AR Guide by Leonard Dickens - Revised: 21st September 2000
Author: Leonard Dickens
Revised: 21st September 2000
ARs Compared to Normal Races
ARs work very differently than the other races in Stars. This section touches on each significant difference, and its effect on race design or game play.
- Starbase Habitat: ARs live in starbases. Thus, while they use the planetary compat as do normal races for purposes of determining the growth-rate modification of the pop growth formula, they do *not* have the same planetary maximum populations. Instead, they get a fixed maxpop determined by the starbase type:
orbital fort 250000
space dock 500000
space station 1000000
ultra station 2000000
death star 3000000
The most obvious effect of starbase habitat is that when the starbase is lost, so is the pop. It is thus much easier to attack an AR than a normal race, since you don't need bombers.
- No factories: ARs cannot make factories, but rather, get resources based on other factors (see below). As with any factoryless race, this means that the need for early minerals is modest.
A less obvious effect of being without factories is that settlements are cheap. Though they can be lost easily, they are also fast to set up. If a normal race loses a planet and it gets bombed out, that race may have to spend 10 to 20 turns to get the factories back up. The cost in germanium (G) is large, probably requiring imports. An AR can have a planet back in business in 2 turns, costing only a few minerals.
- AR mining: unlike normal races, an AR gets a small number of "mines" just by being at a planet. The amount is 1/10sqrt(pop); thus, a planet with 250000 pop has the effect of 50 mines. Not very much. Offsetting this weakness is the special rule about remote mining for AR planets: unlike any planet occupied by normal races, an AR planet may always be remote mined.
- Infinite mining: Unlike the normal races, when an AR remote mines a homeworld that it owns (meaning, has colonized), the homeworld floor of 30 concentration applies to *all* of the AR's remote mining fleets at that planet. This means that an AR can mine an almost unlimited amount per turn, each turn, needing only the remote miners to do so. (Remote mining for all players is subject to a limit of 4000 mine-equivalents per fleet; thus, ARs are limited in remote mining by the 512 fleet limit.)
Thus, if the AR makes it to the late game, where tech is maxxed and minerals are tight, he is a great position. Other races will be spending most of their resources doing highly inefficient alchemy. ARs will be spending all of their resources building, assuming they set up their mineral distribution network properly. This is potentially a game breaker.
- Resource Formula: the formula for AR resources is:
resources-from-planet = hab * sqrt(pop * energy / pop-coefficient)
hab = the planetary habitat
sqrt = square root
pop = the pop at that planet
energy = the AR's energy tech
pop-coefficient = the value from the race wizard
There are effects to all of these, so I will discuss each in turn.
hab: unlike normal races, the planetary habitat directly affects the production of an AR. This means that, all other things equal, an AR will do better with fewer, higher hab planets than with more low hab ones. The value for "hab" is never lower than 0.25; if the planet is lower value than that (or red), then 0.25 is used instead. This, combined with the pop limits based on the starbase, make ARs good at working up reds.
sqrt: the effect of this is huge. It means that all the factors inside it -- pop, energy, and pop-coefficient -- are "squelched" in effect. The larger they get, the less they matter. You will see the "square root effect" mentioned several more times in this guide...
pop: with the sqrt, this factor means that an AR gets fewer and fewer resources as a planet's pop increases. Normal races have a tough problem with their planets: do they want to use them to grow pop on, or to run factories at? If they decide to grow pop, they have to hold the planet between 1/4 and 1/2 full; obviously that cuts the possible production by the same amount. ARs grow using the same formula, but the effect on them is much less: if the AR leaves the planet at 1/4 pop, he loses only half the resources (sqrt(0.25) = 0.5). If he leaves it at 1/3 maxpop, which is the level yielding the largest pop/turn, he loses only 42% of the resources.
This effect, combined with the raised planetary maxpop obtained from starbase living, means that ARs are considerably less dependent on a high growth rate (%grow) than normal races. A typical (OBRM) normal race has a planetary max pop that tops out at 1100000. (JOATs get 20% more.) In order to grow on average at a competitive rate, say, 10% or so over the first 50 years, a normal race needs to have a much higher growth rate -- 17-19% is typical. ARs can grow their pop much more easily, especially after they get the larger stations.
Another very important aspect of the sqrt(pop) effect is its effect on the maximum rate of economic growth possible for ARs. If pop is growing exponentially at rate R, a normal race can increase its economy at rate R. (Usually factories will lag pop, so the effect is somewhat delayed in the early game, but over time this is the case.) What holds back a normal race is crowding and low planet values. An AR, though, growing at the same rate R, gains econ at only rate (R/2)! Crowding holds an AR back somewhat less, low planet values the same. But in general, this fact means it is much harder to grow the AR economy than the normal race economy.
