National College first: A test

[Civilised]

Nares: you're wrong in suggesting that I had an initial bias. I genuinely wanted to find out which approach was better - and I still do. I tried to make my testing as unbiased as possible. The answer about NC first or later may well prove to be situational, but it would be really interesting to establish the circumstances that favour NC first or NC later (and how much later). But I want that answer to be established by evidence rather than opinion - that was kinda the point of my original post! I'd really love to see someone (preferably with more experience than me) make a similar test from a different start point even if they come up with a different conclusion.

At turn 50 I completed the NC with an NC before expansion strategy and I'm at 21 BPT.

Monthar, that's great - but for a valid test you need to run two scenarios to an equivalent position. For example NC + 3 satellite cities as in my original post. One can't tell very much from looking at an optimised NC build by itself.

 

[DaveMcW]

Settle good city sites before building the NC.

The definition of "good" is what you're trying to test of course. It might be faster (though less fun) to calculate it theoretically instead of running a bunch of tests.

 

[Monthar]

Monthar, that's great - but for a valid test you need to run two scenarios to an equivalent position. For example NC + 3 satellite cities as in my original post. One can't tell very much from looking at an optimised NC build by itself.

I posted the initial save prior to the post you quoted so everyone could try various strategies and compare results. So feel free to try your method.

 

[vexing]

Vexing: "I love this completely broad conclusion based on one playthrough". Hmm, I took the trouble to detail the playthrough fairly carefully so that everyone reading it could see the basis for my analysis. My conclusion is self evidently valid for the scenarios I described. Your comment is trite and uncalled for. If you have actual evidence, rather than sarcastically expressed opinion, to show I am wrong in my conclusions then I would value seeing it.

my point was your conclusion is a really broad assertion "that on almost all measures it is better to build at least 3 additional cities before NC" based on little evidence. you can really only apply it to a very small sample (this specific map, this specific tech order, this specific maritime partner choosing strategy, etc).

i'd have accepted "on my two playthroughs, using the specific tech order i chose, i ended up slightly stronger by building x cities first then NC."

anyway, i do look forward to getting home and trying out your save.

 

[Airey]

Airey: I've attached a turn 0 file to this response. But I can't do that with no DLC, nor can I change the game speed - it's the game I already played to domination completion, and it has DLC in it. Sorry. But I agree it would be very helpful to the discussion if someone did the same sort of analysis based on no DLC and a standard speed game.

Thanks.

This is going to be a very useful thread, good job

@ Monthar

Spoiler :

nice game, but for riverside caps, from my experience, going for GL to get CS is actually better than NC 1st.
from your game, it still takes you 21 turns to get to CS. taking it early for civil service allows capital to boom very fast, especially with the 2 mining lux

also a side effect of taking GLibrary early on Immort and below is you take away GL from AI.

- isn't that the leader with 50% golden age bonus? Chinchen Itza?

Settle good city sites before building the NC.

The definition of "good" is what you're trying to test of course. It might be faster (though less fun) to calculate it theoretically instead of running a bunch of tests.

These kind of thing have to be very specific in defining parameters: ie. level + map
On immortal+, for example, players have enough $ to rush buy library thus faster NC, which is harder @ lower levels.
- lux sells from satellite cities also speeds up Library builds on higher level
- ppl are only looking at BPT @ the moment they finish NC, which is kind of misleading, because the head room to boom for expansion first starts are larger, not to mention prob loose city spot to AI if NC first on higher level (+ early DOWs).

 

[Monthar]

Going with 3 cities before NC I completed the NC on turn 109.

Spoiler :

This time I did not meet Hiawatha right away. I altered my tech order slightly to take mining first.
Build Order:

1. Worker
2. Monument
3. Warrior for 1 turn to wait for city to grow to 3.
4. Settler
5. Immortal
6. Settler
7. Library

New cities built 1 Immortal then their libraries.

When I finish researching Iron Working, I discovered my 2nd city was built on top of a 2 iron deposit.

While waiting for the 2nd city's library to finish I took out Elizabeth with 2 wars.

For the first war I traded her one of my luxuries for all her gold and used it to ally with Genoa. I was in my first Golden Age for this war, which made Hastings fall to 3 immortals and a warrior very quickly. After taking Hastings for the 2nd silver and spices, she asked for peace, giving me all her gold and GPT. I took it so I'd have enough to ally with Belgrade.

Then after the 10 turns ended and she had 4 workers near Nottingham, I traded with her to get all the gold I could, denounced her then captured all 4 workers with my previously stations immortals and 1 warrior.

