Detailed analysis: # of specialist cities needed per era

[UncleJJ]

Futurehermit, as I see it the problem with your initial analysis was that you, in terms of my science equation, simply added up all the values of S in all your cities expecting that to give you a useful rate for research. That doesn't happen in the early or middle game.

Then other posters criticised your approach and you adapted your analysis to include B. So now you're thinking in terms of S * B and that is an improvement . I want to convince you to adopt the full generalised solution and then all will suddenly become clear as the scales fall off your eyes

Let me restate my equation :

Science = (C * R + S) * B

This equation is hard to apply directly to a game in a theoretical way which is why so many people find it hard to argue the points and get confused about how the numbers relate to each other. The values of C will vary from city to city and will change over time as the city pop grows and works more tiles and as cottages mature and as trade routes develop. The value of R will vary as you expand your empire by settlement and conquest and then take action to increase effiency. And you will be building libraries, monastries and so on to boost the science output in your key cities. So is it all too complicated to analyse?

No. Using Binary Research as advocated by Zombie can help us understand the complex situations very well and to compare how a specialist economy is doing versus a cottage based one. Let me explain that by an example from my game at 200 AD. Simply switch the science slider to 100% and 0% and record the values.


At 100% and (0% ) research my Empire gives (taken from the finacial advisor)

Research 130 (36)
Gold 0 (59)
Expenses 17 (17)
Income -17 (+42)

And my 4 cities give the following research (taken from the cities screen)
100% 88, 27, 13, 1
0% 22, 7, 7, 0
From this you can tell that my capital is contributing more than half the science output and that it has a good value of B. Furthermore the second and third cities are running 2 specialists and only a library. The 4th city is very new and only has an obelisk at present. You can tell that the second city has a small but nice commerce output that is higher than the output from 2 specialists (without Representation) and would equal it even with.


A cottage economy would have higher commerce and hence more research at 100% but almost zero at 0% and that is the primary difference at this stage between the 2 types of economy. Cottage economies are much more affected by the value of R.

At 200 AD in my game I had reached an interesting point where an important decision had to be made. My 3rd GS had just appeared and I had the option of settling him in my capital to increase my value of S there by 6. Alternatively I could move him to my second city and increase the value of B there from 25% to 75%. But instead I chose to cash him in to research Philosophy (worth 1232 beakers and showing 10 turns of research) That founded Taoism in my 3rd city and gave me a missionary...who was sent to my capital and allowed me to build a Taoist monastry (a few turns later) to increase B from 85% to 95% thereby further boosting long term research .

 

[futurehermit]

@ uncle jj

nope, it's for sure not flawed. it assumes a 0% science slider where you get your maintenance money from conquering cities.

i played a specialist game last night on monarch (stepped down a level just for ease of play) and i had the tech lead (running rep w/specialists) shortly after 1AD and never looked back. also, because i was running 0% science i was able to conquer and hold far more cities. this is because cottages take time to mature, whereas a couple food sources, a library, and two specialists comes faaar faster!

i played a game this morning where i went cottages and 100% science slider because louis's capital was smack dab next to mine and i also had copper, lol! i ended up losing the game because i was operating under the notion that i could capture and keep as many cities as i could in my specialist games. it crippled my economy!

at around 850AD my most mature cottages in my capital that i worked from very, very early on were only at villages (3c/turn). two of them = 6c/turn. one specialist with representation is 6b/turn!!! so look at the CENTURIES of beakers LOST by going for cottages over specialists with representation.

the difference is so phenomenal that i can't understand why anyone in a single player game would opt for any other strategy than to beeline for pyramids for specialists with representation.

this whole experience has made me replace aggression with industrious as the trait that i consider the most powerful.

 

[futurehermit]

Something I've been thinking about:

Running representation...

In a newly captured city, once you chop or pop rush a library, you only have to farm/herd one food resource or farm two grasslands tiles in order to run two scientists for 12 beakers per turn.

How long would it take in a newly captured city to get cottages up to a comparable number? A long time methinks...

This is why you can capture soooo many more cities, run 0% science, and yet maintain a very respectable, even powerful, economy!

