The "Monster" Pantheon

[Barry Kearns]

I'd like to start a new thread, designed to show players the specifics
of races in different PRT categories, all of which can do very well.
I'll start the ball rolling with what some will likely find an
unexpected surprise.

I ran a test yesterday on an AR race that I had been working on
earlier (thanks to someone for reminding me that I hadn't followed up
on that... :). The results confirm my feeling of overall balance
between the PRTs.

Alternate Reality can also be firmly placed into the pantheon of PRTs
which can be turned into a "monster", by which I mean that they can
achieve in excess of 25k by 2450, as well as having good weaponry to
show for it along the way.

The test that I ran was in a small, packed universe with standard
minerals, and Accelerated BBS play. There were no opponents. I
wanted to test the general viability before tuning it for play against
opponents.

I performed a reasonable (but not excessive) level of MM for the first
35 years, and then just let it sit-and-spin from there, doing nothing
more than researching, upgrading the orbital hulls, and building
remote miners. I was quite happy with the results. I'm sure better
results could be had by 2450 if I was willing to go back and continue
the MM, but I was interested in the general concept, rather than
devoting a bunch of time to the test.

The resource results:
2331 in year 2420 Tech 10-5-4-5-3-3 (the first good sign)
3949 in year 2425 Tech 10-10-4-7-3-3
5527 in year 2430 Tech 10-10-4-12-3-3 (Ultra-stations)
10k in year 2435 Tech 14-10-4-12-3-4
16k in year 2440
22k in year 2445 (Death Stars in 2444 IIRC)

31k in year 2450 Tech 18-16-5-17-8-4
53k in year 2460 Tech 22-24-17-11-4 (Armageddons in 2457)
68k in year 2469 Tech 26-26-12-26-12-11 (Omega Nubians!)

Minerals were also plentiful, once I got Robo-Ultras in the
mid-to-late-40's I think, and built a bunch of them. Note that this
one made what I consider to be my own most important benchmark for a
really nasty monster: over 50k and Armageddons before 2460.

The race design? (Race name ARvids)
PRT = AR
LRT = IFE, NRSE, ARM, ISB, LSP
Grav immune, Temp 4c to 84c, Rad 66 to 86 ( 1 in 21 hab )
Growth rate 19%
Coefficient of 10
Energy, Const, and Weaps all -50%
Elect, Prop, and Bio all +75%
+75% start at level 3
zero points left over

The hab range looks scary until you actually play it. Keep in mind
that the AR doesn't have to build factories, so can invest in
terraforming instead, and each point of terraforming is worth a big
boost in hab % to this race. And the terraforming tech becomes
available naturally as a byproduct of energy and weapons research.
This one would do really well if allied to a CA.

The other "advantage" to having this narrow of a hab range is that
your empire isn't thinly spread over a very large number of planets,
making it a bit easier to defend. You also don't raise the "red
flags" as much by gobbling up a very large number of planets. In this
test, only 37 of the planets were ever actually used for anything, (9
other undeveloped colonies still yellow), and only 23 of those 37 were
major contributors.

ISB went a long way towards helping to keep the pop as "right" as
possible in the early going, and once Death Stars became available.

Fire it up and give it a try. You just might find that AR is not
quite the pushover you might have thought they were.

I look forward to hearing the offerings of others, to show how well
all of the PRTs can do.

(Note: I'll be collecting "Best-Of-Breed" races to include in the
Official Stars! Strategy Guide, v2.7. Here's your chance to shine...)

Barry Kearns

[Jason Cawley]

Barry Kearns wrote:

>
> Fire it up and give it a try. You just might find that AR is not
> quite the pushover you might have thought they were.
>

Well, I tried them out. My first attempt was somewhat disappointing,
perhaps because of mistakes in how I did things. I got about the same
tech year 21 that you had year 20 - I had 10-5-4-4-3-3, just missing one
level of construction from what you had a year earlier. But I only got
half the resources you did - 1171 in 2421.

The main reason was probably that my early scouting efforts weren't
aggressive enough. These guys really have to do that right. I also, as
chance would have it, got minimum minerals on the homeworld, and iron
was a problem. With the low hab, the runs are pretty long, so the iron
demand for lift is high. That's the hardest to ramp up with remotes as
well. Some other chance stuff might have been involved - by year 8 I
had found 2 yellows and one green - the green was a little over 250 LY
away. That was the closest one after all were scouted, too.

The yellows do come on line surprisingly quickly, however. And there
are enough of them (I found 20 or so by turn 20). I found 5-6 greens by
year 20, many in the 300-350 LY range. IFE sure is a must. I think the
packed density also helps enourmously with such low hab; I'm not sure
how they'd do without it.

They certainly seem to have potential - the three cheap fields, high
planet values with a little terraforming, and no other resource costs of
any magnitude, are quite a combo. I can tell I could do a lot better
than I did with the first try; of course you got much better and that
shows it can be done.

Just a question about them - how did you handle the scouting? Also,
what lift/movement did you do, when you went to yellows and such (I did
right away, 50000 to each or so)? Anyway, it might help out people like
me who managed to screw it up ;-)

Anyway, thanks.

Sincerely,

Jason Cawley

[Jason Cawley]

Barry Kearns wrote:

>
> The race design? (Race name ARvids)

Well, I tried a variant of your ARvids for my second attempt (I called
mine B-kars - for Barry Kearns ARs :-), having mucked up with your
design as my other post details. All I did to the design was widen the
two narrow ranges to 30% each (from 20%), which raised the habitat ratio
to 1/10 (also centered a little to keep all the terraformables). I paid
for it with 5 clicks of pop efficiency, knocking that down to 15 from
10. I also did the scouting right this time, and had better starting
mineral concentrations on the homeworld (iron 58 rather than 30 the
other time). Voila, over 2000 (on 15 planets) in 2420 and about the
same tech you had then; plenty of expansion room, etc. Even with me
playing them :-)

I figured the hab trade ought to be worth it (at least, makes things
easier for me) because of the ^.5 thing. 1/1.5 on the pop eff reduces
the resources only 18.35% (.667^.5 = .8165). With full terraforming,
the wider ranges give .6 of the two narrow ranges, vs. .5 with full
terraforming with the 20-wide. So there should be 1.44 times the greens
plus yellows combined (1*.6*.6=.36, vs. 1*.5*.5=.25; .36/.25=1.44).
Also get twice the initial greens (which I sure found useful), and all
the planets that will end up 100% with full terraforming start out
green. Don't get quite the bang for the buck on terraforming, but
close. Having a few more planets later on also increases the "spread
efficiency" the ARs get, so I figure I should get the resources back and
maybe more too.

With the better hab, I found them very smooth to play - the spacedocks
are affordable at just the right time (% of cap on the starter colonies,
I mean) and the full bases in two years also work out about right (with
a little mineral shuffling needed, but not much). The denser settlement
early helped with the iron shortage; the shorter runs made looping
easier and faster so I didn't need as many freighters. And then the
spacedocks were ready to kick in for the second wave outward, etc.
Also, getting a decent empire in a fairly compact space might help if
there is more opposition or without packed. My average travel time
early was a lot lower; got good planets in 1-2 year fuel mizer distance
which took the early pop growth. My colony autobuild was one unit of
autobuild "max terraform" - doesn't get much simpler than that :-)

Now, I didn't keep going as long as you did (yet), so I can't claim to
have duplicated the longer-run performance you got, but it ought to be
at least close with decent play. Certainly should be competitive with
the other PRTs.

Anyway, that was my take on your great design idea, and it looks like a
fun sort of race. Thanks for sharing it with us :-)

Sincerely,

Jason Cawley

[Paul Hager]

I find the design interesting but I don't trust the low hab -- I
haven't tried it yet but probably will at some point. One item of
note is that the design is somewhat similar to my AR with 1 immunity
with the exception that the hab is 1:5 and the growth is 15%, not
19%.

The key in AR, in my view, is to have a larger number of colonies.
Given the way resources are calculated you are MUCH better off having
4 colonies with 100,000 than one colony with 400,000. This is not
true for other races where there is a linear relationship between
number of planets and resources.

When I get a chance, I'll post more info on my own design.

--
paul hager hag...@cs.indiana.edu

"The most formidable weapon against errors of every kind is reason."
-- Thomas Paine, THE AGE OF REASON

[Mark A Shieh]

hag...@cs.indiana.edu (Paul Hager) writes:
> The key in AR, in my view, is to have a larger number of colonies.
> Given the way resources are calculated you are MUCH better off having
> 4 colonies with 100,000 than one colony with 400,000. This is not
> true for other races where there is a linear relationship between
> number of planets and resources.

What's the popular belief for this? Is it better to have lots
of colonies when you're attacked, so each one lost hurts less, or
fewer colonies that are better defended? I'm aware of the resource
calculations giving twice? as many resources for the situation
described above.

Mark

[ElCabalero]

In article <5o1mqn$f...@dfw-ixnews9.ix.netcom.com>, bkea...@ix.netcom.com
(Barry Kearns) writes:

>I look forward to hearing the offerings of others, to show how well
>all of the PRTs can do.

Okay, here's my humble offering. What I call a "Sprint Mode" CA,
Version 5. The race started as a thought experiment on Friday night, and
thanks to the rapid "real world" testing afforded by IRC has matured very
quickly, until it is about as well tuned as I can make it. An earlier
version is entered in "Feeding Time at the Zoo", and is doing fairly well
as 4th of a field of 8 in 2454.

Race name: Superons
PRT = CA
LRT = TT, NRSE, CE, OBRM, LSP (yes, CE is a pain, but I needed the points,
and NAS is worse. OBRM is a no-brainer, for reasons that will become
clear later)
Grav .31 to .71, Temp -120c to -40c, Rad 20 to 40 ( 1 in 93 hab[!] )
Growth rate 20%
Pop eficiency 1/1000
Factories 14/8/20 (G-Box checked)
Mines 10/3/21
Weaps and Bio -50%
Energy, Elect, Prop, and Const all +75%

+75% start at level 3

I can't find my notes for resources at particular years, but I can
tell you that this race has consistently benchmarked at 29K in testbeds
(Walking over even expert AI like they weren't there), and has hit just
barely short of the 25K mark in two IRC blitzes (One of which it won, the
other is an "extended blitz" and is still ongoing).

Some of you have tried this race design concept before, I made it
work. The 1 in 93 Hab seems like (and is) quite a handicap, but the
points from it let you tune your economy like an Indy car. At G levels
above 50, this race will hit max facts/pop on a planet after 15 years, if
all the G it needs is dumped in from outside it will do it in 10-12
(depends on how much pop you dump with it). This rapid single-planet
development is critical to the race's approach, because it has so few, it
has to squeeze those for all they're worth, as fast as possible.

Basicly, this race takes the "Fewer but better planets" theory as far
as it can go. You can squeeze as much out of a planet as almost any
monster (4180 max on a 100% world), making up in pop resources what you
lose in factories (most monsters run at 1/2500), and you can do it much
faster on each individual planet, because your rapidly growing, hard
working pop yanks up the mine/factory development, rather than letting it
compound on its own.

