3D Action & Classic Games/NEW.iso/gamedoc/disk#3/ascend1.sol

Text File | 1996-01-25 | 40.0 KB | 800 lines

ASCENDANCY ANALYSIS
===================

 12/20/95 version

 Mike Fay, Ph.D.
 [e-mail] or CompuServe 73717,250
 Atlanta GA

These are the results of a fairly deep analysis of how Ascendancy by Logic
Factory works under the hood, followed by a discussion of the game in
general. Comments invited; two heads are always better than one. Also, if
anybody wants to summarize all this, have at it; my writing can be pretty
thick! Just be sure to credit me.


 DISCLAIMER
 ==========

Like a lot of other folks, I found Ascendancy certifiably lame as a
strategy game. The producers made a nice, pretty shell for a game, and
could have done a lot more--but didn't. The AI is a certifiable moron, and
the game is wall-to-wall micromanagement. The Antagonizer has made the AIs
begin to be tough, but their poor management skill still leaves them in the
dust.

Still, I like to analyze numbers. And I suppose I'm just desperate for a
good space strategy game. Thus, I did this analysis.


 DISCLAIMER # 2
 ==============

This report was largely written in mid-November, before the Antagonizer was
released. Although I have stuck in a few notes here and there, not all
changes due to the Antagonizer are likely to have been incorporated. In the
meantime, while trying to polish this doc, I got into numerous other
projects (as usual) and pretty much lost interest in Ascendancy. Therefore,
you will have to read this entire document keeping in mind that some things
*might* no longer apply. I would be willing to bet that few or none of them
have, however.


 BEFORE WE GET STARTED
 =====================

What about the red, green and blue squares? How do they work?

Unlike what I thought (prior to analysis), they do not *multiply* the
effects of the structure on them, they simply add one point if an
appropriate structure is on them. So it makes no difference, for example,
if you put an Industrial Megafacility, Metroplex, Factory, or Colony Base
on a red square: As long as the structure produces industry, it will add
the one point from the red square to your raw total Industry Points.

As a result, it makes a lot of sense to put your Colony Base directly on a
red or green square. This way, you get its benefit right off the bat. I
recommend a green square, if you have a choice, because Prosperity is
generally the more challenging aspect of colony growth.

The exception would be planets with a lot of black spaces--a small or tiny
planet with any black spaces, a medium with a fair number of black, or a
larger planet with PILES of black spaces. Here, the more pressing problem
is simply a lack of space, so grab a little right off the bat by putting
your colony on a black square. You'll get to the colored square soon
enough.

Note: Red squares do not increase the *multiplicative* effect of a
Hyperpower Plant. It might still add one point; I didn't test enough to
tease this out. But I know it doesn't multiply. So, there is no particular
advantage to putting a Hyperpower on a red, an Internet on a blue, etc.

Note #2: In the following, I frequently use the terms "raw", "observed",
and "actual" when referring to Research, Industry, and Population Points.
To make sure there is no confusion: "observed" is always the same as
"actual"; both mean what you actually see when you click to see your
Industry IPs, for example. "Raw" is what you get when you total up how much
your people and structures input. Read on...

 RESEARCH, easiest and therefore first
 =====================================

Research simply adds up the research points for your research structures
(see below). As usual, one point is added for each research structure on a
blue square.

Internet adds 50% more points to your raw research.

Example:

  #  Type                Points  Adds
  2  Labs on white        2x1      2
  2  Labs on blue         2x2      4
  2  Res. Camp. on white  2x3      6
  1  Res. Camp. on blue   1x4      4
                                  --
                         Total    16

            x 1.5 for Internet    24 total RPs for this planet

** RESEARCH BUILDINGS

 RP=Research points, IP=Industry points, PP=Prosperity points:

                      Cost   $/RP   Points
 Laboratory             50     50    1 RP
 Research Campus       160     53    3 RPs
 Metroplex             200    200    1 RP, 1 IP & 1 PP; +2 spaces
 Engineering Retreat    80     80    1 RP & 1 IP
 Logic Factory          80     80    1 RP & 1 PP

Obviously, the Research Campus will let you pack the most RPs into a planet
in the long term. Other buildings are shown just for completeness
(everything that makes an RP); not necessarily because I recommend them. I
never make Retreats or Logic Factories.


