SGOTM 09 - Xteam

What has happened in BtS is that the hammer to beaker conversion ratio is changed from 3:1 to 1:1. This means that in the late phase of a game you will often get more beakers for your hammers when building science directly rather than building infrastructure in your cities or units.

Coming up with a decision rule would be a fun exercise for me. I would guess that it's not possible to get both bonus multipliers for the hammers and the research applied together, though.

I can imagine four cases:

Case A (neither bonus)
raw hammers ==> actual beakers

Case B (research bonus)
raw hammers ==> raw beakers
those raw beakers * research multiplier ==> actual beakers

Case C (production bonus)
raw hammers * production multiplier ==> actual hammers
those actual hammers ==> actual beakers

Case D (both bonuses)
raw hammers * production multiplier ==> actual hammers
those actual hammers ==> raw beakers
those raw beakers * research multiplier ==> actual beakers

Does anyone know offhand which case it is?

Grover
 
Before we get to the real game I think it would be good if SCT or Fred could detail how the "hammer research" process that one of the other teams used effectively in the last SGOTM works. In general we should have a startegy from the start of what key techs are required to get to the max research level. So even though Radio may be available for free from Liberalism there maybe a better tech that will speed us to Corporations and Assembly Lines faster.

As Fred said, hammer research turns hammers into beakers or gold at a 1:1 ratio. When a city builds wealth (available with Currency) or research (available with Alphabet), the hammers receive the hammer bonuses from forges, factories, plants, Ironworks, Bureaucracy, and State Property. Beaker bonuses and gold bonuses do nothing. So a city with a lot of workshops, mines, lumbermills, and watermills (and Mining Inc) can produce a lot of beakers or gold by building hammer multipliers buildings.

The question of whether to use a hammer economy in a city should arise during the midgame when we start capturing mature cities and can either build cottages or workshops. If a captured city already has a lot of villages and towns, then it doesn't make much sense to bulldoze them. But what of a city with none? With State Property, Caste Systems, Guilds, and Chemistry, workshops add 4 hammers to a tile with no loss of food. With State Property and Electricity, watermils add 1 food, 2 hammers, and 2 commerce. Compare that to cottages, which start off adding only 1 commerce, but eventually can grow to add 1 hammer and 7 commerce. A mature town with Printing Press, Free Speech, and Universal Suffrage is better for research than the best workshop or watermill. But cottages take a while to become towns (67 turns on Quick?). For each city, we want to calculate whether a hammer economy or a cottage economy will produce greater total benefit through the modern era.

The main buildings for a hammer city (forge, factory, coal plant) cost about the same as the main buildings for a science city (library, university, observatory, lab). However, a hammer city can build its needed buildings faster. On the flip side, being Philosophical will let us grow a lot of GSs, so we can build academies in a lot of cities, giving a commerce economy a bonus.

Other factors to consider are that 1) adopting Emancipation doubles cottage development speed and that 2) corporations don't work while in State Property. A cottage economy can overcome the initial advantage of the hammer economy fairly quickly with Emancipation. If pursuing this avenue, taking Democracy with Liberalism is a strong play. On a map with a lot of gold, silver, copper, iron, and coal, Mining Inc can be a powerful addition to both the hammer economy and the cottage economy. But heavily-workshopped cities may face a food deficit when not in State Property. Sushi/CerealMills can sometimes make up this deficit.

I have become a fan of the hybrid hammer-cottage economy for space colony games. In this gambit, one wants to have enough hammer cities to build enough wealth to keep research at 100% from the midgame through the spaceship launch. As we saw in SGOTM08, a typical economy can convert 1 gold to more than 1 beaker, because of the greater prevalence of beaker bonus buildings. Maximum research speed is acheived by using wealth building and GM trade missions to allow deficit research with minimal gold reserves. The hammer cities are ideal places to build units in the midgame and spaceship parts at the end. So one may aim to have enough hammer cities to BOTH build enough wealth for 100% research AND build units and spaceship parts. As in SGOTM08, we can pre-build workshops over towns so that at the end, we can build a lot of SS parts simultaneously.

The one immutable truth in any space colony game is that no matter what type of economy one chooses, more land and more cities = faster launch. A top space race game should start out as a controlled domination. By the time the modern era rolls around, one should be nearing the domination limit. I'm happy to say that we have the master of controlled domination on our team: Cactus Pete.
 
