Detailed analysis: # of specialist cities needed per era (Part II!)

Ok, so many of you may remember my original thread for this work. If not, here it is:

http://forums.civfanatics.com/showthread.php?t=177506

Part II is dedicated to analyzing the final 3 eras, as the first thread only looked at classical and medieval.

Now, the reason I stopped at the first 3 eras was that I assumed specialists would start to fizzle out late game and that instead one should transition to a cottage economy post-democracy.

If you share this assumption, these numbers may in fact surprise you.

I will start by rehashing the first three eras. I've worked out a better strategy for accounting for the great library/super science city, so the numbers may be a bit different.

I will then present my findings for the final three eras.

And then I will offer my conclusions and recommendations.

All of my numbers in this thread assumes running representation early via the pyramids. I won't bother comparing to non-representation as the numbers are vastly inferior. I may come back and do a part three comparing to a cottage econ late game if I can figure it out.

Without further ado...(keeping in mind is that the goal is to sustain a tech rate of 1 tech/5 turns throughout the game, starting in the classical era)

Classical era

Average tech cost: 315.4

Great library/super science city beakers produced in 5 turns: 210

-The math for this assumes 4 scientists (2 from library, 2 from GL) and a rate of 10.5 beakers/turn/scientist (6 base, 3 from academy, 1.5 from library). Thus 4*10.5=42*5 turns=210

Average tech cost minus beakers produced by super science city: 105.4

Rate for scientists not in super science city: 7.5 beakers/turn (6 base + 1.5 from library)

So, the number of scientists needed to get 105.4 beakers in 5 turns is: ((105.4/7.5)/5)=~3

Therefore, the number of science cities--in addition to the super science city--required is 1.5 or 2 if you want to round up (since each science city can only run 2 scientists via a library). Running 2 science cities + the super science city w/great library and an academy will give you in excess of a rate of 1 classical tech/5 turns.

Medieval

Note: in the medieval era I'm including one monastery per city in my analysis. obviously, if you have say 3, the results are going to be better. but for simplicity, i am only doing 1. it's not always easy to spread 3 religions to all of your cities when you're focusing on other things.

Average medieval tech cost: 733.33

Great library/super science city beakers produced in 5 turns: 305.25

-The math for this assumes 4 scientists (2 from library, 2 from GL), 1 settled great scientist (GS) [with a rate of 9 base + 4.5 academy + 2.25 library + .9 monastery = 16.65], and a rate of 11.1 beakers/turn/scientist (6 base, 3 from academy, 1.5 from library, .6 from one monastery). Thus (4*11.1=44.4)+(1*16.65=16.65)=61.05*5 turns=305.25

Average tech cost minus beakers produced by super science city: 733.33-305.25=428.08

Rate for scientists not in super science city: 8.1 beakers/turn (6 base + 1.5 library + .6 monastery)

So, the number of scientists needed to get 428.08 beakers in 5 turns is: ((428.08/8.1)/5)=~11

Therefore, the number of science cities--in addition to the super science city--required is 5.5 or 6 if you want to round up (since each science city can only run 2 scientists via a library). Running 6 science cities + the super science city w/great library and an academy will give you in excess of a rate of 1 medieval tech/5 turns. You will want to note that you need 6 universities for Oxford university, which comes along in the next age and is a huge factor. I'm saying here you need 5-6 cities + your super science city to get the medieval tech rate. This puts you in the ballpark for preparing for Oxford.

Renaissance

Ok, so here we are at the new material. I stopped here before because I assumed we'd be transitioning to a cottage econ. All I can say, as you'll see below, is that I would seriously recommend keeping your specialist econ because the renaissance age is where it shows its true power!!!