 INDUSTRY, a little tougher but not too bad
 ==========================================

Industry starts off with your raw IPs (Industry Points), much as for
research, but modifies them using an exponential function:

  Observed IPs = Round ( (Raw IPs)^0.85 )

This exponential leveling causes the amounts of IPs produced to drop off as
the raw IPs increase. You get a 1-to-1 raw-to-observed only up to three
IPs, then it starts slacking off.  The curve drops off quickly then is less
severe. For example, when going from 1 to 2 raw IPs, you only add 0.803
observed IPs (80%, but rounding obscures it); from 3 to 4 raw IPs adds only
70.3% observed (or actual) IPs (and observed stays at 3), from 9 to 10 raw
IPs adds only 0.606 (observed goes from 6 to 7), by 30 raw IPs each raw IP
adds only about 0.50 actual IPs (now in the 18 observed IPs range), by 140
raw IPs each new one adds only 0.40 (in the 67 observed IPs range), by 360
raw, each adds 0.35 (the 150 observed range). So your facilities always add
something, but it is progressively less. IMFs already add only two actual
IPs instead of their three raw IPs by the time raw is at about 5, and are
down to 1.5 actual at around 30 raw/18 observed. It approaches 1 observed
way out past 400 raw/163 observed. (Note, all the preceding are WITHOUT
Hyperpower; see below--multiply observed by 1.4 for Hyperpower.) See
Appendix A for a detailed tabulation of Raw vs. Observed IPs.

As usual, red squares just add one Raw IP to your total, if an IP-producing
structure is on them. It does NOT multiply what is on the square; there is
no difference between putting a factory, IMF, or colony base on a red
square.

If you want to calculate your IPs, you have to examine your colony and add
up all your Raw IPs, then run it through the above equation. (Or just use
Appendix A as a look-up table!) I tested this function with dozens of
datapoints from one of my games, and it seems very predictable.  There were
some very minor variations of +/- one point that seemed to be due to some
rounding within Ascendancy's algorithm; they only happened when the decimal
remainder was very close to one half (.47 to .53), and went in
unpredictable directions. It only caused my sample to be off 3 times, so
the correlation seems very good.

** INDUSTRY BUILDINGS

 RP=Research points, IP=Industry points, PP=Prosperity points:

                      Cost   $/RP   Points
 Factory                30     30    1 IP
 Ind. Megafacility     110     37    3 IPs
 Metroplex             200    200    1 IP, 1 RP & 1 PP; +2 spaces
 Colony Base          (120)   120    1 IP & 1 PP
 Engineering Retreat    80     80    1 IP & 1 RP
 Shipyard              240    240    1 IP
 Orbital Docks         170    170    1 IP

Obviously, the IMF will let you pack the most IPs into a planet in the long
term (see Discussion of Planet Packing). As usual, this is everything that
makes an IP, not necessarily something recommended *for* making IPs. If you
are trying to use or check my equation, don't forget to add in IPs from,
e.g., your Shipyard, Colony Base, Orbital Docks, and Metroplexes, including
adding a point for each one on a red square.

 EFFECT OF HYPERPOWER
 ====================

The equation seems to be:

  Observed IPs = Truncate ( (Raw IPs)^0.85 x 1.4 )

In other words, almost exactly Observed IPs plus 40%. It is important that
the 1.4 is outside the ^0.85; you wind up with a bit more (because it's not
"held down" by the 0.85), especially with a high number of IPs.

I didn't do a whole lot of sampling on this one (about a dozen datapoints),
because it seemed very straightforward. However, it definitely suffered
worse from the rounding error (i.e., observed was randomly +/- 1 versus
predicted). For one of the planets I sampled, my predicted was correct if
values were *rounded* instead of truncated, but that threw most of the
other samples off by one. "Shrug."

 SCIENTIST TAKEOVER
 ==================

This appears to be a straightforward calculation:

 *ADD* to existing RPs:  TRUNCATE  (Raw IPs)/4

In other words, ST adds (not replaces) 25% of your Raw IPs to your Research
IPs.

If you have Hyperpower:

 ADD:  TRUNCATE  1.5 x (Raw IPs)/4

In other words, Hyperpower adds 1.*5*, not 1.*4*, to ST, unlike how it
usually affects IPs.

We don't see the 0.85 that enters into the observed IPs equation here; just
your raw IPs, which means your IPs are used better (linearly)--30 to 50%
better, depending on how many IPs you have (see discussion of observed vs.
raw IP drop-off, above).

[If you don't want to add up your raw IPs, you can always use Appendix A as
a sort of "look up table"--find your Observed IPs (with or without
Hyperpower) and slide over to the left to get a raw IP estimate. (It's an
estimate because more than one raw IP can have the same observed IP, after
rounding.)]

Also note, Internet does NOT help (i.e., increase) Scientist-Takeover RPs.