Grover22 said:
Coming up with a decision rule would be a fun exercise for me. I would guess that it's not possible to get both bonus multipliers for the hammers and the research applied together, though.

There are some decisions that can be made based on calculations - e.g. the return of building a Courthouse can be estimated fairly accurately if you have an idea of the total length of the game and the turn number when State Property is adopted (If this is going to happen). In many cases we will probably have to rely on approximations and maybe even intuition (This is where CP comes into the picture :D). Any effort to put numbers on our actions would be most welcomed since intuition is not always good enough - particularly in a game that has been altered because this invalidates experience from standard games to some extent.

ShannonCT said:
For each city, we want to calculate whether a hammer economy or a cottage economy will produce greater total benefit through the modern era.

Another reason for getting an estimate of the end turn number. Can HOF help with that?

ShannonCT said:
The one immutable truth in any space colony game is that no matter what type of economy one chooses, more land and more cities = faster launch. A top space race game should start out as a controlled domination. By the time the modern era rolls around, one should be nearing the domination limit. I'm happy to say that we have the master of controlled domination on our team: Cactus Pete.

This sounds very reasonable. The question is how to determine the "best" rate of expansion.

It seems quite clear that most cities will contribute commerce/beakers when we sum up over the whole game length. The first few cities will have very low maintenance and will thus provide commerce from the moment they are founded. This is particularly true if there are workers ready to improve city tiles immediately. As you get more cities costs go up for two reasons: One is the linear growth due to having one more city and the other is the quadratic growth due to increasing "Number of cities" maintenance that go up for all cities. The quadratic growthy may be saturated at some point - I don't know the specifics. Anyway, later cities will loose commerce in the beginning due to maintenance and it will take longer time before they produce more than they cost. Captured cities are more favorable because they come with a population and if large enough they may contribute commerce immediately. If some of you play test games, could you record the jump in maintenance when founding 2nd, 3rd, 4th... etc city.

Still, the answer to the question of "best rate" expansion is not answered by this. Getting late - I will come back to this issue another time :)
 
If some of you play test games, could you record the jump in maintenance when founding 2nd, 3rd, 4th... etc city.
With the 2nd city - +1 to Distance and +1 to Number of Cities
With the 3rd city - +1 to Distance and +1 to Number of Cities
With the 4th city - +2 to Distance and +2 to Number of Cities.
With the 5th city - +3 to Distance and +3 to Number of Cities.
With the 6th city - +3 to Distance and +3 to Number of Cities.

+20 Maintenance per turn, ouch!! :(
 
I've had my fun finding a formula for the break even point for building a building to increase the research bonus vs just dedicating the hammers directly to research. The calculation was disappointingly simple, which may mean I got it wrong, and which probably means it's uninteresting/inanely trivial to you all. Probably you've all figured this out for yourselves already. The calculation was actually a little more fun in the other bonus cases (both research and production bonuses, or just the research bonus applied to the raw hammers.) But here it goes, anyway.

Let...
T1 = the number of turns spent building the building
T2 = the number of additional turns after the building finishes until the break even point is reached.

The total number of turns to break even then is the sum T1 + T2.

More definitions...
B = raw beaker rate (beakers per turn)
C = cost of the building in hammers (total cost)
E = extra bonus to research rate from the building (expressed as a decimal, 25% = .25, for instance)

I considered two cases, and the answer came out the same both times.

Case A:
build the building for T1 turns then dedicate hammers to research for T2 turns thereafter
vs
build research for T1 turns and research for T2 turns thereafter

Case B:
build the building for T1 turns then dedicate hammers to something other than research for T2 turns thereafter
vs
build research for T1 turns then something other than research for T2 turns thereafter

In either case, T1 = C/H and T2 = C/BE. The total number of turns to break even is T1 + T2. We're getting behind for the first T1 turns, and catching up again for the next T2.

Details for anyone who cares:
Spoiler :

One more definition...
R = the research multiplier to convert raw beakers to actual beakers, expressed as a decimal (25% bonus = 1.25, for instance)
It'll turn out that R drops out, anyway.