Some of you are probably wondering how it manages to find any planet
but the HW, with that hab. However, a CA with TT starts with TT5, that
makes its effective hab 1 in 28, still very low, but you will be able to
find two or three planets in your hab range within 300 LY, unless your
luck is truly awful (in 20+ testbed universes, I've yet to find less than
two, and I think only one wouldn't hurt too much). These become your
first generation colonies. And you will have the resources to send out
enough scouts in 2404 to cover all that territory by 2416 (I usually have
14, the one I started with plus 13 built in 2404). Do it, or die on the
vine later.

From there, the early game is basicly a race to get the Bio to get the
planets to get the resources to get the Bio.... Every time your Bio gives
you another TT level, your effective hab range goes up (At TT10, it is
about 1 in 12) and the previous generation of planets hits 90-100% (most
of the new generation start at 70%+). Although this means lagging behind
in other techs, this isn't as bad as it seems, because reasonable levels
in "War" tech can be gained quickly, I have twice gone from nearly nothing
to Jihadi or Epsilon cruisers in 4 years, then built large enough fleets
in two years to slap down the WM quick-start that went for the throat in
IRC blitzes.

One interesting thing about this race is its degree of "Disposable
Income". This is resources that are not needed to be reinvested in
creating more resources directly, by building mines or factories, but can
be spent on ships, research, whatever. Basicly, the figure in the
Research screen when nothing but mines and factories are in the build
queues of all your worlds. Where for traditional monsters, this figure is
usually about 1/3 of the total resources, for this race it is closer to
2/3.

Basicly, at any given time, this race has a smaller proportion of
developing worlds, and a greater one of "Mature" worlds, where mines/pop
and facts/pop have maximized. It also has an almost unique category of
"Adolescent" worlds, where you have maximum mines, but not factories, and
more resources than you can spend to improve the economy, because your
factory building is G limited. Only the very highest G worlds will not
spend time in this phase, unless you dump in around 6000 kt of G (Which
you will do, later, for low-G worlds).

This means that where a traditional monster at 60K in 2450 would have
20,000 in disposable income, this race would have 30K in 2450, but the
very same in immediately usable resources. Obviously, this is not that
simple in absolute terms. The traditional monster can choose to "dispose"
of a much larger percentage of that income than that, at the cost of
delaying his economic growth. But, if at this phase you choose a
defensive stance, and have been careful in your diplomacy, you should have
a fighting chance at weathering the storm, especially if you have started
your minefields early.

And here is where the Superons part company with traditional races.
By 2450, most races have thoroughly settled their territory, expansion has
to come at someone else's expense, which means fighting a bloody,
expensive war, which you may lose, or be so weakened by winning that
somone else picks you off.

In the 2440-70 era, the Bio = Hab equation really goes to work for the
Superons. At TT 15, you double your number of available planets, to 1 in
6. At TT 20, it is 1 in 3. At TT 25, it is a wide 1 in 2, really 2 in 3.
At TT 30, it is virtually all planets. And nearly all of these planets
are at or very near 100%

You can literally expand inward, squeezing as much out of your little
mine-infested pocket of space as most races can squeeze out of the whole
universe. And with your single-planet ramp-up tuned as fast as possible,
colonizing is a matter of throwing 100K colonists and an equal weight of G
into a planet, and baking for 15 years. Presto, instant production
center.

By 2470, a monster who has knocked down everybody else in his path is
facing an opponent that is stronger economically, at least equally
advanced technically, and is operating from a much denser area of
maximized worlds, filled to the brim with minerals begging to be converted
to a high-tech war fleet (bringing a new level to interior lines).

Again, obviously, it is not quite that simple. In this case, it is
because you have to get that far, and you have a period of high
vulnerability in the late forties and early 50's, where your research
effort is still primarily focused on Bio, your number of maximized worlds
is small enough that losing one or two will really hurt, and you are
likely to get advanced scanners just in time to find that your worst enemy
is thoroughly settled in "your" space. I wish I had a magic bullet for
that, but I don't, yet.

Omnivore from IRC has combined elements of this and of the ARvid with
his Valheru, and created a truly scary quick-start, that cleaned the
Superon's clock in a Tiny/dense matchup (Tiny dense, and he had Jug
cruisers in the 40's). We both felt that in a larger universe it would
have been a different game, I was simply too close to him to have much
hope of protecting myself in that vulnerable period, and he was more
prepared than most to exploit that (And in no position to stop the
monstrous packets that blew half my worlds to dust).

Obviously, the best ally this race can have is an SD. Protection
through that critical 2440-50 period will let you hit the 2460's running,
and nobody punches through an SD in a hurry. Fortunately, you have
something to offer, quicker and better terraforming than would normally be
possible, unless his hab is really wide you can take all his greens and
most of his yellows to 100%. And best thing you can do, diplomatically,
is to stay at peace with any neighbors with the ability to hurt you in
that period.

Unfortunately, everybody will know what you are, hiding your race
style is near impossible in a public scores game. When your resource
levels are consistently keeping pace with the leaders, but your number of
planets is only a third of theirs, isn't hard to see what's going on.

I feel somewhat smug about this race, more than one Stars! guru said
it couldn't be done, or would be so marginal it would have no chance of
survival in a real game. A unstoppable juggernaut it isn't, a contender
it is.

--Dave

[Jason Cawley]

Paul Hager wrote:
>
> I find the design interesting but I don't trust the low hab -- I
> haven't tried it yet but probably will at some point. One item of
> note is that the design is somewhat similar to my AR with 1 immunity
> with the exception that the hab is 1:5 and the growth is 15%, not
> 19%.

Well, with a somewhat similar idea in view, I tried it with 1/10 hab
(two fields 30 wide, plus the immune) and 15 pop eff. In packed, I
settled 13 greens and 2 good yellows in 20 years. The farthest was 300
LY away; 3 were 250 LY or so, the rest 200 LY or less. I then did the
"sit and spin", stopping the development completely year 20. 15 planets
was a little less than needed for everything to go smooth, with the
lower pop eff - 20 would have been fine. With only 15 and no
follow-through after year 20, I still hit 20k in 2450; all planets had
death stars by 2451 (plus Juggernaut missles). Even with no further
effort, I had weapons 24 and the supporting tech and such by 2465.
That's with only 15 planets.

I also tried it in dense, in a huge universe with 15 expert AIs. I am
not very far along in that test, but things are going fine. The spread
is harder, no question, but going 300-350 LY in each direction I have
about 15 greens (all will be 100% after full terraforming) and about as
early/good yellows. Three AI neighbors encountered in that area, on the
borders. I can get the space needed and they can't stop that. Resource
stuff on track, maybe a year or two behind with all the extra scouts
needed (I built 30 or so). Understand, this is with 1/10 not 1/21 (paid
for by the 15 pop eff).

The narrow hab has certain excellent things about it. When you
terraform a world, you raise the hab % and for AR that raises the
resources. This gives you extra "growth rate". The planet values are
quite high pretty fast. With, e.g. 70% starting hab and 19%, the growth
under 25% is 1.133 per year. ^.5, that gives 6.44% growth. But if 1%
of terraforming gives 2% in hab, you go from 70 to 72 - which is a 2.86%
increase. Together, the planet gets both multiplied, or 9.5% in this
example. Bigger planets get more terraforming done and have faster pop
growth; when at 100 they still get 9.09% from the 1.19 pop ^.5 (under 25
or course - which does get blown through somewhat). If you get energy
26 by say 2460 or so, that gives 5.099 times over the whole period,
which works out to 2.75% annual average rate. So you can get pretty
high growth rates from all working together - pop, energy tech,
terraforming hab. And the resources are needed quickly to "ramp up"
right, because you want the new station tech in time to build them
before the pop gets too wrong for too long. e.g. going from starbase to
ultrastation, a world at 66% of cap drops back to 33%; you stay in the
better part of the pop growth area. Part of the point is that AR is
better in the "capacity"/space available department than the other
races, when it becomes required.

When I played them, with 19% growth these things went pretty smoothly,
and most of the capacity (I stopped expanding) was realised by 2460.

I'm not sure the 15% and wider hab would work as well (haven't tried it,
but the 19% looks awful tough to beat to me). You don't "power-up" to
the higher pop levels as fast, and you don't get quite as much per
resource invested in terraforming. Using our example figures from above
with 1% hab per terraforming and 15% growth rate, I get 6.62% rather
than 9.5%; in the energy takes a bit longer because of this (say to year
70 to get level 26) that component drops to 2.35%, and overall for the
sample planet (under 25%) I see 9.13% growth vs. 12.5%. 3% faster to
year 60 is 6.2 times. You would get some of that back from spreading
out, but not too much. e.g. 1/10 green gives .36 yellow; you won't get
6.2 times 36% of all planets ;-)

Anyway, too abstract no doubt. I do suggest trying Barry's version; if
the hab doesn't suit you, try widdening it just enough that you can
stand it and see how that works for you. (Can be paid with via a few
pop eff or maybe 1 growth or so - the idea being to keep Barry's basic
idea).

>
> The key in AR, in my view, is to have a larger number of colonies.
> Given the way resources are calculated you are MUCH better off having
> 4 colonies with 100,000 than one colony with 400,000.

Sure. But with 19% vs. 15% you aren't going to have the same number of
people; you are going to have a lot less. If you really get 4 times to
planets, OK. 19% vs. 15% at 1/3 of capacity on 80 average hab gives
2.34% extra growth; in 60 years that's 4 times the pop. Even Barry's
hab gives 1/4 yellow, so more than 4 times the planets doesn't seem
likely ;-) But a little wider might be worth it, to have more planets
too (without giving up the growth rate or the immunity). Anyway, I
recommend the trial.

Sincerely,

Jason Cawley

[Iannick Daniel]

Paul Hager wrote:
>
> The key in AR, in my view, is to have a larger number of colonies.
> Given the way resources are calculated you are MUCH better off having

> 4 colonies with 100,000 than one colony with 400,000. This is not
> true for other races where there is a linear relationship between
> number of planets and resources.

In fact:
4 worlds of 100,000 pop equals 1 world of 1,600,000 pop
and
2 worlds of 100,000 pop equals 1 world of 400,000 pop
but also
2 50% worlds of 100,000 pop equals 1 world of 100,000 pop

So high habitability is better than high growth rate, because low
habitability reduce reproduction AND production for AR races.

--
Iannick Daniel
(Bizmut)
--
E-Mail address:
i...@id.ml.org
--

[Jason Cawley]

Iannick Daniel wrote:

>
> So high habitability is better than high growth rate, because low
> habitability reduce reproduction AND production for AR races.
>

But they also get a bonus of sorts here - lower hab does not reduce
planet capacity - that is determined solely by which type of station is
up. Which, it seems to me, is one of the things Barry is exploiting.
The point being that narrow hab and one immune gives more space/capacity
to an AR than to any other race. Then the terraforming rapidly
increases the hab, growing the economy. The growing economy buys more
capacity (better starbase hulls) as the pop/capacity ratio rises, so it
stays good for a much longer period of time. That number being better,
the terraformed value being very high, and the pop growth rate being
high, give lots of people. Even with only 15 planets to grow on (in one
test with 1/10 hab - two ranges 30 wide - colonizing arbitrarily stopped
year 20) by year 60 this gave mostly filled deathstars everywhere, and
the resources to get the energy tech. Those planets then gave resource
totals comparable to other races - 2000-2500 each. Of course, with more
managing and spreading out later, a lot more might be done - but
35,000,000 plus pop and 25k resources/year make that a lot easier.

