CAs and Terraforming, by Leonard Dickens
By Leonard Dickens (leonard@dc.net)
Jason's recent discoveries about the computation of habitat inspired me to revamp an old article I wrote about CAs and the effect of terraforming in the long run.
A guy once wrote:
- > What is it about CA that makes them so good at getting resources?
- > I've heard all these ???k by 2450 strings with CA, but no explanation of
- > why CA works so well.
Because of the free terraforming.
Recall that there are just two possible sources of resources: pop, and factories. And it takes pop to run factories. So, in fact the main limit to resource growth is pop growth. Everything else is secondary to that.
Now consider the limits to pop growth. There are three: the pop growth percentage (obvious), available space, and the habitibility percentages. Any race can boost the first just by paying in the wizard, and most do, since it is cheap. The second and third are related. A wide hab gets you both more planets and better average hab per planet. But cost of wide hab is strongly correlated with the pop growth rate. Play with the wizard for a bit; this should be obvious.
What this means is: most races can either have high growth, or high hab, but not both. So, a 19% growth race will buy as wide hab as possible, but it will still end up something like 20 clicks in from either endpoint on all three variables. This initial habitat is not that great (good thing we get home planets!).
In principle, the details of the initial (and terraformed) habitat should be calculable precisely, but I am not the man to do the math.:) I did write a perl script to compute the effects using a randomized model, with hab computed using the formula recently posted by Jason. The result was an average habitability of 52% or so on the initial greens, which are 29% of the planets. Once you get over 25% maxpop on the homeworld, you should expect *much* worse pop growth; 29% of planets habitable is not going to be a problem very soon (there should be plenty of greens), but the 52% average hab is a problem. This means only 9.9% average pop growth. 9.9% over 50 years gives a multiplier of 114. A good player will do somewhat better than that, even without terraforming, by going to the best worlds first (not to mention starting on a perfect world). But that is still not going to be enough to be a monster.
Terraforming has a very strong effect on habitability. Consider the same race as before, but assume 5% terraforming has been accomplished everywhere. What is the effect of that? First off, you get 14% of the original planets yellow. Second, the average habitability for fully terraformed greens and yellows jumps to 60%, on average. This may not seem like much, but look at the effect over 50 years: a 10.64% rate compounds the initial amount 229 times! I.e., *double* the economy than would be possible without terraforming. Much better; here there be monsters!
Of course, there is a downside here: you have to pay for all that expensive terraforming with the planet itself, which is small, and thus poor. It needs to grow to get rich, but it cannot grow much because of low habitat -- a bootstrapping problem. Unless you are CA. Then the terraforming is essentially free and instantaneous. And THAT is a big part of why CAs currently rule the monster roost.
Just to complete the analysis, let's look at all the terraforming levels available. Note that only the first few digits in the following numbers are significant; look at the %Green to see the kind of variance I got. (Could get better numbers by running longer, but the point here was to get a general ballpark idea of what was going on, not to use my CPU all night.)
| %Green | %Yellow | %Red | AvHab | AvGYHab | CG50y | |
|---|---|---|---|---|---|---|
| 0 | 29.09610 | 0.00000 | 70.90390 | 15.23105 | 52.34737 | 114.54 |
| 3 | 28.84108 | 7.91508 | 63.24384 | 21.14148 | 57.51820 | 178.71 |
| 5 | 29.02616 | 13.84920 | 57.12463 | 25.92117 | 60.45702 | 229.71 |
| 7 | 28.95196 | 20.47096 | 50.57708 | 31.03403 | 62.79278 | 280.19 |
| 9 | 29.11368 | 27.67077 | 43.21555 | 36.91849 | 65.01514 | 338.23 |
| 10 | 28.81147 | 31.42024 | 39.76830 | 39.64542 | 65.82151 | 362.08 |
| 11 | 29.17977 | 35.37134 | 35.44889 | 43.16574 | 66.87064 | 395.59 |
| 15 | 29.02604 | 53.00995 | 17.96401 | 57.38234 | 69.94777 | 512.38 |
| 20 | 28.71315 | 71.28685 | 0.00000 | 74.51433 | 74.51433 | 750.35 |
| 25 | 29.08123 | 70.91877 | 0.00000 | 82.59113 | 82.59113 | 1462.94 |
| 30 | 29.29371 | 70.70629 | 0.00000 | 89.11691 | 89.11691 | 2492.92 |
|
One other thing to note about CAs: that extra 1% shift every ten years adds up. You can see, above, the effect of moving all three variables from 10 to 11%; fairly dramatic for such a small shift. Throwing on this extra terraforming over time makes the CA great in the long run, as well as the short. Very powerful.
The CA is definitely the most advantageous race in the game. To balance it, I advocate reimposing the cost for CAs to terraform (if bought on planets). I believe it used to be 50% as much as normal races (i.e. 50 or 35 resources). CAs would still have the cheap terraforming from their orbital adjusters, so this should not affect them much in the long term or even the midterm, but it would cut down quite a bit the current CA ability as best hypergrowth race. Also, I would remove the auto-unterraforming when a CA leaves a planet.
-Leonard
See also: "The CA Resource Potential" by Leonard 1997 v2.6/7