Case A:
hammers go to building for T1 turns and to research for T2 turns
total beakers: BRT1 + B(R+E)T2 + HT2

hammers go to beakers for T1 + T2 turns
total beakers: BRT1 + HT1 + BRT2 + HT2

equating those two expressions
BRT1 + B(R+E)T2 + HT2 = BRT1 + HT1 + BRT2 + HT2

simplifying a little,
BET2 = HT1
T2 = HT1/BE

but HT1 is just the cost of the building, in hammers, which is C
C = HT1,
T1 = C/H
T2 = C/BE

Case B:
hammers go to building for T1 turns and to non-research for T2 turns thereafter
total beakers = BRT1 + B(R+E)T2

hammers go to research for T1 turns and to non-research for T2 turns thereafter
total beakers = BRT1 + HT1 + BRT2

equating them,
BRT1 + B(R+E)T2 = BRT1 + HT1 + BRT2

and simplifying,
BET2 = HT1, as before

T1 = C/H
T2 = C/BE

 
Shannon has already dealt with the most important parts of any race to space. Everything I add is on the assumption that the map resembles something that might be generated by the standard map generator.

State property hammer economy is not optimal for speed. Mining inc will almost always provide more benefit.
Push the dom limit. More land means more resources and more cities. There are times when it might make more sense to vassalise an AI and demand their mining resources, can claim the same number of resources for only half the landtile points.
With enough resources fueling mining inc it is worthwhile to spam cities in the most rediculous locations just to have them build courthouse+forge and then nothing but wealth for the rest of the game. The math can be dealt with later but it's something to keep in mind.


Maximise capitals commerce
This usually means the obvious academy+bureau+oxford trio asap. Cottage as many tiles as possible (15+)
Use every trick in the book to run at 100% research for as many turns as possible.
We definitely will want at least 3-4 other heavily cottaged cities too though. It's complicated a lot by all of these changes with the additional techs so it's hard to be specific without more testing (which I'll be working on soon).

One of the biggest problems I encounter with my games is getting the great engineer in time for mining inc while still pushing out a suitable number of GP (it's either a case of slowing down the GP farm just to let a GE pop elsewhere, or pollute the genepool and hope for the best), Hopefully this will be less problematic with a philo leader and players with more experience managing GP.
 
ok, I played around with the start and everything got a lot more complicated than I expected. There really is a very strong case for electricity+environmentalism windmills+watermills as the primary focus across the entire empire. Lumbermills don't seem all that potent so there's room to turn the non river/hills tiles into cottages
There's a question to be asked about a possible free market switchover later, but for now nothing seems stronger than the electricity slingshot, after which we can tech through to Civil Service manually.

Is it too early to be discussing the first scout move/possible capital locations?

There are a million subtleties to space colony that could be discussed but they're all hypothetical until we have more info.
 
In either case, T1 = C/H and T2 = C/BE. The total number of turns to break even is T1 + T2. We're getting behind for the first T1 turns, and catching up again for the next T2.

So if T1 + T2 < (number of turns until last SS tech), build the building
and if T1 + T2 > (number of turns until last SS tech), don't build the building.

I think I intuitively make this kind of calculation when I build a building. The one factor you need to add in though is the fact that B is not a constant function. City growth and cottage development increase commerce over time. And increases or decreases to the tech slider affect B as well. Also, buildings have additional effects - forges and markets can bring in happiness that can increase B; forges, markets, and libraries allow specialists, which can generate some GP that gives am important global effect.
 
ok, I played around with the start and everything got a lot more complicated than I expected. There really is a very strong case for electricity+environmentalism windmills+watermills as the primary focus across the entire empire. Lumbermills don't seem all that potent so there's room to turn the non river/hills tiles into cottages
There's a question to be asked about a possible free market switchover later, but for now nothing seems stronger than the electricity slingshot, after which we can tech through to Civil Service manually.

Is it too early to be discussing the first scout move/possible capital locations?

There are a million subtleties to space colony that could be discussed but they're all hypothetical until we have more info.
Nice work PK and G22! :goodjob:

I think it is fine to begin the Scout move discussion. One of the things we have done in the past is to move the Scout and post a screenshot of what is found. Then we can build a quick test save and work out what we think the best solution may be.