Average renaissance tech cost: 1528.57

Great library/super science city beakers produced in 5 turns: 1206

-You'll notice this is a HUGE increase over the medieval era, and here's the math...The math for this assumes 9 (!!!) scientists (2 from library, 2 from GL, 1 from observatory, 1 from running mercantilism, and 3 from Oxford), 2 settled great scientists [with a rate of 9 base + 4.5 academy + 2.25 library + .9 monastery +2.25 observatory +2.25 university +9 oxford= 30.15], and a rate of beakers/turn/scientist (6 base, 3 from academy, 1.5 from library, .6 from one monastery, 1.5 from university, 1.5 from observatory +6 Oxford =20.1). Thus (9*20.1=180.9)+(2*30.15=60.3)=241.2*5 turns=1206.

Average tech cost minus beakers produced by super science city: 1528.57-1206=322.57

Rate for scientists not in super science city: 11.1 beakers/turn (6 base + 1.5 library + .6 monastery + 1.5 university + 1.5 observatory)

So, the number of scientists needed to get 322.57 beakers in 5 turns is: ((322.57/11.1)/5)=~6

Therefore, the number of science cities--in addition to the super science city--required is only 1.5 or 2 if you round up (!!!) (since each science city runs 2 scientists via a library, 1 via mercantilism, and 1 via an observatory). Now, based on the development done in medieval, you'll have 6 science cities at this point in addition to the super science city. YOU ONLY NEED TWO TO GET A TECH RATE OF 1/5 TURNS!!! You're going to be flying through the renaissance faster than you can blink!!!

The key here is your super science city. Get Oxford up asap--i.e., get universities up asap--in your great library city with a couple (or more!!!) settled GS. Don't forget observatories--my recommendation is getting astronomy with your free tech from liberalism. Your super science city becomes a juggernaut and away you go!

Industrial

Well, after cruising through the renaissance era, we hit a bit of a snag: scientific method! This costs us the monasteries AND the great library However, the nice thing is we should have ensured we got Statue of Liberty toward the mid-to-late Renaissance, so that softens the blow a little bit of losing the great library...

Average industrial tech cost: 2366.66

Super science city beakers produced in 5 turns: 1365

-You'll notice this is NOT a huge increase over the renaissance era. As I noted, the loss of the monastery and the great library hurts. But, here's the math...The math for this assumes 8 scientists (2 from library, 1 from observatory, 1 from running mercantilism, 3 from Oxford, and 1 from statue of liberty), 4 settled great scientists--YOU GET ONE FOR FREE FROM PHYSICS, MAKE SURE YOU GET IT FIRST -- [with a rate of 9 base + 4.5 academy + 2.25 library + 2.25 observatory +2.25 university +9 oxford= 29.25], and a rate of beakers/turn/scientist (6 base, 3 from academy, 1.5 from library, 1.5 from university, 1.5 from observatory +6 Oxford =19.5). Thus (8*19.5=156)+(4*29.25=117)=273*5 turns=1365.

Average tech cost minus beakers produced by super science city: 2366.66-1365=1001.66

Rate for scientists not in super science city: 10.5 beakers/turn (6 base + 1.5 library + 1.5 university + 1.5 observatory)

So, the number of scientists needed to get 1001.66 beakers in 5 turns is: ((1001.66/10.5)/5)=19

Therefore, the number of science cities--in addition to the super science city--required is 4 (since each science city runs 2 scientists via a library, 1 via mercantilism, 1 via an observatory, and 1 via the statue of liberty). Not as good as renaissance, but still better than medieval! You still only need to run 4 science cities in addition to the super science city to get a tech rate of 1 tech per 5 turns. [and, you'll notice this doesn't even include the potential of caste system, which i'll talk about in my conclusion]

Modern

Well, here we are in the modern era, and I certainly feel like you should be here before your AI competitors! Here's the breakdown...