The net effect is that ST is a very poor substitute for "real" (Research
Campus) research. Each colonist devoted to an Industrial Megafacility (IMF)
generates 3 raw IPs which converts to 0.75 RPs vs. 3 for a Research Campus.
Best case: an IMF on a red square with Hyperpower generates 1.5 RPs
((3+1)x1.*5*/4), but a Research Campus on a blue square with Internet
generates 6 RPs. So Scientist Takeover is definitely just something
optional to do with otherwise wasted IPs; it is *not* somehow a better way
to make RPs.

Actually, I suppose it is marginally better if you have IMFs and/or
Hyperpower before you get Research Campuses and Internet (so you're stuck
with Labs), but at that point in the game I'm usually furiously building
stuff, and otherwise just putting Labs on all blue squares.

Note that I did not test whether having industry on blue squares helps you
with Scientist Takeover. I would guess that it wouldn't, though. And--"what
does it matter". :)

 PROSPERITY POINTS (pretty messy)
 ================================

First things first: it takes 50 Prosperity Points (PPs) to give birth to a
new colonist.

 HOW PPs ARE CALCULATED
 ======================

Unlike Research and Industry, where I could just take off structures one by
one and generate a long list of datapoints, most of my colonies only had a
few to a dozen Prosperity points, no matter what their size, so there
wasn't a whole lot I could take off. Neither was there much room to avoid
rounding ambiguities, since all observed PPs were low (max 12 or so; most
in 1-5 range). I wasn't able to make my predicted algorithm real pretty;
there may be a number of roundings compounding within Ascendancy's calcs,
or I may have my algorithm slightly off.

That's the problem with black-box reverse engineering: There are a dozen
ways to do anything mathematically. For example, their program might walk
through each of the structures on a planet, compounding something as it
goes--this could make an exponential-like function, while possibly also
introducing subtle rounding differences, depending on their level of
precision. While it makes a reverse engineer wince, they had no obligation
to make their algorithms precise and predictable to the Nth decimal for
reverse engineers; they just had to make the game playable. So, who knows.

Anyway: The best I can figure the Prosperity Points equation is:

 PPs = Round( (Raw PPs)^0.85 - Round( (0.4 x Pop)^.85 ) ) + 1

 (Pop = Total Population, used and unused)

In other words, add up your raw PPs on one hand, but subtract a PP for
every 2.5 people, too -- and muddle both with 0.85 exponentiation. If you
compare your planets' populations and raw PPs, you will see that you only
get observable PPs if you have more raw PPs than people/2.5.

I had 63 datapoints (planet results from the game), and this equation got
all but 9 correct; for 7 of these, my calculated value was one less than
the observed value, the eighth value was one more, and the ninth, two more
than the observed. (Yes, I triple-checked the latter planet to make sure I
got its data right, sigh.)

With this kind of equation, there are any number of ways to play around --
round or truncate before or after exponentiation, add 1 inside the first
exponent, subtract it inside the second, etc. etc. etc. I feel confident in
my general approach toward Ascendancy under the hood, since the IP equation
was such a good match. But this is the closest I could get to PPs before
getting tired of trying yet one more variation (I tried dozens). See
Appendix B for a table of PPs, but be warned, 10% of the time my result was
one lower than I observed, and a couple of times it was off by more.

FWIW, I did test for effects from Industry, number of buildings, and used
vs. unemployed people. I could not see any real effect (try it; just take a
saved game and whack off lots of stuff). Some things are quite difficult to
test, though, such as the number of free spaces left in a planet. If you've
got the time, let me/us know if you can pin it down better!

In summary, Prosperity points are like a "tip of the iceberg" effect: Your
raw PPs have to rise to the point where they are higher than people/2.5
(since both have the 0.85 factor). Then, raw PPs suffer from the same
exponential reduction as IPs. Thus, if you have 9 raw PPs, a Hydroponifer
only adds approx. 1.8 observed PPs (3 x 0.6 raw PPs).

** PROSPERITY BUILDINGS

 RP=Research points, IP=Industry points, PP=Prosperity points:

                      Cost   $/RP   Points
 Agriplot               30     15    2 PPs
 Art. Hydroponifer     100     33    3 PPs
 Metroplex             200    200    1 PP, 1 RP & 1 IP; +2 spaces
 Habitat               160     80    2 PPs; +3 spaces
 Colony Base          (120)   120    1 PP & 1 IP
 Logic Factory          80     80    1 PP & 1 RP

As you can see, the Agriplot is a *much* better deal than the Hydroponifer.

