qualify).

Please note that watching a lot of numbers move at one rate then start moving abruptly at a different rate following an update does not leave much room for applying reasonable doubt; I say that the default position should be that the update caused a change, and that I am the one to demand an explanation how it could be that so many pilots just ceased activity in the middle of a cycle at three systems otherwise projected to complete quite easily. The answer "they were all fatigued" does not meet that credibly.

Actually—

So in short, I think there are other explanations than "FD reverted the adjustments made last month".

If I take your example wording there and append "accidentally", that sounds ever so much more likely to have been the case. It fits what we see numerically, and an accident is unannounced by definition. Note also that I am also not the only number-watcher calling it, as I noted in my feedback post about restricting such changes during a cycle.

 

[Ned Flandalorian]

 despite announcement does not appear to have included Conflict zones in Control systems (or if they exist, I would quite like to know why Tougeir and HIP 24329 do not qualify).

I'm not able to check a map, are they "warfront" systems on the edge of the thargoid bubbles? Fdev said it would only be the outlying systems getting CZs and they are definitely present in some others.

(They also said salvaging rescue items and tissue samples would count towards progress but those actions aren't listed as possibilities for uninhabited alerts. So those systems still remain as indefensible as ever, meaning we are fighting a war where the enemy can take territory at will...)

 

[[VR] Mad Mike]

I was in pathamon last night (making an embarrassing show both times I tried to land thanks to new controller set up). yet again it was.bugged out . the whole session it had the 2 hydra bars full but they never showed. a few players would not leave the instance as they were happy with it. I don't blame them but it's frustrating that FD have still not fixed the instance issues.
I have still yet to see any ship beaten beyond a medusa due to this bug and insure as hell can't fight them alone so have to run when they turn up in an empty instance

 

[Kayvan]

I was in pathamon last night (making an embarrassing show both times I tried to land thanks to new controller set up)

Yeah don't blame yourself, that would be mostly the high gravity on those planets. It drags you down real hard, making it much more difficult to line up the pad and touch down instead of crashing full smack in the ground.

--

@Aleks Zuno : hmm, I see. Well then let's hope FD fix that soon. :/

 

[Ned Flandalorian]

Lodged an issue to try and get control systems working again to get this war back on track: https://issues.frontierstore.net/issue-detail/57612

 

[Aleks Zuno]

Obamumbo returns, and joining are Orong and Hyades Sector AV-O b6-5.

Top targets at 09:10 3rd February 3309:
Putas Invasion 36% — Hadad 23 Ly, 2 ports, 861 Ls outpost attack
Pathamon Invasion 26% — Leigong 22 Ly, 3 ports, 193 Ls planet + 193 Ls outpost attack, 104 Ls planet + 2695 Ls outpost damage
Ebisu Invasion 12% — Taranis 21 Ly, 5 ports, 105 Ls planet + 105 Ls outpost attack, 63 Ls outpost + 1797 Ls planet damage
Huile Invasion 12% — Oya 30 Ly, 1 port, 1543 Ls starport attack, 3853 Ls planet damage
HIP 20491 Invasion 8% — Indra 23 Ly, 1 port, 1713 Ls planet attack, 3194 Ls planet damage
HIP 26688 Invasion 8% — Taranis 21 Ly, 1 port, 230 Ls planet attack

I'm not able to check a map, are they "warfront" systems on the edge of the thargoid bubbles? Fdev said it would only be the outlying systems getting CZs and they are definitely present in some others.

HIP 24329 and Tougeir seem to be as outlying as any could be, on the edge of M. Oya, next to the defended Jeng:

I could not see any Conflict zones in those systems; either there are some and my filter selection took matters upon itself to decide to change, or there are none and we have a new mystery to resolve, that of what defines a Control system bordering Human space.

 

[Ian Doncaster]

I could not see any Conflict zones in those systems; either there are some and my filter selection took matters upon itself to decide to change, or there are none and we have a new mystery to resolve, that of what defines a Control system bordering Human space.

Jeng is at 23.04 LY from Oya, Tougeir at 16.22 LY, HIP 24329 at 18.93 LY ... but Oya's reach extends just over 30 LY total.

Could just be a "fraction of radius" calculation, in which case those two are quite some way "inside" the sphere, but HIP 115777 or Cephei Sector ZE-A c12 which are around 30 LY out should get them.

It will be important to observe progress this evening before judging, though at present I suspect that both the Control and the Invasion requirements were increased on Monday–Tuesday.

Invasion doesn't seem to have shifted massively out of line with what it was before for the systems receiving heavy support. Control, yes, definitely.