Average modern tech cost (not including future tech): 4185.71

Super science city beakers produced in 5 turns: 1732.5

-Well, this is a bit better than industrial era, thanks to the extra scientist made available from a laboratory. Here's the math...The math for this assumes 9 scientists (2 from library, 1 from observatory, 1 from running mercantilism, 3 from Oxford, 1 from statue of liberty, and 1 from laboratory), 5 settled great scientists [with a rate of 9 base + 4.5 academy + 2.25 library + 2.25 observatory +2.25 university +2.25 laboratory +9 oxford= 31.5], and a rate of beakers/turn/scientist (6 base, 3 from academy, 1.5 from library, 1.5 from university, 1.5 from observatory +1.5 laboratory +6 Oxford =21). Thus (9*21=189)+(5*31.5=157.5)=346.5*5 turns=1732.5.

Average tech cost minus beakers produced by super science city: 4185.71-1732.5=2453.21

Rate for scientists not in super science city: 12 beakers/turn (6 base + 1.5 library + 1.5 university + 1.5 observatory + 1.5 laboratory)

So, the number of scientists needed to get 2453.21 beakers in 5 turns is: ((2453.21/12)/5)=41

Therefore, the number of science cities--in addition to the super science city--required is 7 (since each science city runs 2 scientists via a library, 1 via mercantilism, 1 via an observatory, 1 via the statue of liberty, and 1 via a laboratory). So, finally, we have to consider adding 1 additional science city. However, I would recommend easily combatting this addition by finally adopting caste system--if you didn't do it earlier, see below--to add the extra specialists required to reach the tech rate.

Conclusions

Ok, first of all, you'll notice I left a few things out. You could possibly get an additional couple monasteries to help you up until the industrial era.

Secondly, caste system. Obviously, if you adopt caste system, you can take your number of scientists beyond the limit afforded by libraries. This means that your tech rate will be EVEN FASTER than 1 tech every 5 turns, which is already an incredible pace!!! I would recommend adopting caste system once you hit civil service, that way you can combine irrigation with the additional scientists.

Now, I think this analysis makes a HUGE case for philosophical leaders. First off, they will have additional settled great scientists, which are HUGE beakers, especially later in the game! They also get cheaper universities, which means FASTER OXFORD, which as you've seen is HUGE in terms of its impact on your tech pace, especially in the renaissance era.

I think it also indicates that you can get away with 6 science cities, including the super science city--especially if you do adopt caste system. This is the exact number you need for oxford, hint hint. On top of these 6 science cities, you'll need a few production cities to pump units and probably the statue of liberty. Beyond that, you don't need any more cities!

Clearly, this strat lends itself to a small-empire ultra-fast science victory. However, I think the tech lead can also be put toward a domination victory as well. Just watch your maintenance costs. A good candidate here is Mao since he's organized and so you can pop down cheap courthouses to manage maintenance and also get forbidden palace sooner.

Finally, I feel confident that the switch to cottages is not required at all, meaning you can use biology to boom your food econ late game, which can be very helpful for domination, obviously, but also diplomatic if you want another option

As always, if I made any errors in my analysis, I would appreciate if people could point them out. And I look forward to any discussion


EDIT1: Ok, my first error was pointed out, thanks to Mutineer. You can't assign a scientist w/mercantilism in addition to the two afforded you by the library because you don't have any slots available. So, definitely, I would insist on making the switch to caste system once you hit civil service for irrigation, which will come prior to banking and mercantilism.

The math for this assumes 9 (!!!) scientists (2 from library, 2 from GL, 1 from observatory, 1 from running mercantilism, and 3 from Oxford), 2 settled great scientists [with a rate of 9 base + 4.5 academy + 2.25 library + .9 monastery +2.25 observatory +2.25 university +9 oxford= 30.15], and a rate of beakers/turn/scientist (6 base, 3 from academy, 1.5 from library, .6 from one monastery, 1.5 from university, 1.5 from observatory +6 Oxford =20.1). Thus (9*20.1=180.9)+(2*30.15=60.3)=241.2*5 turns=1206.

Click to expand...

That is incorrect, as Mercantenilism does not give you scientist, it just give a free specialists but assigning them follow useal rules.

So, maximum scientist you can have in one city in this age with out castle system will be:

2 library, 2 GL, 1 Observatory, 3 oxford = 8

Gah, that sucks!!

