Powerplay Faction: Aisling Duval What is the long term goal?

Basically?

What are we trying to do? What is the aim?

And is this not a case of the bigger you get the worse it gets?
 
Which is the basic issue.

No one in charge, no one direction and even if there were, many would still just be doing their own thing.

I still believe that there are many, many high paying, non contested systems, very close to Cubeo and within the Aisling bubble and if we work at bringing these on-board, we should have a much higher CC to play with. At the same time, PP is not working, as well as it could be.
 
Last edited:
Leverage Player Groups

There seem to be a number of player groups that are all pledged to Aisling (Prismatic Imperium, etc). Do they have a common goal? That seems like the best way to focus player actions. If their representatives posted their goals for the week, then other players outside those groups who also want to see this power progress could help as well.
 
The long-term goal is for other powers to manage to become as successful as Mahon.

Just like every empire strives to become Rome.
 
There seem to be a number of player groups that are all pledged to Aisling (Prismatic Imperium, etc). Do they have a common goal? That seems like the best way to focus player actions. If their representatives posted their goals for the week, then other players outside those groups who also want to see this power progress could help as well.

That's what the stickied thread "Powerplay Objective for Week xx" is about.
 
Our long term goal is the same as any other power; to make our space a better place. The means to achieve it are mainly twofold:

- Improving our CC-balance by expanding into good systems and losing bad systems that are dragging us down. What makes a good or a bad system is mainly up to three factors:

  1. The income or profit of said system; obviously it has to give us more than it would cost us. However there is an added overheads cost of 62.1CC that you have to take into account. Thus the true cost of a system is its Upkeep+62.1. That means we're looking for systems with a Potential Value of more than 62.1CC (or Base Income > Upkeep+62.1), if we want to make any profit - which we do ;). If none can be find, we can start looking into systems with a Base Income of more than 62.1CC, as those can still make a profit when fortified - since a fortified system has an Upkeep of 0, however this doesn't affect the 62.1CC of overheads.
  2. The distance between said system and other controlled systems; as if another control system (CS) is less than 30Ly away, some exploited systems (ES) might be shared with this other CS (the ES would then give its income to only one CS) or even contested if the CS belongs to another power (the ES would not give its income to either). You need to be wary of the numbers displayed on the galaxy map, as this Potential Income doesn't take into account the shared ES.
  3. The distance between said system and Cubeo; indeed after 95Ly it becomes easier to undermine a system than to fortify it. For example at around 125Ly, the undermining trigger is only half of the fortification one. Which means we have to be careful when considering expansions that are really far away.

- We also want our space to eventually be filled by what we call CCC governments, or Communism, Cooperative and Confederacy. This work is called bgs or bgsm, for background simulation manipulation.
Since Aisling Duval is "strong" against these CCC type of governments, if they represent the majority of a CS and its ES population then the cost to fortify said CS is greatly reduced. With fortifications being a weekly endeavour, it quickly adds up to a lot.
On a similar note, we want our worst systems, such as Panganau, to have a majority of Feudal, Prison Colony or Theocracy; governments Aisling is "weak" against. This would increase the fortification cost of those systems, making them less likely to be fortified and thus more easily lost.

All this (and more!) is discussed on the Slack, which I strongly suggest you join if you are interested in helping the Princess.
 
Last edited:
The goal is quite simple: Keeping Aisling in the top three (and Arissa out of it) long enough until the devs are forced to accept it and construct a lore event that makes Aisling Emperor. To that end we have to, obviously, get into the top 3 and stay there, which is by geting a whole lot of profitable systems and alot of CC in the process.
 
Last edited:
Our long term goal is the same as any other power; to make our space a better place. The means to achieve it are mainly twofold:

- Improving our CC-balance by expanding into good systems and losing bad systems that are dragging us down. What makes a good or a bad system is mainly up to three factors:

  1. The income or profit of said system; obviously it has to give us more than it would cost us. However there is an added overheads cost of 62.1CC that you have to take into account. Thus the true cost of a system is its Upkeep+62.1. That means we're looking for systems with a Potential Value of more than 62.1CC (or Base Income > Upkeep+62.1), if we want to make any profit - which we do ;). If none can be find, we can start looking into systems with a Base Income of more than 62.1CC, as those can still make a profit when fortified - since a fortified system has an Upkeep of 0, however this doesn't affect the 62.1CC of overheads.
  2. The distance between said system and other controlled systems; as if another control system (CS) is less than 30Ly away, some exploited systems (ES) might be shared with this other CS (the ES would then give its income to only one CS) or even contested if the CS belongs to another power (the ES would not give its income to either). You need to be wary of the numbers displayed on the galaxy map, as this Potential Income doesn't take into account the shared ES.
  3. The distance between said system and Cubeo; indeed after 95Ly it becomes easier to undermine a system than to fortify it. For example at around 125Ly, the undermining trigger is only half of the fortification one. Which means we have to be careful when considering expansions that are really far away.

