Distance-based restrictions

[ThinkSome]

I would like to file a complaint about distance restrictions for certain buildings. They serve no purpose and are in my opinion completely unnecesary. The only "value" they add is to be completely annoying for everyone. They have a detrimental effect on everyone's gameplay and enjoyment of the game as (at least I) spend an hour+ drawing on the system map in GIMP to optimally layout all the buildings.

Most of the pain comes from Interspecies design centre and the new regional administration. The latter is also a major pain in the rear because you have to do spreadsheeting to figure out if it is better to build them early on or later when they are more expensive. You also have to plan your policies 50 turns ahead and spreadsheet all the species and their stabilities and locations of IRAs to obtain optimal industry/research bonuses (which are mostly clustered around 10 stability). FO at this point does not feel like a game but as a job. It is no longer fun.

What I want:

1) distance restrictions are dropped completely
2) IRA cost is fixed.
3) IRA PP cost can be substituted by ongoing and FIXED influence cost. (or not fixed but based on planets in its area. But then the planets need to have fixed influence costs.)

((The growing colony influence and colony building/ship production cost is also a sign of poor game design. Namely, artificially limiting empire growth))

[Oberlus]

No. Granting that would completely unbalance IRA.

Just don't do 50 turn simulations, instead play onevgame at a time and find out slowly.

[ThinkSome]

I'm sorry, but introducing complexity for the sake of complexity will net you max 2 turns per day. You reap what you sow.

EDIT: ok, this comment is a bit antagonistic, but you get the point. I'm not going to sacrifice optimal gameplay.

[LienRag]

Thinksome's proposals are senseless, but maybe a better UI could help figure out where to put one's buildings ?
It's true that it becomes a chore when one wants to do it well (and since there's real in-game consequences to not do it well, it's very reasonable to want to do it well).

[Daybreak]

It's true that it becomes a chore when one wants to do it well (and since there's real in-game consequences to not do it well, it's very reasonable to want to do it well).

Then, doesn't that mean, there is a problem.

[Oberlus]

introducing complexity for the sake of complexity
[...]
I'm not going to sacrifice optimal gameplay.

I understand the optimal gameplay point here.
I enjoy puzzles and the such. I play sudokus on maximum dificulty for a reason: the fun comes from having to think. I spent tones of hours optimizing my cities in Lincity-NG.
There are two sides here:
- Enjoying optimality.
- Enjoying problem solving.
If the problem is too easy to optimize, then there is little joy in solving it. If the problem is too hard to optimize, then perfectionists suffer from our obsesive-compulsive disorder (be water, my friend...).

If we remove the distance restriction from IRAs, then we remove the allocation problem: no need to optimize a removed problem, but also no joy from solving it.
Also, the IRAs are meant to be spatially constrained to avoid making it a must-have for stability. Although that can be achieved with haaving a single restriction (if no spatial, then exponentially increasing PP cost, but at least one is a must).

As LienRag commented, having tools embedded in the UI to make it easier to find out good allocations is IMO much better than simplifying IRAs.

[Oberlus]

Then, doesn't that mean, there is a problem.

No, in my opinion. ThinkSome is special regarding optimization. He can't move a finger without first calculating the lowest energy gradient.
The problems he is referring to, that takes up hours of his time, I am solving them after a few minutes of looking at the map and the objects window. I for sure am not so optimal, but also don't feel the issue.

[Geoff the Medio]

Would things be better or worse if, instead of a fixed minimum number of starlane jumps apart to produce a Regional Administration, there was a substantial cost increase if another Regional Admin or the imperial capital are closer than some threshold? This could possibly be exploited by building multiples in parallel closer than the threshold distance, so the production restriction could be strengthened to one per empire on the queue.

Similarly, what about also adding an ongoing influence cost when two Regional Admin are too close? That could also replace the destruction of other-empire-produced Regional Admins.

[LienRag]

Would things be better or worse if, instead of a fixed minimum number of starlane jumps apart to produce a Regional Administration, there was a substantial cost increase if another Regional Admin or the imperial capital are closer than some threshold?

Not sure, it seems worse to me (more complicated, and quite confusing for beginners).