One final aspect of the sqrt(pop) factor is spreading. A normal race gets no immediate benefit for spreading its pop out, as long as all planets are below maxpop. (Spreading can increase the rate of growth, though, for planets above 1/4 maxpop. Thus, it is certainly a good idea, but not imperative.) For ARs, though, because of the sqrt(pop) factory, spreading immediately increases the economy, as well as potentially increasing the growth rate. For example, assume an AR has two 100% planet available. If he puts 1M pop on one, and zero on the other, he gets sqrt(1M*energy/eff), or 1000sqrt(energy/eff). If he puts 500K on each planet, he gets 2*sqrt(500K*energy/eff), or 1414sqrt(energy/eff). 41% more, just for rearranging where the pop is!
Note, though, that the amount an AR gets from spreading drops off, again as a result of the square root. Spreading his pop from 1 planet to 4, an AR can double his econ. But to quadruple it, he needs 16 planets. And to octuple it, he needs 64. Obviously, this process is only good for a relatively small amount of growth; getting 16 planets is possible in most typical games. Getting 64 is unusual. Getting 256 is very unlikely -- if you are playing in a galaxy that large, you are an idiot.
(In theory, a 6% AR could spread his initial 55000 pop out into a large packed (910 planets), and get a 30x boost of his economy. Of course, this would be less than 100 pop per planet. With some early growth, though, there would be enough pop for 100/planet; with energy 10, say, this would have an economy of around 7900 resources! Growing at 3%/year, and assuming energy 26, this would get to 55000 resources by year 50.)
energy: obviously, this one means that taking energy cheap or normal will be a good idea. Which is generally true; note, however, that the sqrt involved means that the AR gets diminishing returns the higher his energy goes. Going from energy 1 to 2 means a dramatic 41% increase in resources (and the cost is very low). Going from 10 to 11 means a modest 4.9% increase (and the cost is also higher). Since the cost increases and the return decreases the higher you go, cheap energy is not as important as it would initially seem. Note, though, that the cost of energy research is a fixed amount per level, whereas the benefit applies empire wide. So, if your total economy is 1000 resources, getting energy 11 probably does not make sense; you are paying over 1000 resources for a gain of 49/turn. If your economy is up to 10000, though, you pay 1000 for a gain of 490/turn. That's a very good deal.
pop-coefficient: you drop this from 1/10 to 1/25, you get 600 points. That's a lot. But you lose 37% (sqrt((1/25)/(1/10)) = 0.63) of your econ... that is also a lot.
- Movement losses: 2% of the AR pop on a freighter dies when it moves. This is annoying, but a typical AR has so much pop that it is hardly noticed. It can be bad for low growth ARs, though. In this case, there is an important feature (or defect, perhaps) to be aware of: the 2% is rounded off. Thus, if you have 2200 or less pop in a fleet, you lose zero. Each 3300 (or fraction thereof) added to that, you lose 100/turn.
The grand strategy of all ARs is the same. You are strong in two parts of the game: the very start (where you get big resource gains from spreading and energy tech, and when you have resources to spend and nobody else does), and the end (where you have minerals, but nobody else does). Thus, the game that an AR hopes for is to use the first period to get established on a good number of planets without angering his neighbors. Then, he plays defensively, using diplomacy to avoid conflict as much as possible, building up his economy and tech, playing to survive until the end. Once minerals become scarce, he can go forth and conquer.
Game Parameters Discussed
Before you begin to design a race, or to adapt a design you already have, you should sit down and think about the game you are going to enter. Naturally, the game parameters and special rules can (and should) influence your design. Probably the most basic decision is: should one play an AR at all? If so, what sort of AR? Following is a list of universe parameters and special rules, with discussion of how they affect the AR.
- Universe Size and Crowding: ARs do best in uncrowded circumstances. Of course, every race can do better with more planets to work with, by scouting actively and "cherry picking" -- colonizing the best planets first. But in addition to this effect, as previously mentioned ARs increase their resource totals by simply spreading out. Finding four 100% planets does not immediately help a normal race; an AR can benefit as soon as he can a colonizer there. A second aspect of larger galaxies is that the more planets there are per player on average, the more likely tech is to go to the maximum before serious conflict starts. ARs benefit more from max tech than other races, since it means they should have high energy for resources, and death stars for growing pop.
The less crowded an AR is, the narrower he can make his hab. In the extreme, an AR can be a one-immune with both other fields of minimum width, for about 1/21 initial greens. Barry Kearn's "ARvids" is of this ilk; it was the first public AR "monster" race. Such a race can afford very high %grow, 1/10 pop efficiency, and good tech. In a very uncrowded place (such as an unopposed small packed testbed), the race monsters easily. However, in any reasonably constrained area, such as a tiny normal, the ARvids flop. I call the artificial sense of security people get from testbedding in absurdly large places with narrow hab races "the ARvid effect".
One more point to add about choosing to play in uncrowded games: although uncrowded places are good for AR and help them relative to most other races, there is one PRT that benefits even more than ARs: CAs, especially narrow-hab TT CAs. So, if CAs are banned in an uncrowded game, that is the best possible situation for an AR.
- Game Definition: some of these would certainly affect my decision to play an AR. Let's look at them.
