*Spoiler5* Gotm21-Special Topic:Must have Submitted

[Dianthus]

Open 1.21f (Euro)

I've just performed a little wonder investigation. Here's a table of all of the wonder build dates (those with a green background were owned by the Greeks at the end of the game ) :

I didn't even try for the early wonders, instead concentrating on expansion. The 1st Wonder I went for (and got) was Sun Tzu's, so the early wonders were un-influenced by human meddling.

I had a little look at some autosaves from just prior to the building of The Oracle and The Pyramids. At 1200BC (a little before the Oracle was built in 1075BC) I didn't know that many Civs, but investigated the following :

It's interesting that the Egyptians/Minoans/Hittites hadn't started building any wonders (they had the techs). Is there a number of cities they have to reach first, maybe (6)?

The following are the results of investigating known Civs at 710BC (a little before the Pyramids were built in 650BC) :

There is a lot more wonder building going on at this point, with every Civ (except Greece ) building 1 or more wonders. I also had a little look at the terrain for the quickly expanding Zulus/Ottomans (with 12 cities each), and they both had lots of available grassland around their start position. The Hittites (with just 7 cities) had quite a lot of forest to clear, and were also pressed in by France/Egypt. I'm still not sure which came 1st, did the AI's stop expanding because they focussed so much on Wonder building, or did they run out of room for expansion, so start building wonders? They obviously aren't investigating competing Civ's Wonder cities, looking at the number that are building the Pyramids above!

 

[mad-bax]

Dianthus, this is great, as is the point you are making. The AI appear to start building wonders even when they have zero chance of completing them. The Hittites for instance trying to build the Collossus when the Zulu are 6 turns from completing it.

Is it true then that the AI doesn't investigate each others cities? The Hittites would not have even had to perform an investigation as the Zulus were building in their capital, they just needed to establish an embassy whilst the zulus were building the Collossus. Is this just another example of the stupid AI, or didn't the Hittites have an embassy with the Zulu?

 

[Dianthus]

Originally posted by mad-bax
Is it true then that the AI doesn't investigate each others cities?

I'm just surmising that from the data. Let me put it this way, if they do investigate, then they don't use the information gained!

 

[DaveMcW]

Open PTW 1.21

1375BC - Colossus - Greece (Athens)
875BC - Oracle - France (Paris)
610BC - Pyramids - Monoans (Knossos)
570BC - Great Library - Greece (Athens)
490BC - Great Wall - France (Paris)
390BC - Great Lighthouse - Germany (Hamburg)
380AD - Hanging Gardens - Spain (Madrid)


While this list of completed wonders is interesting, I think there are too many variables to draw any meaningful conclusions.

A better way would be to list all the dates when the AIs start building wonders. You could then compare empire size and city productivity to determine when the AI governor decides to start a wonder.

But then you still have to worry about random build decisions and civ-specific governor preferences.

 

[LKendter]

In addition, there is a chance of an AI rushed wonder.
I know for a fact that it happens - I lost a battle - saw a leader generated. The next turn a nearby city completed a wonder. I don't think it was pure luck that it happened that way.

 

[DaveShack]

Originally posted by LKendter
In addition, there is a chance of an AI rushed wonder.
I know for a fact that it happens - I lost a battle - saw a leader generated. The next turn a nearby city completed a wonder. I don't think it was pure luck that it happened that way.

Indeed yes, I've seen this happen many times. I had an Ottoman leader returning home from a battle with someone else walk right through a heavily guarded area, and totally surrounded it with units to make sure it didn't get home before my wonder build finished. IIRC we were allies at the time and I didn't want to just destroy it...

 

[Vlad Dracula]

From what i remember it wound up like this.Berlin oracle Sogut gl Athens lighhouse only one left Colossus persia forgot city have to check a save for the others

 

[SewerStarFish]

PTW 1.21 Open:
Pyramids:Ottomans-Sogut-630BC
Colossusersia-Pasargadae-950BC
GLH:Minoa-Knossus-310BC
GLibrary:Russia-Minsk-110AD
Wall:Greece-Sparta-130AD
SunTzu:Hittites-Hattusas-650AD
Oracle:Germany-Berlin-950BC
Hanging Gardens-Ger-Berlin-130AD

Rome was gone early circa 1000BC, Carthage seemed to be slow to initiate GW construction usually in Utica which seemed okay for shields.

