Specialist Economy - are you guys really buying this?

[UncleJJ]

It's a convincing argument if all you're worried about is how many total beakers does one city produce over N turns, but in a real game, that's not the only concern. Your approach might produce more beakers overall, but Wodan's will produce them sooner. That acceleration of your research might translate into all kinds of indirect advantages.

Maybe you found a religion that you would otherwise be beaten to. Maybe you complete a wonder that you would otherwise be beaten to. Maybe you build your advanced military units sooner and win a war more decisively. Maybe you win a race to a bonus technology like Liberalism or Economics. And so on ...

I accept what you say as minor qualification to my main point. Indeed I am well aware of the race situation you bring up and enjoy winning those . It is very unlikely that I would be whipping away substantial numbers of specialists in the middle of a race like that, rather I'd be at 100% research (and losing money), and many specialists would be working in starvation mode

It is not possible for Wodan's method to produce either more beakers or more GPPs in the middle or long term so therefore by definition he will be behind in research. Soon I'll be so far ahead any situation that he considers a race I will already have won Seriously, all you are suggesting is that sometimes short term considerations will overide the longer term efficiencies of my approach. I know that and acknowledged there could be exceptions with ...

It is generally superior in a SE to work food tiles before running specialists while regrowing. There are a few exceptions when I would still be working specialists and have farms unworked, while re-growing from a whip, but that would be to do with manipulating the GPP pool rather than simply producing beakers or whatever.

I used the word "generally" and alluded to manipulations of the GPP pool (and type of GP) and perhaps I could add the research race condition you mention ... but as I said above why would I be whipping away research assets if they were more important than the thing I am whipping them for? I don't do that and I assumed others would think like me...

Your approach probably completes buildings earlier (because your growth will come earlier so you can therefore whip earlier), but Wodan's should produce great people earlier, though it might produce fewer great people in total. Which is better? I don't think we can know.

In no way will Wodan's approach produce GP faster than mine. He might produce the first one at the same time but after that he will consistently fall behind. And the more often he uses his approach the quicker he will drop behind in production of beakers and GPPs since it is an inherently less efficient use of resources.

Let's just quantify the long term losses he might expect. Over the 100 turns we looked at let's say Wodan works a single specialist for just 30 of those turns when he could have worked a grassland farm instead. He will have lost 30 food and that is simply 15 specialist turns which is equivalent to 45 base beakers and 45 GPPs that his city lost. Not much you might say but is a simple and avoidable loss in almost all circumstances. A small consideration but it does make a difference.

Civ is a game of tradeoffs and in a real game you need to analize the complete situation and decide what approach is best. I would bet that most of the time, some combination / hybrid between your two methods is the true "best" strategy.

Civ is indeed a game of tradeoffs and the complete situation does need to be considered. But a good player needs to be aware of fundemental mechanisms that drive the game and "Food first, specialist later" is a good maxim for the middle and long term and could only hurt in a very few short term cases that you have made much more of than they deserve IMHO. There is no need to hybridise the two methods since I already acknowledge there might occasionally be a situation where keeping a specialist in the short term can be beneficial... but those situations are rare and in no way are the two approaches equal in general efficiency as you seem to be saying.

EDIT: On reflection I might have initially underestimated the losses Wodan's method will suffer. If I work a grassland farm for a turn and he works a specialist the farm will obviously give 3 food more per turn while he gets the beakers and GPPs from his specialist early. But the beakers and GPPs he gets early must be recovered later due to the faster growth making me more specialists much faster. That surprises me. The inefficency is 3 times worse than I first thought and I have to edit my numerical claims. Have I got that part of the argument right... it seems too strong

It now seems that each turn Wodan works a specialist when he could have worked a grassland farm will cost him 3 food in the long term. And 3 food is worth 1.5 specialist turns (not 1 food and 1/2 specialist turn as I initially thought). That makes the long term inefficiency 3 times what I stated at first. So taking 3 beakers and 3 GPPs in the short term seems to lose 4.5 beakers and 4.5 GPPs long term. That is a very significant loss.

n.b. I have edited this and my previous post to account for my revised thinking.

EDIT 2 : DRAT, I was right in the first place. Here is the argument: Wodan's specialist gives him 3 beakers and 3 GPPs now, while my farm gives me 3 food. As my city regrows faster the 3 food I gained during that turn will support 1.5 specialists and so I get 4.5 beakers and 4.5 GPPs later. The net difference in the long term is thus 1.5 beakers and 1.5 GPPs which is equivalent to 1 food.

n.b.2 I have edited this and my previous post back to the original thinking

 

[Wodan]

While I agree with much of what you're saying, you're greatly exaggerating the advantage of Slavery to an SE. As you'll see when you look through my game logs (I've posted them now, I'll see about finding somewhere to upload the saves later), there is virtually no difference in the effectiveness of slavery under SE and CE. Since slavery (especially if we're talking about whipping things like libraries) works best for small city size, the vast majority of the food surplus is from a couple of tiles which will be present in both economies.