And if you choose to run caste system when the captured city sizes start hitting size 6, then you don't even need the library (of course it's still good to get one for the 25% bonus). just get the food sources and bam your city is pumping full-tilt beakers.

Is there a more powerful strategy? I am extremely, extremely sceptical

 

[UncleJJ]

Interesting futurehermit Can you tear yourself away from the new method to supply some statistics for the conquest fuelled game? Stats like how many cities and what income and costs are. Perhaps use the format I outlined above. How many specialists were you running, were they all scientists, and how many GPs did you make?

We all know the AI is weak at war once you get a significant technical lead and the only limit to expansion then is the number of cities you can maintain after which you have to raze any newly captured ones. You seem confident you have a method that can fuel that expansion which is very interesting... will it work at higher levels or is the difficulty in getting the Pyramids the limitation?

 

[futurehermit]

^^Difficult getting the pyramids is a limitation, no question. Industrious is obviously important, and stone is huge. With stone and industrious pyramids is no problem. With industrious and no stone, it's difficult. With stone and no industrious, a bit easier. With no stone and no industrious, probably impossible. (All of these I'm talking about emperor.)

How many cities: During classical I was running around 6 cities or so with about 4 of them running specialists and 2 of them production centers (Built 2 and there were some nice tasty barb cities nearby). I also had the great library and used my first GS (I didn't get a GE that game for once) on an acad in my GL city. I settled a second GS later in this same city.

All the specialists in my game were scientists.

Income and costs, I can't tell you because I don't attend to this. I just run 0% science and manage my maintenance through conquest.

I was running 2 specialists per city during classical era.

As I made my way through the classical era I was waging war on Louis (I HATE that guy...). I ended up wiping him out by the time I was into the medieval era, keeping all of his cities, and I think he had around 8 cities, so this would've given me 12-15 cities or so. At this point I was still running 2 scientists per city and was considering moving to caste system to up it to 3 or 4. However, it was at this point that I ended the game to go to bed.

I clearly had the tech lead, it was around 800AD. I was 2nd in military power to Roosevelt who was beating up on Capac. I was planning on letting Roosevelt (who was Pleased with me) finish Capac while I launched a war on Kublai who was Friendly with me (sharpens knife). Kublai had the Hindu holy city and shrine and me, him, and Roosevelt were all Hindu (nice infusion of cash). Once I took out Kublai (easy as I had a huge tech lead on him), I would've had a much larger army than Roosevelt, who would've been the last civ on my large continent. Finish off Roosevelt, and cruise to whatever victory condition (minus cultural) that I wanted...

 

[DarkFyre99]

UncleJJ,

the main advantages of a specialist economy are:

1) Farms don't need to mature. Cottages take 10 turns to mature into hamlets, another 20 turns to mature into villages, and finally 40 turns to mature into towns. All this time, specialists are putting out an equivalent of 6 commerce per turn with Representation.

2) As a result, it's very easy to get a sizable tech lead, and keep it, with a smaller empire.

3) Farms aren't quite as vulnerable to pillaging. A pillaged town takes 70 turns to return to full production. A pillaged farm is at full production as soon as it's rebuilt. And it doesn't allow your enemy to live off your land.

4) Specialists produce Great People Points. The specialist strategy produces slightly more Great People than a dedicated Great Person farm, it produces them sooner, and it gives you better control/idea over what type of Great Person you'll be producing.

5) Specialists are more precise than cottages and the slider, allowing you to fine tune your economy, and aggressively specialize your city. Going negative? Swap a scientist for a merchant. You control which city this will be in, allowing you to take advantage of building wealth infrastructure in that city alone. Unless you stay at 100% science, in a cottage economy, all cities except your one wealth city aren't taking full advantage of the commerce they're producing, or you're wasting hammers on buildings that are underutilized or not fully utilized.

6) A specialist economy allows you to use the Culture Slider to produce happiness without it crippling your economy. Need five happy faces? It'll cost you, on average, 3 gold per city, rather than 50% of your economy.

7) Specialists = a large population = more votes in the UN. Great for an early diplomatic victory.