***News Flash:  I initially wrote the Prosperity section without having
realized how easy Automation can make things, if you have it and want to
spend the bucks. So, read this all with a grain of salt if you lean toward
Automation. A discussion of Automation has been put at the end of the
Prosperity section.

 FERTILIZATION PLANT
 ===================

By this point, I was getting quite tired of sampling data from the game and
playing with it, only to find something always off by one <g>. Anyway,
based on nine planets, the effect of the Fertilizer Plant (FP) appears to
be that it increases the raw exponential PP half of the equation by 40%,
much as Hyperpower does. Then, the Population factor is subtracted. If you
*really* want to see the best (god-awful) function I could come up with,
here it is:

 TRUNCATE  (  ROUND( (Raw PPs)^0.85 ) * 1.4 -       [Raw PP part]

              ROUND( ROUND( 0.4*Pop ) ^.085 ) + 1  )  [Pop part]

With this equation, my predicated PPs were one less than observed PPs for
two of the 9 samples; the rest were dead-on. Here, though, the frustration
of tweaking can be seen clearly. There must be dozens of permutations to
try, and however I flipped these around, a different set of datapoints
would be off by one. So, the general observation is that the FP increases
PPs by 40%, but exactly how it rounds, I can't say.

Throughout this article, I make no claims to have totally figured out what
the algorithm is doing. I just know in general what's going on, and may
have actually hit a few of them pretty closely <g>.

The above shows why you can see something strange with the FP: It has
little apparent effect on some planets, but lights a few planets up like
Christmas trees. A big change means you had a large PP base (i.e., a large,
developed planet) that was just barely at the Population/2.5 hurdle. When
the FP was added, "the rest was gravy"--the 40% shot the large PP base way
up over the Population factor. On small planets there is not enough PP base
to have a big effect (or observed PP would have been large already, given
the small Population), and it will have no observable effect on large
planets that still don't have enough raw PPs to make it to the
Population/2.5 cutoff.

Although I haven't looked at it closely, I would imagine that the FP starts
being a viable proposition just like the Hyperpower plant: When raw PPs are
around 9, the FP is equivalent to adding another Hydroponifer, and above 9
raw PPs, the FP does better (for the one colony space utilized) than a
Hydroponifer. But, since your Pop may be masking your PPs, you will have to
look at your colony to estimate how many raw PPs you have. Appendix B can
also be used to estimate it, *if* you have observable PPs.

 CLONING PLANT
 =============

What it says: It just increases your Pop by two, instead of one, when each
new colonist is born. An alternative way to build your pop. It still
requires enough raw PP base to get your observed PPs above the Pop/2.5
cutoff, so it is not a "magic bullet" for the problem of a highly developed
planet with no PP base that needs to increase its Population.  Its
advantage is that it does not require ever more and more Prosperity
buildings to double what little you might have on a large planet; it just
needs the one clone plant, and all your Pop increases are doubled. It's a
bargain in anyone's book; indeed, it may even be "too powerful" (wasting
excess generated Pop) in some circumstances--but, who cares?

A Cloning Plant is probably best used for situations where rapid growth is
needed, such as very large planets with lots of space to fill.

 ENDLESS PARTY
 =============

If I thought Fertilizer Plants were bad, Endless Party (EP) was the worst.
Like a drunk who couldn't remember what he did the night before (but was
sure it must have been dumb), by this point I had swum through so many
numbers I didn't want to see another one--and I still had EP to do.

I took 15 data points for EP. The equation is something like:

 ROUND ( (raw PPs)^0.85 + (Observed IPs)/3 -

          ROUND( (0.4xPop)^0.85) ) + 1 )

If you look closely, it's simpler than it seems: It's pretty much the
regular PP equation, but adds in PPs equal to observed IPs divided by
three. (After all, the description of EP basically says it converts IPs to
PPs.) You still subtract out the Pop/2.5, so you can still have masking.
This equation was correct for 10 of the 15 datapoints, and was off by +/- 1
to 3 for the other five. (This was for observed PPs, with EP, ranging from
0 to 23. That both the highest value and many of the lowest ones were
precisely correct tells me I'm on the right track.) So it's close, but not
totally there. You are welcome to tweak the equation all you want. Just
make sure you get all the input data right! <smile>

You will remember that Scientist Takeover is based on Raw IPs, as opposed
to EP, which uses Observed IPs. It's consistent, though, with how RPs are
generated linearly, but PPs attenuate by ^.85.