For cities with 2-3 food resource, I agree. Also, if the CE builds some farms in all cities, then yes I agree for those too; if the CE does not, then I don't agree, because there's a huge difference when the city has +1 food per tile worked.

That's for the small city size situation. I think you're also forgetting the health/happy limit situation. Generally both CE and SE will start whipping again when unhappy citizens appear. The SE will regrow those citizens a lot faster than the CE will.

Now, what you did in your games might not necessarily correspond with what's possible or advantageous... hard to say without seeing your logs.

There's also the point an SE can run all its specialists, or have a greater food surplus than the CE, but not really both at the same time.

How do you come to that conclusion?

+1 food per tile (before Biology) means 1 specialist per 2 working citizens. With no Caste System, that's only size 4 to run 2 scientists. So, any city of size 5 or larger will have a bigger food surplus than the CE city working all cottages, even with the two specialists.

Wodan

 

[Wodan]

I hope that is a convincing argument dealing with a very complicated situation. Total food production of a city over a period of time can be analysed in those overall terms. A properly run SE should attempt to maximise its food production at all times.

Dr. Elmer (thanks, by the way, I had a busy day) answered this. I would have replied with the same argument... speed and when is of great importance in this game.

In the end, however, there's probably no one answer to this question. Sometimes, it's better earlier. Sometimes, it's better to have it later in somewhat greater amount.

In addition, I might point out one more thing...

Take two identical cities in a SE and work one of them as Wodan is suggesting and work the other I as advocate. Let's follow these two cities over say 100 turns of a game. They start with the same size and end with same size as each other and there is the same amount of food in the two granaries before and after the experiment. We will the two cities the same way to produce the same buildings and units so the same amount of food is converted into hammers in both cases.

The latter assumption is where we differ.

If you noticed, I advocate two things:
1) starting a science specialist early (before max growth, when you apply 2), yes, this slows down the application of the second scientist
2) assuming it is appropriate to whip after max growth is achieved, I advocate leaving 1 or 2 scientists even after the whipping is done; yes, this means that max growth will be re-achieved later than if all scientists were removed at that point in order to max growth at earliest opportunity.

What does this mean? Well, you get science (and GPP) earlier. We all agree on that. It also might mean you get less in the long run. That's debatable (the earlier argument does not apply because of the invalid assumption (see below please).

It also definitely means you get less whipped hammers. The hypothetical situation posed by Uncle JJ has as assumption that equal whipping is done. This patently is not the case. Since max city size is not re-achieved as quickly, then it is not feasible to whip as frequently.

Is that bad? Maybe. Whipping at max city potential is not ideal (as we all have agreed). If anything, it's making the best out of extra food. It's basically using the "unhappy" citizen to kick start the whip and get one free guy out of it. Most buildings etc by that point will cost more than 1 citizen (presumably the cheap stuff such as Granary and Library were whipped when the city was small) so it's not a great situation no matter how we look at it.

In any event, the point is, at that point in the game it's of supreme importance to keep both scientists working (and not whipping them away). That's why I advocate whipping the citizens working tiles but keeping the scientists employed. Yes, it'll be a while before the city regrows. To me, that's of secondary concern. Most important is to keep the research coming in.

It is not possible for Wodan's method to produce either more beakers or more GPPs in the middle or long term so therefore by definition he will be behind in research.

Well, I'm not sure the hypothetical situation is valid.

Above, I replied to the comparison for whipping at city max size. We still have to discuss the comparison between starting a scientist early. I'll give it shot.

SE scientist cities will have a food resource or flood plains (if they don't, they're production cities). So, for simplicity, let's look at fp. Say max size (due to happiness) is 7. Each worked tile has a farm and produces +2 food. Great, we love those cities.

Say for city A at size 4 we assign a scientist. I am still working 2 tiles (let's ignore city tile) and getting +4 food. At size 5 we assign a 2nd scientist, thus are still getting only +4 food.

Meanwhile, in city B we could be working 3 tiles and getting +6 food, and at size 5, 4 tiles for +8 food.

Let's just suppose it takes double the time for city A to get to max size than it takes city B. (It doesn't, it's less than that, because at size 6 they both increase, A to +6 and B to +100.) Let's call that X (in turns) for city A.

City A will be getting a little less than 6x in beakers (12x if Representation). City B doesn't get squat.

Now if not running Caste System, then City B never catches up. Period. There is no way city B ever gets any more beakers. What city B gets, however, is additional whipping opportunities. (See above.)

On the other hand, if running Caste System, when city B reaches max size (7), he can now afford to have 4 scientists. So, city B gets 12x beakers (24x if Representation).

Long winded way to work my thinking around, but I think I may have just concluded that if you don't have caste system, it's better to start early in all cases. With caste system, it depends on whether you want some early or more later.

Wodan

 

[MrCynical]

How do you come to that conclusion?

+1 food per tile (before Biology) means 1 specialist per 2 working citizens. With no Caste System, that's only size 4 to run 2 scientists. So, any city of size 5 or larger will have a bigger food surplus than the CE city working all cottages, even with the two specialists.

