I've only recently picked up the game, and read a lot about various strategies. I like the all-labs/facs approach, but couldn't help wondering: if - like everyone and their grandma points out - it is really so much of a drain to the economy to support the increased level of production that you gain, shouldn't you consider the amount of tiles required for econ buildings when comparing the different strategies, rather than just comparing X labs + Y facs against X+Y facs (or labs)?
What I mean: When you have N free tiles to spare for your productivity buildings, shouldn't these N tiles include the tiles required for supporting that infrastructure? Free tiles is a limited commodity! And when I say support, I don't just mean Stock Markets, but also farms and morale buildings that are required for the optimal tax-income / tiles-used ratio on econ focus planets.
I tried to put everything in relation, and find out just how much of an effect this has. I am using the following symbols:
X = number of labs; MX = lab maintenance per turn; PX = lab production per turn
Y = number of facs; MY = fac maintenance per turn; PY = fac production per turn
Z = number of tiles to support the production; MZ = average maintenance for support buildings per turn; PZ = average production (i. e. tax income) per turn from support tiles
N = total number of available free tiles: N = X+Y+Z
C = total cost per turn, including production and maintenance: C = X*MX+Y*MY+Z*MZ+X*PX+Y*PY
I = total income per turn: I = Z*E
I've considered the different strategies: all labs, all facs, and mixed facs and labs with 0/50/50 sliders. The goal was to find out the optimal amount of sustainable production I could yield from N free tiles, using a mix of labs/facs and support buildings. Based on the symbols and equations above I determined the effective productivity P per tile, dependend only on PZ and MZ. To cut an already log posting short, these are the equations when using manufacturies (MY=3, PY=12) and Discovery Speheres (MX=18, PX = 4):
a) all labs: P1 = 18*(PZ-MZ)/(21+PZ)
b ) all facs: P2 = 12*(PZ-MZ)/(14+PZ)
c) mixed: P3 = 7.5 *(PZ-MZ)/(10,5+PZ)
When you calculate the differences, you'll find that strategy a) always yields the best results, and c) always the worst. Moreover, for very high values of PZ, the productivity per tile tends towards the constant factor at the front of each of the equation: 18, 12, and 7.5 respectively. The only effect that the maintenance cost of the individual buildings has, is just how fast those asymptotic limits are approached.
That said, I did some very rough calculations to estimate realistic values for PZ, and found them to be somewhere between 15 and 30 (for DL - probably considerably lower for DA/TL).
Even though I have found no post mentioning the impact of requiring additional tiles to support the higher productivity, it appears that this doesn't change the fact that all-x is still a superior strategy. Also there doesn't appear to be any dependency on the number of available tiles, i. e. size of the galaxy/ number of planets, contrary to some statements I've seen (although a larger galaxy and more planets may implicate other difficulties that are easier or harder to overcome depending on your strat). The only apparent dependency is the efficiency of the tiles you dedicate to income, both directly (markets) and indirectly (farms, morale). The differences between the strategies will become more distinguished with higher economic efficiency.
I wonder what everyone else thinks about this, or if I maybe have forgotten yet another factor that I should include in this total-infrastructure model.
I've only played a few games so far, no more difficult than 'Tough'. I realize I still lack experience, but I did put some effort into reading up on game mechanics. I am currently playing the DL version only, so it may be a bit different for DA/TL. I still struggling to find the best strategy for me, and this is my attempt to find a good guideline for an all-x type strategy.