Ok, well caste system it is then

The switch to cottages isn't required. However I don't think that you're going to do better with scientists. It all depends on how much margin for error you have at your given difficultly level. Because cottages will still do more. And this strategy is very wonder reliant.

Also your analysis is flawed in that it doesn't examine how these numbers compare to the alternative. Mainly a typical cottage fueled commerce setup with the appropriate civics. You put together a lot of numbers, use bold, and it all seems impressive. But without a base of comparison they don't tell us much other than its feasible to tech using primarily scientist specialists.

How deep are labs in the modern tech tree? I don't recall them being around for me before its pretty much pointless (unless I'm building spaceship parts).

The switch to cottages isn't required. However I don't think that you're going to do better with scientists. It all depends on how much margin for error you have at your given difficultly level. Because cottages will still do more. And this strategy is very wonder reliant.

Click to expand...

It only requires 3 wonders: pyramids, great library, and statue of liberty. And I can get these pretty much every game. Great library is easy because the AI doesn't go for alphabet early. Pyramids you target. And statue of liberty you get because of your tech lead.

Also your analysis is flawed in that it doesn't examine how these numbers compare to the alternative. Mainly a typical cottage fueled commerce setup with the appropriate civics. You put together a lot of numbers, use bold, and it all seems impressive. But without a base of comparison they don't tell us much other than its feasible to tech using primarily scientist specialists.

Click to expand...

As for a comparison, I'll try and do one, but really 1 tech/5 turns--probably even better with caste system--is a great rate. Will cottages be vastly superior to this? I doubt it.

How deep are labs in the modern tech tree? I don't recall them being around for me before its pretty much pointless (unless I'm building spaceship parts).

Click to expand...

All you need is radio and computers, which you can target as your first techs in the modern era.

futurehermit said:

As for a comparison, I'll try and do one, but really 1 tech/5 turns--probably even better with caste system--is a great rate. Will cottages be vastly superior to this? I doubt it.

Click to expand...

Honestly my instinct after many games is that cottages are markedly (if not "vastly" superior.) The beaker numbers for your cities seem low for the eras compared to my experience and I usually do better than one tech per five turns.

I'm really curious to see your analysis, though.

Being used to Marathon speed*, 5 turns per tech looks really weird to me, I must tell you. During one long expanding campaign (read: "wars") I was at 0% science rate in order to keep up with my increasing maintenance costs and had around 190 turns to research Astronomy. Now THAT's more likely. I eventually got courthouses in my new cities, built the Forbidden Palace and Versailles at opposite ends of my empire and it all came back to more normal numbers. (phew)

---

*combined with Huge maps really gives me the "real world" feeling, though I'm starting to get tired in the late stages...

I've been finding it ironic that it's been easy for me to get the Pyramids/GL but not easy at all to cash in on them. Getting the extra science cities going is a big deal. Assuming no food bonuses are available then you're looking at a stagnated size 4 city that can perform no useful tasks besides making the science. Supporting these scientists while trying to do something else is a huge drain on the city's economy. And taking the time to grow to have non-science citizens available, while short (6-8 turns with good surplus per growth point), still means that I'm not contributing enough science for a couple of techs. At what points to you consider your economy as needing to reach each of these levels? The breakdown you give seems directed more towards if you're doing horizontal research up the tree instead of vertical. If I'm trying to work towards an advantage (by rushing to Liberalism, say) then I'm often going to be researching the most expensive tech available at any moment. So it's unclear to me how I would tell what city setup I'm expected to have at any point. It looks like you want to find a small area of your empire to spam out 6 junk cities that only need to run the science, and they produce everything with slavery. I'm curious how you deal with these issues. I'd extend your guide to include a vague timeline of what you're doing at each of the major milestones.