OK, firstly you're only supporting 2 scientists in a size 4 city at +1 food per tile if you stagnate the city, which a cottage city on the same terrain wouln't be. You'd need a size 6 specialist city to match the cottage city growth rate and run both scientists. Hence only a city of size 7+ would have a greater food surplus by that argument. Even at relatively low levels, the happiness cap is looming at this size.

Secondly you're forgetting that unless your cities are stupendously badly placed they will have one or two food resources, which the vast majority of the surplus is provided by, even in the SE. It is only if the city is still size 7+ after it is whipped that it still has a higher food surplus under the SE. The regrowth phase in both cases should be so short if you're using the whip properly that you'll only pick up half a dozen extra food while regrowing even with a fully farmed SE city.

Thirdly, this is only valid if you're assuming the SE is only running the two scientists per city, and no other specialists whatever. If you read my logs, I've pointed out that the contribution of two scientists per city to the economy is minimal. I could see an argument for scrapping them completely and running with just the food surplus, and then slavery would be significantly more effective. However that isn't in any sense a specialist economy.

Fourthly, to run the two scientists the SE city must be at least two pop points bigger than the CE. Since each pop costs more food, you're getting a poorer food to hammers exchange rate than the CE city which is whipping at two pop smaller, which counterbalances the tiny amount of extra food the SE might pick up in the regrowth phase.

Fifthly, this is only valid until representation, and only then if you don't run Caste System. After that the SE needs to run all the specialists it can, so slavery is actually slightly less effective due to the higher population.

As someone who's just played through a game as both SE and CE, I'm entirely serious that I found no advantage whatever for the whip in an SE.

 

[Wodan]

OK, firstly you're only supporting 2 scientists in a size 4 city at +1 food per tile if you stagnate the city

I said assign 1 at size 4 and assign the second later.

You'd need a size 6 specialist city to match the cottage city growth rate

Why do we need to match the cottage city growth rate?

Secondly you're forgetting that unless your cities are stupendously badly placed they will have one or two food resources

I said, for simplicity, let's look at the flood plains case.

And, I daresay that a city with flood plains and no food resources is hardly "stupdendously badly placed". As long as you have the unhealthiness in check and are using slavery, that's a wonderful spot.

Thirdly, this is only valid if you're assuming the SE is only running the two scientists per city, and no other specialists whatever.

Correct. I was talking about research only.

If you read my logs

Did you post them? I missed seeing that somehow.

EDIT: I see them now. Will try to look at them later today if I can to see how they went. We're grilling out so I won't have much time.

I've pointed out that the contribution of two scientists per city to the economy is minimal.

How so?

Fourthly, to run the two scientists the SE city must be at least two pop points bigger than the CE.

Why?

BTW this is exactly my point. A specialist city does NOT have to be bigger than a corresponding CE city, in order to run specialists. The only requirement is to have the food necessary to support said specialists.

Fifthly, this is only valid until representation, and only then if you don't run Caste System.

Why does this become invalid with Representation?

After that the SE needs to run all the specialists it can, so slavery is actually slightly less effective due to the higher population.

Agreed. Said that myself.

As someone who's just played through a game as both SE and CE, I'm entirely serious that I found no advantage whatever for the whip in an SE.

I'll be interested in seeing your logs!

Wodan

 

[mzprox]

I don't know the "classic" SE strategy, but what I am doing:
using many specialist in all my cities (merchants if i need money). In the capital I use only scientist specialist. Every great person is put in the capital where I build the national epic and Oxford University. In this way I could get 20-30+ GP "builded" into the city and it produces around 800+ RP itself after biology... very powerfull strategy .

 

[MrCynical]

I said assign 1 at size 4 and assign the second later.

Without wishing to nitpick, you didn't:

With no Caste System, that's only size 4 to run 2 scientists

Ditto for:

I said, for simplicity, let's look at the flood plains case.

I can find a reference to looking at +1 food per tile, which I assumed was farmed grass, but nothing whatever about flood plain. If you're referring to floodplain as the +1 food per tile without farming it's a rather strange comparison since the SE then has no additional food over the CE. If you mean that it's in practice +2 food per tile with farmed floodplains then your example works, but that isn't what you originally said.

BTW this is exactly my point. A specialist city does NOT have to be bigger than a corresponding CE city, in order to run specialists. The only requirement is to have the food necessary to support said specialists.

Then you're hitting the problem of losing commerce and hammer output because you're working fewer tiles, and there are plenty of tiles that give 3+ food and some hammers or commerce. Also if as you suggest you're capping your city at size four your whipping is hampered by limit of only whipping half your population. Again I find being able to whip 3 at a time a big step up, and again a reason to prefer size 6+ cities.

Why do we need to match the cottage city growth rate?

Well if you're growing slower than the cottage city then by definition the food box is filling slower and you can't whip as effectively!

Why does this become invalid with Representation?