Max Minerals: good for factory based races. Hurts ARs, though, since their main ace in the hole -- the infinite minerals, eventually -- is much less competitive the more "normal" minerals everyone else gets.
AccBBS: good for hypergrowth races. Hurts ARs, since with the square root aspect of resource generation, starting with 25000 pop does not hurt their production that greatly.
Slow Tech: puts a premium on resources, thus, hurts ARs. Depending on the galaxy size, may make it completely impractical to play AR.
Galaxy Clumping: little effect.
No Random Events: helps ARs, slightly, since it means nobody will be getting the Alien Miner or Genesis Device, and thereby partly or completely nullifying your late-game mineral advantage. The existence of other MT parts in the game is also probably bad for ARs, as many of the parts tend to make BBs much better fighting ships, whereas AR tend to have cheap con and therefore will often be the first to get nubians. Also, ARs often have very few spare minerals until the early BB era, later than other players, so will often miss the first few MTs.
Public Player Scores: generally, hurts ARs. People already know you are weak, but because they cannot count your pop, they don't really know how weak. Without knowing where your economy is, they cannot know your tech. Thus, you have the diplomatic flexibility to lie a little bit. Public player scores takes the guesswork out: your neighbor will know exactly what your tech is, and what your econ is, and how many ships you have.
- Special Game Rules:
Team games: ARs are nice to have in team games, since with teammates you have built-in trustworthy trading partners; you offer minerals and can get tech and protection.
Jump games: sometimes hosts do games in which there is an initial "jump" of 500 to 1000 years where players can do tech but not interact. You must examine these very carefully. If you can enter one of these games and end up after the jump on equal footing, AR would be a very strong race to play. Typically, the mineral shortage won't happen for a while, but on the other hand the AR ability to grow *fast* and cheap will put you in good stead. However, in my experience most of these games are dreamed up without considering ARs, and thus have no special rules for them. And given that ARs only generate about one half the resources of a normal factory race, playing an AR would be suicidal. So, be very careful about playing a jump game with AR: ask yourself: if I were a IT, what could I do with this setup? (The reason for IT, is that by starting with two planets you can rather easily get twice the resources of any non-IT or PP.) You should also consider what a narrow-hab TT CA might do in the setup.
- ISB: Almost mandatory. Very low %grow ARs may avoid it.
You get two things from ISB, both handy. First, the space dock. These allow you a very cheap way to have up to 125000 pop on a planet with full growth. They also give fuel, can build ships, etc., as they do for normal races.
The main reason for taking ISB is ultrastations. Con 12 costs around 13000 resources with cheap con; thus it can be obtained fairly early by almost any AR. Con 17 costs around 70000 resources (again with cheap con); this amount is fairly hard to obtain by most ARs before severe crowding sets in.
- IFE: Recommended. Very wide hab ARs may be able to get away without it.
Generally, ARs will want IFE more than normal races. Unlike a normal race, ARs need to be moving as soon as possible. It is true that ARs have resources to spend, but they need every iota invested into growth: energy, con, and terraforming. You need at *least* prop 5 to move far without IFE, and even that is pretty slow.
On the other hand, in denser universes with ARs that have very wide hab and low growth, there should be enough planets very close by so that the early engines are acceptable for the initial colonization.
If you do not take IFE, then taking prop cheap or at least normal is a good idea.
- CE: sometimes. Yucky for any race.
CE sucks at all times, but of all the races to take it with, AR would have to be among the most likely, second after IT. Several reasons. First, taking CE means you can start at prop 2 for mizers. For other races, not a big deal, but AR want to be moving immediately, not even spending a small amount on prop. Second, early on with AR, you are on the defensive. Building gateable horde style ships as specific countermeasures to enemy designs can be quite useful. CE is never as bad on defense as on offense.
CE is bad once you get later in the game, though, so avoid it in larger places, team games, or other places where you have good reason to think you will survive later.
And of course, if you are like me you would rather lose than play the game with the micromanagement induced by CE, and never ever take it. :) But this is not how some people feel. The points are pretty good.
- NRSE: rarely. Better in larger places.
The downside of NRSE is, obviously, losing the ram engines other than the fuel mizer. Generally, rams are cheaper than normal engines in resources, which makes ARs like them more than others. Also, rams cost more G than other engines, meaning that ARs, needing no G for factories, can afford them better than most races.
The upside of NRSE is mainly the points. ARs don't have that many places to scrape up advantage points.
Secondarily, NRSE provides the interspace-10 (the earliest warp 10 engine). However, it is very expensive, and you won't have the resources or the minerals to really exploit it. Furthermore, as an AR you don't plan to be on the attack in the midgame, and warp 10 is much less useful for defense than attack. If I took CE, I would feel a lot better about NRSE. And of course, I would always want IFE for the mizer, but it is practically mandatory with NRSE.
Thus, I would recommend either IFE/CE/NRSE, or just IFE.
- TT: sometimes. Recommended for 1/25 ARs.