Early days yet and that might change as we get more data, but if Invasion difficulty did change it was a much more subtle change than Control got.

 

[Aleks Zuno]

Jeng is at 23.04 LY from Oya, Tougeir at 16.22 LY, HIP 24329 at 18.93 LY ... but Oya's reach extends just over 30 LY total.
Could just be a "fraction of radius" calculation, in which case those two are quite some way "inside" the sphere, but HIP 115777 or Cephei Sector ZE-A c12 which are around 30 LY out should get them.

It could indeed! It would be a quite bizarre definition though, or at least a simple definition with a bizarre result, and if that appears to be the case then it is something I will be mentioning in the official feedback thread.

Summary:

• By that (quite possible) measure, we would have to travel around to the opposite side of a sphere to evict control from some empty systems in the wrong direction before we get any signs of resistance on our side of it.
• Given the preference for acting in Invasion systems, populated by definition and therefore also naturally on our side of the sphere, that antipodal side will become the perpetual "front".
• A better definition, to be suggested if the above appears to be the case, is that Control systems are at the war-front by default unless all nearby systems are Control or Invasion.
• The definition of "nearby" may need to adapt such that some minimum number of systems are considered to be nearby. Alternatively, equate it with the Alert range limit.

 

[Ian Doncaster]

Simple with bizarre result I think is most likely - but I suspect giving any formal definition of "on the edge of" a region defined purely as a set of points is tricky and will always give some sort of bizarre result, especially since outside of 20 LY from any Maelstrom there's a lot of holes in the expansion in general.

I can't think of a formal definition of where the edge of a Maelstrom's region is that is both reasonably cheap to calculate and doesn't give odd results for things like a few stray systems off on a line (like Leigong has a few of) or where there's a lot of holes in the sphere (like every Maelstrom's 20-30LY zone right now)

"How close is it to the bounding sphere" does give odd results, but at least it would give predictably odd results. Anything more complex than that might give more intuitive results in some circumstances but it'd also be much harder to analyse and predict which exact systems would be "front" without knowing the exact details of Frontier's algorithm.

Given the preference for acting in Invasion systems, populated by definition and therefore also naturally on our side of the sphere, that antipodal side will become the perpetual "front".

Theoretically possible yes, but with the Thargoids having close to a 90% success rate for captures, the chances of them ending up significantly non-spherical in practice feels very low.

Oya has had one of the more consistently successful defences and the average position of all its Control systems is closer to Sol than Oya itself is (135.1 LY rather than Oya's 136.2 LY ... but still closer!)
Indra has also seen a lot of successful defences, and its average Control system is 169.6 LY from Sol while Indra itself is 174.5 LY away - that's surprisingly far off-centre, and not in the good direction.

I'll have to check the others when I get more time.

 

[Ned Flandalorian]

"How close is it to the bounding sphere" does give odd results, but at least it would give predictably odd results. Anything more complex than that might give more intuitive results in some circumstances but it'd also be much harder to analyse and predict which exact systems would be "front" without knowing the exact details of Frontier's algorithm.

Perhaps any control system that hasn't yet thrown an alert would be considered a front.

 

[Ian Doncaster]

Perhaps any control system that hasn't yet thrown an alert would be considered a front.

While it's only theoretical, if we did start winning and pushing them back in a region, that would leave the next systems in that direction as "still not the front", though. They seem to be able to generate in their first week of existence, too, so it might result in very few systems total being considered. It'd also have the side-effect that if you abandoned a system deep within Thargoid territory as becoming too costly to defend, the resulting control system would be unable to generate anything and so would be the front.

The other way round - control systems which generated an alert this cycle are the front - doesn't have that problem but I think still ends up very unpredictable about exactly where the front is, especially since Alerts/cycle seems to be roughly constant but the number of "edge" control systems is increasing rapidly.



Other ideas I came up with and abandoned:
1) Refine the sphere to the smallest possible convex polyhedron. Doesn't cope at all well with Leigong-style spikes (neither does a basic sphere, of course)
2) Allow some concavity to the polyhedron. Needs some extremely precise definitions and constraints to stop it ending up as a 100-spiked-star exactly containing each system and a narrow tube to the Maelstrom, or a "travelling salesman" tube route between the controlled systems, either way defining every system as "front", and I can't think of a way to word those constraints that's both mathematically precise enough to be programmable and sufficiently flexible to stop a thin spike de-fronting everything around its base
3) Various things involving assessing whether there's a "Thargoid side" and a "non-Thargoid" side to the system, which again don't deal well with concave areas.
4) A more local definition: "a system is on the front if >60% of the other control systems within 10 LY are closer to the Maelstrom than it is" or similar ones would probably work in most cases but doesn't work for the intermediate systems on a peninsula, or for the systems on the edge of a successful inverse-spike recapture attempt. Once the intuitive surface of the Thargoid region becomes locally concave, I think it's very difficult to satisfactorily say precisely where it is.