 

[Ambiorix]

Lots of examples, but I don't see the discussion going anywhere :
Does the AI go after a wonder that can trigger its GA ?
Can we actually influence the AI's choice of which wonder to build, or just support them in whatever they happen to be building ? And is the latter a valid strategy (gifting workers, luxuries, ...) ?

For the upcoming gotm it would be nice to see some reports of which civ started on which wonder, and when, and what factors may influence that decision (relative strength ? city position ? improved tiles ? luxuries ? ...).

 

[Jove]

Just to add data, in my game the Ottomans built the pyramids in Sogut and later the great wall in some other city. The great wall triggered their golden age.
I tend to buy workers willy-nilly from anybody in the ancient and middle ages. Perhaps the Ottomans were the best suited to withstand this negative worker pressure and this is the reason they got both of those useful wonders.

 

[AlanH]

I copy each autosave automagically as it is created, and so I've built some embassies and done city investigations. I can only look at the civs I knew, of course, so I can't comment on what any of the southern civs did during the early game. They were way behind the north in techs when we met up.

In my game the early wonders were built as follows:

650 BC Pyramids - France, Paris
630 BC Colossus - Russia, Minsk
630 BC Oracle - Keltoi, Entremont
550 BC Great Library - Ottomans, Sogut
150 BC Great Wall - France, Marseilles
150 BC Great Lighthouse - Minoans, Knossos
150 BC Hanging Gardens - Keltoi, Entremont

The first three builds were independently started and were not cascaded. Paris and Berlin started the Pyramids race in 1725 BC. Both were building settlers when I first met them in 2140 BC , and these both completed in 1950 BC when the cities grew to pop 3. They then prioritised spears, both completing them in 1725 BC before starting the Pyramids. Paris' early rate was 4 spt with Berlin at 5 spt, but Paris increased its rate to 10 spt by the end. Berlin had fewer tile improvements and only reached 8 spt, so Paris won the race. I didn't buy any workers from Germany that would have slowed him down, so maybe he just had some bad luck with barbs.

By the end of the Great Pyramids Race there were five known contenders. I had decided not to bother, and Keltoi were not competing. It's interesting that the Keltoi went for the Oracle and the Colossus, both playing to their religious trait. Brennus had five cities when I met him in 1450 BC, and I'm gessing that he didn't have enough cities to warrant dedicating a third one to a Wonder. The finishing order was:

Paris - the winner
Sogut and Berlin 6 turns to go
Moscow 22 turns to go
Zakros (Minoan) 85 turns to go

Sogut, Berlin and Moscow switched to the Great Library when they were beaten to the Pyramids by Paris, and Sogut completed it in 550 BC. Marseilles and Knossos then switched from the Library to the Lighthouse, displacing Gortys which had been building it. The other known Wonder builders fell off the cascade at this point because none of them were coastal cities. So in 510 BC there were only two cities building the Lighthouse. Knossos won that one because it had been going longer than Marseilles.

Entremont won the Hanging Gardens competition in 150 BC leaving 6 other cities eating dust. The Keltoi were at war with the Ottomans in an alliance with me at the time, and they captured Sinop that turn, so I can only assume they won a Great Leader to complete the Gardens, as they still needed 18 more turns at 11 spt. In the same turn Knossos seems to have done some intelligent micro-management as their rate was 3 spt and they still needed 7 or 8 shields to complete the Great Lighthouse. However, there was enough shield production in the hills and mountains to complete it in one turn if they turned off food temporarily. That cascaded Marseilles to the Great Wall, completing the ancient wonder set.

The only mystery for me is why did Moscow and Zakros even bother to compete for the Pyramids. I guess every AI civ has a ration of Wonder cities and will try to use it no matter how hopeless the cause.

On a final gloat-note, after the Ancient Age Germany and Russia got to build just one more GW each and my often-gracious buddy Hannibaline built and kept the Sistine Chapel and Newton's Uni. So I finished up with 23 of the 25 Great Wonders.