TT is a very nice thing for an AR to have, because of the way terraforming works in the resources equation. That is, that the value of a planet directly produces resources, unlike normal races. Thus, terraforming is a straight investment for ARs, in addition to a means of increasing the rate of growth. As with any investment, making it cheaper is good. Consider an average green for a one-immune: it might start at 50% hab with each point of terraforming increasing the value by 3%, say. Let's say there is enough pop there to produce 100 resources/turn. Spending that 100, you could increase the planet to 53% and thus the resources to 106. Without TT, the rate of return from the terraforming is 6%: not great. With TT, the rate is better: 8.6%. By comparison, normal races invest in factories. A typical 12/9/x race gets 13.3%, given G.
TT also helps the AR work up yellows. Yellow colonization is typically much more important for ARs than other races, since they are so desperate for resources.
With all that good stuff to say about TT, why only "sometimes" take it? Because it is very expensive. I recommend TT more readily to 1/25 ARs in part because they have fewer resources from pop, but also because they have the extra points to invest.
Also, unlike normal races ARs have a hard time taking cheap bio, because they need cheap energy and con (everyone needs cheap weapons). Thus, the higher levels of TT will be out of reach of the AR until the very late game.
- LSP: OK for high growth ARs, though they should consider dropping %grow instead. Not good for low growth ARs.
Some people swear by LSP for ARs, because the square root effect: you lose 30% of the pop, but it is only 17% of the resources. But the fact remains that you are two years back of where you would otherwise be, or three for a midgrowth (13%) AR. This means there are planets that you might have gotten, but will not, because you will be too late. Is it worth it? As always, maybe. In a less crowded galaxy, LSP (and high growth) seems like the right way to go. If you lose one planet out of 50, it only reduces your econ by 2%. You may well be able to buy 2% more econ with the points, perhaps with a wider hab. On the other hand, if you are only getting 10 planets, then losing a planet is 10% of your econ, and you will probably want to think twice about LSP. Remember your grand strategy: get out of the gate quickly, grab space and hold it.
- OBRM: never.
The one mining module you get with OBRM is initially the equivalent of about "mines cost 40", if that were possible. With full miniaturization, at con 21/elec 20, the OBRM miner is about like "mines cost 10". At the midlevels where you have to be starting to buy miners, it is about mines cost 20. All of those are way, way too much.
- ARM: Often. Good for any AR, but not necessary.
ARM provides several things beyond what you get with "normal" mining (neither ARM nor OBRM). First, you get the two potato bugs to start with. These are nice, but not necessary except perhaps for very low growth ARs. You also get the ultraminer hull, which is the cheapest way to buy mines, but not gateable. And there is the miner hull. Not amazingly useful.
The main thing you get with ARM is superbugs: the midget miner hull and the ultra miners. Superbugs are gateable, and very cheap.
Having miners be gateable has three benefits. First, you get the minerals in places other than the homeworld. The minerals at home are cheap, but the planet will quickly drop to its 30 min con. And so at many places the minerals will be even cheaper. It also costs to distribute the minerals out from the homeworld; you must buy freighters or packets.
Second, gateable miners allow you build them anywhere, then gate them home to mine. This means you can increase your mining very quickly, if need be, by having all planets build miners.
Finally, and perhaps most importantly, is the fact that *other* players can use the minerals in your planets quite nicely. They just have to kick you off of the planets. Thus, one aspect of the gateable superbugs is defensive: to quickly drain out any new planet, so as to deny the minerals to enemies or would-be enemies. I would advocate leaving a fleet or two of superbugs on every planet until it is considerably below 30 con, for this reason.
- NAS: often.
NAS is a good point mine. Not having penscans is not as great a problem on defense, and the doubled range can be helpful. That said, penscans are a great benefit to any race, and you should always think twice before giving them up. For ARs the choice is perhaps a little harder than most, since they need allies and friends in the midgame, and trading penscans is one very helpful way to keep friends friendly.
- RS: sometimes.
This LRT is nice to have in the very early game, and in the nubian era, but weak in between. This is about the same zone as where ARs are weak, so they should avoid RS if the midgame is likely to be long. But, if the midgame is likely to be long, ARs should probably avoid the entire game. That is, if you are considering AR at all, this may be a good LRT.
RS will mean you cannot effectively armor your stations, so if you like huge deathstars it may not be for you. It works well, though, with the late game Mao-style war that ARs excel at.
RS also makes it a bad idea to armor your ships. This is not necessarily a bad thing, just different; it means you will be building twice as many BBs to get the same armor. On the plus side, you get twice as many shields and they are up 40%, so your shield/armor balance will be great. (Normal BBs tend to have much less shielding than armor, making them fairly good targets for capital ship missiles.) On the minus side, you effectively pay twice as much for for the engines and electronics on the BB; thus, if you are thinking about NRSE you should probably avoid RS.
- GR: never. You need energy and con fast, not the others. Going at 65% speed is going to hurt ARs more than any other race.
- BET: never. You have cheap con. Don't waste the opportunity to be the first to nubians by making them double price forever.