8) Last but not least, you can pillage your enemy's cottages to your hearts content, without worrying about crippling your economy. You're supporting specialists with farms, which don't take forever to mature.

There are a few disadvantages with the specialist economy, of course:

1) It's most powerful in the early game. By the late game, cottages come into their own, and a specialist economy can't keep up.

2) Representation is a must. Given that this strategy is most powerful in the early game, the Pyramids are also a must... and you can't count on conquoring it from a neighbor.

3) Caste System is extremely useful with this strategy... meaning no slavery and no whipping.

4) The circumstances where it can be used are more limited than a cottage economy.

In one of the games where I was running a specialist economy, if I remember correctly, in the 14th century, I had six cities with at least eight scientists, two cities with at least six merchants (and one was still running a few scientists), four production cities, and I overran Isabella with my riflemen, cavalry, and cannons, who were opposed by longbowmen and knights.

This was on Noble, though. I have yet to start in a good position for a specialist strategy on Prince. I play with random everything except the choice of continents.

 

[Eqqman]

With stone and industrious pyramids is no problem. With industrious and no stone, it's difficult. With stone and no industrious, a bit easier.

I'm curious as to why you assigned this ordering and not vice-versa. With Industrious you get half production from the get-go, with just Stone you have to spend 6 turns on the Quarry and possibly 2 or more turns on a Road, so you'd expect the Stone route to be slightly slower and therefore not easier than simply being Industrious.

 

[Eqqman]

At this point I was still running 2 scientists per city and was considering moving to caste system to up it to 3 or 4.

What do you think of an 'early' Caste System. Equally important to the unlimited specialists (you might think) is that fact that you can place the specialists without the required buildings. So you can start Merchants and Artists without having to stop to get Drama and Currency, and you don't have to delay your war production by making Libraries and Markets all over the place. Especially when you found a new city, you can just run an Artist for 3 turns to pop the border and then move on... no need to waste time on Obelisks or Stonehenge.

 

[futurehermit]

Stone gives 100% production bonus while industrious is only 50%. You can pre-build the road, so that leaves only 6-8 turns (depending on game speed) for the quarry. 2x the bonus means stone will be superior imo. It takes many turns for the pyramids to be built so 6-8 is negligible.

1) It's most powerful in the early game. By the late game, cottages come into their own, and a specialist economy can't keep up.

i disagree. not in the sense that cottages aren't better fully matured, but rather i would argue that if you run specialists the whole game AND consistently warmonger, your huge empire full of specialists won't require you to make the switch to cottages. you can support a larger empire than the cottage-builder throughout the game.

3) Caste System is extremely useful with this strategy... meaning no slavery and no whipping.

true caste system is useful. however, you can switch back and forth. slavery during war time and caste system during peace time. also, you can use markets and grocers and then later observatories (only 1 sci, but still) if you still want specialists and slavery.

i prefer caste system myself because i find that once you get over the hump where you have one of, if not the, largest empire and the tech lead that whipping is less of a priority.

you also have access to police state which means that whipping military isn't as big of a deal. and you can also use organized religion for buildings.

all of this civics stuff obviously indicates that spiritual is useful (ind-spir = gandhi)

What do you think of an 'early' Caste System.

i'm in favour of it. of course at the higher skill levels you won't get both pyramids and oracle-->CoL, but you should be targetting CoL anyways to support your growing empire...

slavery is great, no doubt, and i think it should be used in conjunction with caste system. use representation and caste system when you want to achieve a tech lead. then once you have it, switch over to slavery and police state (and vassalage and theocracy) and pump out your military. then back.

 

[Eqqman]

...the difference is so phenomenal that i can't understand why anyone in a single player game would opt for any other strategy than to beeline for pyramids for specialists with representation.

The main idea is probably that if you fail to get the Pyramids, you might be screwed, and to play it safe with this strategy you're locked into picking Industrial civs at higher levels to make sure you don't run into problems with a lack of Stone.

 

[Eqqman]

Stone gives 100% production bonus while industrious is only 50%.

Ahhh... this whole time I've been thinking that it's 50% off the cost of the Wonder, not the production rate.

