Detailed analysis: # of specialist cities needed per era

I just tried this with Mao. Monarch.

I pulled it off! BUT, only at 700BC, which is pretty late imo. Ideal is 1200 BC. Good is 1000BC.

I admit though that I diverted for Masonry and The Wheel early on because I had marble in my Oracle (2nd) city. I used the marble to also get Parthenon, since I researched polyth instead of medit. Philosophical, parthenon, plus pyramids for specialist strat, PLUS early metal casting and preventing the AI from getting oracle. OMFG!!! this is INSANE, lol!

i'm going to try again, despite the great start, because I want to try to get it earlier. i'm going to try just going: bronze-myst-medit-priest-pottery-metcast (free)-masonry. in this game i went: bronze-the wheel-masonry-myst-poly-priest-pottery.

if i can get this consistently by 1200 BC i'm going to be going insane!!!

i've been playing gandhi for the fast worker and industrious trait, since i've felt industrious to be essential. but if i can guarantee pyramids w/out industrious trait, that would be HUGE, because i'd really love to have phil and there's no ind-phil :( Mao is also has the best starting techs imho. plus i'm a huge fan of organized because of the cheap courthouses and cheap lighthouses. courthouses you build in every city and lighthouses in every coastal city. that's a LOT of hammers over the course of the game.

man, this was a great idea, good job!

edit: oops, i forgot to mention that i don't think the 8turn window is accurate, i think it's much shorter! i had two workers sitting waiting on two forests for when oracle was done. as soon as it was done, they chopped (2 turns, forgot to prechop). then i 2-pop rushed the forge (1 turn). then i set the engineer (1 turn). this *tied* my forge city's and oracle city's gp rate! fortunately, the ge still popped! so this is a window of 4 turns, with 3 being better. this was normal speed.
 
ese-aSH said:
late game (after biology) :
isnt Caesar the best leader for that strategy ? organized --> many cities ; expansive --> big cities + praetorians for early war (and early huge empire ;) )
:p

I usually play caesar on marathon on prince/monarch <depending on if I'm using HOF mod or other mods>, but recently switched to ghandi after using more specialists over a cottage based economy. It works much better. the pret's are great for early war, but i find a axemen/swordsmen rush can take out one or 2 neighbors, by that point you can switch to caste system. But, with caesar you can't switch between slavery and caste system as needed and he is not industrious. So, with ghandi you are much more likely to get the wonders you want, and with the tech lead I usually have, the roman UU isn't that big a deal. I find I over-expand with rome, and even at 0% science still end up with units on strike. And since its so early most of my cities arent large enough to support enough specialists to keep the tech going.
early war = good, too much early war = disaster.
With ghandi, I can switch to slavery right before war to pop rush some units <like right after getting the required tech> and go to war with almost no waiting. Then 4 turns later <the amount you have to wait after one revolution on marathon speed before you can have another one> right back to caste system.

Although, I do have a quick question. Is it better to build 2 acadamies? I always have tons of great scientists, last game my capital had around 9-12 super specialists + 2 free from great library + 4 or 5 normal specialists by 1200AD when I won dom victory. So, should I build 2 acadamies and split my super specialists between them, or just add um all to once city. I find that since I tend to build most of my wonders in my capital and always build the GL and an acadamy there, I get tons of GP from it, but would it be better to split these?
 
futurehermit said:
edit: oops, i forgot to mention that i don't think the 8turn window is accurate, i think it's much shorter! i had two workers sitting waiting on two forests for when oracle was done. as soon as it was done, they chopped (2 turns, forgot to prechop). then i 2-pop rushed the forge (1 turn). then i set the engineer (1 turn). this *tied* my forge city's and oracle city's gp rate! fortunately, the ge still popped! so this is a window of 4 turns, with 3 being better. this was normal speed.

I think this was probably due to the Parthenon. Without the Parthenon, on Normal speed, you need 50 points to spawn your first great person. The Oracle will do that in 25 turns. The engineer specialist will do it in 17 turns. That should leave an 8 turn window of opportunity plus or minus one turn for rounding and/or other surprises.
 
futurehermit said:
I just tried this with Mao. Monarch.

