(May 18th, 2016, 22:42)El Grillo Wrote: Ah, of course. Sometimes it's boggling how much information you can get out of what the game gives you. I wonder if the developers ever intended detailed C&D, KTB checking before Alphabet, production tracking through Espionage, zooming into fogged cities to check buildings, etc.
T57:
Here's the timing on the attempted Currency bulb plan. I finish Math in 7t at 50%, and my first GP will be born from Onramp in 7t. There are currently 9 Artist points in the pool, so the final odds will be 13% for an Great Artist and 87% for a Great Merchant. If I hadn't messed up the specialist assignments last turn, it would've been a 91% Great Merchant, but it is what it is.
Axes 2 and 3 are in production, with Tellefsen Hall building Impi for the new cities and to send north. Culturally, Khan's borders will pop next turn, which means he should keep the Sheep for now, but my plan is to keep Marchup Cadence at size 1 working an Artist to pop 3rd-ring borders in 9t, while its 2 Workers head east to help improve tiles for backline cities. Part of this is really needing the Worker labor elsewhere, and part of it is worst-case-scenario planning for a Khan invasion. I don't yet have units that can attack into Skirmishers well, so if I improved tiles only for him to pillage them on T63, I'd feel pretty silly. Perhaps I'm going overboard, but Khan's power is still rising, and I think the biggest mistake I could make at this point would be underestimating his forces and losing a swathe of land to the northwest to him. I'd be curious to know how he's thinking about all of this right now.
To the south, I have my settling party on its way to the Stone island. I'm assuming that Ginger has Sailing, but I haven't seen any boats from him just yet. One Impi is perhaps vulnerable to a 2-Axe landing, but I'm getting a friendly read from Ginger, and if an attack comes, I can keep my Galley in position to rush over a defender from Silent Walk. I went back and forth on growing vs. popping borders at Tossing&Turning, but decided to pop borders so that my Worker can Pasture the Cows, then add some more improvements at Initial 64.
Demos still reflect regrowing. I'm curious about the Population values, and I'm going to look into doing some calculations to convert from Demo-screen Population to points.
Graphs:
I get Adrien's graphs in 3t.
Population is done on city size - so a size 10 city has a much bigger pop than 10 size 1 cities. However for the score it is only city size that is taken into account. You can see what poputaltion the game thinks a city is by hovering the mouse over the city name in the city view window.
The one thing I think is a bit too much is seeing extra buildings in cities you have had visibility on. That has always seemed like a little extreme to me.
I couldn't sleep, so I used Solver in Excel to check the real population distribution. I found out that the rival worst is not Ginger, but Khan! In hindsight, this seems obvious, that Ginger has fewer, larger cities, and Khan's newer cities would be size 1. But that's why we run the numbers
Note that the City # doesn't correspond to the actual order of founding. I also had to try a couple of numbers to pin down the rounding in Ginger's population.
My numbers, for contrast. I have a real pop of 21, so I'm indeed in last place. Adrien's far in front, but I think I can catch up to him.
As an aside, I find it rather silly that the convoluted long-form population that BTS uses enables this kind of calculation. In my mod's straight-up demographics, you'd get the total pop right away, but have no way to determine how it was distributed.
My friend and I took stock of our finances and decided to accelerate the Math/Currency plan. We put this into motion by both increasing our tech spending and hiring two more Merchants at Onramp. All else equal, we'd prefer 3t of Currency to 3t of an Axe in play, even considering that we're leaving some 6-yield tiles unworked to do so.
In the north, Khan's outpost popped borders, and will reclaim the contested Sheep. Meanwhile, we've pulled the Workers as planned, leaving a lonely Artist to paint the barren landscape for a few hundred years.
In the south, dropping my units on the island revealed a Ginger city heading toward the middle of the map. I kept the Galley in a defensive posture, so hopefully he doesn't get too touch about it. For the same reason, I'm content to leave Open Borders signed without actively scouting his land, I'd hate to lose his trade routes.
Demographics don't look too bad, considering how approximately a quarter of my people are specialists.
Graphs:
Populations:
I'm now in 3rd place with 23 pop. Now that I have the tool set up, it takes no time at all to keep tabs on the others' population distributions. Now, better players would actively find ways to use the information, I'll just keep an eye on trends, whipping rates, and our relative pop counts.
Adrien: 5->6
Ginger: 6->7
Khan: 4->5, 5->6, 6->7 (data doesn't fit, but alternatives make even less sense)
Mechanics question - is it possible for a city to grow two sizes in one turn on Quick speed, given a large enough food surplus? I ask because my model indicates that that happened in one of Khan's cities, so now I need to figure out whether this unexpected result demonstrates the model's power or its flaws, or if I'm giving it the wrong inputs somewhere :D
EDIT: How about backing into it? Khan's score increased by 33 points between turns, and we know each pop point is worth 7 points. He lost some tiles to culture, but that shouldn't be updated yet. If he grew 4 pop for 28 points, I have no idea where the extra 5 would come from. So let's say he grew 3 pop for 21 points, and gained a tech for 12 points.