This process may continue as often as we need to update things as we move along. If you wish to read this page of our SGOTM08 thread, it may help to better understand. :)

Sorry I'm a bit useless as I'm thinking about heading to Maine this afternoon and tomorrow. Should be back and ready by Wednesday evening... :rolleyes:
 
Maximise capitals commerce
This usually means the obvious academy+bureau+oxford trio asap. Cottage as many tiles as possible (15+)
Use every trick in the book to run at 100% research for as many turns as possible.
We definitely will want at least 3-4 other heavily cottaged cities too though. It's complicated a lot by all of these changes with the additional techs so it's hard to be specific without more testing (which I'll be working on soon).

Yes, the availability of full-strength watermills and windmills (and forest preserves) may require a tweak to the usual strategy of cottaging everything in sight. The immediate benefit of a watermill w/ Elec is 2h and 3c while the benefit of a cottage is 1c for 7 turns, then 2c for 13 turns, then 3c for 20 turns, then 4c, until other techs and civics push that up to 5c, then 1h and 7c. The watermill is dominant for 40 turns, and never clearly surpassed until FS/US civics are adopted.

One of the biggest problems I encounter with my games is getting the great engineer in time for mining inc while still pushing out a suitable number of GP (it's either a case of slowing down the GP farm just to let a GE pop elsewhere, or pollute the genepool and hope for the best), Hopefully this will be less problematic with a philo leader and players with more experience managing GP.

Forges in every city to allow an engineer specialist in every city would give a Philo leader a very good chance of generating one by the time Railroad and Corp are researched. Or we could try for a Mids/HG/Hagia/forge city that would guarantee a GE.
 
ok, I played around with the start and everything got a lot more complicated than I expected. There really is a very strong case for electricity+environmentalism windmills+watermills as the primary focus across the entire empire. Lumbermills don't seem all that potent so there's room to turn the non river/hills tiles into cottages

Hills tiles should be windmills I guess. Grass forests can be chopped and cottaged, unless they are riverside, in which case they can be chopped and watermilled if possible, or lumbermilled for +1h and +1c, or preserved for +3c and +1happy.

Is it too early to be discussing the first scout move/possible capital locations?

I've been moving my scout northwest and west to get a better view of the coast. The thinking is that there would have to be something amazing up there to entice me to settle on the coast. Assuming there isn't, I would settle 1S to free up the hill for a windmill and to allow for the possibility of more watermillable tiles. I think we want to minimize coastal tiles in our capital.
 
leif erikson said:
With the 2nd city - +1 to Distance and +1 to Number of Cities
With the 3rd city - +1 to Distance and +1 to Number of Cities
With the 4th city - +2 to Distance and +2 to Number of Cities.
With the 5th city - +3 to Distance and +3 to Number of Cities.
With the 6th city - +3 to Distance and +3 to Number of Cities.

+20 Maintenance per turn, ouch!! :(

From these numbers it looks like the maintenance increases are going up linearly suggesting a quadratic growth of the total maintenance since the derivative of a quadratic function is a linear function.

ShannonCT said:
Yes, the availability of full-strength watermills and windmills (and forest preserves) may require a tweak to the usual strategy of cottaging everything in sight. The immediate benefit of a watermill w/ Elec is 2h and 3c while the benefit of a cottage is 1c for 7 turns, then 2c for 13 turns, then 3c for 20 turns, then 4c, until other techs and civics push that up to 5c, then 1h and 7c. The watermill is dominant for 40 turns, and never clearly surpassed until FS/US civics are adopted.

Since hammers can be converted to commerce at 1:1 ratio I guess that we can add hammers and commerce into one "yield" number, y. This means that the watermill is yielding a constant y=5 while the cottage is yielding:

y=1, t<=7 (Cottage)
y=2, t<=20 (Hamlet)
y=3, t<=40 (Village)
y=4, t<=T_PP (Town)
y=5, t<=T_FSUS (Town)
y=8, t>T_FSUS (Town)

With Emancipation the numbers 7,20,40 will be only half. T_PP is the time when Printing Press is discovered and T_FSUS is the time when Free Speach and Universal Suffrage is adopted. It should be noted that adopting US comes at the cost of missing the extra specialist beakers from Representation.