To have any hope at all of keeping up with a CE an SE has to get to Representation as soon as possible, and beginning running many specialists, some of which will be non-scientists. This removes the limitation of two specialists per city (which is utterly inadequate compared to a CE at this stage anyway). See my logs for my reasons for thinking this, and also why I feel that scientists prior to Representation may actually do more harm than good. The small amount of the total science (often less than 15% even with max scientists) seems inadequate to compensate for the huge amount of food being burnt, and the extra maintenance costs of the larger cities. My impression was that I made it to Representation in the SE despite, not with the aid of, running all the scientists I could.

 

[UncleJJ]

I don't know the "classic" SE strategy, but what I am doing:
using many specialist in all my cities (merchants if i need money). In the capital I use only scientist specialist. Every great person is put in the capital where I build the national epic and Oxford University. In this way I could get 20-30+ GP "builded" into the city and it produces around 800+ RP itself after biology... very powerfull strategy .

That is what a SE needs to do in the middle game to keep its research rate up. It should produce GPs and settle them in the best city to make use of their different outputs combined with the best multipliers. I prefer to split the function of the Science City (Academy, several Monastries, Oxford U and settled G Scientists) from the Gold City (Holy Shrine, Wall Street plus G Prophets and GM). That gives more control over what else to build and what sort of specialists to run. My Gold City usually gets the Iron Works to make use of the hammers from the G Prophets.

 

[UncleJJ]

Well if you're growing slower than the cottage city then by definition the food box is filling slower and you can't whip as effectively!

I agree with you. Wodan seems to work specialists before his city has regrown from the loss of pop from the whip. That is a sub optimal strategy in most circumstances. A SE city should maximise its food production over time, while still taking the hammers needed for units and buildings, and that means working farms as often as possible within the happiness constraints.

[Wodan: I am preparing a properly laid out argument supporting my assertions here so please wait a day or two and I'll post in this thread. It is effectively a reply to your post 183]

To have any hope at all of keeping up with a CE an SE has to get to Representation as soon as possible, and beginning running many specialists, some of which will be non-scientists. This removes the limitation of two specialists per city (which is utterly inadequate compared to a CE at this stage anyway). See my logs for my reasons for thinking this, and also why I feel that scientists prior to Representation may actually do more harm than good. The small amount of the total science (often less than 15% even with max scientists) seems inadequate to compensate for the huge amount of food being burnt, and the extra maintenance costs of the larger cities. My impression was that I made it to Representation in the SE despite, not with the aid of, running all the scientists I could.

I don't agree that the Pyramids and early Representation are needed in a SE for it to be competative with a CE. The SE does, however, have to be run in a very particular way, revolving around the concept of a Science City. In my SE games my capital is the Science City and is producing more than half the beakers when I have 8 cities.

The important idea behind this economic strategy is to concentrate the settled scientists in a city with the best science multipliers (short term and eventually long term). There is effectively a transfer of food from the other supporting cities to the Science City. The supporting cities running 2 scientists in a library turning food into GPPs to make a GS that is settled in the Science City. It doesn't really matter that they only produce 3 beakers each I'm after the 3 GPPs. It takes 17 turns with 2 scientists and hence only 68 food to make the first GS and then 134 and 200 food for the 2nd and 3rd GS. The first GS becomes an academy and thereafter others are settled (giving 11 beakers/ turn with the +95% science multiplier in the Science City) or used to lightbulb key technologies.

Also if any new religions are collected (by research, conquest, or spreading from the AI) it is a high priority to pass that on to the Science City for another monastry and perhaps a temple. Once Education is researched (high priority) building a university in 6 cities is a very high priority and then Oxford U is built in the Science City. The science multiplier is then over 200% and the settled GS give 19 beakers/turn.

As capital this Science City runs Bureaucracy (+50% commerce) and has the palace (+8 commerce). I will work any good sources of commerce and build several cottages (early in the game) in the capital to boost the commerce and try to run the research slider at 100% as long as possible. There are many ways to get gold to finance an early economy (and hence use commerce to provide beakers), and I will use any or all of them. Some might call this a hybrid economy, but I don't care, and even if it was I certainly run it just like a SE and build farms in all other cities. After Liberalism I often shift research towards Constitution and then switch from Heriditary Rule to Representation. At that stage of the game the number of specialists in all cities (scientists, merchants, engineers and priests) makes the change worthwhile.

It surely is an effective strategy and can easily match a SE using the Pyramids in many games and also a CE (depending on leader traits and map and even AI opponents). Unfortunately I can't see any other strategy for a SE without Pyramids. I am not sure how quickly MrCynical managed to build his Science City or how effective it was. It will be interesting to see his savegames and comment constructively on his play style .

 

[Wodan]

Without wishing to nitpick,

For not wishing to do so, you certainly did. I will, nevertheless, try to answer your questions.

you didn't:

I said, "Starting a science specialist early (before max growth, when you apply 2),..."

Ditto for:

You're taking that out of context. I said that when replying to what the minimum food situation is required to support two specialists. My very next sentence after that went on to talk about size 5 and above, saying how the situation got better the more you go beyond the minimum.