I pulled it off! BUT, only at 700BC, which is pretty late imo. Ideal is 1200 BC. Good is 1000BC...

edit: oops, i forgot to mention that i don't think the 8turn window is accurate, i think it's much shorter! i had two workers sitting waiting on two forests for when oracle was done. as soon as it was done, they chopped (2 turns, forgot to prechop). then i 2-pop rushed the forge (1 turn). then i set the engineer (1 turn). this *tied* my forge city's and oracle city's gp rate! fortunately, the ge still popped! so this is a window of 4 turns, with 3 being better. this was normal speed.

I've been trying this myself, also coincidentally with Mao/Monarch. 1200 seems a tad slow (unless you mean that's the date you have the Engineer make the Pyramids). It's not unusual for me to see the Oracle appear in an AI's hands in the 1300s BC and to have the Pyramids go up in the 900s BC. From what I can tell the planning on this is very tight, so tight that you're almost doing just as well to make the Pyramids yourself. However I like this alternate plan a lot better since you have good options left if you fail to get the Pyramids. Lose the Pyramids building them yourself and you have a busted plan and some cash to pick up the pieces. Lose the Pyramids with the Engineer plan, and you can still use the one you get to make the Hanging Gardens if you want to boost further GE production, or use him on a Machinery slingshot. The latter is nice since Machinery is at the top of his list of things to get and you have all the prereqs you need from Metal Casting. Just make sure to research or trade/demand Iron Working and Archery while you're getting the GE out. Of course the Hanging Gardens are nice too in higher levels when your health limit gets closer to the happiness limit, and if you're willing for a little delay you can use the first GE on the Gardens and your second on Machinery.

The hardest part seems to be getting the Forge done in a timely manner. You either need plenty of forests prechopped or a size 4 city to whip if you don't have enough trees in your second site. Figuring out the proper time to found the 2nd city and if/when to get a 2nd Worker to speed up the chops is the greatest aspect of pulling this off (for me anyway).
 
Eqqman said:
The hardest part seems to be getting the Forge done in a timely manner. You either need plenty of forests prechopped or a size 4 city to whip if you don't have enough trees in your second site. Figuring out the proper time to found the 2nd city and if/when to get a 2nd Worker to spped up the chops is the greatest aspect of pulling this off (for me anyway).

I was doing some quick mental calculations on this earlier. Let's see if we can formalize it. The examples below are for Normal speed, but the results are the same at any game speed. The scale factor winds up in both the numerator and the denominator, so it cancels itself out.

forge cost = 120 hammers
chop value = 20 hammers (assume Mathematics is not yet researched)
whip value = 30 hammers

The forge needs to be built within 8 turns, so we can calculate the required production in hammers per turn (hpt) under various circumstances.

  • 0 chop, 0 whip -> 120 / 8 hpt = 15 hpt
  • 1 chop, 0 whip -> (120 - 1 * 20) / 8 = 100 / 8 = 13 hpt
  • 2 chop, 0 whip -> (120 - 2 * 20) = 80 / 8 = 10 hpt
  • 3 chop, 0 whip -> (120 - 3 * 20) = 60 / 8 = 8 hpt
  • 4 chop, 0 whip -> (120 - 4 * 20) = 40 / 8 = 5 hpt
  • 0 chop, 1 whip -> (120 - 30) = 90 / 8 = 12 hpt
  • 1 chop, 1 whip -> (120 - 1 * 20 - 30) = 70 / 8 = 9 hpt
  • 2 chop, 1 whip -> (120 - 2 * 20 - 30) = 50 / 8 = 7 hpt
  • 3 chop, 1 whip -> (120 - 3 * 20 - 30) = 30 / 8 = 4 hpt
  • 0 chop, 2 whip -> (120 - 2 * 30) = 60 / 8 = 8 hpt
  • 1 chop, 2 whip -> (120 - 1 * 20 - 2 * 30) = 40 / 8 = 5 hpt
  • 2 chop, 2 whip -> (120 - 2 * 20 - 2 * 30) = 20 / 8 = 3 hpt

At this stage of the game, you're probably looking at 5 to 10 hammers per turn in your city. Ten if you're lucky and got a nice food resource and some hills and metal. Five should be an easy minimum to reach (1 city center, 3 mined grassland hill, 1 forest) even if you need to temporarily stagnate growth to do it. You might have a little more in an extreme situation, but 10 maximum is probably about right.