Anyway, IMHO Barry has found a design type that is definitely strong.
Whether there are other approaches to the AR PRT that will prove just as
strong is more than I know.

Sincerely,

Jason Cawley

[Dominik Mayer]

On Sun, 15 Jun 1997 23:08:10 -0500, Jason Cawley
<jason...@worldnet.att.net> wrote:

>Barry Kearns wrote:
>
>>
>> Fire it up and give it a try. You just might find that AR is not
>> quite the pushover you might have thought they were.
>>
>
>
> Well, I tried them out. My first attempt was somewhat disappointing,
>perhaps because of mistakes in how I did things. I got about the same
>tech year 21 that you had year 20 - I had 10-5-4-4-3-3, just missing one
>level of construction from what you had a year earlier. But I only got
>half the resources you did - 1171 in 2421.
>

Ok, I give up.

I have only reached 920 resources in 2423 with Barrys design. <sniff>
What am I doing wrong? I have 4 planets (HW included; hab is 100%,
100%, 77%, 14%), each one got a colonizer immedeately after discovery.
The HW is around 250000 pop, the others around 60000 pop; I am
shipping the excess to the other worlds, in loads of two privateers
with fuel pods only. 5 scouts are zipping around for new worlds; so
far none to be found.

Same thing, i.e. those low resources, happened to me with other
proposed monster races, too (namely monster CA, JOAT and IT). I am
really clueless. My testbeds are small standard universes as with all
options set as stars! proposes.
I build one scout per turn (beginning a bit later in turn 2404 for non
AR monsters) up to five or six to find new planets in time; then I
just colonize and send excess pop over 250k (sometimes I use the 330k
borderline) away to my new colony. I never ever reached more then 14k
with any race in 2450. I never ever reached 2000 resources in 2420.
Perhaps someone could send me the first 50 files of a testbed? Or is
someone willing to examine a testbed of mine? How are you playing
(refered to this sample AR)?
It is now sure that my bad performance is not the fault of the race
design. But I do not see no flaw in gameplay yet, and I sure do follow
this newsgroup. HELP!

RU
Dominik

[ElCabalero]

In article <33A5D5...@id.ml.org>, Iannick Daniel <i...@id.ml.org>
writes:

>So high habitability is better than high growth rate, because low
>habitability reduce reproduction AND production for AR races.
>

High planet habitability, not high race Hab. They not only aren't the
same thing, but high race Hab decreases the capability of terraforming to
improve a world, and does more to increase the number of soso worlds than
it does to improve the rest. As I am discovering, a few 100% worlds are
better than a greater number of lower hab worlds, regardless of PRT. CA
and AR are uniquely positioned to exploit this, since CA don't have the
drag of terraforming costs, and AR don't have to worry about building an
economy, so they can both concentrate all their efforts on the other side
of the equation.

This is what the whole marginals issue stemmed out of, and
pop-management in general. As it turns out, squeezing the most out of
50%- worlds really is more work than it is worth. The ARvid trade quality
for quantity, and the trade seems to be a good one for AR's, which were
supposedly tilted in the other direction.

Why? Because Higher-Hab/Lower Growth is actually a much worse trade
than it looks (Yes, I used to think it was a good idea. So what, I was
wrong). Not only are low growth rates disproportionate to their absolute
values, because of the pop compounding having less to work with at each
step, but the disproportionate number of so-so worlds grow even slower
than that because of habs impact on growth, and for AR, capacity. Do what
you like with pop-management, what it comes to in the end is robbing what
could be strong worlds to prop up cripples, and the strong worlds are what
really drives your empire (And it's a heck of a lot of work, along the
way).

I have found that the real relation between nominal growth values is
not their absolute value, but the ratio of the squares of those values, as
near as I can calculate it. I found this out for certain when I thought
knocking one point off PGR and using it for cheaper/more efficient
mines/factories would let the Superons grow faster. The result was that
they actually grew slower, the loss of 1/20th of my nominal pop growth
rate lost me a full 8/th of my real growth (measured in resources, of
course, and in the time it took to max the planetary economy), even though
I had tried to compensate by strengthening the mine/factories side of the
equation (Remember, in the Supes I was explicitly trying to max
single-planet growth).

Now, the Supes are an extreme case, because they have such a narrow
hab that PGR point was only worth 40 advantage points. But more testing
with less extreme cases found similar effects, everywhere I looked. For
races that otherwise would have low mine/fact economy, it can be a good
trade, but for a race that already has both high, it's a bad move.
Explicitly trading PGR for Hab is always a bad trade, trading it for
something else, specifically cheaper tech, good LRT's, or the lack of bad
LRT's may have real world benefits that outweigh this, especially at the
19-20% cusp.

What does this mean, in real terms? It means that a 50% world is only
worth 1/4 as much as a 100% world, unless you play the Pop MM game, and
the more "marginal" the worlds you go after are, the harder you have to
work for it, and the less benefit you will gain for your efforts.

But it gets worse, if you explicitly trade half the growth rate for
twice the worlds. Now, you have increased the number of good worlds
hardly at all, but you sure have a whole mess of the 50%- crowd (And
*surprise*, their "real" growth rate is only about 1/16 that of a 100%
world's would be at the higher nominal growth. In other words, you go to
a great deal of trouble, and you never had a chance of gaining anything
for it). So, your ability to exploit any world has been severely
compromised, in order to get a bunch of worlds that will never be good for
much at all. Indeed, it is entirely possible to design a race so that it
will have basicly nothing but good worlds, but only a few, and achieve
"real" results as good as any.

What I'm still working on is how the Hab/Economy trade works. Since
the mine/facts side is only indirectly affected by planet hab, through
real pop growth, it may turn out that there is a trade to make there.
But, the experience of Valheru vs. Superons indicates that it becomes an
issue of "peaking", whether the race is aimed at the early, mid, or late
game, or hedging your bets by having more planets, and thereby not being
as hurt by the loss of any one of them.

Oh, and the final reason that four 100,000 worlds don't equal one
1,600,000 world in the long run is a "real world" concern. A 1,600,000 AR
"world" is a well equipped Ultra-Station or Death-Star. Four 100,000 AR
worlds are probably Orbital Forts, possibly Stardocks, just maybe poorly
armed Starbases. Which is harder to kill? Which is easier to deploy
mobile assets to properly defend?

--Dave

[Jason Cawley]

Dominik Mayer wrote:

> Ok, I give up.

Naw, just want a pointer or to - your addicted :-)

>
> I have only reached 920 resources in 2423 with Barrys design. <sniff>

First thing - if you find that with your play you can't make the 1/21
work (we know it can, but that doesn't mean we can do it) make these
changes - widen the two narrow hab ranges to 30 each (5 clicks above
minimum). Pay for it by reducing the pop eff/coefficient to 15 from 10.
Then slide the narrow ranges so that you have 15 left at least on the
right - you want all the possible terraforming, in other words.

You should find this easier to play. It gives 1/10 hab; you'll find
livable places closer. I will go through the other things this race
needs (in my limited experience) as well.

> What am I doing wrong?

Let's find out. I also didn't do too well in my first try with Barry's
design; with a second attempt, and the above changes, I got it to fly.

I have 4 planets (HW included; hab is 100%,
> 100%, 77%, 14%), each one got a colonizer immedeately after discovery.
> The HW is around 250000 pop, the others around 60000 pop; I am
> shipping the excess to the other worlds, in loads of two privateers
> with fuel pods only. 5 scouts are zipping around for new worlds; so
> far none to be found.

OK. Two things right away. One - these are ARs, so they don't need
Germanium for factories. They do need Ironium for shipping. The
privateer costs lots of iron and little germ - that can make it good for
germ-hungry races but is bad for AR. Use medium freighters instead. If
they don't have enough fuel, add a scout to the fleet to "boost" them.

Second - scouting. These low hab races need hyper-aggressive scouting.
With 5 scouts hitting a world a year and 1/21 green, you'll only find
one every 4 years on average. You found 3 in 20 years, right? Imagine
you have 21 scouts working, or 25 for the two-year jumps, etc. Then you
might find 1 per year, right? Try it first with the 1/10 hab; it's
easier to see how it works and luck will play less of a role. You want
20 scouts. And with so many, you have to plan their use correctly to
get the benefit from them.

Here's how you do it. Whenever you have new scouts (including the
starting one) give them orders for *20 years*. Don't circle around -
have them go straight away from the homeworld, hitting a planet a turn,
going at whatever warp you need. Occasionally there will be two year
jumps, or three years for the later ones getting out. Keep the fleet
paths on. You want no crossing, as little side-to-side moving as
possible. Pick a line for the first one that includes a one-year jump.
The next one goes in another direction. Every one year jump from the
homeworld has a different scout assigned to it. Whatever they miss in
the two-year range the follow-on guys go for, then keep trucking away
from the homeworld. Build scouts as needed to fill gaps in the "spoke"
pattern (having the next 20 years of orders helps see these things
immensely). Go fast.

If you have 20 scouts built over a period of time, you get
"scouting-years" like this - 20 for the first, 19 for the second, 18*2
next, etc. Maybe 2/3rds, maybe more are planet-sighting jumps. They
should all leave in the first ten years. This will let you scan lots of
planets; you'll find greens and decent yellows. You want 20-25 planets.
With 1/10 hab and maybe as many decent yellows, that isn't too much
scouting - 100-200 planet-sightings will do it. Ideally, you want to
scan an entire small galaxy (or close to it) by year 30 or so. Really
aggressive scouting, in other words.

Also, I should say something about your spending plans. Doing this
right takes getting used to - the ARs are very different from the other
races.

Your first few years just research energy. This is like building
factories for other races. You want only such scouts as you can afford
and still get the energy levels - don't look just at the next one,
either - look two levels ahead or so. If it would take everything for
three years to get energy 3, just do the research and nothing else. If
you see room for a scout or two, put them in. You want to get about
energy 4 or so. Then hit the scouts hard - your full production.

After that, you need to get con 3 in time to move, using the medium
freighter. Watch your pop - if you plan ahead you can spread it out, a
level a year on the construction, while getting all the scouts you want
(or close to it).

Hold back the initial colonizer until you are almost ready to move. The
turn you put your first freighters in the queue should be the turn it
goes out, to the highest value you found nearby (even a yellow if its
that bad, but it should be - that's why you are holding it back). Then,
if you have the planets sighted and the iron, build a new colonizer a
year. Send the freighters to carry the new pop after it (put fuel pods
on these). You can send them out together if you want, and you don't
have to fill the freighter entirely if the fuel is tough. Have the
first freighters loop back for more pop; you'll need fewer that way,
saving iron. For really long runs (say, over 4 years) you can just
merge them with the colonizer and "crash" colonize.

The new colonies use this autobuild - autobuild max terraforming - one
unit. You can have them not contribute to research to help them out, or
contribute if you aren't getting enough from the homeworld.

After the con 3, go back to energy. Push it up to 5-7.