Grassland forests present a special problem, because the forest preserve is only yielding y=3 (1h2c) but at the same time raising the happy cap of one or more cities. These extra citizens will also produce commerce and you have to factor that in to the value of the forest preserve. If that extra citizen is working cottaged grassland or a windmill/watermill (not to mention a resource) the additional yield seems worth it. Forest preserves inside overlapping fat crosses are particularly strong raising the happy cap in several cities. So when do we not want forest preserves? If there are no decent tiles left to work and no excess food to hire specialists. Or if we are not near the happy cap due to lack of food. So decisions on using forest preserves probably ned to be taken on a per city basis.
 
Grassland forests present a special problem, because the forest preserve is only yielding y=3 (1h2c) but at the same time raising the happy cap of one or more cities. These extra citizens will also produce commerce and you have to factor that in to the value of the forest preserve. If that extra citizen is working cottaged grassland or a windmill/watermill (not to mention a resource) the additional yield seems worth it. Forest preserves inside overlapping fat crosses are particularly strong raising the happy cap in several cities. So when do we not want forest preserves? If there are no decent tiles left to work and no excess food to hire specialists. Or if we are not near the happy cap due to lack of food. So decisions on using forest preserves probably ned to be taken on a per city basis.
Really, isn't this only in terms of Worker turns? Early in the game, we cannot work all the tiles in a city. On the least valuable grassland forest tiles, away from rivers, we could build Forest Preserves as we need them to sustain pop growth to work the more valuable tiles. As we gain happiness, through other resources and buildings, we could exchange the forest preserves for other needed improvements and chopping hammers. If we use a hybrid hammer economy, then Workshops. If not, we could build some cottages, especially if we are in Emancipation? :hmm:
 
Really, isn't this only in terms of Worker turns? Early in the game, we cannot work all the tiles in a city. On the least valuable grassland forest tiles, away from rivers, we could build Forest Preserves as we need them to sustain pop growth to work the more valuable tiles. As we gain happiness, through other resources and buildings, we could exchange the forest preserves for other needed improvements and chopping hammers. If we use a hybrid hammer economy, then Workshops. If not, we could build some cottages, especially if we are in Emancipation? :hmm:

Yes, we can use forest preserves early to get big fast. (We also get +6 health from Environmentalism, so health will not be a problem early.) And they're not bad to work temporarily (2f1h2c) while we grow to the happy cap. Riversides that can't support watermills probably should get cottages. They won't take too long to surpass forest preserves in total yield. When we start getting Calendar resources, forges, markets, odeons, etc., we should replace preserves with cottages. That will still be early enough in the game where towns should outperform workshops in the long run.
 
It seems quite clear that most cities will contribute commerce/beakers when we sum up over the whole game length. The first few cities will have very low maintenance and will thus provide commerce from the moment they are founded. This is particularly true if there are workers ready to improve city tiles immediately. As you get more cities costs go up for two reasons: One is the linear growth due to having one more city and the other is the quadratic growth due to increasing "Number of cities" maintenance that go up for all cities. The quadratic growthy may be saturated at some point - I don't know the specifics. Anyway, later cities will loose commerce in the beginning due to maintenance and it will take longer time before they produce more than they cost. Captured cities are more favorable because they come with a population and if large enough they may contribute commerce immediately. If some of you play test games, could you record the jump in maintenance when founding 2nd, 3rd, 4th... etc city.

Still, the answer to the question of "best rate" expansion is not answered by this. Getting late - I will come back to this issue another time :)

The Empire Size maintenance stops increasing after 14 cities on these settings. Empire Size maintenance also increases in a city as population increases, but it looks like it's capped at 6.0. That's really quite small compared to the commerce a city can make with a few tile improvements. The availability of full-powered windmills and watermills means that new cities can pay for themselves very quickly. Whether by settling or by conquering, there looks to be a strong payoff for early expansion in this game. As soon as we have Currency, we can build wealth and easily stay at 100% research. So we should not fear expansion.
 
ShannonCT said:
Yes, we can use forest preserves early to get big fast. (We also get +6 health from Environmentalism, so health will not be a problem early.) And they're not bad to work temporarily (2f1h2c) while we grow to the happy cap. Riversides that can't support watermills probably should get cottages. They won't take too long to surpass forest preserves in total yield. When we start getting Calendar resources, forges, markets, odeons, etc., we should replace preserves with cottages. That will still be early enough in the game where towns should outperform workshops in the long run.