Looking at those numbers, that means most of the time you probably want to whip 2 and possibly chop one forest if your production is below 8 hpt. If you have an extremely forest-rich city and the right timing on your worker turns, you could conceivably chop 3 or 4 forests and spare the whip, but that's not always going to work. Whipping 1 and chopping 1 or 2 forests is also possible. Whipping just 1 population is usually considered a bad idea, but in this case it could be necessary depending on your city size.
 
The best I've been able to get using this strat is pyramids built by 750BC. On monarch. This is too slow imo, esp for emperor.

Basically I use a 2 city strat. Here it is with Mao:

Build: Bronze-Myst-Medit-Priest-Wheel-Pottery-MetCast(free)-Masonry

Capital: Worker-Worker-Settler-Unfinished Stonehenge-Forge

2nd City: Obelisk-Oracle

Reasoning on using capital for the forge is that it will have some developed tiles and should be at size 4 for the whip. I try and leave 1-2 forests if possible and have them prechopped.

Anybody have any tips on my build?

Anybody get a faster time than 750BC?

If it can't be done by at least 1000BC then I would say it is a great idea, but won't work in practice. With Gandhi I can regularly get the pyramids by 1000BC. If you pull it off, it's faaar more powerful than the industrious build; however, if you don't pull it off you're left with an inferior cottage economy.
 
Dr Elmer Jiggle said:
At this stage of the game, you're probably looking at 5 to 10 hammers per turn in your city...

Ah, but which city? In order not to miss the Oracle you're likely to be building it in your capital. So the Forge is going into a city you have to rush to develop in time. And the fewer production resources you have there the more you'll have to rely on whipping and therefore food improvements, with a high probability you need to rush or chop an Obelisk to get the expansion you need to reach some resource or more forests. Of course in a high forest region it might be best to skip the border expansion if you have enough to make up for the loss of Hammers that results... all the more reason to make sure you have a second Worker out.
 
Eqqman said:
Ah, but which city?

Yeah, that's obviously the critical question. I was assuming forge in the capital and Oracle in city #2, but I haven't actually tried it yet. I play on Prince, so I may not be familiar with the level of urgency required to finish The Oracle on Monarch or above.

It does seem like you'd have a better shot at getting The Oracle if you built it in the capital.

Are you chopping The Oracle or building it from scratch? In other words, how long does city #2 have to develop while you build The Oracle before it needs to be ready to build the forge?
 
On emperor, the fastest I've been able to pull it off was 825BC. I built a worker then straight to a settler in the capital. Second city was settled in a high food spot to grow to 4 pop as quickly as possible while producing warriors. The worker actually followed the settler to start improving City #2, as it was going to be a while before the capital grew into the improvements built while producing the settler. Capital started on the Oracle as soon as Priesthood came in, finishing around 1720 BC. Chopped 4 forests to speed the oracle. Second city started work on the Forge as soon as oracle was complete and finished 7 turns later (1 chop and pop-rushed 2).

There are two difficult aspects of this strategy that must be balanced against one-another. First, the second city must be built early enough and be given enough time to be able to contribute a large number of hammers. Unless you can settle the city in the middle of a lush forest or next to marble, you're best off using it as your engineer city, giving it more time to get up to speed. Second, you must build the oracle suitably early. The reason for the early oracle is quite clear. It's going to be a little under 25 turns between the oracle and your great engineer. That's ~1000 years. If you're shooting for a 1200BC pyramids doing this, it ain't happenin'; I haven't seen many 2200BC oracles in any situation that didn't involve a straight beeline.

I'm still experimenting some; determining the best scenarios in which to try for the engineered pyramids. Off hand, does anyone know a typical completion date for the pyramids by a non-industrious civ without stone in a real game (in a non-beeline scenario).
 
malekithe said:
On emperor, the fastest I've been able to pull it off was 825BC. I built a worker then straight to a settler in the capital. Second city was settled in a high food spot to grow to 4 pop as quickly as possible while producing warriors. The worker actually followed the settler to start improving City #2, as it was going to be a while before the capital grew into the improvements built while producing the settler. Capital started on the Oracle as soon as Priesthood came in, finishing around 1720 BC. Chopped 4 forests to speed the oracle. Second city started work on the Forge as soon as oracle was complete and finished 7 turns later (1 chop and pop-rushed 2).