When your colonies are near 25% of cap, go back to the construction and
get the spacedock hull. Design an empty hull version. Have the
colonies build one when they get to around 25% of cap with the starter
colony. This will give them more room to grow at the best rate. Keep
the new people moving from the homeworld. 33% of cap is fine, because
your travel times will be pretty high. Don't sweat it if it gets a
little out of line (the homeworld capacity will rise later when you get
better stations).

When the colonies are doing their one point of terraforming per year,
let the rest go to research. Energy research. All the way to level 10
as fast as you can go. Only build the shipping you need - colonizers,
freighters until there are enough of them looping (or for longer
colonizing runs), the initial scouts (but stop them at 20 built or year
10, whichever comes first). Also, if iron is tight, you can add some
mining ships. Use the best iron concentration within one year of the
homeworld (though the homeworld itself is better if they are close).
The tech 4 miners on the midget miner hull are best, if you can afford
one in a turn. Only build these to get iron for the ships you need, if
you need to. Everything else goes to energy.

When your better colonies get over 200 resources, design an empty
starbase hull. Check the % of cap on these places, their resources, and
G available and being mined. Check these against the cost of the empty
starbase. You can run G out to them from the homeworld if they need
more - one medium freighter load should do it. You want the colonies to
have one of these starbases up in two years when they get to 25% of cap
on the spacedocks. If you don't make that, fine. Understand that when
the new base is up, the % of cap will fall by 1/2 (500,000 cap from
spacedock, 1,000,000 from starbase) - so even 66% isn't as awful as you
might think :-)

When you get energy 10, switch to construction. Go all the way to
ultrastations. Upgrade to these from the full bases the same way you
did from the docks, starting with the homeworld.

Then move to weapons and get level 10 for more terraforming.
Then back to energy, all the way to 16 - more production capacity and
more terraforming.
Then con to deathstars and upgrade as before.
Then get weapons 16 (more terraforming), and prop 5.
Then go back to energy and leave it there. (You can throw in your
weapons goal at some point - I got Armeggeddons by year 65 - Barry got
them in 57). You can also round out your tech once the economy is big.

I got the deathstar tech in year 49, Barry got it in year 44. Once they
are up, your cap on a planet is 3,000,000. With 19% growth and the one
immunity, and the regular terraforming all getting done from the
autobuild, this should get your planets to the 2000 resources range
(with your high energy tech of course). So 25k isn't hard (I got that
with 15 planets - I stopped expanding because I was being lazy :-); with
20-25 planets (including some yellows worked up) you should be able to
get 40k or so by year 60 or 70.

Now, those are just benchmarks to shoot for, understand. The main thing
is to do the scouting right, find the 20-odd planets, get people to
them, and then keep the tech marching to the plan and the bases upgraded
(which might take a little mineral-juggling, but not much).

Everything works together here. If you don't get enough planets, you
won't get enough excess resources after that point of terraforming to
hit the tech. Then your growth will slow (not enough energy tech), you
won't be able to get the base upgrades soon enough, the pop will get
"wrong", slowing the growth even more, and everything will seem very
hard.

But with the right scouting, you should be able to find the 20-odd
planets you need. Getting some people to them shouldn't be too hard
(they can come from spacedocks farther out to cut down travel time, for
instance). If you have them, the one-terraforming autobuild will march
the hab up pretty quickly. That will make the pop and resources grow.
The resources will get you energy, and more resources. Then those will
get you the bigger stations and keep the pop growing. That'll let you
get more terraforming tech, higher planet values, more resources. And
more energy, more again. When those 20 planets have 1000-2000 resources
each, there you are :-)

When you try this, use a small *packed* universe. That will make
finding the planets easier, and they will be closer (that's what Barry
ran his test in, BTW). Later, when you get the hang of it, you can try
it in dense, throw in some AI opposition, and such. But for now make it
easier (might even use maximum minerals the first time or two, until you
see it can be done).

Anyway, give it another whirl with the scouting right and the tech
research plan in mind. You might just be surprised how things fall into
place.

I hope this is helpful.

Sincerely,

Jason Cawley

[Jason Cawley]

anthony m. vervoort wrote:

> personally, i think it would be valuable for some of the expert players
> to try their favorite designs in normal or sparse density universes
> to compare results.

Sure, density (and also how much opposition there is - the space per
race) makes a difference in terms of both what a race can acheive and in
what design things optimize the performance.

I normally do not use packed for testbeds. I do normally use dense,
because I want to see in a relatively small period of time how well the
race will do in a real game in medium. But I normally use 3 AIs as
well. The idea is to have a "quadrant" of a medium game with many
players, in effect. But I will occasionally use normal or sparse to see
if a design can stand it, and I'll test with close to real-game settings
before a large PBEM starts up, with the race I plan to use.

>
> i believe resource monster type races will still do pretty well with
> fewer planets, compared to other types of races, but i don't believe
> you'd be seeing as many hitting the 25k by 2450 benchmark if they
> weren't being tested in dense/packed universes.

Hm. No. I have done 25k be 2450 in small normal vs. 3 random expert
AIs (with an IS pop-grower - a "quickstart" in Brian's terminology).

But would it be as easy to get the 50k and 75k figures some people have
done? No. There are fewer planets, as well as farther apart, harder to
find, etc. Some things like low hab and pacing of pop movements are
rather different in lower densities, but many things are the same. So
there are probably designs that do quite well in unopposed small-packed
that wouldn't get 25k in 2450 with less space - but there are certainly
others not in that category.

If I take 1/10 hab and expect 2500 resources from a developed world, I
expect to have 10 planets or better to get to the target. I expect to
scan 100 planets if I only go to greens. But if I take 1/3 hab, or a CA
race that effectively adjusts to that, and expect 1500-2000 from a
somewhat developed world, I might need 15 planets but I need only scan
45 or so to get them. The former design might do worse; the latter will
do fine.

BTW, I recently ran a test of a one-immune war-monger with 1/10 hab in
small, normal - with 13 random expert AIs as opponents (Acc BBS, normal
minerals). I also used "slow tech advances". The start looked a little
rough; I wasn't sure I would get space to grow. But they made 17k in
2450, with weapons tech 16; I had "blockades" over 4 AI homeworlds by
then (plus some of their secondaries). Admittedly the AIs aren't so
good defending themselves as humans. But they had to face 200 escorts
with jihads and bludgeons in the 2440's with slow tech - and I think
even humans might find that somewhat hard with such little space to work
with and grow on. 14 races in small-normal is an average of only 9
planets each. I used more like 1/4 of the galaxy to get my 17k, using
yellows as well as greens. But the point of the design was to be able
to grab and control that much airspace quickly.

Want to try it? The design is:

PRT - WM
LRTs - IFE, ISB, CE, OBRM, NAS, RS
Hab - Grav immune, temp 20-140 C, Rad 55-85 mR - 1/10 overall
Pop Growth - 19%
Pop Eff - 1/1000
Factories - 12/9/16/3G
Mines - 10/3/12
Tech - Weapons cheap, rest expensive and start at 3.

Now, that tech looks awful I know, with slow tech. But the starting
stuff is OK and if the growth happens it is livable - you start with 23
levels. I ran the test for team game things; I expect in the real trial
to import much of what the race needs. I put up the first gate by year
20 to reflect this, and built tech transfer ships as though I was
exporting tech.

The point of the test was to determine whether 1/10 hab was livable with
so little space. Try it and see.

Just one man's take on these things...

Sincerely,

Jason Cawley

[anthony m. vervoort]

dom...@nei.franken.de (Dominik Mayer) writes:

[poor dominik is ready to give up on these monster races]

>Same thing, i.e. those low resources, happened to me with other
>proposed monster races, too (namely monster CA, JOAT and IT). I am
>really clueless. My testbeds are small standard universes as with all
>options set as stars! proposes.

i think this is your problem. i think all of the monster tests i've
seen used dense or packed universes, not normal density. if you just
can't get to as many planets as quickly, you're not going to be able
to reproduce the kind of results the monster players are having.

personally, i think it would be valuable for some of the expert players
to try their favorite designs in normal or sparse density universes
to compare results.

i believe resource monster type races will still do pretty well with

fewer planets, compared to other types of races, but i don't believe
you'd be seeing as many hitting the 25k by 2450 benchmark if they
weren't being tested in dense/packed universes.

i await Barry or Jason coming along to show me for the fool i am. :)

-anthony

[Barry Kearns]

While not intending to make you a fool, I'll give you the results from
the ultimate "test" of a monster race: the actual PBEM game. I'm
playing in "Elite" right now, an expert-level PBEM, with an old
version of the Viruses, one of my first monsters.

It's a large-normal universe, with 11 players.

I got 41k in 2450.

'Nuff said.

Barry Kearns

[alex hall]

Well I have a number of WM monster type designs, the more
standard one that I will post now and then a variation which
has a res/colonists of 1/2500 with very cheap factories and
also a 1 immune race.

The following is one based on the design I am currently
using in a blitz game on IRC, the current year is 2494 and
my WM has been in first place for all but 2 years. 3 of the
six players have now retired and the remaining players are
my WM, a CA and an AR. Currently I am at war with the AR
and am about to attack its Home Planet.

The design is not perfected yet, I won't post the latest variation
as I will be playing it sopon, and anyway many people won't
necessarily like this type of design, but it suits the style of
play that I am using in the game.

PRT - WM
LRT - IFE, NRSE, CE, OBRM, NAS, LSP, RS

Gravity - 0.12 -> 1.48g
Temperature - -40 -> 200 degrees C
Radiation - 40 -> 100mR
Habitat - 1 in 4
Growth - 18 %

1 res per 1000 colonists
Factories - 1/8/20/G
Mines - 0.8/3/21

Energy - Standard
Weapons - -50%
Propulsion - +75%
Propulsion +75%
Electronics - +75%
Biotechnology - +75%
Cost 75% start at 3

One other thing to point out, that using someone elses
Monster design doesn't guarantee that you will get
monster scores, also there is a degree of luck involved
as to whether your race happens to be stuck close to
an aggressive warlike neighbour or not, if this occurs then
it can be very difficult to achieve monster type scores by
2450.

For more details on my thoughts of WM see my post on WM
strategy or look at GeneralsRm (Mathias Dilbert) website at
http://homeplanet.home.ml.org

[Randy L King]

Jason Cawley <jason...@worldnet.att.net> wrote:

>Here's how you do it. Whenever you have new scouts (including the
>starting one) give them orders for *20 years*. Don't circle around -
>have them go straight away from the homeworld, hitting a planet a turn,
>going at whatever warp you need. Occasionally there will be two year
>jumps, or three years for the later ones getting out. Keep the fleet
>paths on. You want no crossing, as little side-to-side moving as
>possible. Pick a line for the first one that includes a one-year jump.
>The next one goes in another direction. Every one year jump from the
>homeworld has a different scout assigned to it. Whatever they miss in
>the two-year range the follow-on guys go for, then keep trucking away
>from the homeworld. Build scouts as needed to fill gaps in the "spoke"
>pattern (having the next 20 years of orders helps see these things
>immensely). Go fast.
>
>If you have 20 scouts built over a period of time, you get
>"scouting-years" like this - 20 for the first, 19 for the second, 18*2
>next, etc. Maybe 2/3rds, maybe more are planet-sighting jumps. They
>should all leave in the first ten years. This will let you scan lots of
>planets; you'll find greens and decent yellows. You want 20-25 planets.
>With 1/10 hab and maybe as many decent yellows, that isn't too much
>scouting - 100-200 planet-sightings will do it. Ideally, you want to
>scan an entire small galaxy (or close to it) by year 30 or so. Really
>aggressive scouting, in other words.

