Analysing the Thargoid Simulation

[yttrbio]

I think checking every atmospheric worl in thargoid controlled systems, and especially in maelstrom ones, could be usefull, maybe they are doing something, terraforming?
Maybe we could glearn something interesting, about them.

Problem: probably that would require 2 commanders, 1 pilot, and one taking samples, unmanned ship probably will be destroyed very fast.

Not necessarily. I've gone down to a few atmospheric worlds in occupied systems, including the moon around Taranis's planet. If you land and dismiss your ship immediately, it will only take a few shots, if any, when Thargoids show up before your ship leaves. If it's an interceptor, they've been ignoring my SRV and/or pilot, though scouts don't. I'm not sure what causes scouts to appear. The first 10 or so landings I did, interceptors showed up and I had no trouble, only the last few had scouts show up sometimes. but they were also slow landings, so maybe the interceptor calls them when you're too slow to land and then dismiss the ship. I haven't tested thoroughly enough, but it usually worked fine for me. It might just be luck.

That said, I didn't find anything of note on the worlds I visited, and the exobio seemed unremarkable. But I wouldn't know what I was looking for.

Perhaps explained by cubic vs spherical distance metric? The fact that BGS uses cubes rather than Euclidean distance would make me suspect that the war sim does the same, with the same annoying effects on statistics computed on distance.

I'm not in a position to do much number-crunching for a bit, but I can try that when I get back if no one else will.

 

[sathlei]

I wasn’t aware Canonn Research conducted such extensive research on the Maelstroms and the preceding Unknown Interstellar Anomalies aka Stargoids.

The resulting visualization is amazing and may lead to interesting conclusions / hypotheses.

source: Canonn Research

Let me add one element to this puzzle. I’ve modified my simplistic visualization by adding Power headquarters as well as drawing a surface supported by Maelstroms’ positions.

Maelstroms create a plane between the Human space and probable Thargoid core systems. This plane extends “below” all Human HQs, except Gateway, and is slightly concave, hinting at possible all embracing finale...

Unless...

 

[Ian Doncaster]

This plane extends “below” all Human HQs, except Gateway, and is slightly concave

Though, this would also be true of almost any eight reasonably-spaced points on the edge of the same hemisphere of the bubble, and any set of scattered systems deeper within the bubble (engineers, ammonia worlds, systems named after greek mythology, etc.)

I wasn’t aware Canonn Research conducted such extensive research on the Maelstroms and the preceding Unknown Interstellar Anomalies aka Stargoids.

Also a long forum thread here - https://forums.frontier.co.uk/threa...-0-8ly-per-minute-thargoid-mothership.607506/ - trying to work out the destinations more precisely than "the bubble". There's various tables in there of how close the Stargoid trajectories would pass to various "interesting" systems, if they (had) continued into the bubble {summary: not very close}.

 

[Ralph Vargr]

It's a slow-motion envelopment.

Remember Space Invaders?

 

[]M[]

Well, it seems the basic parameters/algorithm for the wall formation were/was: head roughly for Sol (these insects have grown wings and have been spreading out from there for about one thousand of their home world’s stellar orbits - possible problem due to sudden technological discovery/advancement detected recently), stop at appropriate systems (with Ammonia worlds and related favourable conditions) along the edge of their space (contain this nuisance prevent further advance of hostile species, just in case it gets really irritating…)

…whether this arc/dome grows to form more of a partial sphere, or continues to advance, focussing in on Sol should logically (i mean in terms of game narrative-logic ofc ;p) depend on human (ie: player base) response… fairly sure it’s all on auto/AI-mode for the holiday season, spreading out locally from around Maelstrom points according to the rules of the war system, to keep players entertained while at least some dev staff get to take a break - though what happens in the new year when everyone’s feeling better rested and back at work, remains to be seen…

Happy Holidays, all

 

[yttrbio]

It's worth being careful about confirmation bias. If you assume the target is Sol, then you can find patterns that suggest Sol is the target. But the point Ian Doncaster is making is that you can come to the same conclusions using many other points. Sol itself is a quite poor fit for the target, and way out of position considering the symmetry when you consider alternative systems. The UIA paths don't converge at all in the y direction, and they loosely converge (except for Raijin) when you ignore the y direction, but not on Sol. The midpoint between Sol and Achenar fits better, but even then you have plenty of alternative locations that fit better still.

 

[sathlei]

Though, this would also be true of almost any eight reasonably-spaced points on the edge of the same hemisphere of the bubble, and any set of scattered systems deeper within the bubble (engineers, ammonia worlds, systems named after greek mythology, etc.)

Right, but the key phrase here seems to be "reasonably spaced".