I can find a reference to looking at +1 food per tile, which I assumed was farmed grass, but nothing whatever about flood plain.

#183. I said, "SE scientist cities will have a food resource or flood plains (if they don't, they're production cities). So, for simplicity, let's look at fp."

If you mean that it's in practice +2 food per tile with farmed floodplains then your example works, but that isn't what you originally said..

I said, "Each worked tile has a farm and produces +2 food."

Then you're hitting the problem of losing commerce and hammer output because you're working fewer tiles, and there are plenty of tiles that give 3+ food and some hammers or commerce.

I've said repeatedly that my focus is maximizing research. I think yours is too, because you said "to have any hope of keeping up with a CE the SE has to get to Representation as soon as possible".

Also if as you suggest you're capping your city at size four your whipping is hampered by limit of only whipping half your population. Again I find being able to whip 3 at a time a big step up, and again a reason to prefer size 6+ cities.

Size 4? It's capped at size 7, just like the CE case.

I suggest getting to size 7 or 8, then whip 1-3 and go down to size 5 at worst. Keeping at least one scientist working the entire time.

As I understand it, you advocate getting to size 7, then whip as needed. The only time you assign scientists is when the city is capped and you don't want to whip. Anytime you whip you remove the scientists. Is all that right?

Well if you're growing slower than the cottage city then by definition the food box is filling slower and you can't whip as effectively!

Exactly. I stated as much.

In the first place, whipping at city cap size is not as effective as whipping at small city size.

I'd rather have more research plus a few hammers than less research and slightly more hammers. Hammers at that point are a side-show, because I already whipped (when the city was small) all the buildings I really wanted anyway.

To have any hope at all of keeping up with a CE an SE has to get to Representation as soon as possible, and beginning running many specialists, some of which will be non-scientists.

How does running a non-scientist help you?

I can think of only two things. One is that you use this extra economy to produce more military, and then conquer some CE cities, which could add to your research. If that's it, then that's fine... when gearing up for a war pretty much everything any of us has said is out the window. To me, it's a game of race to better tech and then race to crank out the better units.

The other is that you have many of your cities produce Research; a 30 hammer city would make 15 beakers -- to match this the city would have to be under caste system and running 5 scientists; however, if this is what you propose then you wouldn't be running non-scientists except perhaps 1 engineer, you'd be working mines, so that probably isn't it.

This removes the limitation of two specialists per city (which is utterly inadequate compared to a CE at this stage anyway). See my logs for my reasons for thinking this, and also why I feel that scientists prior to Representation may actually do more harm than good. The small amount of the total science (often less than 15% even with max scientists) seems inadequate to compensate for the huge amount of food being burnt, and the extra maintenance costs of the larger cities. My impression was that I made it to Representation in the SE despite, not with the aid of, running all the scientists I could.

I've read your logs but don't see how you arrive at that impression.

A super-science city can generate the majority of research. For example, in my SE test game right now, my super-science city is making 400 beakers a turn. Meanwhile, my next-most city is making less than 100 beakers a turn. Regardless of the numbers, at any given point in the game (after I started running case system), when added together, all my other cities probably add up to about equal of my super-science city.

You are advocating the removal of all scientists except in the super-science city, to somehow get to Representation faster. That would cut my research in half!! You'll have to explain because I'm not getting this one at all.

The only thing I can figure is that you're assuming no caste system. But, before code of laws, the super-science city isn't really kicking in anyway. Each 2-scientist city will make 3*2 = 6 + Library + Monasteries (let's call it average of 30%) = 7.8 beakers. Meanwhile the super science city (if any) should have 6 + 1-2 great scientists + Academy + Library + Monasteries (let's call it 80%) = 27 beakers. So, it takes only 4 cities running 2 scientists to double your super-science city beakers.

With caste system, you can run more than 2 scientists per city, both in the super-science city (if any) plus in all other cities. The super-science city will be making more, but so will many of the other cities.

Wodan

 

[Wodan]

Wodan seems to work specialists before his city has regrown from the loss of pop from the whip.

Yes.

[Wodan: I am preparing a properly laid out argument supporting my assertions here so please wait a day or two and I'll post in this thread. It is effectively a reply to your post 183]

No sweat.

I'm not claiming my thoughts are Absolute Truth TM, but I'll argue them over a beer. I'll even buy the beers.

I don't agree that the Pyramids and early Representation are needed in a SE for it to be competative with a CE. The SE does, however, have to be run in a very particular way, revolving around the concept of a Science City. In my SE games my capital is the Science City and is producing more than half the beakers when I have 8 cities.

A lot going on in that paragraph.

I, too, believe you don't need Pyramids to be competitive. Regardless of our beliefs, exactly this issue is what the test attempts to illustrate, one way or the other. We'll see what we see, I suppose.