- MA: never never never. Obviously, just build more remote miners.
- UR: rarely. The main problem with Ultimate Recycling is that it is quite costly, and ARs tend not to have points to spend. Also, the big advantage to UR happens when you get to nubians, since it allows a race to win the counterdesign wars. But if AR can make it to nubians, he is already is a very good position. Getting there is the problem, not winning once there. Finally, using UR can be fussy at best, and it is usually possible to avoid building designs that won't last.
Hab and Growth
There are two conflicting imperatives to keep in mind for AR habitat. On the one hand, because of spreading and in general the low resources an AR gets per planet, he wants all the planets he can get. This argues for wide hab variables, which means centered habs or immunities. On the other hand, ARs want the diplomatic advantages of intersettlement, an argument against centered habs. This is one reason immunities work well for ARs: they are wide without being centered.
The following sections discuss various AR hab/growth strategies. They are discussed in order of how strong they are, IMO, with the stronger hab schemes coming first.
- Single Immune ARs
Arguably the best AR hab, for general use, is one immunity, one relatively wide variable, and one narrow and offset. %Grow from 13% up to 17% or so; the higher values will require narrow habs.
Typically, it is done as immune grav, wide rad, and narrow offset energy. This works with other standard AR choices. Immune grav means prop research for terraforming is not needed; IFE, expensive prop, and sometimes NRSE/CE build on this. ARs, like anyone else, need weapons, so you are going to get the rad terraforming in any case. And ARs tend to go high into energy fast; it is not uncommon to get energy 10 by 2415. A slight variant of this scheme uses immune grav, wide centered temp, and narrow offset rad.
A second one-immune scheme is wide grav, rad immune, and narrow offset temp. This has one disadvantage over the previous scheme: it requires at least some prop research, and it "wastes" weapons research. But the problem is less important than it seems: because the temp is much narrower than grav, it is going to be the preferred field for terraforming anyway. Offsetting this disadvantage is the advantage gained from rad immunity: it raises your overall universal hab slightly more than either of the other two immunities would, because the distribution of rad is less centered than grav/temp.
1/25 pop efficiency can work well with one-immunes. In this case, I would recommend spending the points on hab and TT. The reason for hab is obvious: you are giving up 36% of your economy; the only way to get it back is to get more planets and/or better planets. So, widen the hab to get more. The reason for TT is less obvious: in shrinking each planet but taking more of them, you have increased the amount of the economy that will be spend on terraforming. This slows the race down, so that even if you widened the hab enough to get 36% more planets, you are still much slower. Thus, it is still a reasonable idea to take expensive bio and TT with such a race.
- Non immune ARs
Generally, the non immune AR is a less aggressive, more HPish style of play than the one-immune. By not spending on the immunity, you are getting points to put into two wide hab variables and one mid-width one. Thus you will end up, in the long run, with more planets. The downside is that in the early going, you have more low-value greens, yellows and near reds which are not producing much.
Non immune ARs should consider TT for the higher terraforming levels, not just the cheap terraforming. If they do take TT, they should consider cheap or normal bio.
Given that weapons and energy are the two most important fields for any AR, a non immune AR will want to have either temp or rad be the narrowest hab variable, or both equally wide. Grav, therefore, should be wide, no narrower than 15 clicks in from the edge. Wider is OK; this increases hab values; IMO even full width is OK. Generally, I like to take rad as the narrowest field, because it can be most successfully offset.
I do not recommend 1/25 efficiency for non-immune ARs. Generally, they should find it hard to widen their hab by 36% without taking an immunity.
TT is recommended for non-immune ARs. Compared to a one-immune, they are going to have 50% more terraforming expense (three fields to work on instead of two), and they will have to get it somehow using all those small planets.
- Bi-immune ARs
Bi-immune ARs are, IMO, borderline playable. Some of them can reach almost the resources totals of one-immunes. The typical bi-immune AR has growth 13%, 1/25 pop efficiency, and immune in grav and rad, with temp the narrowest possible and offset only somewhat. This gives 23% green initially; with 15 points of terraforming, 56% of planets are green and all of them will be at least 41%. A second hab scheme is to take grav and temp immune, rad narrow and offset, typically 12 points in from one edge or the other. This gives slightly fewer greens, but by being more strongly offset it has a better chance of successful intersettlement and noncompetition with monsters.
The HP version of bi-immunes is to take 9% growth with 1/10 efficiency. This race has better long term production, but it ramps up quite slowly.
TT is optional for any biimmune. With two immunities, the amount of terraforming you need to do is quite reduced.
- The AR triimmune
The triimmune AR is not really viable for high level play. However, it is certainly an idea to keep in mind for team games, or other games with special rules. Generally, the triimmune is 6%, 1/8 pop efficiency. Variants are possible, of course. 7% is possible, for instance, though the points required are considerable, resulting in a race with weak LRTs.
No triimmune should ever have worse pop efficiency than 1/10.