What were the cities like when you managed this? I've been trying to run it based on 'non ideal, so anybody can do it' settings and haven't been able to meet the challenge. Without a 5 Food tile and high Commerce tile (any of Gold/Silver/Gems) available the best I can pull off so far is completion by the 600s. Of course, if you're flexible with your strategy and missed the Pyramids I've already mentioned that you can still find good uses for the Engineer. But if you're just playing for a Pyramids guarantee I'm not sure it's possible without these good tiles at hand. The 1000BC challenge date leaves exactly 75 turns on normal speed. 17 of those you are waiting for the Engineer, and most of the first 25 are lost getting the Worker and Settler out. That leaves about 33 left to make a lot of magic happen.

The biggest bottleneck I've run into so far is simply getting the research done to start the Oracle in a timely fashion. Saladin has a head start over Mao with Mysticism but I've been finding that having to start by researching a food improvement tech negates any advantages you might have had with not starting with one. And picking a Philosopher without Mining gives you a double whammy.

I haven't done any checks yet on the pros or cons of picking a Fishing Philosopher. The advantage is you get to work both Food and Commerce at the same time (and Elizabeth really helps here) so you have that leg up on research but the use of your first 30 Hammers to improve only a single resource probably negates that.

The thing that really gets me is that the unacceptable date of 760 BC comes only 6(!) turns after the 1000 BC challenge. But I get where you're coming from, if you have your heart set on a certain objective then even one turn is too late. You'd think you'd be able to shave off those last 6 turns, but as I mentioned I'm not sure this is possible without a really good starting location.

malekithe said:
I'm still experimenting some; determining the best scenarios in which to try for the engineered pyramids. Off hand, does anyone know a typical completion date for the pyramids by a non-industrious civ without stone in a real game (in a non-beeline scenario).

I assume you mean for human players trying to finish Pyramids? I wouldn't think there's a lot of data on this since I can't imagine somebody with no Stone or Industrious would even start the Pyramids at higher levels- you'd expect it would be an automatic write-off.
 
Yeah, without stone or industrious on emperor and maybe monarch pyramids is nearly impossible (you would have to be very lucky; e.g., all the ai also in the same boat so that you have the advantage of chopping).

With Gandhi I can get the pyramids by 1000 BC even w/out stone as long as I have a respectable number of forests (I always settle my 2nd city in the spot that provides the most number of forests available, even if it kinda sucks otherwise). The thing about Gandhi is that his fast worker saves you all kinds of worker turns early on, which is huge, and spiritual prevents anarchy which is at least one turn, maybe two if an early religion spreads to your empire. Spiritual really shines later in the game of course, so it's nice having industrious which shines early and then when it starts to lag, spiritual starts to pick up.

I really feel that the engineer-pyramids strategy has amazing potential, but we really have to come with a way to get it down to 1000BC consistently. Anybody have any ideas? We'd probably have to really streamline it down, but the question is: Is it possible?

Is everyone in agreement that Mao is best for this strategy?
 
I popped a GP at 1000BC with this strat this morning, meaning the pyramids would've been rushbuilt at 960BC (respectable).

HOWEVER, 1) I popped mysticism from a goody hut which saved around 7-9 turns, which translates into about 300-350 years); and 2) I wasn't able to get my second city up to pop 4 in time because my worker had to start chopping the oracle as soon as the settler popped; as a result, I ended up getting a GP instead of a GE.

I didn't chop the settler though since I was using the time the settler was building to develop 3 improvements.

I'm going to try again, this time chopping the settler. Maybe something will come together...
 
One advantage Mao has but that hasn't been mentioned yet is his starting techs of Agriculture and Mining. Most of the Philosophical leaders also start with Mining, but I suspect that Agriculture is very helpful for this strategy, since rapid growth seems to be a critical obstacle.

I would think that if you want to make this approach truly general, you would need to make sure there's time for a detour to Animal Husbandry if you happen to start with livestock instead of grain. That will obviously cost you a few turns, but it seems like the improved resource ought to make that a worthwhile tradeoff.
 
Dr Elmer Jiggle said:
I would think that if you want to make this approach truly general, you would need to make sure there's time for a detour to Animal Husbandry if you happen to start with livestock instead of grain. That will obviously cost you a few turns, but it seems like the improved resource ought to make that a worthwhile tradeoff.