However, I don't agree that you have to have a super-science city. Several kinds of super-science cities. Regardless of type, it generally has oxford. Types (off the top of my head):
1) GP farm with National Epic, running max farms plus 2+ food resources, Academy, caste system, max scientists, with settled GSs (great scientists)
2) #1 but GSs after Academy used to lightbulb instead of settling
3) Cottage city, max cottages + food resources, must run Bureaucracy, must have research slider at high level, Academy, settled GSs
4) #3 but GSs after Academy are lightbulbed
5) Not predictable or uncommon situations, such as Colossus + Financial on coast, many dye or other commerce resources, Temple of Artemis, etc.

A super-science city will often (but not always) produce more and earlier GSs, which can be settled or lightbulbed. Can you get by without this? Sure. For example, super-science city types (as I defined above) #3, #4, and #5 all don't produce many GSs. In addition, maybe your GP farm is aimed to produce GPs of other types. e.g., great artists to get a cultural victory. We're talking SE, so you'll still get a good number of great scientists from your secondary cities. A bit later in the game, but you'll get them, and later in the game lightbulbing them gives more beakers, so it's not as bad as it might seem.

The important idea behind this economic strategy is to concentrate the settled scientists in a city with the best science multipliers (short term and eventually long term). There is effectively a transfer of food from the other supporting cities to the Science City. The supporting cities running 2 scientists in a library turning food into GPPs to make a GS that is settled in the Science City. It doesn't really matter that they only produce 3 beakers each I'm after the 3 GPPs. It takes 17 turns with 2 scientists and hence only 68 food to make the first GS and then 134 and 200 food for the 2nd and 3rd GS. The first GS becomes an academy and thereafter others are settled (giving 11 beakers/ turn with the +95% science multiplier in the Science City) or used to lightbulb key technologies.

Do you then turn off the scientists in the secondary cities (after they have produced the GS, which was your primary goal)? Just wondering, because I've tried playing that way. It worked pretty well. Maxed my GPP, that's for sure. Whether it was better from a research perspective I don't know. Would be extremely hard to calculate.

It surely is an effective strategy and can easily match a SE using the Pyramids in many games and also a CE (depending on leader traits and map and even AI opponents). Unfortunately I can't see any other strategy for a SE without Pyramids. I am not sure how quickly MrCynical managed to build his Science City or how effective it was. It will be interesting to see his savegames and comment constructively on his play style .

Agreed. You'll see mine too... my super-science city is cranking on all cylinders currently. I do not, however, have a tech lead on the AIs. Some of them I'm ahead of, but Mansa in particular has 3 techs on me. (It's 1850AD or so.) I'm not terribly worried, as I am beelining and thus, while I have fewer techs, I have more strategically critical ones. I have to decide real soon whether I'm going to get tanks and go on the warpath, or go for the elevator and a space victory. It's a bit tardy to be deciding that. Partially a result of my choice in the game. I purposefully played to get a late-game decision... pushed my empire out without a major war, and then built the infrastructure all game. I figured that would give a great comparison to a CE on a long-gametime basis.

Wodan

 

[MrCynical]

How does running a non-scientist help you?

Very simple; once scientist slots are exhausted you can still pick up three research per turn, plus some other stuff, from other specialist types once you have representation.

The only thing I can figure is that you're assuming no caste system. But, before code of laws, the super-science city isn't really kicking in anyway. Each 2-scientist city will make 3*2 = 6 + Library + Monasteries (let's call it average of 30%) = 7.8 beakers. Meanwhile the super science city (if any) should have 6 + 1-2 great scientists + Academy + Library + Monasteries (let's call it 80%) = 27 beakers. So, it takes only 4 cities running 2 scientists to double your super-science city beakers.

With caste system, you can run more than 2 scientists per city, both in the super-science city (if any) plus in all other cities. The super-science city will be making more, but so will many of the other cities.

This kind of double counting is one of the things that winds me up about the SE/CE argument. You can have slavery OR caste system, not both, so don't attribute slavery as an advantage of an SE (and as I've explained it doesn't seem to function any better under the SE in practice), unless you're willing to cap your cities at two scientists each. I was working under this principle of running slavery, not caste system in my test based on your earlier arguments that caste system was unnecessary.

You are advocating the removal of all scientists except in the super-science city, to somehow get to Representation faster. That would cut my research in half!! You'll have to explain because I'm not getting this one at all.

So you're seriously getting half your research from specialists in non-science cities PRIOR to representation? That isn't very plausible even WITH caste system. Without caste system you're going to need a lot of cities at two scientists each (hence only 7.8 beakers even with your rather implausible three monasteries everywhere) to match a super science city which will likely be running off cottages. That's to say nothing of terrain and trade route commerce which is present in both economies. If you have read my logs you'll see I highlight that less than 15% of my beakers were attributable to scientists. The loss of this wouldn't make representation that much later, and would give me a heck of a lot more food to convert to production (It would also save an amount in maintenance almost equal to the research output of the scientists). GP are of no relevance since if you have a reasonable GP farm you won't get any additional GP in the SE than CE before representation (or quite likely for a long time afterwards).