(Equally, while the Thargoid regions remain roughly spherical, a bounding sphere approach gives roughly the right answers anyway and is much simpler)

 

[YoEgo]

The strategy is still bad.

The maximum expansion of the thargoids occurs when more systems are on alert. Liquidate the alerts, you end their expansion. Even FD said so.

I already said it before, but now you can check it. DCoH.

An invaded system is more difficult to recover, and EDO settlements are abandoned, so numerous resources are lost. As if that were not enough, then the system goes into recovery, and it takes something like a month to be operational again.

There are many reasons to shut down the alerts, and let the invasions run their course to our regret, until we control the situation and go on the attack. Right now, we are following the trail of destruction they are leaving, with isolated and recovering systems.

 

[Aleks Zuno]

Great design ideas! Definitely if I was designing the "front", I would want it to be independent of where the Maelstrom is:

• It should be possible to make a push and surround a Maelstrom system with Human systems, leaving a disjoint region which has contested surface systems and protected interior systems.
• More generally, it should be possible to carve arbitrary planes through that space to achieve multiple fronts.

Perhaps during the Alert selection process, define a Control system to be protected if a check to discover whether it could contribute an Alert finds that all of its candidate systems were already involved in the war, or exposed if at least one Alert candidate was possible (regardless of whether said candidate was already identified by another Control system, and of whether it actually became an Alert system).

Speaking of Alert systems, it just occurred to me that a hypothesis for the real definition could just be that the Control system lies within some range of an Alert system. Despite being reasonably our "front", there were no Alert systems in the image I posted earlier, so that could be why. I will investigate this a bit later today!

The maximum expansion of the thargoids occurs when more systems are on alert. Liquidate the alerts, you end their expansion. Even FD said so.

Ongoing/unresolved accidents permitting, would you care to describe how repelling an Alert in an empty system is easier than evicting Control? Also for populated systems, note that you are welcome to make a thread calling for all combat pilots to stop that and all run cargo delivery instead.

There are many reasons to shut down the alerts, and let the invasions run their course to our regret, until we control the situation and go on the attack.

Why wait for an Alert equilibrium? "Expansion" equals Alerts minus completions (of any type), therefore one completion is as good as any other as far as pure totals are concerned. Ironically:

• Before we had any idea how each type of action fared, in my first post here I imagined combat pilots handling Control systems and trade pilots handling Alert systems, perhaps combining the respective surplus at an Invasion system. It does not quite work that way though, for Invasion systems can vary in preferred activity over their duration, depending on which ports are attacked or damaged.
• We were already shifting combat efforts over to enact the latter half of your prescribed strategy, before the Control requirement was increased prohibitively a few days ago.

Please note that you will also need to command trade pilots to become combat pilots for that to work. I will go with not commanding anyone to do anything in particular, watching for and listing evidence of other Commanders attempting to defend particular systems (potentially for reasons beyond the present system state), and helping there if I can.

 

[Ned Flandalorian]

In theory stopping alerts is a good strategy but in practice, uninhabited alerts were already impossible to defend. Stopping inhabited alerts is possible, but that puts the system directly back to regular operation and possible to be targeted again (instead of making use of the invasion period/recovery time to clean out the systems that were the source of the invasion in the first place).

 

[Ian Doncaster]

The maximum expansion of the thargoids occurs when more systems are on alert. Liquidate the alerts, you end their expansion.

It's certainly true that if it was possible for human pilots to deliver enough coordinated effort to defeat ~80 systems a week, then the Thargoid expansion would grind to an immediate halt.

However, the type of system doesn't matter for that - if they place 80 new Alerts and we instead take back 81 of their existing Control systems, then at the end of the week they still have one fewer system than they started it with. The difference is which systems they have and whether it's "better" for them to be constantly fighting over the same ground or if attacked systems then get a breather before the next wave.

As it is, the Thargoids place 80 Alerts and even if you could convince everyone to fight those first we'd defeat about 10 of them, leaving them with 70 still going which would become Control systems uncontested (either because they're inhabited or because the resulting invasion would be ignored). That's basically the same rate their expansion is going at on a "don't bother with the Alerts" strategy.