This is one of the things I've been looking at in the 'non ideal' case. You don't have time to research it, however, if you get livestock next to a fresh water source you could still build a farm (getting replaced with Pasture much later) and end up with a 4F/1C tile. But this is going to require having access to another tile that can quickly give 3 Commerce or more. Without such a tile you're forced to build and work at least 1 Cottage which means getting Pottery ahead of Bronze Working.

I've tried going the Library route to make up for not having good Commerce, but that forces you to get Animal Husbandry if you want a fast Writing, which is a needless delay, plus you have to use up extra trees to get the Library out fast enough to make a research difference (you've already spent time on 2 techs not on the needed path). You could also go Wheel->Pottery->Writing but this leaves it even longer to go after Bronze Working, which you also need quickly to take advantage of pop/tree rushing and pre-chopping.

The real limiting factor may be tree availability. If you can essentially finish the Oracle and Forge in 1 turn each, then you can afford to have a little bit of a research delay. The needed start might be a riverine Agriculture bonus in the middle of a thick pile of forests. If you can get a 5F tile going, then you'll have an easier time saving trees for the buildings.
 
Could someone give me their thoughts on game speed for the overall strategy (running representation and scientists)?

Do you think that normal game speed leverages representation scientists moreso than epic and marathon?

My thinking is that the representation scientists give you 6 beakers no matter what (7.5 with library), and techs cost more on the slower speeds. This would mean that if you go down to slower speeds you gain a military edge, but you lose an economic edge provided by having representation scientists...

I'm thinking that on normal speed, since you'll have a tech lead anyways, the military edge provided by going down to epic isn't as important. One of the reasons imo to go down to epic is to allow your army more time to do dmg before you get outteched and your army becomes obselete. But, since we wouldn't be suffering from the usual drawback of getting outteched when being aggressive, would it be better to stay on normal speed?
 
I just wanted to say that the path discussed of building the oracle getting a forge and shooting for a GE does have a good backup plan. I missed the pyramids by 2 turns, so used the GE to get machinery for free. the resulting pile of cho ko nus were sufficient to wipe out 2 of my neighbors and establish myself as the dominant power. with the lack of pyramids you can fall back to a cottage based econ, and still come out in a strong position

NaZ
 
i'm working on a new strat and i think i can guarantee the pyramids by 1000BC. it doesn't use the oracle. i'm convinced the oracle build takes too long.
 
Have you been beaten to the Pyramids yet in your test games, or are you just worried that it's too close a call? I'm still wondering about whether you have more time than you think. I haven't had a chance to research it in detail, but I did check one old saved game, and Gandhi built the Pyramids with stone in 880BC. This was Prince, so you need to dial that back to a somewhat earlier date (assume he could build it roughly 5% more quickly), but still, the point is that this was the worst case scenario -- industrious, stone, fast worker. I think this would be a reasonable approach even if you can't beat an industrious leader with stone, as long as you can still beat a leader who has one of those advantages but not both. As NaZdReG pointed out, you have an excellent backup plan.
 
Jiggle,

exactly!

normally I start building the pyramid in the capitol because it has the most hammers of my cities at that time period. but this time I used a forge to generate a great engineer. the turn before he arrived the pyramids were built by someone else :(.. oh well what can you do ;) fortunately while a specialist econ is quite cool to work with having a cottage based one of cource will let you survive and keep atleast tech parity with the ai.

I do have to thank you hermit though for getting me to try this strat, as it did show that 100% cottage spamming is actually a bad idea. having atleast 1 farm in the fat cross results in faster growth, to make sure that you can always keep your population at the maximum your happiness levels allow.

I still will use 1 area as a GP farm, and I generally try to split up my GP's by either prophet, enginner, or scientist. I think that popping a tech is worth signifigantly more than settling a Great person, but building academies is generally worth it.

as a warmonger :ar15:

I like to see great prophets spawn occasionally, because I conquer my rivals before they can build the shrine for their religion center. I then use 1 small (useless in terms of military production) city for each religion.. usually conquered territory. to spam missionaries of each religion I acquire the holy city for across my empire, adding gpt from the shrines and setting me up for a great happiness limit when I get to free religion :D

NaZ
 
Top Bottom