Given that I have visibility on 3 cities of his with combined pop of 14, and assuming that last turns' numbers were correct, Khan's remaining cities have 13 pop points divided among them. However, there's no configuration that fits the data: the model says he has 16 pop among those cities. Something's clearly wrong here, but this has suddenly taken way longer than I expected :D
EDIT 2: And just as I said that, I figured I'd try setting the objective to the number of pop, not the longform population. Still doesn't work, but it got me thinking... I'm not ready to throw out all of the results just yet, but I'll keep this tucked away and not rely on it.
On T57, Khan's visible cities were size 6, 4, and 1. On T58, all had grown, to size 7, 5, and 2. If the model's results from T57 were correct, and Khan is the 562,000 player*, then Khan had to have additional cities grow. If we assume the 33-point increase in score was all from pop, and he grew 5 pop, for a total of 29 pop, then no configurations fit the data. There do exist configurations for a total pop of 30 and 31, but that would mean that Khan lost score somehow AND that he had cities growing 2 sizes in one turn. My tests haven't shown any way for these to be true, so we're still stuck. Ideally, I get the save tonight/tomorrow morning, run the numbers, and everything checks out. Then, I could go back to older saves and try them as well. If this turn just happened to be some kind of bizarre fluke, it picked an awfully inconvenient time to show up!
*This is especially frustrating, because Adrien and Ginger's numbers make perfect sense, leaving Khan as the odd one out.
(May 20th, 2016, 01:20)El Grillo Wrote: On T57, Khan's visible cities were size 6, 4, and 1. On T58, all had grown, to size 7, 5, and 2. If the model's results from T57 were correct, and Khan is the 562,000 player*, then Khan had to have additional cities grow. If we assume the 33-point increase in score was all from pop, and he grew 5 pop, for a total of 29 pop, then no configurations fit the data. There do exist configurations for a total pop of 30 and 31, but that would mean that Khan lost score somehow AND that he had cities growing 2 sizes in one turn. My tests haven't shown any way for these to be true, so we're still stuck. Ideally, I get the save tonight/tomorrow morning, run the numbers, and everything checks out. Then, I could go back to older saves and try them as well. If this turn just happened to be some kind of bizarre fluke, it picked an awfully inconvenient time to show up!
*This is especially frustrating, because Adrien and Ginger's numbers make perfect sense, leaving Khan as the odd one out.
Popultation doesn't have an exact cost. This always bugs me but take the pitbosses where there are clear indications of gaining 1 pop per settles etc. They often flit between 2 numbers due to rounding.
That's a bit of a bother. In any case, the takeaway here is that I need to get my new cities growing and contributing soon, and that may mean taking some time off military to fit in more Worker builds in my production cities. Ideally, with Math coming in, I can try to chop in Granaries as well, then swap to Slavery for Courthouses at size 4.
Another idea would be to take advantage of my Stone monopoly, along with OR and Math, to attempt to build the Pyramids. It would fit nicely perfectly with my SPI specialists, and it might let me get away with my low-Worker setup a bit longer, but it might also put a huge target on my back. No save today, so all I can do is muse, but I don't want to just cruise into second place here
In the north, the Workers have pulled back from the frontline and are going to be busy improving tiles at Sproul Steps. The Settler I have exploring will head over to the Red site, which will be a real powerhouse, growing straight to size 6 on resources alone. We'll eventually get to the Fur/Gold site as well.
In the south, we settled The Titanic on the Stone island. Haven't decided yet on whether to whip in a Courthouse or Lighthouse first in. After a brief discussion, we also began work/chops for the Pyramids at Tellefsen Hall, with a cool +125% modifier. The Wonder costs 150 base hammers, TH can work 12 base hpt at size 5, and I have 4 Forest chops lined up over the next 7 turns, which will finish it at ~158/150. No one else has Stone hooked, so I doubt anyone is racing me for it. I'd love to get my hands on it, of course, but the failgold rate is also pretty reasonable, and I only commit the second pair of chops on the last turn. In the middle, my Scout sees a rather confusing absence of units. Khan has a lot more units than what he's showing me, so I have to assume there are Skirmishers either headed my way, headed to attack Adrien, or held in reserve. The good news is that I can redeploy my Impi from Marchup Cadence to Onramp very quickly, but they'll be fighting at a disadvantage without any defensive modifiers. Combat 1 Axes will get odds on flat ground, but the hammer exchange would be pretty poor. Perhaps a quick detour through Archery and HBR is in order?
