[PB60] Amicalola and Vanrober: Gone' Fishing!

[Amicalola]

Nicely done, glad you were able to get GLH at a decent date without it slowing down your expansion too much. I’m curious how tech rates are across the board, this being the first game played at this difficulty in a while. Which classical techs will you be prioritizing?

Thanks! We're very glad too. 

PB59 is played on Emperor.  But the map is so relatively lush that it's probably a very different beast. This will likely be slower in that regard. Broadly speaking, we're thinking of pushing for Currency ASAP to get our PRO trade routes working overtime. But with marble in our borders and one of the better economies (probably), we might hit up Calendar first for a certain mausoleum... 

If we finish Moai in Tuna and Literature is un-researched by everyone, we might try for Great Library with that. But I think that's less likely.

Edit: Oh, Iron Working is the other big priority. Gotta start settling those jungles...

[Amicalola]

T74


Superdeath killed our "Superdeath watching' scout with a war chariot. There's only two explanations for this, neither of which I like.

1. He's planning to attack us with an army that our scout would have seen.

2. He's a dick. 

It's hard to tell if it's the former. He has a small but not huge spike on the power graph. I've retasked our warrior to replace our scout for now, so we can keep an eye out on the north. If it's just the latter, I'm sure we can figure out some retaliation soon enough. 



We also got a natural spread of Judaism!! To... the capital. This does raise the possibility of building a temple in Salmon instead of Tuna for our prophet. It would mean overflowing our anarchy-settler into a (OrgRel-boosted!) temple, and then running a priest for 30 turns. What would you think of that, Vanrober? It would slow down our expansion, but make our prophet risk-free.

[Amicalola]

T75
Lazteuq settled a humbling ninth city this turn!  He's got the lowest GNP in the game (including Stonehenge), so his research must be hurting, but still. As vanrober said, he's going to be a monster in 20-30 turns. 

In the spirit of being humbled, here's our demos - wouldn't want to give you the wrong impression that we're doing super well, or anything like that.  

[Amicalola]

T76-77

We founded our first island city, called Betta. It's behind ol' Captain Ahab, who refused to move for my screenshot. This city is named in honour of my own fish, that I got right at the start of this game.  



Our island city bumps our commerce up by 17. I was planning on heading straight Writing->Maths->Currency from here, but it turns out that with that extra commerce, we can finish up Monarchy on the exact same turn we planned to convert civics (OrgRel). I can't resist that symmetry, so we'll be getting Monarchy to double-convert on T81.

Bettas

are also known as 'Siamese Fighting Fish.' They're very pretty. And angry:

[Amicalola]

T78
There's another island!


I think we want that reasonably soon to make sure Lazteuq doesn't get it. 

Mjmd built the Oracle last turn, and predictably took Metal Casting. The most interesting thing about this is that someone else already has it - probably Jowy?

Here's our demos, for the curious. We're GNP cheating (20% on Monarchy), and about to whip 3 cities, so this is our zenith for a while. The game is really close - everyone is pretty similar in most categories, and we've been bouncing sporadically between good and terrible demos. I guess that's a function of the mirrored map? 

[Amicalola]

T79-81

We converted to Organised Religion and Hereditary Rule this turn, and will have another turn of anarchy for Judaism next turn. We were 3 gold away from not being able to finish Monarchy - at 10% less on the final turn, it wouldn't have completed. Whew!



We haven't had a spread of Judaism since our capital spread on the turn after founding. I'm not sure if we're unlucky now, or were lucky with that first spread.  

In world news, Lazteuq/civac got their Stonehenge prophet, and Jowy has rapidly gone from 4->7 cities. So he's definitely still looking good.

[Amicalola]

T81 Rambling - Serfdom vs. Slavery.

TL;DR - Serfdom is not worth it (non-SPI) until Nationhood and Replaceable Parts, and even then on this map it's optional. We will not be converting to it in the near future.

Since we got Monarchy, I thought I'd discuss Serfdom vs. Slavery. This is a rare CtH feature I feel qualified to talk about - I've probably played with Serfdom more than anyone else in multiplayer. To be clear, I think Serfdom is strong, especially in a vacuum. By the numbers, Serfdom is completely insane; the AI also recognises this, and in singleplayer will convert to Serfdom ASAP, keeping it for the entire game. But for humans, who can do whip micro and build fast armies, Serfdom doesn't just need to be good - it needs to be better than Slavery, which is a pretty high bar. 

The unintuitive thing I've learned about Serfdom is that even though it primarily buffs hammers, it is only good in low-food scenarios, rather than low-hammer. In a high-food map, those windmills and farms are contributing less to your cities, relatively speaking; Slavery also becomes better, as you can whip more frequently. This is doubly true for fishing villages, which have no natural production. But where Serfdom shines is on a low-food map, where you can't afford to work mines or whip constantly. There, those tiles really add up, and while fishing villages become weak, your primary cities make up the difference. 


Late game, with buffed improvements, those primary cities get truly absurd. 

On any map, Serfdom gets a lot better as the game goes on. Whipping gets less efficient as items get more expensive, higher happiness lets cities grow taller, and the Serfdom improvements get upgrades (e.g. Replaceable Parts). The drawbacks of Serfdom are also a lot more critical before drafting, as a player is relatively defenseless until then. The obvious problem then becomes "all my rivers are already cottaged," which is legitimate. A watermill is better than a cottage, but is it better than a village? A town? I'd argue no. 


We were never willing to bulldoze towns for watermills, even post-Replaceable Parts + Electricity.

So the more common benefit of Serfdom is the windmills and farms. After Replaceable parts, Serfdom windmills are literally mines with +1 food and commerce; on paper that's absurd. But the problem remains - how am I going to build an army quickly, when necessary? I think it's important to note that Tarkeel and I in PB58 were not a good representation of Serfdom. We had the statue of Liberty, Angkor Wat, the Apostolic Palace, and IND/ORG forges/factories everywhere. Our research was also carried by GLH, Colossus, Monk Wonders and SoL+Representation, rather than actual tiles; it was truly the best-case scenario. Of course, Serfdom is still nearly always better in the hyper-late-game (and the reverse is true early on). The optimal switching time depends more on the map, and strategy of the player. 


So is this map good for Serfdom? At first glance yes, with all the hills, plains and desert. But there's a lot more food here than PB58, deceptively - the ocean fish are a huge deal, and there's no mountain ranges. There also aren't many river-flatland tiles, and some have resources - what's left will be long-cottaged by Replaceable Parts. The map also has plenty of hammer-poor fishing villages and small islands, which will be near-useless for production in Serfdom. 


This map is actually quite lush, and will allow for many fishing villages. 

To conclude, I believe this map is terrible for Serfdom. I can't imagine wanting it before Replaceable Parts and Nationhood, and in reality we might not want it until even later. Other than Spiritual Lazteuq/Civac, I think it would be a huge mistake for any teams to prioritise early Serfdom here, and I'm really curious to see if that plays out in the game. Maybe I'm totally wrong!

[Charriu]

I've looked into the AI civic switch code recently and I can say that it is not a good evaluation. For example one reason why the AI switches to serfdom immediately is the added yields from windmill and watermill. The problem is that even if the AI can't build mills for a long time - like if serfdom would be available at BW - it still counts those not buildable improvements as a plus and switches because of them.

[Charriu]

For example fo the following from standard BtS:

Put serfdom on BW
Remove everything from serfdom
Add only that it gives +90 to every yield on forest preserve
Start an ancient game
See how the AI all go serfdom even though the bonus is thousand of years away

[Charriu]

The AI also overbuilds on farms.