Sorry, but I did not test for the effects of the Fertilizer Plant or
Hyperpower on Endless Party. I know informally that Hyperpower increases
it; I can only imagine, by the 40% it increases observed IPs. I don't know
if the FP increases it, since Internet did not increase the effect of
Scientist Takeover (there was just a straight conversion of IPs to RPs).

PARTY OR PRODUCE, the perennial question (for me, anyway):

Should you build more Prosperity structures or just say "screw it" and
start Partying, if your colony has gotten fairly advanced and you're short
of PPs?

* * BIG MATH WARNING * * -- Skip all the way down to the "Conclusion"
paragraph at the end of this section to avoid a lot of numbers!

It takes 100 observed IPs to make a Hydroponifer, which provides 3 raw PPs
on a continuous basis. These 3 raw PPs will be decreased by exponentiation
to approx. 50 or 60%, depending on the size of your raw PP base. In other
words, the Hydroponifer will actually contribute approx. 1.6 observed PPs
on a continuous basis.

(I'm discussing Hydroponifers instead of Agriplots because, using
Agriplots, your colony is probably well advanced before you start running
into Population problems. At that point, you will probably have plenty of
Industry but little space. So the question will probably be whether to make
more Hydroponifers, not Agriplots. Increase the 1.6 PPs to 3.3 PPs per 100
IPs, if you are considering whether to make more Agriplots to increase
population.)

By comparison, with Endless Party, 100 observed IPs would contribute 33
actual PPs (before subtracting Pop) on a one-time basis.

Obviously, Partying is a *very* cost-effective way to increase PPs in the
short term. It would take approx. 21 turns (33/1.6) before the Hydroponifer
has "broken even" with Partying, and generated more than 33 actual PPs. Ten
turns if you're talking Agriplots. To understand the difference, let's look
at a colony in progress.

Colony Murgleblast is an enormous planet (73 spaces) with about 30
colonists, and about 40 left to go. The player has all relevant research
(IMFs, Hyperpower, Hydroponifers, etc.) done and has developed a strong
industrial nucleus. How to strategize the rest of the planet's population
development? Let's say this is a highly industrialized planet (best case
scenario), and the player has 80% of his Pop in IMFs at any given time.

If he used Endless Party exclusively, it would take 6,000 IPs (40 colonists
x 50 PPs/colonist x 3 IPs/PP) at face value, *but* this would be reduced by
the growing population. Here we go; skip to the next paragraph to avoid
this math: At 30 pop, subtract 8 observed IPs ( (0.4*30)^0.85 ); with 80%
in IMFs, he's making 53 IPs (30x.8=24; (24x3)^0.85 x 1.4 for Hyperpower).
So he puts 45 IPs (53-9) into EP, generating 15 PPs/turn, and gets a new
colonist in 3.3 turns. When he's up to 50 colonists, he's generating 82
IPs, subtracting 13 for Pop, leaving 69 for EP, or 23 PPs a turn: a new
colonist every 2 turns. When he tops out around 70 colonists, it's 109 IPs
minus 17 equals 92; divided by three is 30.7 PPs/turn, or a colonist every
1.6 turns. Doing a weighted average of the three snapshots (In case you
haven't gathered, by this point, I'm just writing stuff on paper; this is
not solid modeling off of a spreadsheet. I told you I'm getting tired of
these numbers!): The player devoted 10x3.3 turns of production to go from
30 to 40, 20x2 for 41 to 60, and 10x2 for 61 to 70 (but it doesn't let you
carry forward PPs *when generating a new colonist*, so anything below two
turns is wasted unless you have enough IPs to do it in one turn (thought
that would be 150 observed IPs, equal to 250 raw IPs with Hyperpower, or 83
IMFs--more than an enormous planet can hold). However, all through this
time, the player will also have been planting Metroplexes or Habitats,
which add some PPs already. Let's say that these reduce the needed turns
from 90 to 80 (that two-turn minimum cutoff is a bitch!).

In total, you will have used very roughly 80 turns of production to take a
planet from 30 to 70 colonists using Endless Party alone (with a minimum of
PPs thrown in from structures). Now let's look at using Hydroponifers. Skip
to the next "In total" paragraph to avoid the math. :)

First, the player wants enough PPs to be able to keep building constantly.
50 points are needed for each colonist, and let's say he wants a new
colonist every three turns. (If he's already at 53 observed IPs, he can
soon make a new IMF in two turns, but will also be sprinkling in
Metroplexes, which take 3 or 4 turns (2 turns at the very end of
development), Hydroponifers, Internet, etc.