She took out one immortal on the same turn Belgrade gave me a swordsman. I captured Nottingham for the ivory and had to change techs to actually get to use it even though she'd already built the camp for me. With the money I looted I upgraded my warrior so I'd have 2 swordsmen.

I pushed on, taking the rest of her cities in fairly short order.

I grabbed optics before pushing towards education so I could start sending my troops to the landmass to the south.

Here is the status of the empire on turn 109 with 3 cities before NC.

Capital

Spoiler :

Demographics

Spoiler :

Tech Tree

Spoiler :

Now I'm going to continue my NC first game out to turn 109 to better compare the results of the two different approaches.

Edit: I added labels for each of the screenshot spoilers so it's easier to compare with the other games.

 

[TheMightyQuinn]

One important thing to note is that settlers can be bought whereas the National College cannot. A technique that results from this is to research mining or calender after writing, then develop and sell a luxury resource* to buy a settler** while the library and National College are building. Of course you only found the city once the NC is built.

In this way I have had two cities (including my capital) and the NC by turn 50 (standard speed) on Emperor. You can then tech Iron Working very quickly which allows optimal placing of a third city shortly afterwards. If you build cities quickly before getting the NC then you risk having none of them near an iron resource.

* The importance of flogging everything to the AI cannot be overstated, whether you beeline the NC or not.

** While the library can be bought, for about the same cost, I nomally don't have enough money to do this when I want to. Unless you settle on top of a resource then it takes a good 8 turns to move a worker on to it and develop it, by which time most of a library could already be built. (How I miss earlier versions of CIV where the amount you paid for rush-buying decreased if you'd already built some of it!)

 

[Monthar]

I've extended the NC first game out to turn 109 to better compare to the 3 city then NC game. Here are the results.

Spoiler :

I built the Great Library while researching Philosophy, Animal Husbandry and Trapping. This timed out perfectly to complete Trapping on the same turn the GL completed, allowing me to slingshot Civil Service.

I built the first settler, a 2nd worker then the 2nd settler. I researched The Wheel and had my workers build the road to my 2nd city, then improve it's silver and and cows while the 2nd settler finished and moved to found the 3rd city. During this time I upgraded my warrior to a swordsman and had my first golden age.

The 2nd city's first build was a pikeman and the 3rd city's first build was a work boat. They both built a library as their 2nd build to help increase my BPT.

I then built the Chichen Itza in the capital to take full advantage of the Persian UA. Followed by a swordsman.

I've refused RAs with both Elizabeth and Hiawatha, because I didn't want to give them the additional 100 gold they asked for. Instead I used my funds from the GA and selling one excess silver to ally both Genoa and Belgrade.

My workers recently finished the 2nd iron and 3rd silver mines. I'm prepping to go take out Elizabeth, after selling that 3rd silver to her for some additional pre-war funds.

I have 2 swordsman and a pikeman with 3 more swordsmen cooking.

My BPT is slightly lower (45 BPT instead of 53 BPT) and the demographics aren't as good (because I haven't taken out England) as they were when I went with 3 cities before NC, however:

1. I will have a stronger military when I engage the English.
2. I'm #1 in literacy instead of #3 and am poised to accelerate my BPT as soon as I build the universities. Thanks in part to the GL-CS slingshot.
3. The Chichen Itza will extend the next GA, which should hit either while taking out England or when I'm poised to take out Hiawatha. That will be a big help with Persia's UA boosting movement speed and unit strength for a longer duration.

Here are the screenshots and saved game.
Demographics

Spoiler :

Tech Tree

Spoiler :

Capital's city screen

Spoiler :

 

[Monthar]

One important thing to note is that settlers can be bought whereas the National College cannot. A technique that results from this is to research mining or calender after writing, then develop and sell a luxury resource* to buy a settler** while the library and National College are building. Of course you only found the city once the NC is built.

In this way I have had two cities (including my capital) and the NC by turn 50 (standard speed) on Emperor. You can then tech Iron Working very quickly which allows optimal placing of a third city shortly afterwards. If you build cities quickly before getting the NC then you risk having none of them near an iron resource.

* The importance of flogging everything to the AI cannot be underestimated, whether you beeline the NC or not.

** While the library can be bought, for about the same cost, I nomally don't have enough money to do this when I want to. Unless you settle on top of a resource then it takes a good 8 turns to move a worker on to it and develop it, by which time most of a library could already be built. (How I miss earlier versions of CIV where the amount you paid for rush-buying decreased if you'd already built some of it!)