Even FD said so.

Their statement could also be interpreted as encouragement to pick off their outer control systems so that they can't reach out further with Alerts on them - that's a week more proactive than just taking out the Alerts when they arise, and could keep them away from placing Alerts in an entire region of the bubble for weeks.

(Also - see Powerplay or Political BGS - I wouldn't rely on Frontier knowing what the optimal strategy is)

An invaded system is more difficult to recover, and EDO settlements are abandoned, so numerous resources are lost. As if that were not enough, then the system goes into recovery, and it takes something like a month to be operational again.

That's certainly true, and any system where losing its stations for even a week or two would be unpleasant should be defended at the Alert stage. But that doesn't apply to the vast majority of systems - if some 100k population refinery isn't producing Indium for a couple of months while it recovers, who's going to notice?

Sure, if it's Meene, or Fujin, or some other system with one-of-a-kind facilities, I don't think it'd be difficult to recruit people for hitting that Alert as a top priority.

 

[YoEgo]

I am referring to the systems on alert with human presence, which in a week become invaded and we lose those resources that I mention, the empty ones almost nothing can be done today: there are hardly any signals, and those that do exist do not go up almost at all.

Also FD said that empty dominated systems are easier... but that doesn't matter if you need around 1000 people to defend a system with human presence, that you need only 500, or 300, or 100 for an empty system... Why would we dedicate so many people to an empty system, unless we knew that from there they can threaten several key systems for humans?

We must understand that the thargoid expansion is exponential, while our efforts are linear, they dominate more territories every day, and we have the same forces to avoid it. And it is clear, we are not going to achieve it, and even less with the current requirements of people per system. That is why it is so important to deal with alerts in human space: they are easier and do not even require a fight to win, only freight transport, but the system, if it is won, remains fully operational.

And there is also the matter of the recovered systems, and even the systems that we saved in the alerts... How long will it take before they are threatened again? Those efforts may be doomed in the future anyway if we fail to clear the surrounding systems of thargoid presence, empty or not. And currently I see it as impossible due to the requirements of people.

It also doesn't help that we don't have a clear picture of what the thargoid expansion front is, we can't figure out how we can clearly stop them, which systems will have to be sacrificed, and which will have to be saved.

Anyway, and according to what FD hinted at, thargoid expansion has a limit, and that as they move away from the Maelstroms, they "weaken", whatever that means. The most obvious thing is that it is easier to move the bar, but who knows, coming from FD... maybe there will come a time when fewer systems threaten each day, until it stops, because they have a fixed limit of systems that they can control.

The only clear thing: they plan to leave the thargoids there for a good while... and I don't know if they have planned that this war can even be won.

 

[Ian Doncaster]

We must understand that the thargoid expansion is exponential

Is it? Since all eight maelstroms arrived, their growth per week has been roughly constant.

I expected some exponential component, but if there is one it's so far too weak to tell apart from noise.

 

[Aleks Zuno]

Putas and Pathamon are doing quite well, and we have a decent second band forming below them, with a present projection of around eight systems this cycle. Awara and HIP 25679 return, and I see fresh Alert activity in Minawara!

Major thanks to everyone reporting the inability to affect Control systems, which has now been noted, and especially to @Ned Flandalorian for the extensive testing and the associated Issue Tracker entry.

Top targets at 19:15 3rd February 3309:
Putas Invasion 50% — Hadad 23 Ly, 2 ports, 861 Ls outpost attack
Pathamon Invasion 38% — Leigong 22 Ly, 3 ports, 193 Ls planet + 193 Ls outpost attack, 104 Ls planet + 2695 Ls outpost damage
Ebisu Invasion 18% — Taranis 21 Ly, 5 ports, 105 Ls planet + 105 Ls outpost attack, 63 Ls outpost + 1797 Ls planet damage
Aowicha Invasion 14% — Oya 23 Ly, 1 port, 4350 Ls starport attack, 1765 Ls planet damage
HIP 26688 Invasion 14% — Taranis 21 Ly, 1 port, 230 Ls planet attack
Huile Invasion 14% — Oya 30 Ly, 1 port, 1543 Ls starport attack, 3853 Ls planet damage
Desurinbin Invasion 10% — Cocijo 25 Ly, 5 ports, 2844 Ls planet + 456k Ls planet attack, 455k Ls planet + 456k Ls planet damage
HIP 20491 Invasion 10% — Indra 23 Ly, 1 port, 1713 Ls planet attack, 3194 Ls planet damage