So, he needs 13 *observed* PPs to make a colonist every four turns (50/4).
Since we're looking at the sum total, by the time he reaches 70 colonists,
he will need to overcome (0.4*70)^0.85 = 17 base PPs, and still generate 13
observed PPs. Based on the Prosperity equation, this will be about 52 raw
PPs ( 52^0.85 - 17 + 1 = 13). That's about 17 Hydroponifers (we're pressed
for space). Let's give the chap a break (and also account for longer
projects that let Pop grow by itself) and say we have a Fertilization
Plant. This adds 40% to the 52^0.85 part; to make 13 observed PPs, we then
need about 36 raw PPs, or 12 Hydroponifers. (This is for *70* colonists;
they will have been constructed evenly over time to keep up with Pop
growth.) By the time he reaches 70 colonists, he will have used 1,400 IPs
(12x100+200 for FP) to make his 12 Hydroponifers and FP. Using the same
53/82/109 IPs/turn weighted average from the previous discussion, the
player will have wound up devoting about 22 turns to Prosperity structure
development. Finally, there will have been quite a bit of Habitat and
Metroplex development that will have added Prosperity, so let's drop that
number to say 15 or 18 turns.

Conclusion: It is *much* better to develop and rely on your Prosperity
infrastructure *for long term population growth*, and much better in the
short term to use Partying to your full advantage.

Sounds like a good recipe for life to me! (ooh, bad)

 ===========
 GROWTH BOMB
 ===========

As far as I can tell, the Growth Bomb adds ten more population slots to
your planet, period. In very informal observations, it does not seem to
have any effect on PPs/fertility or how many black squares you have (IOW, I
didn't actually check this very much, but can't remember it happening).
Growth Bombs can only be used once, as far as I can tell.

One serious drawback to these babies: You CAN'T TELL WHO YOU'VE USED IT ON.
Ascendancy is stupid enough to let you use it again without warning. Your
two options would appear to be (1) keep an exhaustive list of Bombed
planets (holy shit, thanks very much LF!), or (2) build some cheapo thing
on the planet that you wouldn't normally build, that is a "flag" that This
Planet Has Been Bombed.

I make one cheap orbital shield in the lower left space slot of all planets
RIGHT after they have been bombed. This means you have to pay attention and
not fly through your "this planet made." notices real fast. I consider it
pretty important to always make the shield immediately, so that I don't
forget (I make myself a little note in cases where it can't be avoided.
with SO much micromanagement, it's triple easy to forget).

Making something in space means it doesn't take up important ground/IMF
space. Making it super-cheap means it's easy to crank out, and easy to
Automate. Finally, it never hurts to have shields around, especially with
the Antagonizer.

 =================================
 MY PLANET'S MAXXED OUT, NOW WHAT?
 =================================

If your planet is maxxed out with useful structures and you have a lot of
Prosperity infrastructure hanging around, you will ask yourself, do I
really need all those Hydroponifers any more? If it's a tiny colony,
probably not; scrap'em once you've used up all the space you can. Let's
look at the situation for larger colonies a little more closely.

An enormous colony has 73 spaces for people, and you will likely have a
couple of spare bodies hanging around, for a total of 75. (I keep a couple
in my closet, how about you?) I usually have approx. 25 Metroplexes (23-28;
maybe 17 Habitats) on such a large planet. So you already have 25 to 35
base PPs, while the amount subtracted for the Population is 18 (
(0.4x75)^0.85 ). This means just about exactly even ( 30^0.85 = 18 + 1; see
equation). So you're basically already set for instant pop-up of the
Population with either Endless Party or, for a more chronic situation (like
a Ship production planet), a Fertilization Plant.

Smaller planet sizes will work on a similar basis. The smaller the planet
is, the less Prosperity you will have built into Metroplexes (MPs) or
Habitats, but you will also have fewer Population. A large planet has 45
spaces + 2 = 47, and I generally have about 16 MPs on full Larges. This
works out to about 12-12+1=1 observed PP, for a similar situation. So,
scrap all your Agriplots and Hydroponifers when your larger planets are
maxxed.  This will be a little tighter if you use the Growth Bomb because
it doesn't require you to build more structures that provide prosperity.
(It might be severe if you are the Mebes; I don't know how much their
special ability increases the max colony size).

 PROSPERITY WRAP-UP AND ETC.
 ===========================

As your Pop goes up, you must continually build your Prosperity base.
Agriplots are *much* better deals for the buck *early on*, so use them
liberally until available planet space is maxxed. I watch my observed PPs
and if my next Factory (or whatever) will be ready before my next person,
bam: another Agriplot. (A.k.a., "farm in advance" so it's working for you
while you construct something else.)