OK!!!!! That's what I've been doing wrong. I always come back for
fuel. Make about 30 total and call them expendable??!! I think I can
live without those resources.

Although this assumes you have IFE. In what situations would it be
practical to have expendable scouts with a non-IFE race???

Thanks in advance.

Randy
<grey...@sojourn.com>

************************************************************************
The Grey Wizard walks amongst us... spreading his magic... sharing his
wisdom... freeing the world of its fear and pain.
************************************************************************

[ElCabalero]

In article <33a80a3a...@news.sojourn.com>, grey...@sojourn.com
(Randy L King) writes:

>Although this assumes you have IFE. In what situations would it be
>practical to have expendable scouts with a non-IFE race???
>

If you have some other way of getting Prop 4, and the Long Hump 6
(Like expensive prop and CE). The LH is a good enough engine to scout
everything within 300 LY, which is about your maximum colonization range
anyway.

--Dave

[Jason Cawley]

Randy L King wrote:

>
> Although this assumes you have IFE. In what situations would it be
> practical to have expendable scouts with a non-IFE race???

Well, you can always treat them as expendable; put a fuel tank on them
and let them go. But they won't be very fast or effective beyond a
fairly short range without decent speed.

Tiny galaxy? IT, starting with tech 5 in prop? Maybe a cheap engines
JOAT with expensive prop? (start at prop 5 plus penscan)?

Other than those, I can't think of one. So the moral is - don't take
1/10 or lower hab without IFE, unless you are one of the exceptions
mentioned. And maybe even then. Low hab needs good speed, to settle
places as well as to find them. If you don't have the speed, don't
trash the hab. If you trash the hab, get the speed.

I hope this is useful.

Sincerely,

Jason Cawley

[ElCabalero]

In article <33A85F...@worldnet.att.net>, Jason Cawley
<jason...@worldnet.att.net> writes:

>Well, you can always treat them as expendable; put a fuel tank on them
>and let them go. But they won't be very fast or effective beyond a
>fairly short range without decent speed.

300ly radius at warp 6+ on a long hump is enough, if you're willing to
push them till they die, then let them rot where they fall. Takes a fair
number to scout that much radius, but it can all be done by 2416.

>Tiny galaxy? IT, starting with tech 5 in prop? Maybe a cheap engines
>JOAT with expensive prop? (start at prop 5 plus penscan)?
>
>Other than those, I can't think of one. So the moral is - don't take
>1/10 or lower hab without IFE, unless you are one of the exceptions
>mentioned. And maybe even then. Low hab needs good speed, to settle
>places as well as to find them. If you don't have the speed, don't
>trash the hab. If you trash the hab, get the speed.
>

Unless you have enough resources that you can afford to literally
throw away a dozen or two scouts in the first decade, yes. Every game the
Supes have played, they've gotten by with Long Hump 6 and plenty of tanks
until 2436 or so, when they usually jump straight to Prop 9 and the
Trans-Galactic drive, or even further to the InterSpace 10, probably the
best deal in Prop besides the Fuel Mizer itself. I just build a *lot* of
scouts, and don't expect each one to get to more than 5 or 6 planets, and
don't expect them to come home at all. And this is in real blitzes, not
just testbeds.

Then, once I have both the T-G and the Ferret, I build about 25-40
scouts and shoot them all over, blanket the galaxy. Even the most
determined space-interdicter has a hard time stopping that flood before
they send me back a lot of data. And whatever space was supposed to be
covered by the now-dead scouts gets swarmed by the survivors. Wasteful as
hell, but effective. Also an arrow pointing at my home space, but they
probably already know that, I'm the one who wasn't scouting.

I have the feeling that the shoulder-to-shoulder scouts lined up as
far as you can see has made more than one player decide it wasn't worth it
and let them fly by.

Keep in mind, this works well in the Dense/Packed universes common for
blitzes. In a Normal/Sparse I'd probably bite the bullet and either take
IFE or drop NRSE. However, even IFE doesn't increase your effective
colonization range that much, most feasible speed drops pretty fast past
180 ly. A fully loaded Privateer with 3 tanks can't go much past that at
better than warp 5 or 6, even coming back empty, although you can do a
little better with a Crash-and-Burn fleet, that doesn't help you support
it.

--Dave

[Jason Cawley]

ElCabalero wrote:

(some fun stuff on scouting methods and such)

>
> In article <33A85F...@worldnet.att.net>, Jason Cawley
> <jason...@worldnet.att.net> writes:

>
> >Well, you can always treat them as expendable; put a fuel tank on them
> >and let them go. But they won't be very fast or effective beyond a
> >fairly short range without decent speed.
>
> 300ly radius at warp 6+ on a long hump is enough, if you're willing to
> push them till they die, then let them rot where they fall. Takes a fair
> number to scout that much radius, but it can all be done by 2416.

That sounds fairly reasonable. I don't know that you scout enough
planets for low hab in that range with other than packed, though. I
have had trials in packed with 1/10 where that was about the range I
went to (meaning colonized, not scouted) by year 20 (this was with IFE),
but it yielded 15 planets or so. OK, but not a lot.

I tried looking at some numbers on this. 200 LY, 300 LY, 400 LY search
radius, each density, center start. Just roughly.

Reads "expected planets"

Range Scouted

200 300 400

Sparse 21 47 84
Normal 25 57 101
Dense 32 71 126
Packed 41 92 163

Now, an edge start will cut those numbers in half, and a corner start
will cut them to 1/4. As one can see, the number reached basically
doubles for an extra 100 LY searched. But to me the interesting
question is the absolute total, and its relation to the habitat
fraction. 300 LY in packed should find a 1/6 hab race 15 greens.
That's fairly decent early, but 20 to 25 is better for an empire.
That's more like 1/4 hab, or a little under the 400 LYs.
With e.g. 1/10 hab and an edge start, even 400 LY should average only 8
greens (but perhaps as many yellows under -7 or so). And that's in
packed. If you are CA of course, the modified hab or "plus yellows"
figure is what matters. But the basic relation, it seems to me, is
pretty simple - if you halve the hab plan on going another 100 LY.
And that's what I recommend IFE for (and ISB, too, if at all possible -
for the fuel and/or for the follow-on colonization from the spacedocks).

I don't usually expect the colonization fleets from the HW to go 400 LY
direct. There are exceptions (IS, SD sometimes), but usually that will
be follow-on from secondary worlds. But having them scouted early can
help a lot. Helps direct efforts, and deal with diplomacy/border
things, intelligently.

Low hab needs good speed, to settle
> >places as well as to find them. If you don't have the speed, don't
> >trash the hab. If you trash the hab, get the speed.
> >
> Unless you have enough resources that you can afford to literally
> throw away a dozen or two scouts in the first decade, yes. Every game the
> Supes have played, they've gotten by with Long Hump 6 and plenty of tanks
> until 2436 or so,

Seems pretty late to me. But what do I know?

when they usually jump straight to Prop 9 and the
> Trans-Galactic drive, or even further to the InterSpace 10, probably the
> best deal in Prop besides the Fuel Mizer itself.

The tech 9 ram is a great engine. You don't get that extra 25% speed,
which I agree is big, but they are cheap and their range effectively
unlimited, even with cargo and at speed. Very forgiving, too.

I just build a *lot* of
> scouts, and don't expect each one to get to more than 5 or 6 planets, and
> don't expect them to come home at all.

Right. I never expect scouts to come home. They only do if they scout
out a whole area in one direction. Then they come back for fuel and go
out again, never to return :-) But that's only a couple.

And this is in real blitzes, not
> just testbeds.

Like I say, what do I know? I have been in PBEMs in medium dense where
I have gone as far as 500 LY or more. That was because of open areas
and politics - a quick alliance with a weak race on a favorable border,
that sort of thing. I had IFE and was SD, so getting there was not a
problem (though it took a while, even at warp 9). The aggressive
scouting helped a lot with this - it set up the political deal and let
me stake a claim to territory so far away. The other guy didn't know
how far away I was; it was relatively easy to bluff. Once a few
privateers made it and then a few minelayers, it wasn't really
reversible anymore - he was too weak and needed the alliance I offered.
Then I filled in the area behind.

>
> Then, once I have both the T-G and the Ferret, I build about 25-40
> scouts and shoot them all over, blanket the galaxy. Even the most
> determined space-interdicter has a hard time stopping that flood before
> they send me back a lot of data. And whatever space was supposed to be
> covered by the now-dead scouts gets swarmed by the survivors.

I've had things like that used against me effectively, by JOATs
especially (so cheap for them). But when I play SD, it doesn't work for
them :-)

Wasteful as
> hell, but effective. Also an arrow pointing at my home space, but they
> probably already know that, I'm the one who wasn't scouting.
>
> I have the feeling that the shoulder-to-shoulder scouts lined up as
> far as you can see has made more than one player decide it wasn't worth it
> and let them fly by.

I just love going for those "sprays" :-) I'm like a kid in a candy
store. Does that mean they don't get the info they need? No, they
usually do get some. But I'll kill those things for the sheer pleasure
of it... I suppose that also has a bit of deterrence value - they find
out whether I am prepared.

>
> Keep in mind, this works well in the Dense/Packed universes common for
> blitzes.

With what hab? I'd imagine it would work fine for say 1/4 or better
(including CA adjusts if any); ITs with two launch points might go lower
still. But with really narrow hab I still have my doubts that the tech
3 standards are good enough. I could be wrong, though.

In a Normal/Sparse I'd probably bite the bullet and either take
> IFE or drop NRSE. However, even IFE doesn't increase your effective
> colonization range that much, most feasible speed drops pretty fast past
> 180 ly. A fully loaded Privateer with 3 tanks can't go much past that at
> better than warp 5 or 6, even coming back empty,

I don't find this to be true. Sure, the best speed with medium
freighters and fuel tank and mizer are 192 LY or less (3 years at warp
8). But the privateers can go 250 easy. And you don't have to fill up
every ship in the fleet. Also, ad hoc fuel ships increase the range -
just scouts with fuel pods for most races. IS and SD are much better at
this - one reason I like them so much. IS gets the fuel transport at
the same time as the privateer, the average weight on the haul is lower,
and the lost growth not much of an issue. SD gets the "SD fuel ship" -
mini-minelayer hull, fuel mizer, fuel pod, nothing else. Costs pennies,
carries 650 fuel, adds almost nothing to fleet weight. I used those and
the privateers for the long range colonization in the game I discussed
above. I even ran smart bombers (to the homeworld of a guy that dropped
out) 450 LY away at warp 9. I just added as many "SD fuel ships" as I
needed to make the run.

although you can do a
> little better with a Crash-and-Burn fleet, that doesn't help you support
> it.

True. But with many early colonization issues, if the race is built for
it, that isn't much of a problem. If the pop growth is high and the
mining good, they get big enough to take care of themselves pretty fast.
Once prop 11 is reached, they all tie in with gates, up to 600 LY (500
LY allows missle cruisers at con 13). And other things can be overgated
in a pinch if light enough - once that place is big enough for that.
Again, I usually play IS or SD (if given half a chance :-), and they can
defend themselves decently once they have any kind of base. Also, I
tend to do such long-range things where diplomacy has cleared a path, as
it were.
A few wandering early DDs with rams can help too (with enough fuel to go
forever).