I forgot to take a screenshot of the Demos, but there's nothing exciting to report there.
At EOT, we get Adrien's graphs, confirming that he is indeed ahead in the core demographics. I'd currently put Adrien in the lead, us in second if we play tight, and Khan and Ginger trailing a bit, but definitely threatening.
Maybe it's the weekend or the tracker or pure coincidence, but we actually got another turn today!
T60:
Khan reinforced his border with a third Skirmisher, but there's still no sign of any units in the middle of the map. I parked my Scout on a hill that gets visibility on any units coming my way.
Here's my Pyramids situation at Tellefsen Hall. It's a great coincidence that those 5 tiles yield a multiple of 4 hpt, and Stuntsheet One is borrowing the pair of Sheep to grow rapidly. Elsewhere, I've adjusted tech and specialists to finish out the Math/GP plan.
If all goes to plan, I'll be the first to Currency by a reasonable margin, and then I can fire my specialists and focus on growing and developing.
After EOT, my GP is born, and it is indeed a Merchant at 87%. I also got a free Confucianism spread and noticed that Khan had declared war on Adrien - that certainly would explain where his units are! Needless to say, I'd love for Khan to hurt Adrien here.
Three cheers for turn pace! Got the save early enough to play it before heading out. And what a turn it was.
T61:
As mentioned earlier, we bulbed Currency this turn and immediately fired all of our specialists but the Artist in Marchup Cadence, then dialed down tech to finish Archery. Time to see if this was all worth it.
I checked the others' treasuries:
Adrien: 43g + 27gpt
Khan: 60g + 3gpt
Ginger: 95g + 24gpt
We also learned that Khan had razed Adrien's city in the center, and on top of that, he has a Settler ready to settle on top of the Marble. Bold move, Khan - luckily for him, my Scout is not an Impi, and I wouldn't mind staying out of this conflict altogether for now. In the near future, with a Great Artist and a stack of 2-movers, though... anything is possible.
At EOT, Buddhism (Adrien's religion) spread to The Titanic by way of OB with Ginger. We haven't reached 10t so I can't renegotiate the agreement with him, but he's been a good friend so far, and we will consider letting him keep the deal as is as a sign of goodwill. In any case, that means The Titanic can build a missionary or two to add another Monastery to our capital.
My friend played the turn this afternoon, and it seems like we were an hour or two short of a 3-turn day! First off, Adrien offered OB, which we rejected. We also have some pressing concerns in the north:
Khan has a Galley in the sea between us, and it's presumably just dropped off a Settler and a Skirmisher. We wanted this spot for the amount of threat it projects and its ease of defense, in addition to providing another ICTR. However, we had originally planned to settle it 6 turns from now, using a Fort to make use of our first Galley, and Khan has beaten us to the punch. Or has he?
We redirected a chop at Sproul Steps into a Galley and swapped into Slavery nationwide, setting up some whips and not costing us any specialists except for the Artist at Marchup Cadence. We're going to get some Work Boats and a Settler now, then some Courthouses before the next swap opportunity, when we might have the Pyramids to enable Representation/Caste.
Next turn, we're going to whip the Galley in Sproul Steps and load a 1-promo Axe and a 1-promo Impi into it. If the first lands two or more hits, then the second gets odds. The plan is to raze the city immediately before Khan can reinforce it, at which point capturing it becomes almost impossible, and then keep the Axe fortified there to prevent a repeat landing while we bring our own Settler into position. Then, if we can get 3 units into the city, the math favors us and it'll become almost impossible for Khan to take.
This plan is admittedly a bit at odds with our previously ongoing plan to consolidate and grow while the others contest the center, but I have reason to believe that Khan has already committed the bulk of his forces there - Adrien's no slouch in power, and he lost a city two turns after it was founded. I can't imagine that Khan will feel secure in his position there against two players who have units nearby and would love that city for themselves, so he'll have to position most of his reinforcements there and not north. Furthermore, since we don't have a city in the center, Khan can't realistically retaliate, except at the heavily fortified Marchup Cadence. My friend also mentioned that Khan has no Gems hooked up, which means that Adrien settled their midpoint, so that's another front he has to watch. It's also worth noting that this raze would disconnect and deny his Stone, if he's also planning on building the Pyramids.
In summary, we find unacceptable the threat presented by a Khan naval base on the inner sea, we're going to nip the problem in the bud, and we believe that Khan will find it impractical to retaliate
![[Image: 2mYrCt9.png?1]](http://i.imgur.com/2mYrCt9.png?1)