Also a long forum thread here - https://forums.frontier.co.uk/threa...-0-8ly-per-minute-thargoid-mothership.607506/ - trying to work out the destinations more precisely than "the bubble". There's various tables in there of how close the Stargoid trajectories would pass to various "interesting" systems, if they (had) continued into the bubble {summary: not very close}.

Thanks a lot for reminding it.

 

[TobiasFuzz]

I was looking around in HIP 38235 (invasion state). Non of the ports offer any evac mission. No passengers, no wounded. Searched in attacked and not yet attacked stations and surface ports. Is this a bug or intentional? Looking at INARA it seems that it is not an issue for me personaly, as there are no evacs reported at all.

Could it be that there is no rescue ship assigned to the systems, so there are none of this missions generated?

 

[yttrbio]

Another exciting tick. Only 51 alerts (that we know of, see below), so it's not growing exponentially at the moment. Thor, Raijin, and Hadad have repeated the behavior of Oya, Cocijo, and Leigong in not sending out any alerts their second week. Since the set of control systems is the same on the first two weeks, it makes sense that the landscape hasn't changed sufficiently for there to be any reason to put out new alerts on week 2, but more investigation is needed to nail that down.

Unfortunately, Leigong had another exciting week of invisible expansion. The list of invisible control systems is:

Arietis Sector ZE-R b4-3
Hyades Sector KN-K b8-2
Hyades Sector KN-K b8-0
Arietis Sector NX-U c2-20
Hyades Sector LX-U d2-95
Arietis Sector BQ-P b5-0
Arietis Sector XJ-R b4-0
Arietis Sector XJ-R b4-3
Arietis Sector BQ-P b5-2
Arietis Sector NX-U c2-19
Arietis Sector XJ-R b4-2
Hyades Sector KN-K b8-1 (new)
Hyades Sector KN-K b8-3 (new)
Arietis Sector XJ-R b4-1 (new)
Arietis Sector KM-W d1-93 (new)
Ceti Sector XJ-A c18 (new)
Ceti Sector ZU-Y b1 (new)

There could be any number of invisible alerts around Leigong. So far, the invisible systems have not been uniformly distributed around Leigong. They are generally in the +x direction from it.

 

[MacHurto]

Unfortunately, Leigong had another exciting week of invisible expansion. The list of invisible control systems is:

I think I missed this. What do you mean by invisible expansion? Like those systems do not show in the galaxy map but they are under Tharg control?

 

[yttrbio]

Exactly. Details here

 

[Ian Doncaster]

Looking at the recent Alert systems, it seems that the ones ~30 LY from their Maelstrom are taking a bit less effort than the ones ~20LY out. Only four data points since Frontier's rebalance announcement so it's hard to say if it's a real pattern or just coincidence.

No obvious patterns in difficulty for Invasions, but pretty much every Invasion so far can (very) loosely be described as ~20LY out; there's not quite the range of options.

It'll be interesting to see if the pattern continues for Alerts and appears for Invasions once there's more of them ... or whether it disappears for Alerts once more points are considered.

 

[yttrbio]

I've run another experiment on the difficulty of control systems this week, and, at least for the outlying ones, they may not be as bad as it was in HR 1737 last week.

Basically, I was able to almost single-handedly add 3 total progress pips across 2 systems (Hyades Sector GW-M b7-5 and Hyades Sector PN-T c3-8), with a combined kill total of 371 scouts and 15 cyclops, which, compared to AXI throwing themselves at HR 1737 for 2 days and only getting 4 pips, suggests that hitting outlying systems is MUCH easier.

We don't fully know the specific causes of difficulty from this experiment. It could be distance to maelstrom, number of control systems nearby, or whether the system is populated or not. But it seems clear that not every system is HR 1737.

Details here:

Thargoid Control System Difficulty Experiment

Summary: In trying to retake Thargoid-controlled systems, outlying uninhabited systems have proven to be easier to make progress on than inhabited systems right next to the maelstrom. A single commander, with very little help, has managed to generate 3 pips of progress across 2 different systems...

docs.google.com

 

[Ian Doncaster]

The new Alert number stays in the 50-80 range this week, though it's still not clear what causes the variation in it. Still, it's a constant enough rate that it seems very likely that the number of Maelstroms is what determines it.

So ... if the Thargoids want to take over the bubble quicker than the next 20 years, favourable plot events aside, they might need to get some Maelstroms down. Given where the initial eight were placed, systems with Ammonia Worlds seem the likely targets - with uninhabited ones perhaps preferred.