Anyway, a lot of this longer range stuff may not arise in small (I take
it small, packed is the biggest people go on IRC, no?). But PBEMs tend
to be bigger galaxies. They seem to start at medium normal and go up
from there :-)

Anyway, some interesting issues. Thanks.

Sincerely,

Jason Cawley

[Scott Phelps]

On 18 Jun 1997 20:08:20 GMT, elcab...@aol.com (ElCabalero) wrote:
>... within 300 LY, which is about your maximum colonization range
>anyway.

Maybe for YOU....

Scott Phelps

[Paul Hager]

With regard to Jason's comments and insightful analysis...

I ran a test for 30 turns with my own design and Barry's side-by-side.
One observation right off. Barry's design would be improved if prop
were +75% and construction were -50% -- the opposite of his setup.
Grav immune is a natural for the 1 immune variant of AR so there is
no need for rapid development in this area. There is always a need
for rapid construction tech to get first the Ultra and later DS
stations. Also, there is a need to get the Ultra-miner. All of this
argues for cheap construction as an absolute necessity for AR.

In the side-by-side test, I used a medium packed universe with 6 AIs.
My design was 15%, grav immune, with -50% energy and construction,
normal weapons, rest 75% starts at 3. I did not select IFE which meant
that the race started with w-6 engines. Hab was 1:5.

What I found was that the two designs actually tracked pretty closely
with my design pulling slightly ahead early on and staying there through
turn 30. This surprised me. It turned out that although my variant
did settle more planets, I had a much harder time transporting population.
This was slighly mitigated by the development of a couple of other
production sites from which I could launch additional colonizers/explorers.
I believe that at turn 30, which was about the time that the AI was going
to start coming into play, my version had something like 50% more planets
and more "greens". Barry's design had fewer but larger greens.

The key factor for Barry's design was the fuel mizer. I knew that it
is particularly important for AR to have minimal residence time for
population in unproductive transports. The fuel mizer cuts transit
time in half. I think that Barry's design would not have trailed mine
if it used my suggested adjustment because then the fuel mizer would
be available from the start.

What I have concluded is that a canonical AR design has the following
features:

grav immune
cheap energy, construction, weapons
fast transport
decent hab
fairly high growth

What is needed is a mini-max solution where either density/number of
planets is high enough that transit is less a factor, or sufficient
numbers of good planets are available when the fuel mizer is used.
Given what I observed, I suspect that there is a range of acceptable
solutions but the optimal one probably lies somewhere between my
design and Barry's (with the tweak suggested above).

[ElCabalero]

In article <33A8A1...@worldnet.att.net>, Jason Cawley
<jason...@worldnet.att.net> writes:

>ElCabalero wrote:
>(some fun stuff on scouting methods and such)
>> In article <33A85F...@worldnet.att.net>, Jason Cawley

>> 300ly radius at warp 6+ on a long hump is enough, if you're willing
to
>> push them till they die, then let them rot where they fall. Takes a
fair
>> number to scout that much radius, but it can all be done by 2416.
>
>That sounds fairly reasonable. I don't know that you scout enough
>planets for low hab in that range with other than packed, though. I
>have had trials in packed with 1/10 where that was about the range I
>went to (meaning colonized, not scouted) by year 20 (this was with IFE),
>but it yielded 15 planets or so. OK, but not a lot.
>
>I tried looking at some numbers on this. 200 LY, 300 LY, 400 LY search
>radius, each density, center start. Just roughly.
>
>Reads "expected planets"
>
> Range Scouted
>
> 200 300 400
>
>Sparse 21 47 84
>Normal 25 57 101
>Dense 32 71 126
>Packed 41 92 163

That fits with what I've found, near enough. Part of it is the way
that blitzes get set up, 4-6 people usually, and we want to get into the
meat of it quickly. So we use smaller universes, and you're lucky if you
can get a 300 ly perimeter. Mediums only if people feel they have the
time, and there are enough players. And almost always dense or packed, so
we get things moving faster.

>
> Now, an edge start will cut those numbers in half, and a corner
start
>will cut them to 1/4. As one can see, the number reached basically
>doubles for an extra 100 LY searched. But to me the interesting
>question is the absolute total, and its relation to the habitat
>fraction. 300 LY in packed should find a 1/6 hab race 15 greens.
>That's fairly decent early, but 20 to 25 is better for an empire.
>That's more like 1/4 hab, or a little under the 400 LYs.
>With e.g. 1/10 hab and an edge start, even 400 LY should average only 8
>greens (but perhaps as many yellows under -7 or so). And that's in
>packed. If you are CA of course, the modified hab or "plus yellows"
>figure is what matters. But the basic relation, it seems to me, is
>pretty simple - if you halve the hab plan on going another 100 LY.
>And that's what I recommend IFE for (and ISB, too, if at all possible -
>for the fuel and/or for the follow-on colonization from the spacedocks).

By that formula the Superons ought to be going out about 600 ly, even
allowing for the initial +/-5. But the supes are a special case, they are
banking on finding 2 or 3, and leveraging that into a high Bio, which
means a *lot* more planets in the space they've already claimed. When you
are trying for that 25 worlds, how many of those are 80%+? Maybe 5.
Well, I'm after the same five, it's just that those 5 are *all* I'm after.

>
>I don't usually expect the colonization fleets from the HW to go 400 LY
>direct. There are exceptions (IS, SD sometimes), but usually that will
>be follow-on from secondary worlds. But having them scouted early can
>help a lot. Helps direct efforts, and deal with diplomacy/border
>things, intelligently.

Nailing down a border for the Supes is crucial, because their interior
is mostly empty (It's also what they will expand into later). It isn't so
much important where that border is as that it gets set.

>
> Low hab needs good speed, to settle
>> >places as well as to find them. If you don't have the speed, don't
>> >trash the hab. If you trash the hab, get the speed.
>> >
>> Unless you have enough resources that you can afford to literally
>> throw away a dozen or two scouts in the first decade, yes. Every game
the
>> Supes have played, they've gotten by with Long Hump 6 and plenty of
tanks
>> until 2436 or so,
>
>Seems pretty late to me. But what do I know?

Again, special case. Keep in mind, until then they only have maybe 6
colonies, even in that radius. I certainly wouldn't leave it that long
otherwise. But I need Bio more than I need anything else, so Prop gets
left until a single rounds worth can take me to somewhere really useful.
When you're only concerned with a handful of planets, its easier to juggle
fuel concerns. When you have resources to burn early, it's easier to
build tiwce as many haulers and just fill them halfway. By the scale of
what the Supes have to put into Bio, even large, wasteful fleets of
haulers don't come to much.

>
> Wasteful as
>> hell, but effective. Also an arrow pointing at my home space, but they
>> probably already know that, I'm the one who wasn't scouting.
>>
>> I have the feeling that the shoulder-to-shoulder scouts lined up as
>> far as you can see has made more than one player decide it wasn't worth
it
>> and let them fly by.
>
>I just love going for those "sprays" :-) I'm like a kid in a candy
>store. Does that mean they don't get the info they need? No, they
>usually do get some. But I'll kill those things for the sheer pleasure
>of it... I suppose that also has a bit of deterrence value - they find
>out whether I am prepared.
>

Blitz conditions again. Tracking down all of your interceptors, and
trying to give them the orders to chase down that many scouts is a
laborious project, most won't take the time.

>>
>> Keep in mind, this works well in the Dense/Packed universes common
for
>> blitzes.
>
>With what hab? I'd imagine it would work fine for say 1/4 or better
>(including CA adjusts if any); ITs with two launch points might go lower
>still. But with really narrow hab I still have my doubts that the tech
>3 standards are good enough. I could be wrong, though.

1 in 93. 1 in 28 with the +/-5 adjustment. Omni's new Valheru are
making it pay with 1 in 28, and only normal terraforming.

>
>In a Normal/Sparse I'd probably bite the bullet and either take
>> IFE or drop NRSE. However, even IFE doesn't increase your effective
>> colonization range that much, most feasible speed drops pretty fast
past
>> 180 ly. A fully loaded Privateer with 3 tanks can't go much past that
at
>> better than warp 5 or 6, even coming back empty,
>
>I don't find this to be true. Sure, the best speed with medium
>freighters and fuel tank and mizer are 192 LY or less (3 years at warp
>8). But the privateers can go 250 easy. And you don't have to fill up
>every ship in the fleet. Also, ad hoc fuel ships increase the range -
>just scouts with fuel pods for most races. IS and SD are much better at
>this - one reason I like them so much. IS gets the fuel transport at
>the same time as the privateer, the average weight on the haul is lower,
>and the lost growth not much of an issue. SD gets the "SD fuel ship" -
>mini-minelayer hull, fuel mizer, fuel pod, nothing else. Costs pennies,
>carries 650 fuel, adds almost nothing to fleet weight. I used those and
>the privateers for the long range colonization in the game I discussed
>above. I even ran smart bombers (to the homeworld of a guy that dropped
>out) 450 LY away at warp 9. I just added as many "SD fuel ships" as I
>needed to make the run.

Okay, I meant colonizing past 300 ly. Again, blitz conditions, if you
don't plant the flag by 2428, somebody else is already going to be there.

>
>Anyway, a lot of this longer range stuff may not arise in small (I take
>it small, packed is the biggest people go on IRC, no?). But PBEMs tend
>to be bigger galaxies. They seem to start at medium normal and go up
>from there :-)
>

Small packed and dense is most usual for IRC. We do get some mediums,
but those have a lot of players. I dunno about PBEM, all my hosts keep
quitting before the first 10 turns are done.

--Dave

[Randy L King]

Jason Cawley <jason...@worldnet.att.net> wrote:

> Well, I tried them out. My first attempt was somewhat disappointing,
>perhaps because of mistakes in how I did things. I got about the same
>tech year 21 that you had year 20 - I had 10-5-4-4-3-3, just missing one
>level of construction from what you had a year earlier. But I only got
>half the resources you did - 1171 in 2421.

I got about the same.

>
> The main reason was probably that my early scouting efforts weren't
>aggressive enough. These guys really have to do that right. I also, as
>chance would have it, got minimum minerals on the homeworld, and iron
>was a problem. With the low hab, the runs are pretty long, so the iron
>demand for lift is high. That's the hardest to ramp up with remotes as
>well. Some other chance stuff might have been involved - by year 8 I
>had found 2 yellows and one green - the green was a little over 250 LY
>away. That was the closest one after all were scouted, too.

Another possibility on this one, Jason, is that he found a large green
close within 100 LY and was able to colonize it early. That would
explain the near doubling.

I do concur the problem with ramp up with my test, for I had a HW with
only a 32 concentration in Ironium. With my first green 410 LY away,
it was hard to work with. :/

[huge snip]

> Just a question about them - how did you handle the scouting? Also,
>what lift/movement did you do, when you went to yellows and such (I did
>right away, 50000 to each or so)? Anyway, it might help out people like
>me who managed to screw it up ;-)

Enquiring minds want to know. GRIN

[Andrew Turner]

Jason Cawley <jason...@worldnet.att.net> wrote in article
<33A8A1...@worldnet.att.net>...