As with normal races, it is typical to only consider 1/10 or 1/25 for an AR. The reason is similar: consider the drop from 1/10 to 1/11. Here, you are sacrificing roughly 5% of your economy for 50 points. Now consider the drop from 1/24 to 1/25. Here, the same 50 points means losing only about 2% of your economy. 1/9 gains you 5%, but for *200* points. ARs are desperate for resources, but usually not *that* desperate.
For most players, and certainly beginners, I would recommend the 1/10 AR. Compared to the 1/25 AR, it has fewer, better planets. That's generally easier to run in MM terms. It is also easier to take a more offset hab, when needing fewer planets. A typical 1/10 AR will use perhaps 60% of planets with full terraforming. This gives diplomatic advantages, which any AR needs.
1/25 ARs are viable, though. They tend to need better play, though, because of their need for many more planets: with full terraforming, a 1/25 AR will use essentially all planets.
Tech is more important for ARs than ordinary races, since they need at least some energy to get resources, and con to get the advanced space stations. Like other races, they need weapons for defense. Therefore, the traditional AR takes good tech -- cheap energy, weapons, and con. Two or three of the others expensive; sometimes normal prop or elect.
That said, ARs are also fairly desperate for advantage points from any source, in order to get the growth, hab, LRTs, etc. that they will need to survive. Thus, sometimes players want to mine the tech for points.
For the AR, the game may be broadly broken down into three main sections: the early, mid, and late games. These correspond to the AR's situation vis-a-vis other players. In the early game, the AR is expanding outward into unoccupied space. In the midgame, the AR tries to build upward, getting energy tech and con to increase his planet's sizes. Finally, in the late game the AR goes on the offense, using his mineral advantage achieve victory.
Early game: Tech Strategy
This is a sketch of how I play ARs. First, a general idea of what you should be doing in terms of tech. Here is the basic sequence of what techs you want to get, in what order.
energy to 4+, higher is better...
[homeworld getting close to 250K]
con to 3, prop to 2 (if IFE), else 5 (if not IFE).
[medium freighter pop mover now possible]
[first colony getting near to 62500 colonists]
con to 4, for dock
energy to 10
cheap terraforming tech (weaps/prop 5 if you aren't immune in
con to 12, for ultrastations
bio 4 (for minelayers)
energy to 11-14
con to 17, for deathstars
elect to 8 for ultraminer
energy to 16
That's the basics. Of course you need to get low bio levels in there (for terraform techs), but don't do them until you have at least one colony that will actually use the tech. Another thing to keep in mind is the sequence above is only the *economic* part of your tech game. You may well be doing other tech within that sequence for reasons of defense, mainly weapons but also electronics and even prop. For instance, depending on the pressure from other races you may want to get weapons 16 before death stars.
Regarding the very early production, generally you will do fine building only scouts and colony ships before you get con 3. The scouts are there to find good planets, the colony ships to get a toehold. Many AR players don't start colonizing until they get their home planet to 250000, but that is a mistake. You should colonize any green, and even some yellows, well before then. Remember the square root effect: your most efficient colonists are the first few. However, don't send more pop (just the 2200 in a colony ship), until the home planet is nearing 250000, unless the planet is in the high 90's.
With a very low growth race (9% or less), I sometimes build one or two "minitrucks" -- small freighters with FM and fuel pod; they only move 7000 pop, but that is enough to double the production of a colony with 2200 pop... minitrucks are typically not needed with high growth rate races, though, since they will get to 250K fast enough. For a triimmune AR, you never want to move more than 2200 pop per fleet until the midgame, and so you can make a slightly better minitruck by using the colony ship hull.
Once you get near 250000 pop at the HW, get con 3. Then you can start building 2 or 3 medium-freighter "trucks" each turn, until you have enough out to continually carry off all pop above 250000 on the HW. The "standard" medium truck is a medium freighter hull, a mizer, and a fuel pod. It can just get 162 ly at warp 9 and return empty, or 192 ly at warp 8; beyond 192 ly you need to add fuel somehow. Building docks in strategic places can be very useful in this regard, since they provide gas. For more distant planets, you will have to augment trucks with fuel-pod scouts, that is, a scout with just FM and a fuel pod. These are quite cheap, and will allow you to do warp-8 or 9 runs to any distance and back, if you add enough. Some players use fuel DDs instead of scouts; they are a bit more expensive but can be armed, which may be very helpful.
Unlike normal races, you should spread your pop exports. Generally, I try to use them to keep all planets at the same pop level, excepting an extra load or two early to high value planets.
Mid game and later game considerations
After the early game, there is no cookie-cutter prescription for what to do as AR. The situations will vary far too much for that.
Probably the best situation for you (that is likely, anyway) would be to be part of an alliance against a distant leader, one of a group of more or less peers. Your allies need you too much for them to turn on you or let you be killed, allowing you to get through to the late game where your minerals will dominate. In this case, you need to play a very machiavellian game. Let the enemy pound your allies if they are starting to become dominant; keep the balance of power stable. If the situation becomes unstable (i.e. your alliance beats the enemy and takes their space), then your alliance will fracture, and you don't want that to happen until you are ready.