Similarly, what about also adding an ongoing influence cost when two Regional Admin are too close? That could also replace the destruction of other-empire-produced Regional Admins.

Maybe...
Will beginners understand it ?

[Daybreak]

Will beginners understand it ?

I doubt it, unless they have the ability, and the ability needed appears to be very high.

The Pedia has nothing on what you have been talking about. I originally thought that the IRA was a way to extend supply and provie stability, and although that may be true, it appears that is is more complicated than that. The incorrect placement of an IRA sounds like it will be detrimental in some way, which does not make sense to me - it is already detrimental to your empire if building one is in the wrong place already wastes PP. Is that not enough.

But its not just the IRA.

When I play the game I seem to do well overall, but I am missing all the little nuances / calculations that you all seem to take for granted.

How long would it take a completely new person who does not even have the advantage of playing previous versions, to understan all of this complexity, considering a few of you are struggling or have concerns.

Maybe I am beeing too hard on myself and the game, and playing a multiplayer game may help.

Note - I do like the new game, but from what I am reading, I also understand there is a lot that I am missing or not understanding.

[ThinkSome]

Would things be better or worse if, instead of a fixed minimum number of starlane jumps apart to produce a Regional Administration, there was a substantial cost increase if another Regional Admin or the imperial capital are closer than some threshold? This could possibly be exploited by building multiples in parallel closer than the threshold distance, so the production restriction could be strengthened to one per empire on the queue.

Similarly, what about also adding an ongoing influence cost when two Regional Admin are too close? That could also replace the destruction of other-empire-produced Regional Admins.

I'm against what I perceive as arbitrary and unnecessary restrictions/costs, under which I'd also file these two proposals. I think that having a fixed ongoing cost per IRA or a variable cost based on number of colonies where the IRA in question is the closest one (but then the colonies themselves need to have fixed costs) would be OK. But the costs shouldn't depend on proximity to other IRAs.

[ThinkSome]

Yes, one side effect of introducing arbitrary costs and complexity for the sake of complexity is that it is impossible to effectively play the game without constant browsing through the source code. I also have a python program to simulate decisions.

[Geoff the Medio]

I'm against what I perceive as arbitrary and unnecessary restrictions/costs, under which I'd also file these two proposals.

The Regional Administration is supposed to be regional, meaning not too close together. The restriction on location is designed to enforce that. The restrictions also ensure that an empire can't easily produce as many as needed to avoid any penalties for Bureaucracy or disconnected systems. How or why this is "arbitrary and unnecessary" is unclear. Replacing the limit on minimum jumps to produce at all with an increased cost when too close would hopefully have a similar effect, but be a more flexible and thus easier to work or plan around.

I think that having a fixed ongoing cost per IRA

Fixed per-turn IP costs costs for a building could be OK, but what would they achive, especially without additional restrictions / limits on placement?

...or a variable cost based on number of colonies where the IRA in question is the closest one

Variable how, and what would that achive? If the "variable cost" of a building is proportional to the number of colonies that it is the closest such building to, then producing an additional Regional Administration would just change which is closest to which colonies without changing the the total cost.

(but then the colonies themselves need to have fixed costs)

I don't understand what this means in this context.

[Ophiuchus]

Replacing the limit on minimum jumps to produce at all with an increased cost when too close would hopefully have a similar effect, but be a more flexible and thus easier to work or plan around.

i think this is (almost) a good solution for the regional administration center. And certainly better than what we have now.

it would be better if the UI could communicate inflated costs. E.g. "orange-to-red" flagged price in production queue when you are too close. would have to introduce some kind of "base cost" in FOCS and backend though. a tooltip on the item would show the cost difference (both realtive and absolute) and in the best case an explanation.

without better UI, exponentially increasing costs also helps communicating this (e.g. 4hops: x2, 3hops: x4, 2hops: x8) as a player hopefully notices the high costs (at least if it matters because of current PP budget).

per-turn-IP costs could maybe balance this better, but are worse in terms of UI (effect is shown after the building is finished) and gameplay (it does not remove the optimisation problem).


also i think that we should remove/change the Beraucracy having-as-many-regional-administrations-as-possible benefit , i dont have too much data on real gameplay here. but it gives an incentive to build the administration center far away from your colonies just so the number of adminstration centers is increased, which seems wrong.