Good point about buying settlers and positioning them while the building the NC. I keep forgetting to do that. So I think I'll try this for the next comparison using the same Persian game I posted earlier.

 

[Monthar]

@ Monthar

Spoiler :

nice game, but for riverside caps, from my experience, going for GL to get CS is actually better than NC 1st.
from your game, it still takes you 21 turns to get to CS. taking it early for civil service allows capital to boom very fast, especially with the 2 mining lux

also a side effect of taking GLibrary early on Immort and below is you take away GL from AI.

- isn't that the leader with 50% golden age bonus? Chinchen Itza?

The only problem with this approach is there isn't enough time to build some Immortals to upgrade to Pikes before unlocking the Pikes. This means we can't have Pikes with the fast healing promotion, which can slow down the warmongering some.

However, I did go with this idea when I extended the NC first game. See the spoiler post above.

 

[TheMightyQuinn]

* Tangent Alert*

Monthar, if you're playing as Persia then make sure you time your golden ages properly. You want to have it start at the same time you declare war to benefit from the extra movement (3 move catapults!) and combat. Trading luxuries and leaving coliseums half built will help you do this.

 

[Monthar]

* Tangent Alert*

Monthar, if you're playing as Persia then make sure you time your golden ages properly. You want to have it start at the same time you declare war to benefit from the extra movement (3 move catapults!) and combat. Trading luxuries and leaving coliseums half built will help you do this.

Nice tip. I'll try to remember this, since I'm also trying to get the 5 GA achievement completed.

 

[Monthar]

Next I tried NC 1st, but with buying the first additional settler before it finished. I stopped at turn 109 again, for accurate comparisons with the two other strategies.

I also went with the idea of building a monument first, worker second. As with the other games, I placed the first 3 cities in the exact same locations. Building the Monument before the worker caused me to complete the NC a few turns later, even though I got to Aristocracy before I even started it. Probably because I wasn't able to get as many farms up with the first luxury mine as quickly, which slowed the capital's growth. So I think Worker first is the better route.

Spoiler :

Build Order:

1. Monument
2. Worker
3. Library
4. NC
5. 2nd Settler

Tech Order:

1. Pottery
2. Writing
3. Mining
4. Bronze Working
5. Iron Working
6. Animal Husbandry
7. The Wheel

After this I don't recall the exact order, but I worked on getting Coliseums and Longswords before heading towards Civil Service. This seems to have worked out quite well. Especially since the first unit Belgrade gave me was a LS.

I timed it to start the war with England when my first GA kicked off. However, I delayed 1 turn into the GA to have enough cash to Get Belgrade as an ally. I already had Genoa as an ally. This 1 turn worked out even better because my deals with England expired, so I was able to fleece Elizabeth for all her cash then start the war.

Unlike the first game of NC first, I had a few more Immortals, my warrior was already upgraded to a swordsman and I had built a 2nd swordsman.

I still only captured Hastings, but I was about to take Nottingham when Elizabeth offered it, Coventry GPT, cash and 2 luxuries for peace. I managed to get her to give me even more GPT and only had to give her one of my 3 silver. Since I had no-one else to trade the excess silver to, yet, this worked out great. I already had an active trade with Hiawatha for a silver.

In the screenshot of the Capital you'll notice I've built the Pyramids. I only built them because of a quest from Belgrade. Genoa asked me connect spices, which I did with the worker I captured when I took Hastings. So I've got some nice leeway with these two CS's before I have to feed then cash again.

Capital

Spoiler :

Demographics

Spoiler :

Tech Tree

Spoiler :

My tech tree isn't as filled out, I don't have as much BPT as the two previous strategies. However, I do have a much stronger military with a rating 38,419 (2nd place) vs "NC First with Great Library & Chichen Itza" 23,238 (6th place) and "3 Cities then NC" 30,659 (2nd place). Literacy is ranked lower than the 3 City, but I've got as much of the tech tree filled out, 22%, as that one and only 2% less than the NC first with Great Library to slingshot Civil Service.

Overall I think, at least with this game's start, that going with NC first and the build order of worker-monument-library-NC and buying at least 1 settler would give the best results. In fact, if I can make a trade in the first try at NC first to buy a settler, using that turn 50 save, I could probably have that one exceed all the previous strategies in BPT, Literacy, Military Strength and still be able to take all of Elizabeth's cities by turn 109.

I'm really enjoying this challenge of comparing different strategies and working out which route is best with this Persian game's start. I'm very much interested in what others can pull off in these first 109 turns.

 

[Celevin]

I like this thread so far, though I'd want a more regular start. Your food sucks!