A more likely scenario would be that you are on the front line versus the enemy. Your allies do want you alive, just small enough that their resources are the ones that turn your minerals into a win -- for them. This situation must be played very carefully, giving your allies minerals but driving a hard bargain in terms of getting tech and planets from them. Generally it will be easier to get tech, especially the less useful techs (prop and elect), so concentrate on that. Remember that to you, resources are precious, much more than to them.
The midgame is the most dangerous era for an AR. This is the period others have many more resources than you, and have caught up with your earlier tech lead (at least in weapons, and maybe other fields). They can therefore field more ships than you can.
Packing planets -- similar to the case of normal races, ARs should first pack out very low value greens. However, it is for different reasons; normal races should always pack their high value planets last because of their pop growth rule. ARs pack out <25% greens because of the minimum value on hab for the purposes of resource generation. Unlike a normal race, it does not matter how you pack higher value planets (>=25% greens, that is) with an AR; from low to high value planets or from high to low. Just do what is convenient in terms of where the planets and minerals are.
Reds -- ARs use reds better than most other races, so grab any free planet you can get. Bring four large freighters of pop, if possible, and build a dock in the first turn. Then queue a death star with a 300/500 gate (use a dock if you don't expect to keep the planet long), and forget about the planet for a while. When the base completes, find a free fleet of superbugs and gate it over to mine the planet out as fast as possible.
Tech Trading -- You should always trade tech if you can, be you AR or not. Most non-AR races will take cheap weapons and other tech expensive, so you can very frequently trade energy and/or con for weapons. Even races with good tech will rarely have superior energy to yours, so you can almost always trade energy for something.
Late in the game, with an AR, you can expect a time to come when you can outproduce the rest of the galaxy combined, since they have no minerals left and you do. Even once this happens, you still may be in danger -- they have their fleets in being, and may be able to stomp you before you can build up enough to match theirs. But beyond point, the game will be more or less won. You just need to mop up.
I am not discussing this part of the game much, because the warfare is much like the standard Stars war. "Mahan" style, you take your massive fleet and go to the enemy, destroying his planets until he is desperate enough to commit his fleet, or until you have so much force that you can beat his fleet even spread out.
What I am more interested in, is the style of warfare before this point in the game is reached. That is, for an AR that has made it into the midgame. Your tech might be 18/20/12/17/12/9 or so. You have starting making superbugs in many places and have minerals to spend, but so do your enemies. People are starting to make noise with BBs.
ARs are, by their nature, factoryless. They grow pop better than any other PRT. They use reds well, and should typically be designed so that with high terraforming tech, between 50% and 100% of planets will be yellow or green. Furthermore, until the very late game when their mineral edge finally becomes dominant, they will always have fewer resources/lower tech/fewer ships than normal races. This means they fight at a disadvantage, typically, in terms of fleet size, though they can hope to be part of an alliance so that combined fleets are more or less balanced. But that does not always happen.
All of these things point at one style of operational warfare for the AR: a spread out war on the map, denying the decisive battle to the enemy while fighting him in a population/resource war. This strategy of warfare has been dubbed the "Mao" strategy by Jason Cawley, in a excellent article on it that can be found on the net (search google for "Epworthian"), and in a modified form in the official strategy guide.
(The funny thing about this article, is that it is really fairly useless for most races in planning strategy. Most races rely on factories for the majority or vast majority of their resource production; for these races, factory sites need to be defended. Cawley's race in the article in question happened to be perfect for a Mao strategy, which is why he did it. But it is not something you will get to do every day in stars, unless you play factoryless races including ARs.)
Rather than repeat everything that Cawley says in his article, I suggest that you find it and read it now. Go on. And now, I will merely comment on some differences between how the AR fights a Mao style war, and what Cawley did with the Saxons.
First, the Saxons had a big advantage that ARs don't have: they did not die with their battlestation. So, the Epworthians had to bring lots of bombers, and even then had to spend a lot of time bombing. This slowed their attack. And it allowed bomber-killing kamikazi actions against the bombers, by the Saxons and his ally.
As an AR, you should expect to lose planets to a superior enemy very quickly. The enemy should be able to kill at least one planet per turn, and probably more if he guesses well, sweeps aggressively, and is willing to take losses. Therefore, you need to be able to colonize at least two planets a turn, to make up for your losses. However, you can do that. Keep a reserve of one or two colony ships at each planet. And keep a few large freighters at each planet near the front.
Whenever the enemy looks like he might destroy one of your starbases, take inventory of the local situation. First, determine if you are likely to win the battle. If not, the freighters should be used to remove any minerals, especially germanium, from the planet, and population after that. Deny germ to the enemy especially, and he will have a hard time consolidating his advances.