Once the initial planet space is maxxed, you probably have a fairly
powerful Industrial base (25 observed IPs, possibly much more) that can
crank out the more expensive Hydroponifers and other Prosperity structures
much better, when the squeeze is on space. And, as each old Agriplot is
abandoned, a new person is freed up for another Prosperity structure (or
whatever). As shown, 12 observed PPs is a great clip to keep your
Population rolling along--possibly a bit too good. Just look at what you're
building with a planet; how long it takes to build the Campuses or IMFs
you're concentrating on (with Metroplexes, Hydroponifers, and/or Habitats
thrown in), and have enough observable PPs that you will be generating 50
PPs every x number of turns it takes you to make whatever you're making, on
the average.

As your planet maxxes out and you want to cut out most of your Prosperity
structures, you might find the Fertilization Plant a better choice than the
Cloning Plant. If your observed PPs (without an FP) are hovering right
around 0 or 1, as shown above, a Cloning Plant could mean a long time is
taken to make another person, although it will be two when it is made. A
Fertilization Plant on top of PPs hovering around 0 or 1, though, means
about 6 or 7 observed PPs, for Pop growth every 8 turns or so. Buttressed
by a planet-wide Party and it will likely just be 2 or 4 days. (WHAT are
they DOING when they PARTY?)

For maxxed-out Ship-producing planets, try to leave *two* unused colonists
around (via Metroplexes, etc.). Then, when each Ship leaves and takes a
colonist for crew, you will immediately have another available--no waiting
for the PPs to add to 50--and you will meanwhile be producing the
replacement colonist while working on the ship, assuming you have a modicum
of PPs. Much smoother than the other way around. (Why don't they roll the
50 forward and let somebody appear as soon as one leaves, if a planet is
maxxed??)

Another recommendation is to put your Colony Base on a green square,
whenever possible. This gets your Population building faster right off the
bat (*while* you work on other things), and continues to contribute for the
rest of time.

Unlike IPs, PPs can be masked and carry over from previous turns. Thus, a
brand new colony always says it will take 50 turns to make a new colonist
(25 if the Base is on Green). But, later in the game, you might drop to
zero PPs, then later come back up. You might then see something like "3 PPs
per turn; population will grow in 2 turns". In this situation, previously-
generated PPs are being carried forward.

The thing that struck me after this work is how strongly Prosperity is
likely to take a hit if you don't develop it smoothly. Remember, Prosperity
works *against* what it is *for*--every 2.5 colonists eat another PP. Any
time you stop building Prosperity, growth will eventually halt. Indeed, in
my first game (before this analysis) my highly developed planets were often
waiting on Pop (Partying their fool heads off) so they could increase. By
using Agriplots liberally at first, then keeping a better eye on it, it's
not nearly so much of a problem this time around.

 ==========
 AUTOMATION
 ==========

After having written all the above, Viper on CompuServe made me realize
that I had overlooked the benefit of automation. I had gotten it late in
the one game I had completed, and I have yet to reach it in my current
game, so I hadn't really had much chance to play with it.

In a nutshell, automation is a simple way to stop worrying about making
space for more people (so you can put structures down on all the squares of
a planet). With Automation, you can simply automate another structure when
your population maxxes out. There is no reason to have as many "population
holes" as there are squares on a planet.

Automating a structure costs the same amount as the structure does (e.g.,
an IMF costs 110 IPs to automate). Therefore, it makes beaucoup sense to
automate your cheaper structures first. For the successful emperor, this
usually means IMFs. And, while you might automate very cheap structures
that you are going to remove later (such as agriplots), you wouldn't want
to automate expensive structures that you'll remove later (assuming there
are any cheaper structures around!). In general, though, there will
probably be plenty of permanent structures.

Let's look at the economics.

The preferred population structure is the Metroplex (MP). If you were to
build a planet to full capacity using MPs, you could fit in two IMFs for
each MP. That's 200 plus 2x110 = 420 observed IPs per three squares, plus
some IPs sunk into population growth, one way or another. Final cost,
approx. 450 observed IPs. This configuration produces seven raw IPs and
approximately four actual IPs (six with HPP) for every three squares in a
largish colony.

In contrast, you can use automation to get three IMFs on those three
squares. That's 3x110 plus 3x110 for automation = 660 observed IPs. These
IMFs will make nine raw IPs or something over five observed IPs (approx.
eight with HPP) in a large colony.