I'd go worker->monument->National College, stopping the monument if my techs are fast enough. I'd go mining->pottery->writing for the techs. I would buy my library, and hopefully buy a settler as well if I have the cash left. The 2 mines make this a possibility. The worker would start with a farm obviously, then move onto building the first mine.

 

[Monthar]

I like this thread so far, though I'd want a more regular start. Your food sucks!

I'd go worker->monument->National College, stopping the monument if my techs are fast enough. I'd go mining->pottery->writing for the techs. I would buy my library, and hopefully buy a settler as well if I have the cash left. The 2 mines make this a possibility. The worker would start with a farm obviously, then move onto building the first mine.

I know the food sucks, but that actually helps with testing this strategy since a start with better food would mean growing the city much faster, thus much faster research and build times. In other words, this is a typical start. Sure, there are better starts, such as 2 wheat on a river with a mined luxury and marble near enough to expand to quickly, but this isn't about having optimal starts. It's about comparing variants of a strategy in a typical game.

Yes, mining could be first so you can sell the resource. However, unless you're playing on Immortal or Deity, the AI will not have enough gold to sell the luxuries to them. As it is, with mining after writing, they still could only afford half the gold with the other half coming from GPT trade. When taking 5 GPT with cash, the AI never gives 150 gold. Instead they typically only give 143 gold with the 5 GPT when friendly, so you lose out on 7 gold. That may not sound like much, but can cause a delay of a turn or two to make the purchase you wanted.

Even if the AI somehow managed to have enough gold to buy both luxuries at 300 each, that's only 600 gold. The library and Settler cost 420 and 440 respectively. So you'd need a total of 860 gold to do so, which means you have to generate 260 gold. in that same time frame. Sure, that is possible, but again, this Persian game is only on Emperor difficulty, so the AI doesn't make that much cash that early. Thus, you have to use that 260+ gold you generate to cover just buying either the library, or the settler. To even get enough for the purchase, you'd have to wait long enough to have built the library instead of purchasing it anyway.

 

[Celevin]

True, I forgot that it's Emperor. It's a bit weird that in some ways you're actually disadvantaged on lower difficulties.

Even without the settler (immediate) buy, I think my build order's pretty good for the start. You could buy the library, start saving for the settler, and start building one the moment the NC finishes. You'll still get both, just not immediately after hooking up resources.

 

[Civsassin]

After beating Emperor with most of the civs - I stayed at Emperor to play around with the civs and experiment - I am now playing more in immortal. Obviously, the AI is getting much better bonuses than it did in Emperor. I've started with a new strategy that ultimately ends in NC, but seems to so far be working well. The only problem I've been noticing is that the AI is out expanding me, but I am getting a strong capital.

- Build order depends on available terrain. If near or on a river and have available grasslands, I will go straight for a worker, so I can get some capital growth going.

- Research order is beeline to Writing

- I usually have some time left before I get to writing, so I will get a monument out to help with those first couple of SP.

- Tradition, Aristocracy, others within the tradition branch (this is mainly since I am forgoing expansion until the NC is up)

Once the monument is done, I head straight to the Great Library and tech Bronze Working. I take Iron Working as the free tech (unfortunately, the last time I tried this, I had a one turn mismatch and chose Philosophy instead).

As soon as the GL is completed I build a Library and the NC.

I haven't followed this strategy to complete a full game, since I have been doing it more experimentally.

I've been at or near the top of the tech leaderboard, but my lack of expansion seems to be problematic. I am going to try it with a follow up ICS to see how well it works in various maps and starts. Anyway, I think there is promise in the strategy. It just needs to be refined and proven.

 

[vexing]

i loaded up the save... and now i'm especially curious, did you restart from the beginning or use a save some amount turns in for each run?

 

[Tabarnak]

You can always take Liberty for 50% settler discount to get these cities very fast after NC. It's only one policy in the tree after all. Combine this with Tradition 50% excess food and you will pump settlers like mad. Use the huge overflow to put hammers in things you will need (worker,units,etc.). Then choose a settler again for next build.

On quick speed i have rarely the time to build a monument before a library and i need an extra warrior to protect settlers but on slower speed it's ok i guess. No bonus for worker production, so keep cash form sold luxuries to buy workers instead.

 

[Monthar]

i loaded up the save... and now i'm especially curious, did you restart from the beginning or use a save some amount turns in for each run?

Each run I started from the initial save. I only continued the first game of NC before expansion from its turn 50 save to expand it out to turn 109 since that's the turn I completed the NC on the 3 cities before NC strategy. That was just to get an accurate comparison on the same "final" turn of the test.