- We also want our space to eventually be filled by what we call CCC governments, or Communism, Cooperative and Confederacy. This work is called bgs or bgsm, for background simulation manipulation.
Since Aisling Duval is "strong" against these CCC type of governments, if they represent the majority of a CS and its ES population then the cost to fortify said CS is greatly reduced. With fortifications being a weekly endeavour, it quickly adds up to a lot.
On a similar note, we want our worst systems, such as Panganau, to have a majority of Feudal, Prison Colony or Theocracy; governments Aisling is "weak" against. This would increase the fortification cost of those systems, making them less likely to be fortified and thus more easily lost.

All this (and more!) is discussed on the Slack, which I strongly suggest you join if you are interested in helping the Princess.
Great reply, thanks.

Who dreams this stuff up? Where does this 62.1CC number come from and what is that all about. It may answer my continuous question, of why not take on the profitable systems, 100s of them, laying sort of dormant within our power bubble. I see a lot of 'profitable' systems here, that are waiting to be 'cleaned up?'. However: if I have to add, 62.1CC, it becomes insane, as most I have seen, only have a profit of around 20CC. So what is this; in my opinion, hidden, 62.1CC all about?
 
Last edited:
If you go to Aisling's power page, under overview you can see "3788 CC overheads". This is the added cost of controlling our systems.
If you divide it by our number of controlled systems which is 61, you find 62.1. This is the added cost of controlling a single system.
Each time our total number of controlled systems is increased by one, this "CC overheads" is increased by another 62.1. We can translate this as each time we get a new system, for it to have a positive influence on our total command capital (CC), it has to have Base Income > Upkeep+(this added 62.1), or Potential Profit > 62.1.

- For example if you take the case of Nyalayan with a potential profit of 63CC, if it were to be added to our system list the overall effect would be a gain of 0.9CC when left unfortified.
- On the contrary a system like Chnumar displaying a potential profit of 33CC would actually give us -29.1CC overall. And it's even worse than that, since it is in the middle of our power bubble it actually contests a lot of our exploited system, and thus its true profit when you take that into account (and remove the arleady exploited ES) is more in the range of 8CC. Or -54.1CC taking the overheads into account. And with it being so close to our Cubeo, it would be nigh impossible to ever get rid of it (really low forfication trigger, really high undermining trigger, and it would attract grinders), which is why we were forced to enter turmoil this cycle; so we could make the expansion attempt fail.

If you want to know more, here is the formula that is being used to calculate this added overhead cost for each new system. I don't know by whom it was originally found, but you can find it on a few reddit guides and it has been accurate so far.
I replaced * by x for easier reading.
Code:
[B]Total Overheads = Min[N³ x (23/84)³ ; N x 62.1][/B] where N is the total number of CS before acquiring a new system. (this is the formula that you can find on reddit)

The shift between the two parts of the formula occurs between 54 and 55 systems. Thus we have;
- Overheads (Number of CS ≥ 55) = [B]62.1 for each CS.[/B]
- Overheads (Number of CS < 55) = [B]N² x (23/84)³ for each CS.[/B]
This is the total overheads divided by the number of CS, so the average overhead cost of each CS.

And for each new CS; let a be (23/84)³;
- Overheads Increase (Previous number of CS ≥ 54) = [B]62,1 with each new CS.[/B]
- Overheads Increase (Previous number of CS < 54) = a(N+1)³-aN³
                                                  = a[(N³+3N²+3N+1)-(N³)]
                                                  = 3aN²+3aN+a
                                                  = [B](3N²+3N+1)x(23/84)³ with each new CS[/B], which leads the 54th system to add an overall extra 182.9(!!) CC to the total overheads.
Overall it's really simple when you have more than 55 systems, it only gets a bit more complicated when you don't.
So all you need to know for us is each of our CS has an added cost of 62.1, and each of our new CS will have an added cost of 62.1.
As for why this cost isn't displayed on the galaxy map, or anywhere easily accessible is up to anyone's guess...
 
Last edited:
Back
Top Bottom