 

[malekithe]

1) It's most powerful in the early game. By the late game, cottages come into their own, and a specialist economy can't keep up.

This is something that I'd like to see disproven (or at least given more context) at some point. I think it is mostly based on the assumption that, by running a specialist, you're giving up working two cottages (you've got to work an extra farm to feed the specialist). What most people seem to ignore, though, is that specialists are not limited by the tiles available to the city. Instead, after biology, they are limited by health (not happiness, as you're using the culture slider). So, in any situation where you could be working 8 grassland towns, you could just as easily be running 8 specialists.

Now, the 6 beakers from a scientist still doesn't quite compare to the 7-8 from a fully-developed town, but it's fairly close and is considerably more flexible (you can run merchants or even priests in a pinch). You have a little more flexibilty in some of your civics, as well. You can realistically pick anything out of the legal, labor, economy, or religion columns. The only thing that is required is representation. Nationhood is especially useful and serves as a nice substitute for production if you choose Caste System over Slavery. I'd even go so far as to argue that in a fully-developed specialist economy, you'd never want to run slavery for two reasons. a) Your population is so high that it becomes inefficient and b) you want to be able to run as many specialists as possible (opting for Caste System).

So, completely ignoring Great People, a late-game specialist economy is not altogether much worse off than a cottaged one. Especially when you consider the ability to withstand a pillaging campaign. I'm unsure what kind of effect a fully-developed specialist economy has on late-game great person generation. To be honest, I've never played it out that long and most of this post was just theory.

 

[DarkFyre99]

i disagree. not in the sense that cottages aren't better fully matured, but rather i would argue that if you run specialists the whole game AND consistently warmonger, your huge empire full of specialists won't require you to make the switch to cottages. you can support a larger empire than the cottage-builder throughout the game.

The key words are "Huge Empire." Any huge empire will have a better economy than a small one. However, keeping that early tech lead requires a huge empire once cottages come into their own.

The other key words are "consistently warmonger." I had a game where I was running a specialist economy on a continent, all by my lonesome, so I had to build all my own cities. By the time I could successively invade a neighboring continent, they had begun to catch up on my tech lead. If I had been smart, I would've tried for a cultural win...

This is something that I'd like to see disproven (or at least given more context) at some point. I think it is mostly based on the assumption that, by running a specialist, you're giving up working two cottages (you've got to work an extra farm to feed the specialist). What most people seem to ignore, though, is that specialists are not limited by the tiles available to the city. Instead, after biology, they are limited by health (not happiness, as you're using the culture slider). So, in any situation where you could be working 8 grassland towns, you could just as easily be running 8 specialists.

It's fairly common to compare the two economies tile for tile. But in reality, you have to examine the economies population for population.

Before the late game, thanks to being able to use your culture slider to produce happiness, you can "cheat" and compare them tile for tile. You can buy the happiness to maintain those specialists because it only costs you a couple of commerce, rather than a huge chunk of your economy.

After the late game, population (not happiness or health) becomes the limiting factor. In my (admittedly limited) experience, most cities have a population cap of around 32. After that, between the food penalty due to unhealthiness, and the ever increasing cost of a new citizen, your city's growth crawls to a stop. Cities with a lot of floodplains or food resources can support a larger population, of course, but not that much larger.

By the late game, if you're supporting eight specialists, you could be working sixteen cottages... at a minimum of seven commerce a each.

 

[malekithe]

By the late game, if you're supporting eight specialists, you could be working sixteen cottages... at a minimum of seven commerce a each.

While I'll grant that growing to the massive populations required for a specialist economy to compare to cottages is a non-trivial task, I feel that the statement above is mostly untrue. For any number of reason, for instance, you may not be able to work 16 cottages. You might not have enough grassland available, you might not be able to attain that level of happiness, etc. Additionally, those last 6-8 cottages will most likely be completely immature for a long period of time. That makes up for the extra population growth required to run the specialists. I'll not argue that the two are equal, just that cottages are not the clear winner in all scenarios.