With caste system, you can run more than 2 scientists per city, both in the super-science city (if any) plus in all other cities. The super-science city will be making more, but so will many of the other cities.

Are you running caste system or not? If no, then remember you're restricted to two scientists per city. If yes, throw your claims that the SE has any advantage over the CE on slavery in the nearest dustbin. The SE has a disadvantage to slavery with the SE because it isn't available. You can have it one way or the other, but I'm sick of seeing these arguments trotted out together.

 

[Wodan]

Very simple; once scientist slots are exhausted you can still pick up three research per turn, plus some other stuff, from other specialist types once you have representation.

Two things...

First, I misunderstood you. You said, "To have any hope at all of keeping up with a CE an SE has to get to Representation as soon as possible, and beginning running many specialists, some of which will be non-scientists."

I thought you meant "A and B" whereas you meant "get A to do B". That is, I thought you meant the SE has to run non-scientists before getting Representation. So, I asked how non-scientists help you get research in order to get Representation, because that makes no sense.

Anyway, my apologies for misunderstanding.

Second, after Representation, I think the goal is still to run as many scientists as possible, in order to maximize research. Representation after all helps scientists just as much as non-scientists. So, in order to run non-scientists you have to give up some research, and personally I give research a priority over hammers or gold (with some obvious exceptions such as wartime).

If running Representation but not running caste system, then I agree it will be necessary to run as many specialists as possible, no matter the type.

This kind of double counting is one of the things that winds me up about the SE/CE argument. You can have slavery OR caste system, not both, so don't attribute slavery as an advantage of an SE (and as I've explained it doesn't seem to function any better under the SE in practice), unless you're willing to cap your cities at two scientists each. I was working under this principle of running slavery, not caste system in my test based on your earlier arguments that caste system was unnecessary.

Where in the world did we talk about whipping under caste system?

We've talked about whipping, and we've talked about caste sytem. Yes, we've gone back and forth, but I just reviewed the last couple of days' posts and I don't see where anybody did what you're talking about.

So you're seriously getting half your research from specialists in non-science cities PRIOR to representation? That isn't very plausible even WITH caste system.

Why not?

Without caste system you're going to need a lot of cities at two scientists each (hence only 7.8 beakers even with your rather implausible three monasteries everywhere)

I didn't suggest 3 monasteries... I suggested 1 Library + 1/2 of a monastery as a reasonable average. That equals 30%. 6 * 1.30 = 7.8.

to match a super science city which will likely be running off cottages.

Who cares what it's running off of, what matters is what beakers are generated. And, prior to code of laws, 27 beakers is a reasonable expectation.

That's to say nothing of terrain and trade route commerce which is present in both economies. If you have read my logs you'll see I highlight that less than 15% of my beakers were attributable to scientists. The loss of this wouldn't make representation that much later, and would give me a heck of a lot more food to convert to production (It would also save an amount in maintenance almost equal to the research output of the scientists).

That's fine, but we need to point out assumptions such as "a super science city which will likely be running off cottages".

Many SEs run at low slider numbers to get happiness from culture. Also, many SEs have a super science city running off a GP farm; heck, many CEs have this too. Finally, many SEs (and some CEs) do not have cottages in the capitol. In all of these cases, your conclusions don't necessarily apply because the assumptions aren't in force there.

Are you running caste system or not?

Some civs run caste system, some don't. This is true of SE as well as CE.

If no, then remember you're restricted to two scientists per city. If yes, throw your claims that the SE has any advantage over the CE on slavery in the nearest dustbin.

Any general comment about slavery being always better under SE is misleading. I agree with you there, though I wouldn't go so far as to say the comment is totally false under caste system. Even under caste system, whipping is possible before you get code of laws. We've all agreed that whipping is most efficient when the city is small.

Slavery is most effective for SEs who aren't running caste system, or who are spiritual and spend a LOT of work going back and forth. Something like that is probably a more accurate statement for people to be using.

You can have it one way or the other, but I'm sick of seeing these arguments trotted out together.

You know... I think you punched your own buttons.

If you are "sick" of people talking about slavery and caste system in the same discussion, all I can say is that they're related. It's perfectly natural for someone to say "here's what would happen under slavery, and to compare here's what you could do under caste system".

If that bothers you, I'm sorry, but I personally see nothing wrong with it.

Wodan

 

[CivScientist]

...I then graphed the progression of the city's economic output following each economy. The graph I made is below:



Here's a graph plotting the difference between the two versions of the city for the first 200 years:


...

Sorry it took me a while to respond. I've been rather busy as of late but I am still interested in this topic so bear with me I guess. I'll try to cover some of the key points addressed in regards to my post and discuss how I plan on expanding my model.

First off, I did neglect to consider slavery in my model. My original model actually assumed that no buildings were built whether they were libraries, markets, or graneries. My assumption was that, if these buildings were built at the same time in either the SE or CE situation, the effect of these buildings in both the SE and CE versions of the city would be identical because these buildings serve to multiply economic output (well, maybe not graneries). The only difference would be the difference in economic output between the two versions of the city would be amplified. That assumption isn't unreasonable IF the buildings were built the same time. However, it stands to reason that an SE city that whips judiciously and optimally would be able to build these buildings faster than the CE city that does the same.