At the moment, this is what they have

Maelstrom​	System​	Distance​	Populated​	State?​
Oya​	HIP 15415​	9.10​	Y​	Control​
Hadad​	Col 285 Sector US-Z b14-5​	9.98​	N​	Control​
Indra​	HIP 20350​	12.72​	Y​	Control​
Thor​	HIP 19894​	16.21​	Y​	Invasion (1 week left)​
Leigong​	HIP 9137​	22.06​	N​	Control​
Taranis​	Hyades Sector QN-T c3-11​	22.99​	N​	Alert​

They're all pretty close to the existing Maelstrom, of course - the first three were automatically captured on arrival. The Leigong one was one of its first Alert wave, and has been sitting at Control for four weeks now without apparently doing anything.

Still, perhaps something to keep an eye on in future.

 

[yttrbio]

There are also ammonia worlds nearby that they've left alone. Wangakwan (Taranis), Col 285 Sector EA-Q c5-6 (Thor), HIP 21125 (Thor), Pegasi Sector HH-U b3-2 (Raijin). All are within 10 ly of a control system last week (and that 10 ly range continues to hold up).

 

[yttrbio]

A couple general observations about the new week:

1. Leigong does not have any new invisible control systems, but I was able to confirm one invisible alert system this week (Hyades Sector MY-I b9-0) and there may be more.

2. All new alerts are still within 10 ly of a control system

3. We had the first repeat alert this week. HIP 20485 (Indra) was alerted 4 weeks ago, and that alert was beaten back. It is now alerted again. This suggests a lockout period of at most 4 weeks. It could be lower, but if it is, then whatever reason it had to alert HIP 20485 had to disappear for 4 weeks and then reappear. Unless it's all RNG, in which case we're wasting our time trying to figure it out. The system to watch next week is Vukurbeh (Taranis), which was alerted 3 weeks ago and that alert was also beaten back. If the pattern holds, it will be alerted next week. If not, we know there are other factors at play

4. We also had an alert in Muruidooges. This system was one of the initially-invaded systems around Hadad, and the invasion was stopped the first week, putting it in recovery for 1 week. It is now alerted, 2 weeks after the recovery. Note, though, that it was never alerted in the first place, so it's not unthinkable that the lockout programming was only for repeated alerts. If this is the case, then this is the first week that it wasn't in a thargoid state the previous week, which may be enough for an alert to trigger. The equivalently-positioned system for next week is HIP 20527 around Indra, which had its one-week recovery last week, and would alert next week if the pattern holds.

5. The maelstrom alert counts follow an accordian-like pattern, where large numbers of alerts go out every other week. Taranis and Indra are exceptions, but it's worth noting that the process alert->control takes 2 weeks, but alert->invasion->control takes a variable amount of time. If the alerts are based on the positioning or count of the control system, this accordian-like expansion is exactly what we would expect, and it would be eventually averaged out as invasions complete.

 

[Disemboweled Ego]

1. Leigong does not have any new invisible control systems, but I was able to confirm one invisible alert system this week (Hyades Sector MY-I b9-0) and there may be more.

I haven't checked the Leigong area fully yet, but Ceti Sector XJ-A c18 was in invisible control or alert too (not sure how to tell the difference).

 

[yttrbio]

Ceti Sector XJ-A c18 is invisible control. You can tell by the "Thargoids" control notification in the lower right in control systems, and its absence in alert systems (in addition to interdictions and the signals in the system that differ). The control list from here should still be accurate.

 

[Ian Doncaster]

Interesting change on the invasion lengths - all current invasions appear to have a length of station_count + 1 weeks.

Maybe it's only the initial Maelstrom invasions which get the more aggressive triangular pattern, or maybe it's just changed?

 

[Ian Doncaster]

Taking a bit more of a look at the distance-difficulty relationship now that there are more data points, and some more distant invasions. X-axis on the graphs is distance in LY to the nearest Maelstrom, Y-axis is the number of FSD jumps recorded on EDDN between the start of the week and completion (or, if incomplete - including most of this week, the total scaled up to a projected completion, with only systems above 20% used to extrapolate to reduce errors)

Alerts first - there are only 5 data points since Frontier's rebalance announcement and they actually form a pretty good line

Too good, in my opinion - FSDJump data can be incredibly noisy from system to system - I suspect this wouldn't look anywhere near as clean with more data ... but there does seem to be a definite difference between the ones at around 21 LY and the ones at around 30 LY.

For Invasions there's a lot more data

and the correlation is barely present. A couple of systems have been contested in invasion both this week and last, and this week's totals are about half of last week's for the same system, to give an idea of the sort of random noise that this difficulty measure involves [1].

However, while the "16-22 LY" zone is basically just an uncorrelated blob, it is notable that the "26-30 LY" zone is all at the lower end of that range (the mean difficulty is around 650) and it includes the second-lowest measure.


These patterns may of course disappear entirely once we get more data, but it does seem plausible for now that there's some sort of difficulty/distance relationship.


[1] Alternate theory: the systems are half as difficult this week as last. Since total victories and progress looks like it's going to be similar, I don't think this is the case.