 

[DarkFyre99]

While I'll grant that growing to the massive populations required for a specialist economy to compare to cottages is a non-trivial task, I feel that the statement above is mostly untrue. For any number of reason, for instance, you may not be able to work 16 cottages. You might not have enough grassland available, you might not be able to attain that level of happiness, etc.

True. However, if a city can't support 16 cottages normally, it probably can't support a large population in the first place. That city is probably better as a production city than a science or wealth city.

Additionally, those last 6-8 cottages will most likely be completely immature for a long period of time.

That doesn't seem right. When I'm using cottages for commerce, it's not unusual to have size 15-16 cities by the time democracy rolls around. Since the remaining five or six tiles are usually sub-optimal for cottages, (or farms, for that matter) maturing those cottages aren't much of an issue.

That makes up for the extra population growth required to run the specialists. I'll not argue that the two are equal, just that cottages are not the clear winner in all scenarios.

I agree (obviously). I'm just saying that if you haven't won the game using specialists by the time of the late game, you're going to have a harder time at that stage of the game than someone who built cottages right from the start. On the other hand, you're much more likely to win the game early using the specialist strategy, if you're able to use it in the first place.

 

[NaZdReG]

I think the analysis is flawed because you do not factor in the % magnification of your commerce.

current game I'm playing pangea, prince, as qin shi. currently its 1650 in game
this is at 70% science +25 gpt
my capitol has library, univ, academy, observatory, oxeford
***producing 211 beakers per turn***
I do run 2 scientists there, but only to deal with the food overflow
2 other cities have the same setup -oxford and are producing
136 and 141 beakers respectively

my civilization's total output currently is 879 beakers per turn from 15 cities

15 cities of specialists... even with representation you would need 10 scientists PER CITY assuming you were running caste system instead
with libraries and universities adding in there it would be closer to 6 scientists per city. but still can you find 6 4+ food tiles around every single city??
***not to mention the happiness and health diminishing returns with high populations***
you are also ignoring governmental bonuses to fully upgraded towns of
+1 production and +2 commerce and another +2 from techs..
if I had to get rid of the towns I would be down atleast 5+ hammers per city.

I definately think the specialist econ has its places.. especially with a philo leader and pacifism.. but I am more for running a SINGLE specialist city.. if you want commerce go that way and build market grocer bank and wall street.

if you want priests either mass temples or ankor wat.. which is a whole different discussion because 6 priests can crank out 12 hammers per turn with that wonder.

or even artists in a city with the sistine chapel. 6 artists would be producing a base of
36 culture per turn with lots of ways to add %%% to that to magnify it greatly.

etc.

NaZ

 

[malekithe]

I think the analysis is flawed because you do not factor in the % magnification of your commerce.

Did you even bother to read the whole thread? This criticism has already been voiced and addressed.

EDIT: You're playing a financial civ in your game. That alone throws any comparison right out. No ones going to endorse running loads of scientists when you're financial.

 

[NaZdReG]

yes I did read the thread I was contributing on that side of the arguement. the specialist econ is of cource better for a non financial civ to work with. any financial civ though is MUCH better off running towns rather than specialists.

I still stand by the point of having a single specialist city.. as the other civics can get you further than caste system.

NaZ

 

[UncleJJ]

Did you even bother to read the whole thread? This criticism has already been voiced and addressed.

I don't think the problem has been addressed at all. Simply saying the science slider is set to 0% means that there is an assumption that whatever commerce is floating around (since no particular attention has been paid to gaining any commerce) meets the running costs of the newly expanded empire. That might be true for a time but you can't keep going on that way for long. Eventually the costs will exceed the commerce and if you run the caste system you can then switch to running some merchants to get more gold... but that reduces the amount of beakers. I am interested in when this happens and what can then be done to extend the run of conquests.

I would like to see anyone advocating the conquest and specialist economy to provide some actual numbers rather than giving us some euphoric statements and hand waving the economic problems away. This thread is called a "detailed analysis" afterall . The only analsys of a real game I've seen so far has been my own... so back up your claims with numbers please.

 

[Eqqman]

And if you choose to run caste system when the captured city sizes start hitting size 6...

I forgot to ask, why is size 6 a magic number in relation to Caste System?