One note, if you use Caste Sytstem, the number of Required 'science Cities' Drops (although you still need 6 cities to get universities)

What it sounds like the best thing to do is once you get Caste, start reducing your number of Science Cities (since each can hold more)... and once you get Education, build the needed universities for Oxford in your High Production cities

(the university in the Science Cities will be slower to build)

... but after you get Oxford built in the Super Science city... then you can convert it to pure food+Science... the Observatory will be nice, but it won't make that big an impact if you get it slow.

Actually once you have Mercantilism and Caste, then you might not need any large Science Cities besides the Super one... Each city you have can contribute 6 raw flasks to science

Krikkitone said:

Each city you have can contribute 6 raw flasks to science

Click to expand...

True, but aren't the numbers based on having every possible beaker enchancing building in the city? If you offload scientists from certain cities to spread out the number of specialists, you need extra to make up the difference from not having these buildings in every city.

I'd like most of all to hear game dates, which are roughly similar across game speed, and should present a good idea of where you are. What date do you get Great Library? What date do you get Liberalism? How many beakers do you have at 0 AD? 500 AD? 1000 AD?

I have great library around 400AD (give or take). Liberalism around 1000AD (give or take). 0 AD I'm in classical era, 500 AD I'm in medieval era, 1000 AD I'm in renaissance era. You can look at my post to get a sense of the number of beakers being produced.

Edit: Note that these numbers are approximations. The actual dates vary from game to game. I usually don't dwell in classical. I just get alpha-lit, then trade for iron and math, then get construction, then i head to medieval, trading for the rest of the classical techs as needed. I try and acquire the required # of cities for medieval research via war in classical.

I've been finding it ironic that it's been easy for me to get the Pyramids/GL but not easy at all to cash in on them. Getting the extra science cities going is a big deal. Assuming no food bonuses are available then you're looking at a stagnated size 4 city that can perform no useful tasks besides making the science. Supporting these scientists while trying to do something else is a huge drain on the city's economy. And taking the time to grow to have non-science citizens available, while short (6-8 turns with good surplus per growth point), still means that I'm not contributing enough science for a couple of techs. At what points to you consider your economy as needing to reach each of these levels? The breakdown you give seems directed more towards if you're doing horizontal research up the tree instead of vertical. If I'm trying to work towards an advantage (by rushing to Liberalism, say) then I'm often going to be researching the most expensive tech available at any moment. So it's unclear to me how I would tell what city setup I'm expected to have at any point. It looks like you want to find a small area of your empire to spam out 6 junk cities that only need to run the science, and they produce everything with slavery. I'm curious how you deal with these issues. I'd extend your guide to include a vague timeline of what you're doing at each of the major milestones.

Click to expand...

Cashing in on the pyramids: In classical you want 3 science cities, including great library. For me, that usually means I want 5 cities: 3 science, including my capital, and 2 production for military. It doesn't matter if the science cities aren't doing anything other than producing science. That is their function. It's not a huge drain on your economy, because you're not trying to run 100% science, you're running 0% science. In terms of horizontal vs. vertical, here's the thing: Focus on horizontal until you get the required number of cities/buildings in place to go vertical. I can assure you that you will have the advantage anyways and therefore you don't need to stretch to get it.

Besides that, I'm not sure what I can tell you. I see that everyone is very critical.

However, in my games I have the outright tech lead by 1000AD, as well as the largest empire. Because I run 0% science, I can put my commerce towards maintenance without hampering my economy. Cottage econ has to battle the tug-of-war between maintenance and science slider.

If this isn't enough to whet your whistle then I don't know what is

Honestly my instinct after many games is that cottages are markedly (if not "vastly" superior.) The beaker numbers for your cities seem low for the eras compared to my experience and I usually do better than one tech per five turns.

Click to expand...

Well, as you can see from my post, it's certainly possible, especially in renaissance era, to go waaay faster than 1 tech/5 turns, as long as you run caste system.

Keep in mind I'm only giving you the MINIMUM # of specialists required to reach this rate. In my games, I have many more cities than the minimum, running as many specialists as I can!! If you want to play peaceful, then the minimum is certainly fine. But I don't play peaceful...