[Oberlus]

The incorrect placement of an IRA sounds like it will be detrimental in some way

Let's specify how detrimental it is. First what are the game effects depending on IRA's allocation.

IRA gives a bonus to stability: +5 for colonies in the same system, going down by 1 for each jump away.
Minimum distance is 5 jumps: IRA <--> s1 <--> s2 <--> s3 <--> s4 <--> s5 <--> ÌRA.
Stability bonus in spaces s1 and s5 is +4, in s2 and s4 is +3, and in s3 it is +3.
If you manage to have all your pairs of IRAs at that distance, and
If you minimize the average distance to an IRA (that is, place the IRAs where planets are more dense, so that most of the non-IRA planets are in an IRA or s1 or s2 spaces, and place IRAs close to each other to help this), you can get close to +5 stability in most of your planets.
Here, "incorrect placement" of an IRA means it doesn't let you place another IRA in a good system with many planets. To avoid this, you need to browse the galaxy, locate the dense clusters of planets, try (on paper) different allocation of IRAs to check out stability bonus of the planets nearby, and pick the allocation that is gets you the highest average stability.
OR you can apply local restrictions/objectives to each planet and look for an IRA allocation that manages to fulfill all of them: "Late game, with the policies I plan to have, this huge inferno for Egassem will need +2.5 more stability to get to stability 20 and harvest all the industry bonuses, so it needs to have an IRA at no more than 2 jumps. This Chato world will need to stay at stability 0.5 to maximise RP bonus from policy Necessity, and it's going to have a +2.5, so I want it to have a -2 stability bonus, and so I want it to be at exactly 7 hops from an IRA or the capital... (this for all the planets that you plan to colonize or invade in the whole galaxy). So, if I place this IRA here, planets 1 to 6 meet the requirements, but 7 and 8 won't... if I move this IRA one system to the left, then it's No, can't move it there unless I move this one also, but then planet 14 won't comply..."
All this are very interesting resource allocation problems, partition problems, which are NP-complete. The amount of calculations needed to solve it grows exponentially with the number of elements to partition, in real life they are solved with computationally intensive combinatorial optimization methods that approximate the optimum. I understand ThinkSome is tired and frustrated because he doesn't like aproximation algorithms, he is hell-bent on exact optimization, which is unsuitable for NP problems.

Cost of an IRA depends linearly on number of planets: if IRA with one planet costs 12, with N planets it costs 12*N.
Cost optimality here (minimal total building cost) is about colonizing first the planets that will have an IRA, all of them, and only then start getting other colonies. To plan this ahead is much easier than the above. However, doing this is suboptimal for empire growth if it delays your growth (you don't get those juicy planets near your homeworld because you are first getting more IRA colonies far away, so your PP and RP grow slower and you colonize and build IRAs slower than what you could have done), which is the most common case. So in the end what is optimal regarding order of colonization and IRAs allocation is not being strict on getting only IRA worlds, but it is hard to determine in advance the best equilibrium, you have to play the game several times doing different choices (or develop some skilled spreadsheet for growth simulations, I tried and it is a nightmare), then reload and do the optimal sequence of colonizations and IRA buildings.

You put the two problems together and the combinatorial explosion is just impressive (I really like this stuff, very related to my job, hence the brick).


Now how detrimental is not get the optimal placement of IRAs? Probably not much.
If you place IRAs randomly, you could still get an average of +2 stability per planet. If you put some thought on it (a few minutes with the map). you can get +3 easily. If a species is giving you problems in a given planet (should be +X stability to get some bonus) you can just put a different species there (solve the problem without messing with IRAs). Applying optimization algorithms and game simulations in a Ryzen Threadripper would get you +1 stability and a few hundred of PPs saved. And your game mates having to wait a few hours each turn.

In the end, being optimal here implies not spending too much time in optimization.

I remember playing Civilization. In single player, I saved at start, explored a bit to see the nearby map, sketched the map cells with terrains and bonuses in a paper, and looked for the optimal allocation of cities for my playstyle. I enjoyed it, although it was a bit frustrating sometimes when realizing some old decision means this one city among twenty won't have optimal surrounding. In the end, it was never that kind of optimization what helped me win the game.