> I tried looking at some numbers on this. 200 LY, 300 LY, 400 LY search
> radius, each density, center start. Just roughly.
>
> Reads "expected planets"
>
> Range Scouted
>
> 200 300 400
>
> Sparse 21 47 84
> Normal 25 57 101
> Dense 32 71 126
> Packed 41 92 163
>
> Now, an edge start will cut those numbers in half, and a corner start
> will cut them to 1/4. As one can see, the number reached basically
> doubles for an extra 100 LY searched.

Well you said that's a rough estimate. But I'm going to nit-pick anyway
;)

Of course the *real* relationship between range and number of planets is
the are of a circle: pi r squared. If you double the radius from 200 to
400, you get 4 times the area and 4 times the number of planets.
Unfortunately, if you try to extrapolate beyond the ranges you give, your
rule of thumb breaks down rather quickly. Range 100 would give you 25%
(not 50%) the number of planets of range 200. And going from 400 to 500
increases the number of planets by a "mere" 56%.

While trying to play the ARvids, I tried the scouting methods you
suggested, 20 scouts in a hub-and-spoke configuration and I managed to
scout almost the whole of a small-packed universe by 2420. Unfortunately
that left me with a crippling shortage of ironium, so I was unable to
exploit my discoveries. The one time I managed to achieve the target
resources, was when I built only 10 scouts. By allowing a bit more
sideways movement, I was able to reduce the number of turns my scouts spent
in-between worlds. In that game, the universe was completely scanned
around 2450.

That's all the scanning I needed, because the ARvids, with their 1/21
habitability, look a lot better once you get some terraforming ability.
With +/- 11% in Temperature and Radiation they can live on 1 world in 5.
That's a lot of yellows, but half of them end up with good values. So I
was able to get away with scanning "only" about 100 worlds in the critical
early stages. Instead of buying the extra scouts, I bought a quick level
of weapons tech, to give +/- 3% improvement in radiation. I think my
research also took a detour up to +/- 7% radiation terraform a little while
later, but I didn't take notes.

Now I have to work out how to get the same results in a proper game. That
one time I made the resource target, I was completely unopposed, so I never
bothered to build any weapons! ;)

Andrew

[Paul Hager]

With regard to Jason's analysis, I'll offer the following additions...

I decided to run a side-by-side test to see how the two races (my 15%
and Barry's 19%) faired under similar conditions. I put them in a medium
packed universe with 6 AIs. I have a suggested improvement to Barry's
race, BTW. Instead of expensive construction and cheap (-50%) propulsion,
that should be reversed. Cheap construction is an absolute necessity for
AR, in my view. AR needs to get the Ultra station and Death Star ASAP.
Also, construction 15 is needed for the ultra miner. With +75 propulsion,
Barry's design would also start with the fuel mizer instead of having to
research for it. There is no other advantage in getting more prop seeing
as the AR design is grav immune.

My 15% design for the test was slightly modifed from my standard 15% AR.
I shifted weapons cost from normal to -50% and narrowed the hab slightly.
I also selected LSP. (I think my original design was better and if I
were to run a comparison again, I would use that version against a
tweaked version of Barry's.) In any case, my version was grav immune
ISB, NRSE, ARM, NAS, LSP, 15%, 1:5, energy, weapons, const = -50%;
prop, elec, bio +75%, starts at 3.

To my surprise, the 2 designs were very close in performance. The 15%
version found more planets both yellow and green. After a few turns,
the 15% took a slight lead and held it until turn 30 when I stopped the
test -- the AIs were reaching the point that they were going to become a
factor and I basically got tired. In any case, I think I found out what
I needed to know. At turn 30 my design had about 50% more planets,
was slightly ahead in resources, and both were at identical levels (though
differing in details) of tech and leading the AIs.

The key factor was that, although the 15% found more planets and was
able to bring some additional producers on line ahead of the 19% version,
transport speed was a killer. The fuel mizer cuts travel time by at
least 50%, maybe more, for the longer distances. This is critical for
AR, which can't afford to have population languishing unproductively.
The only reason that the 15% held a slight edge in my test was that
I waited until I had energy 6 before getting prop 2 (actually 3 in one
turn) and the miser with Barry's design.

I suspect that if I use my preferred design and the tweaked 19%, they
will remain close.

I conclude from this preliminary test that a canonical AR has the
following characteristics:

grav immune
cheap (-50%) energy, construction; cheap or normal weapons
fast transit
decent hab

Given that it appears that 1:5 with slow transit and 15% plays about
the same early on as 1:20 and 19% with fast transit, it would appear
that there may be some optimal design between these two versions.

[Iannick Daniel]

Alex Hall wrote:

(Sniped introduction text about following race)
(Sniped begining of race description)

> Factories - 1/8/20/G

BA-BA-BA-A-A-A-D-D-D! VERY BAD! :(
(And I mean it...)

Your factory configuration:
- 1.0 Res_Prod/Fact
- 8 Res_Cost/Fact
- 2.0 Fact/1000Pop
- 3 G_Cost/Fact
cost 307 points,
it's efficiencies:
Res_Prod/Res_Cost = 0.125
Res_Prod/1000Pop = 2.000
Res_Prod/G_Cost = 0.333

The new(better) factory configuration
- 1.2 Res_Prod/Fact
- 9 Res_Cost/Fact
- 1.7 Fact/1000Pop
- 3 G_Cost/Fact
cost 278 points,
it's efficiencies:
Res_Prod/Res_Cost = 0.133
Res_Prod/1000Pop = 2.040
Res_Prod/G_Cost = 0.400

The new factoriy configuration is better than your factory configuration
in ALL categories...
Watch-out when you post races on the newsgroup....

(Sniped end of race description)

[William Butler]

Ok Barry I tried your design and I have to admit...I LIKE IT.

I have to admit that I am an AR junkie and have tried many different
versions of the race from the 6% triimune to a high growth noimmune. From
my previous experience I found that the thing that killed them from a
monster perspective was the fact that the resources were proportional to
the hab% and a low hab world took forever to get to a high% world. I
also"knew" that the AR needed to expand to grow(not unlike the HE). When I
saw your race I said "THAT'S A MONSTER?". Because with a 1 in 21
habitability you would shrivel up and die due to a lack of habitable
worlds.

So I decided to play a game with it and WOW did it work. The low
habitability was an issue but not the crippling blow that I thought. And
every world that was habitable was at near 100% in just a few years. The
narrow ranges ensure that a unit of terraforming is VERY efficient. The use
of many scouts was very important and there was a big need to transport big
chuncks of population over fairly long distances, but the net result was a
fast growing monster. Population management went out the window because I
couldn't ship the people out to enough worlds, but that's where the yellows
came in. I just dump a chunk of people on a yellow and in a few years I
have a decent green world. As a matter of fact my results were actually
better than the results that you posted .
By 2425 I had:
13 worlds
4460 res
11-10-4-9-3-3

Now I haven't played past this point yet, but I see no reason why the
growth won't continue in the same fashion.

These results were so spectacular I decided to try it again at home(the
first game was on company time) so I set up a similar game and gave it a
try. This time things didn't go nearly as well. There wasn't a single green
world within a 250ly radius and the best yellow world in that range was at
-6%. OUCH!! I had to do some serious travelling to get even a secondary
world. And most of the green worlds that I did find were in enemy
terratory. Well, I got what greens I could and went for the yellows with
oodles of population in a scramble to get up to speed. Now in contrast to
the first attempt I had the following in 2435(sorry I didn't write down the
2425 data):

20 worlds
6156 res
11-10-4-12-5-4

Note that these results are significantly behind yours. I'm sure that it
will still go on to be a monster...just not as impressively.

So be careful with a race that has a low habitability range because a
little luck( or lack there of) can greatly effect the outcome. This can be
very important in a small/medium universe where you run into the enemy
sooner.

On the upside if you get a poor selection of worlds at the beginning you
will still do well but the scores won't reflect it quite as soon. This can
actually be a good thing in a world full of monster hunters.

Anyway, Great race Barry. I bow to you as the cheif monster designer. May
we call you Dr. Frankenstein? After all you did breath new life into this
race that most people believed to be dead<g>.

Thanks
Bill Butler

[Barry Kearns]

On Tue, 17 Jun 1997 17:52:52 GMT, dom...@nei.franken.de (Dominik
Mayer) wrote:

>On Sun, 15 Jun 1997 23:08:10 -0500, Jason Cawley
><jason...@worldnet.att.net> wrote:
>
>>Barry Kearns wrote:
>>
>>>
>>> Fire it up and give it a try. You just might find that AR is not
>>> quite the pushover you might have thought they were.
>>>
>>
>>

>> Well, I tried them out. My first attempt was somewhat disappointing,
>>perhaps because of mistakes in how I did things. I got about the same
>>tech year 21 that you had year 20 - I had 10-5-4-4-3-3, just missing one
>>level of construction from what you had a year earlier. But I only got
>>half the resources you did - 1171 in 2421.
>>

>Ok, I give up.

>
>I have only reached 920 resources in 2423 with Barrys design. <sniff>

>What am I doing wrong? I have 4 planets (HW included; hab is 100%,

>100%, 77%, 14%), each one got a colonizer immedeately after discovery.
>The HW is around 250000 pop, the others around 60000 pop; I am
>shipping the excess to the other worlds, in loads of two privateers
>with fuel pods only. 5 scouts are zipping around for new worlds; so
>far none to be found.

First big error. You need WAY, WAY more scouts. 25 to 30 scouts is
usually adequate. You've GOT to hyper-aggressively scout the universe
with this race. If you're not basically DONE scouting an ENTIRE
small-packed universe by 2430, you probably weren't aggressive enough.
The worlds will be rare, you MUST find them as soon as possible.

>
>Same thing, i.e. those low resources, happened to me with other
>proposed monster races, too (namely monster CA, JOAT and IT). I am
>really clueless. My testbeds are small standard universes as with all
>options set as stars! proposes.

Another big problem. This race almost *requires* a density of packed.
Dense might be viable, but just barely, I'd imagine.

>I build one scout per turn (beginning a bit later in turn 2404 for non
>AR monsters) up to five or six to find new planets in time; then I
>just colonize and send excess pop over 250k (sometimes I use the 330k
>borderline) away to my new colony. I never ever reached more then 14k
>with any race in 2450. I never ever reached 2000 resources in 2420.
>Perhaps someone could send me the first 50 files of a testbed? Or is
>someone willing to examine a testbed of mine? How are you playing
>(refered to this sample AR)?
>It is now sure that my bad performance is not the fault of the race
>design. But I do not see no flaw in gameplay yet, and I sure do follow
>this newsgroup. HELP!
>
>RU
>Dominik

These are the two main problems, as I see them. What are you doing
research-wise in the meantime?

Barry Kearns

[Barry Kearns]

On 19 Jun 1997 12:38:33 -0500, hag...@cs.indiana.edu (Paul Hager)
wrote:

>With regard to Jason's comments and insightful analysis...
>
>I ran a test for 30 turns with my own design and Barry's side-by-side.
>One observation right off. Barry's design would be improved if prop
>were +75% and construction were -50% -- the opposite of his setup.
>Grav immune is a natural for the 1 immune variant of AR so there is
>no need for rapid development in this area. There is always a need
>for rapid construction tech to get first the Ultra and later DS
>stations. Also, there is a need to get the Ultra-miner. All of this
>argues for cheap construction as an absolute necessity for AR.