In summary, the MP route costs 450 IPs and makes six, while automation
costs 660 and makes eight. So, the automation route has about 200 observed
IPs to amortize. Since it makes a surplus of two IPs, it will take about a
hundred turns to break even.

Given the nauseating length of this game, probably automation is the better
buy in the long run. :) Just be sure to automate your cheap, permanent
structures first.

 ========
 OUTPOSTS
 ========

On CompuServe, Viper (among others) argued with me that Outposts did have a
use in the game. Perhaps I am dense, but I simply see no usefulness for
Outposts, and _never_ make them. Here's my reasoning.

Outposts give you another population slot (period), but they take up a
surface square and cost money. Usually at this point in the game, you are
hard up for all your bucks; meanwhile, there is always something more to be
done with your IPs, while waiting for better population-control technology.
You'd undoubtedly want to rip up your Outposts later. To me, it smacks of
IPs and time poorly spent, at a critical growth time. Therefore, I
concentrate on other stuff and don't make them.

 ===========
 AUTOMANAGER
 ===========

The "automangler" does exactly that: Mangles what your colony will build. I
never use it. What a shame this is not better; not even in the Antagonizer,
I hear. as usual, LF won't tell the full poop on what they did in the
Antagonizer's planet AI, although folks say it will use colored squares
better now. The AI would probably be much more of a challenge if only it
could build worth squat.

I have heard some folks say that they will set their smaller, unimportant
planets to Automanage, and concentrate their attention on better ones.
Sounds good to me.

 =============================================================
 END OF MATH & FINDINGS, BEGINNING OF GENERAL STRATEGY SECTION
 =============================================================

 GENERAL OBSERVATIONS
 ====================

** ALERT ALERT **

Most of this FAQ was originally written prior to the release of the
Antagonizer, and I simply don't have the oomph to play this game more (I
have six recent blockbusters sitting unopened on my desk!). So, take it all
with a grain of salt. Most of it is still relevant.

 ASCENDANCY, A STRATEGY GAME? I THINK NOT
 ========================================

Like most of the folks on Internet c.s.i.p.games.strategic, I have found
Ascendancy an absolute pushover. The AI is a total moron; by around Turn
1,000 of a regular-density game, I hold most of the planets in the galaxy
while the AIs only have 3 or 5 each, and half of my planets individually
produce more than all of any given AI's planets. While you may not do quite
this good (or may do better!), suffice it to say, You may not be a very
experienced strategy gamer if you lose to the Ascendancy AI <smile>. (They
have promised an update to the AI; we'll see that when we see it.)

That having been said, I have still gotten pleasure out of playing the game
as a software toy, like SimCity. I like to play with numbers (aw, who told
you??), and it lets me do that. For those who are still hanging in there,
here are my notes on playing the game. Discussion welcomed!!

 GENERAL STRATEGY
 ================

 PLANET DEVELOPMENT
 ==================

One of these days, I may write a little program to model the ideal planet
development. It's hard to guess exactly how everything works together.
Until then, here's what I do.

Build up Industry as soon as possible, but *keep laying down Agriplots to
keep pace*. Usually, an Agriplot is the first structure I put down, even if
the Base is on a green square. (I'm not totally sure this is the best
route, though.) For an all-white planet, I build something like: Agriplot,
Factory, Agriplot, Agriplot, Factory, Agriplot, etc., gradually edging over
toward industry, but making sure I always have enough of the cheapo
Agriplot Prosperity.

Switch over to IMFs (Industrial Megafacilities) as soon as you get to 3 or
4 IPs. (From here on out, I mean Observed Points--what you see on the
screen--unless otherwise indicated.) Once you get to about 9 or 10
(observed) IPs, make a Hyperpower.

When you make your first IMFs, there will be a long lull where your
population gets a chance to grow. I usually find myself plopping down one
more Agriplot to get PPs up to 3+ before plunging into IMFs, so the little
population dudes will make plenty of babies during construction. (Reminds
me of the Pepsi commercial with the office girls watching the construction
worker.)

ALWAYS be laying down your structures so as to build toward special
(colored) squares, even if you don't currently need them--be there for them
when you do. ("I was there for ya, man! Where WERE you??")

During lulls in production (waiting for Pop to increase), plop down a Lab
or Research Campus if blue squares are available. Both take longer to build
and so will let your Pop grow in the meantime. Once you have Research
Campus technology, only put *them* down, *once* your planet has decent IPs
(at LEAST 10, preferably 20+). Prior to Research Campus technology, make
Labs for blue squares at the slightest hint of a wait for Population. (You
need all the research help you can get if you don't have Campuses yet! Use
those blues!)

If