I can tell that the only way I'm going to be able to perhaps convince people (if they're open to it), is to do a comparative analysis between specialists and cottages.

You'll have to give me some time though. It took a very long time to put this together and I anticipate it will take me longer to put the comparative analysis together.

futurehermit said:

It's not a huge drain on your economy, because you're not trying to run 100% science, you're running 0% science.

Click to expand...

When I used the word 'economy' I didn't mean the ability to produce gold. I was thinking of everything a city can produce all together. I have a hard time imagining that only two cities working towards production is enough to eliminate an opponent in a timely manner.

It's only two cities early on. Obviously, you're going to be adding more production cities as you conquest. How many axes do you need to take an AI city early on? 4-6 I'd say, depending on whether or not it's on a hill and whether or not it has a 20% cultural bonus yet. How many cities do you need pumping units to get 4-6 units? i'd say 2 is sufficient. If you want to streamline it with 2 production cities, just build 6 units, declare war, and keep chop/pop rushing units out of your 2 production cities. while your first group is healing up from taking the 1st city, the reinforcements arrive, and then you can continue on to the next city, rinse-repeat.

furthermore, i'll usually stagnate my science cities at size 5: 2 food tiles, 2 specialists, and a mine for some hammers.

Liberalism in 1000 AD is OK if it's an aggressive land game; Great Library in 400 AD seems late, but it doesn't seem that, that crucial anyway. It sounds like your strategy is probably behind a cottage economy (at least my cottage economies) but not far enough behind to be crippling.

However, what I like most about your strategy is the fact that you can turn the luxury slider up up up without repercussions to your tech rate. I suspect that repeated and obsessive slaving in your lux-flooded cities will do a fair job of making up for lost production, and that might be a useful tactic to carry away from this whole thing, if nothing else.

futurehermit said:

Cashing in on the pyramids: In classical you want 3 science cities, including great library. For me, that usually means I want 5 cities: 3 science, including my capital, and 2 production for military. It doesn't matter if the science cities aren't doing anything other than producing science. That is their function. It's not a huge drain on your economy, because you're not trying to run 100% science, you're running 0% science. In terms of horizontal vs. vertical, here's the thing: Focus on horizontal until you get the required number of cities/buildings in place to go vertical. I can assure you that you will have the advantage anyways and therefore you don't need to stretch to get it.

Besides that, I'm not sure what I can tell you. I see that everyone is very critical.

However, in my games I have the outright tech lead by 1000AD, as well as the largest empire. Because I run 0% science, I can put my commerce towards maintenance without hampering my economy. Cottage econ has to battle the tug-of-war between maintenance and science slider.

If this isn't enough to whet your whistle then I don't know what is

Click to expand...

I like ur strat, its quite aggro and seems to solve a lot of the buildup problems. I only have two issues, one: requiring the pyramids. makes me sketchy, so now I need to find copper and stone? or beat an emperor AI to the pyramids the old fashioned way? I try to play every game like it really matters, and not spam games to find the right start situation. so I'm worried that this will simply fail a lot of the time, for a lot of civs.

all this talk about the slider, I fail to see how it matters. specialists add to ur base, worked tiles add to ur base and then ur base is modified by ur multipliers. are u implying its too much work to move ur science slider around or something? seems to me its really about what gives u the most overall commerce at which points in the game. which I think ur right in that ur specialists can give u more earlier, if u of course can get all this setup.

the whole point is to run 0% science so you can use the culture slider and use all of your commerce towards maintanance.. resulting in a larger empire far earlier than if you were running a cottage based econ.

NaZ

NaZdReG said:

the whole point is to run 0% science so you can use the culture slider and use all of your commerce towards maintanance.. resulting in a larger empire far earlier than if you were running a cottage based econ.

NaZ

Click to expand...

the culture slider isnt disabled in a cottage economy. like I said previously it seems to me the test is how much total commerce ur putting out. since it can mostly be allocated how u want it(especially in a cottage econ). and not some mystical "my slider is at 0" thing.