Paul, go back and look at the original post. Here, I'll cut and paste
it directly from the original for you.

The race design? (Race name ARvids)
PRT = AR
LRT = IFE, NRSE, ARM, ISB, LSP
Grav immune, Temp 4c to 84c, Rad 66 to 86 ( 1 in 21 hab )
Growth rate 19%
Coefficient of 10
Energy, Const, and Weaps all -50%
Elect, Prop, and Bio all +75%

+75% start at level 3

zero points left over

Notice anything? :)

You transcribed it wrong. The design you were testing was not mine.

>
>In the side-by-side test, I used a medium packed universe with 6 AIs.
>My design was 15%, grav immune, with -50% energy and construction,
>normal weapons, rest 75% starts at 3. I did not select IFE which meant
>that the race started with w-6 engines. Hab was 1:5.
>
>What I found was that the two designs actually tracked pretty closely
>with my design pulling slightly ahead early on and staying there through
>turn 30. This surprised me. It turned out that although my variant
>did settle more planets, I had a much harder time transporting population.
>This was slighly mitigated by the development of a couple of other
>production sites from which I could launch additional colonizers/explorers.
>I believe that at turn 30, which was about the time that the AI was going
>to start coming into play, my version had something like 50% more planets
>and more "greens". Barry's design had fewer but larger greens.
>
>The key factor for Barry's design was the fuel mizer. I knew that it
>is particularly important for AR to have minimal residence time for
>population in unproductive transports. The fuel mizer cuts transit
>time in half. I think that Barry's design would not have trailed mine
>if it used my suggested adjustment because then the fuel mizer would
>be available from the start.
>

Indeed. That's why I designed it from the start in accordance with
your "suggested adjustment". :)

What surprises me is that even with this transcription error, damaging
at least two of the fundamental design precepts, it *still* kept up.
I didn't realize it was *that* fault-tolerant...

>What I have concluded is that a canonical AR design has the following
>features:
>
> grav immune
> cheap energy, construction, weapons
> fast transport
> decent hab
> fairly high growth
>
>What is needed is a mini-max solution where either density/number of
>planets is high enough that transit is less a factor, or sufficient
>numbers of good planets are available when the fuel mizer is used.
>Given what I observed, I suspect that there is a range of acceptable
>solutions but the optimal one probably lies somewhere between my
>design and Barry's (with the tweak suggested above).
>

[Robert M Ashcroft]

My tweeked version of Barry's AR

I am using version 2.7b and there is slight differences to overall
points available when I created a copy of Barry's. Also since I have
been playing Tri-Immune AR race designs against 15 expert AI's, for
months now there are some changes that I consider some what important
for AR.

The first point I would like to make is that I stuffed the first 5
turns. It only slightly hindered my results. The second is that at
2451 I connected with the only MT in the universe. It came trotting
through at about 2445, then another came on at 2451 but is a bit quick
to catch :(.

The design.

Remaining points 2. into surface minerals.
LRTs, IFE, ARM, IS, GR, NRSE, NAS, LSP (though I am not keen on NAS)
Habitat: G - Immune, T: -16 - 104, R: 61-91, GR 19%
Coefficient 10
Research, Energy, Construction -50% (Manditory for AR)
The rest +75, and all 75 start at 3

Result

2451 - 20K
2452 - 21K
2453 - 22K
2454 - 23K
2455 - 26K Techs 17-10-11-16-10-10

Construction tech 17 (Death stars) in two turns

Most major stars have Ultras, and 10 potato bugs arroung them. I am
filling in remaing yellows and greens

Mistakes made involve colony transfers in the first 4 turns. For
those playing this race you will discover colonist management is
critical to development. It does not pay to stuff it. Make sure that
you check your MM throughly

A minor point of digression for the AR race, is when to do miner
production and when to do terraforming and how much to do. I suspect
that I have to do some more play tuning to get the balance correct as
it is here that my main problem in hitting the 25K by 2450 lies.
Another area would be the population suffel that I ignored when I got
Ultras near 2440 when production was about 15K

What did I buld?. Pintas but only what could be produced in a single
turn, about 15 scouts, you don't really need 30 per se. small
freighters about 40 in all, potato bugs (10 per colony), and 5
galleons to do the trade with the MT.

Some notes about playing AR.

Never try to produce something that will not complete in 2 or 3 turns.
You waste your research potential.

If you are playing lower growht rate designs, particularly 10% growth
never waste your colonists (2200 per colonyship).

Try to get 10 potato bugs per colony. This helps orbital
construction.

ARM. Potato bugs only untill you get the Robo Utra Miner or the MT
miner. Drop two of these on a potato bug hul and build 1 per colony,
then an Ultra miners with the above miners. Better investement return
on miner construction. If you do not think this is the case check out
the real cost of your favoured miner designs.

300/500 star gates. Absolutely critical for the next phase of my game
(that starts at about 2450). You need this to speed up mineral
transport cycles. Never build an orbital with out one of these when
you get it. Also include a mineral packet launcher.

I would consider the above design to be a very deadly opponent by
2480. In the hands of a competent player very hard to defeat. Access
to a gensis device does not bear contemplation.

Thanks to Barry for pointing us all in the right direction.

Regards
Robert.

[Dominik Mayer]

On Sat, 21 Jun 1997 03:54:03 GMT, bkea...@ix.netcom.com (Barry
Kearns) wrote:

>On Tue, 17 Jun 1997 17:52:52 GMT, dom...@nei.franken.de (Dominik
>Mayer) wrote:
>>I have only reached 920 resources in 2423 with Barrys design. <sniff>
>>What am I doing wrong? I have 4 planets (HW included; hab is 100%,
>>100%, 77%, 14%), each one got a colonizer immedeately after discovery.
>>The HW is around 250000 pop, the others around 60000 pop; I am
>>shipping the excess to the other worlds, in loads of two privateers
>>with fuel pods only. 5 scouts are zipping around for new worlds; so
>>far none to be found.
>
>First big error. You need WAY, WAY more scouts. 25 to 30 scouts is
>usually adequate. You've GOT to hyper-aggressively scout the universe
>with this race. If you're not basically DONE scouting an ENTIRE
>small-packed universe by 2430, you probably weren't aggressive enough.
>The worlds will be rare, you MUST find them as soon as possible.
>>
>>Same thing, i.e. those low resources, happened to me with other
>>proposed monster races, too (namely monster CA, JOAT and IT). I am
>>really clueless. My testbeds are small standard universes as with all
>>options set as stars! proposes.
>
>Another big problem. This race almost *requires* a density of packed.
>Dense might be viable, but just barely, I'd imagine.
>

<snip>

>
>These are the two main problems, as I see them. What are you doing
>research-wise in the meantime?
>

First of, a big thank you to all guys out there who helped me. I got
lots of emails and even a testbed for comparison.

I got the point about the scouts. This is for me somewhat surprising,
because in never saw that many scouts in the pbems I am in. But now,
as I am testing with the packed small universe I see that they are
really necessary. So many planets!!!! :)

My research strategy is to set the %age to 0 and control the research
with my production queues. When I need a certain tech, I cancel
production accordingly. Ships are only built when absolutely
necessary. Apart from that my research is done by "maxed" worlds only
(i.e. those 250k breeder worlds which are transporting off like hell).

A pity that I do not have very much time at the moment (mid-exams;
only 2 pbems by now...). But I will keep the ng informed where my
mistake was; it sure is something interesting for other beginner guys
out there.

RU
Dominik

[Iannick Daniel]

ElCabalero wrote:
>
> In article <33A5D5...@id.ml.org>, Iannick Daniel <i...@id.ml.org>
> writes:
>
> >So high habitability is better than high growth rate, because low
> >habitability reduce reproduction AND production for AR races.
> >
>
> High planet habitability, not high race Hab.

Erm, I was saying High planet habitability, but I wrote High habitability! :(
Sorry, my fault... (1 immune rules)

<Sniped the rest of the stuff>

> --Dave

[alex hall]

Iannick Daniel <i...@id.ml.org> wrote:

>Alex Hall wrote:

>(Sniped introduction text about following race)
>(Sniped begining of race description)

>> Factories - 1/8/20/G

>BA-BA-BA-A-A-A-D-D-D! VERY BAD! :(
>(And I mean it...)

[smip of details]

>The new factoriy configuration is better than your factory configuration
>in ALL categories...
>Watch-out when you post races on the newsgroup....

>(Sniped end of race description)

Yes the race wasn't perfect, in fact when I posted the original post
I was already 3 evolutions betond that design, one of the things
that changed was the hab, factories and mines.

However the design does hit 25k by 2450 and I posted it
to help other people playing WM or wanting to play WM
see how I personally went about race design, the idea being
they would adapt the race themselves.
I never post the latest design I am playing with for a number of
reasons... :-)

Anyway.. how viable are these WM races.....

Well in tests in a small and medium universes against AI the
latest ones are hitting 40->50k by 2450 and having good tech
levels. Obviously at the end of the day it is the games that count...
I have played various variations of the WM designs I have on blitz irc
as it is a great way to test new designs quickly. So far I have played
in 3 irc games:

Game 1: A Red Eyed Morning - (it was!!! I got to bed at 4am!)

6 players, small dense distant, acc bss, public scores, galaxy
clumping.

3 of the players had dropped out within the first 25 or so years.
1 because I had destroyed his starbase on his HW very eaerly on.
The 3 people left were all in the top 3 positions... was WM (me)
AR (Pendragon) and CA (Warner). I had an alliance with Warner
which gave me free terraforming.

My race was 1st uptil the end in 2494 except for 2 years where I
dropped to 2nd and pendragon was 1st. I played the game
aggressively. The resource scores weren't very impressive
I only got 13k by 2450 and 40k by 2494. however I was in the lead
with resource all the way to 2490.

Game 2 : A Saturday Afternoon.

Alas Shawnbot went AWOL around 2426. I had been in 1st place
upto 2424 when Karrde took over. This was a medium sized
universe.

Game 3: Karrde's Game

This game used the very latest WM design, that also
gets rid of NAS (not very good for blitz)

Small, Packed, DIstant, Acc BSS, PPS, 6 players (1 was AI)
All human players stayed in to end in 2460 as people then had to
go (was 5am here in the UK!)

This game went much better resource wise, I didn't go to war for
quite a while in order to build up an economy.

2k res by 2420
7k res by 2430
14k res by 2440
21k res by 2450

I was 1st thru whole game.

Alas there was an HE in the game so I ran out of planets to colonise
around 2430 then the next 3 planets I colonised were by force, it
would have been interesting to see how the res would have been
if there had been more space.

around 2446 I began fighting another WM in the game.

in 2460 I had the follwoing.

res = 27k
escort ships = 280
tech level = 65
planets = 19

Anyway I hope that lets people see that WM designs are capable
of competeing in games. However yes my designs tend to
get overtaken in the mid->late game depending on the
opponents skills and starting positions.

:-0

Alex aka Morpheus on starlink.