Exactly, we need the forest preserves early on to maximize city size and then as we get other mean of happiness they can be turned into something with a higher yield. Some planning of where to put the forest preserves should go together with dot-maps since forest preserves in overlapping fat crosses will add happiness to more cities.

ShannonCT said:
The Empire Size maintenance stops increasing after 14 cities on these settings. Empire Size maintenance also increases in a city as population increases, but it looks like it's capped at 6.0. That's really quite small compared to the commerce a city can make with a few tile improvements. The availability of full-powered windmills and watermills means that new cities can pay for themselves very quickly. Whether by settling or by conquering, there looks to be a strong payoff for early expansion in this game. As soon as we have Currency, we can build wealth and easily stay at 100% research. So we should not fear expansion.

This is my intuition as well. Maximum focus on expansion could well be the way forward. The difficult part will be to decide which wonders and city infrastructure (buildings) should be given up or postponed in favor of settlers/workers and military units. Maybe we should make a list of early wonders and buildings and see if we can analyze which ones are crucial to build and which ones are unimportant or at least has lower priority.

EDIT: For starters I would say that Oracle is important and so are granaries.
 
Unless there's something spectacular to the north, I think a inland capital is stronger. If we move the settler south, we might find more hills or riversides for windmills and watermills. A city placed on the north coast later could share the capital's corn and serve as a GP farm. I've been settling my capital 1 south of the settler.

My thinking is the same. My one concern is that the land ends just south of the visible trees, in which case moving the settler south would waste us a turn.

This raises the question of moving the scout south of the settler to check it out first, if it shows up coastline in the fog we'd have to re-evaluate.
I'm not sure that moving the scout onto the blue circle in the NW would reveal anything good enough to justify putting the capital up that way. Having an early GP farm that can share the corn though is defintely a strong move.
 
This is my intuition as well. Maximum focus on expansion could well be the way forward. The difficult part will be to decide which wonders and city infrastructure (buildings) should be given up or postponed in favor of settlers/workers and military units. Maybe we should make a list of early wonders and buildings and see if we can analyze which ones are crucial to build and which ones are unimportant or at least has lower priority.

EDIT: For starters I would say that Oracle is important and so are granaries.

Oracle is the only must-have wonder. Some others that would be nice:

Pyramids: for GE points, happiness, and beakers
Hagia: for GE points and 50% worker bonus (road-building speed is actually doubled)
MoM: for extra 3 turns per golden age (which we could have 4-5 of with our Philo trait)
Hanging Gardens: for more GE points in the Mids/Hagia city (though Mids/Hagia/forge is adequate)

Buildings that we should focus on:

granaries: everywhere
forges: next build after granary in most cities
libraries: at least 6, and especially a quick one where we can grow a GS before the Oracle Prophet pops
academies: first in the capital, and then in other commerce cities
barracks: in 2-4 cities before we rush someone
airport: 1 in a central location that can airlift units to the front lines

My thinking is the same. My one concern is that the land ends just south of the visible trees, in which case moving the settler south would waste us a turn.

This raises the question of moving the scout south of the settler to check it out first, if it shows up coastline in the fog we'd have to re-evaluate.
I'm not sure that moving the scout onto the blue circle in the NW would reveal anything good enough to justify putting the capital up that way. Having an early GP farm that can share the corn though is defintely a strong move.

OK, that makes sense. If there's nothing that could justify settling on the coast, then move the scout south. Though I think that even if the scout sees some coastline in the south, it still may make sense to free up the hill for a windmill.
 
From the maintenance thread:

Just so you know.. the game WILL have barbarians... if they pop or not in the game depends on how your give the techs to the civs...

If you give the techs to the civs in the worldbuilder, then no barbarians will pop during the game.. however if you do it by editing the WB file in a text editor, then they will pop
(the secret is what the StartingEra= line in the individual player sections in the WB file says)

Actually.. barbarians will be a bit meaner in this game than in normal games... so beware... :evil:

When I created the original save in worldbuilder, giving everyone Superconductors automatically started everyone in the modern era, and thus eliminated barbs. Apparently that was wrong. So I editted the worldbuilder file so we start in the ancient era. Barbs are back baby!
 

Attachments

Back
Top Bottom