If you are going to win a battle, but lose the starbase (i.e. the enemy is sending in kamikazis to kill bases), then you should gather all the pop possible into the local freighters but don't move them. Order a colony ship to the planet from a nearby planet (so that you get waypoint 1 colonization). If the enemy does not attack, you lose little, just that turn's resources from the pop you lifted. If he does attack and kills the starbase, then *after* combat, the colonizer coming in will put up a new orbital fort. You can then load up to 500000 pop on it, and queue for the first build a space dock. The next turn, you can put on whatever more pop you can arrange to have there, up to 3M, and queue a death star.
Don't be too concerned with losing population, with most ARs (low growth ARs, though, should give more thought to saving pop). Because of the square root effect, and AR can afford to strip off half the pop from all of his planets if he needs it for recolonization.
One point to note about Mao-style wars with AR: unlike the Saxons, who were truly decentralized, an AR *does* have a central point that is vulnerable: his home planet. And the mining fleets themselves may be vulnerable, especially for a non-ARM AR. (This is yet another reason that superbugs are great.) So it is generally a good idea to play in places large enough that homeworlds are not packed together. But note that until the very late game, you don't have to hold your homeworld; it is nice but not necessary, because there should still be plenty of minerals in remote places.
There is an upside to the "mine vulnerability" gap, though. And that is that a race like the Saxons *does* have the problem of reestablishing mines on planets that get bombed out, when it retakes them. Losing 1000 mines costs 3000 resources. In this sense, an AR (especially with ARM) is again a sort of super factoryless Maoist: more vulnerable, but also faster to recover.
On the other side of the coin, your offensive operations. Your goal here is threefold: to pull the enemy fleet away from killing your planets; to destroy factories; and to gain space for yourself. All of these will be accomplished in the normal mode of attacking. Let's look at that.
First off, to attack you need at least a bit of battle strength. But it does not really take that much. A bunch of light and cheap mine sweeper DDs. Just enough of a warfleet to take an orbit, and enough bombs to kill a planet in a turn or three. As is always the case in these things, the exterior circumstances matter a lot. Does he have penscans? If not, things are easier for you. Does he have light forces present? Mines? Etc.
Your general idea, on the attack, will be to take risks with small forces to try to knock down bases so you can bomb. Your bombs, even against good defenses, cost him serious resources, and germanium. Once you get the defenses lowered, you get to add population to the list of things endangered.
For instance, consider a light force with enough cherry bombs to destroy an undefended planet in a turn: that would be 40 cherries. That would also destroy 400 installations on an undefended planet. Against maximum planetary shield defenses, it destroys 210 installations. For an average sort of planet, with maybe 500000 pop and 12/9/16(checked) factories and 10/3/16 mines, that would be on the order of 90 factories and mines, and 11 defenses destroyed. The total cost to the enemy: whatever the starbase cost, plus 1245 resources, plus 55/55/325 iron/boron/germ. So you see, if you capture orbits, you *can* hurt the enemy quite a bit, without needing to take the planet or even be in orbit for more than a turn. However, to win you have to be hurting the enemy faster and more than he hurts you.
It is much better for you, if you can arrange things to keep orbits captured for long enough to bomb out the enemy. (Of course that is not always possible.) Unlike other races, you have no interest at all in leaving any installations standing on the planet; they can only help the enemy. So, if you decide to go with B-52 bombers, or two designs, then going with LBUs as your main skirmishing bomber is a good idea.
In a larger galaxy, I would recommend using minibombers with LBU-45s as your main bomber, supported by B-52s with cherries to finish off planets with your main fleet, when it attacks. The LBUs can skirmish nicely; they are cheap, they kill lots of installations, and they can be gated about quickly. Spread them out into enemy space with light forces, and force the enemy to concentrate there. If he is putting forces there, they are not in your space killing your planets.
In smaller places, just use a B-52 design with mostly LBUs and one slot of cherries. 7 of them together is sufficient to kill an undefended planet, but should inflict serious economic hurt on the enemy even against full defenses.
One final difference between an AR and other races, even factoryless normal races, on the attack. An AR can very sensibly bring population with him on the attack, and gain ground there. Normal races have to bring a substantial amount of population to a planet to have much effect; i.e. assume you drop 55000 Saxons on a planet in enemy space. Well, they can put up a dock, perhaps, and turn whatever enemy minerals are left (probably little G to use) to make light ships. But they only have 55 resources per turn, initially, to do it with.
Contrast that with bringing 55000 AR civvies to the same planet. With energy 18, they would get 314 resources if the planet was 100%, or as little as 78 if the planet is red or low value green. With even fewer civvies at risk, the AR has an even greater advantage (i.e., with 10000 pop the Saxons get 10 resources; an AR gets 33-134).
Only if the enemy has no bombers or packet launchers around, should a normal race put hundreds of thousands of pop at risk in enemy space. AR can, and should, make much better use of this potential. Every planet withing 81 ly of one of yours that is empty, you should have your eye on to colonize. Send over a colony ship, and a large freighter with 100000 pop and some minerals. (If the enemy has no range-3 beamers with battle speed 2.25+, you can use galleons with overthrusters.) If you get the colony there, great, build a dock (they are very, very cheap), and build light ships until threatened enough that you have to run away.