As far as ac****ing or GP, I guess I didn't include that in my model simply because I rarely find GP being produced in any other city than my GP farm. Mind you, with my style of play my GP farm is normally my first or second city, I get that city up and running quickly, and it usually is a very effective city in producing GP and GPPs. So much so that I rarely see GP produced in any of my other cities (except maybe my capital). Mind you, I also try to max out my city growth as much as is safe and reasonable before creating a single specialist in that city for each of my cities (except maybe my capital again). So, I guess I'm not entirely convinced I have any hope of seeing a GP come from this city. Fortunately, if I'm wrong, it's not hard to predict the bonus that this GP will have.

I agree with UncleJJ in regards to it being optimal to grow your city in terms of population as fast as possible before...well, most anything else including hiring specialists. I did the math back for Civ2 and I'm fairly certain it still figures in Civ4. There are times you need the specialist sooner and, of course, your first beaker is worth more than your last beaker. Still, it's generally a good idea and can be shown to be superior given some reasonable assumptions that it's better to grow first and produce later. I'll let UncleJJ show the proof of that when he gets to it.

At any rate, I think my next step is to figure out what the optimal whipping strategy should be given my city's situation and apply the slavery paradigm to both the SE and CE versions of my city. I will see how each building pertinent to this experiment affects the economic output of each city. I will of course also keep tabs on when these buildings are likely to arrive.

 

[Mutineer]

I allmost think you all simply do not want to hear.

It is GP that give beakers to SE, not research directly. That why merchant specialists are good.

GM running for trade mission can get you 1000-1300 gold early on, letting you put research to 100% insted of my typical 30%.

Because of higher % to sci early on it will convert 1000Gp to about 1500 beakers in average, giving you the same amount of beakers ligth bulbing GS give.

SE run on GREATE PEOPLE, not on there direct production.

 

[Beamup]

Which would be a stronger argument if MrCynical's test games didn't show that the CE produced almost as many GP as the SE. And if an SE actually had a meaningful way to convert gold into deficit research on a large scale.

 

[Wodan]

A CE with a good GP farm should produce almost as many GP as a SE, I agree. Not as soon, though. Early techs gained through lightbulbing are traded to AIs for tons of techs, meaning a huge amount of free research.

If you agree with the above principle, then we can indeed conclude that a scientist-lightbulb SE can get much more research from GP than CE does.

Whether other types of SEs (such as Mutineer's merchant-trade mission SE, or a scientist-merging SE) get as much research from GP, would have to be looked at separately.

On the other hand, I think I'd have to try out your strategy Mutineer, before I buy it completely. Great merchants allowing the SE to run at 100% on the slider? I suppose you have to have a cottage capitol, and another city with a merchant GP farm. Seems like fun, though.

Wodan

 

[Wodan]

This makes me wonder. How many types of SE and CE are there? That would make an interesting thread. People tossing out their economy strategy ideas. I mean, we just talked about several of them:

-- merchant / trade mission SE
-- scientist / lightbulb SE
-- scientist / merging SE
-- artist GP farm / caste system for culture victory SE (that was in another thread this morning)

CE is harder to "categorize" I think.

Wodan

 

[UncleJJ]

I allmost think you all simply do not want to hear.

It is GP that give beakers to SE, not research directly. That why merchant specialists are good.

GM running for trade mission can get you 1000-1300 gold early on, letting you put research to 100% insted of my typical 30%.

Because of higher % to sci early on it will convert 1000Gp to about 1500 beakers in average, giving you the same amount of beakers ligth bulbing GS give.

SE run on GREATE PEOPLE, not on there direct production.

I don't think that 1000 to 1300 gold is that good a deal really. A settled GM will give 6 gold and 1 food each turn, and that will later grow to 12 gold with market, grocer and bank and then finally grow to 18 with Wall Street. That is a much better long term investment. It will also produce 3 beakers if you run Representation. The extra food is always useful and will support half another specialist. So for me a GM gets settled.

 

[Wodan]

I don't think that 1000 to 1300 gold is that good a deal really. A settled GM will give 6 gold and 1 food each turn, and that will later grow to 12 gold with market, grocer and bank and then finally grow to 18 with Wall Street. That is a much better long term investment. It will also produce 3 beakers if you run Representation. The extra food is always useful and will support half another specialist. So for me a GM gets settled.

In the past I've pretty much always settled them too. I think he might have something, however.

In any event, he's talking about how to use a great merchant to get more research, which is eminently valuable (assuming you have the preconditions already in place). To contrast, you point out the benefits of merging. Definitely they are benefits. But if you have happy/health caps (and thus can't run more specialists or work more tiles) and Representation aside, what you get is gold. If you have plenty of gold (because of shrines or whatever), then having yet more might not be so attractive as more research, which you can always use.

Wodan