# SGOTM 15 - Kakumeika

Discussion in 'Civ4 - Succession GOTM' started by AlanH, Jan 15, 2012.

1. ### DanF5771Chieftain

Joined:
Feb 21, 2008
Messages:
1,194
Oh my, WastinTime was right all the time. Hatse's SOD would have teleported to the 'Liz Archer' tile.

I had posted the rules for teleportation back on page 36 (although with suboptimal formulation):

1. A unit determines the nearest home city from its current position - for the SOD this would be Akhetaten, the undiscovered city of Hatse 4S of Memphis. It has a plotDistance = 10 from the SOD's tile ... exactly the same distance as HatsTown, but Akhetaten comes earlier in Hatse's city list and therefore gets chosen over HatsTown.

2. The game calculates the jump-cost for all valid jump target tiles (they need to be revealed and accessible), all distances are calculated as plotDistances.

Jump-Cost = 2*Jump-Distance + Distance2NearestCity

Our Corn tile has a jump-distance = 3 and distance to Akhetaten is 11 -> jump-cost = 2*3 + 11 = 17.
The 'Liz Archer' tile has a jump-distance = 7 and distance to Akhetaten is 2 -> jump-cost = 2*7 + 2 = 16.

Even with Corn revealed the stack would not have jumped there.

Jump-Cost alternatives for HatsTown as nearest city:

Our Corn tile: jump-cost = 2*3 + 9 = 15.
2S of HatsTown: jump-cost = 2*8 + 2 = 18.

Funny, even though the Corn tile gets the 'global minimum' of jump-costs, the 'Liz Archer' tile wins because Akhetaten gets chosen over HatsTown as nearest home city.

2. ### KaitzillaLord Croissant

Joined:
Jun 21, 2008
Messages:
8,349
Gender:
Male
Location:
America!

Thanks for the explanation! Isn't the distance from the Liz archer tile to Akhetaten 2.5? It does not appear to be 2E of the Liz archer tile at first glance.

3. ### DanF5771Chieftain

Joined:
Feb 21, 2008
Messages:
1,194
The distances are integer plotDistances -> everything gets truncated down, no .5s.

4. ### KaitzillaLord Croissant

Joined:
Jun 21, 2008
Messages:
8,349
Gender:
Male
Location:
America!
Thanks for digging up the teleport mechanics. It is one of the most baffling mysteries in Civ 4 .
I forgot that you linked some of the formula in post #704.

I had no idea that city order mattered in a tie, much less how any of the rest worked. At first glance that SoD looked much closer to our corn than the Liz archer tile and it pretty much seemed that the stack would be moved there. Guess not

I will believe the next thing that WastinTime says no matter what it is so he stops feeling like Cassandra.

This explanation also explains the confusion over WastinTime's test results here.

There is no Liz archer tile in this test game. There is an ocean tile there instead. So the distance to the tile 1S1W of London is 2*9+1=19 as compared to 2*7+2=16 in the real game. That means the SoD should have teleported to corn in WastinTime's test game because the distance is 2*3+11=17 in both the test and real games. But it did not because Joao did not have vision of the corn, so the army teleported to the far east.

Mabraham fixed the test game by giving Joao vision of our corn tile and led us to conclude Hatty's army would teleport there. However, if we had given Joao vision of the corn AND added the Liz archer tile, then we would have come to the right answer regardless of Joao's knowledge of corn.

Also, after further playing with WastinTime's test game, it seems a secondary SoD attacked simulated BridgeTown 4 turns after the embargo. The main SoD would have attacked after 5 turns. The real BridgeTown in our game needed to survive 6 turns until T127 to get a boat out and the main SoD was 13 turns away. A secondary SoD was ??? turns away. The main SoD was assembled almost instantly on T115 if you count backwards from where it was T121 so a secondary SoD being quickly assembled and thrown at BridgeTown was probable. A war chariot rush from Giza would have been very bad triggering a war T122 and getting our Hindu Missionaries and settler killed. In the end I would still have voted for (Liberate) and beg for peace. Maybe I'd have felt different if it was 3-4 turns to get the boat out instead of 6.

5. ### mabrahamChieftain

Joined:
Aug 9, 2007
Messages:
4,027
Location:
Stockholm, Sweden
Correct - confirmed by adding land 2S2E of Texcoco. So WastinTime was right, but for what seems to be a wrong reason! At least we've learned some stuff. Maybe we came to the right strategy decision for the wrong reason, too. We'll see later.

Yeah, I'd have wanted to do this kind of testing before attempting to keep Gateway City, as I said somewhere earlier.

So, moving on, I posted a list of things I think are important to do in shulec's next few turns. What do others think?

6. ### bcoolChieftain

Joined:
Jan 25, 2008
Messages:
5,569
Location:
San Francisco, CA
agreed on workers and corn settling plan, but I don't think we will get those forests immediately. Needs testing. I think it makes sense to send the GG here and upgrade the trireme with him, could split the xp 10/10 if we wanted another good unit for defense.

sure
I think 3 spears will be enough.
we will have a trireme in this ocean for upgrading already. I think we can skip this trireme. I would build a worker, and set up a chop if we think we need another caravel in this ocean.
yes
north boat building site is good. Defend it. Send a spear that way? Too late?

AP resolution, yes Holy War on Liz (if availiable) I can't think of another useful resolution.

Other suggestions
Steal Optics!
I want to put espionage on Liz this turn enough (40?--overshoot to compensate for the value going up if Liz is using her espionage slider) so that we can see her research. I think it is highly likely she is researching Optics. We need to know this because it is a lot easier to steal optics! If she is researching it we probably want more espionage and possibly another late spy?

Island city

I think mabraham suggested this in another post, but we want a focused stack going into Hatse direction. I think the northern stack should just be a few units to defend the boat building site we are planning to settle up there.

Send a sword/spear towards Shaka if can get going soon (next turn or 2)
Perhaps send another worker Shaka's way to help that boat building site develop. The worker on the boat coming from Island.

Espionage

We want ideally in order of preference
Machinery (obvious)
Compass (obvious)
Optics (obvious)
Code of Laws for artist running options in boat building sites

I think we want another spy and significant espionage, so maybe sacrifice one more spear for the Hatse front and build wealth instead so we can get more espionage or a spy.

2 spears to defend probably is enough? If we combine the stack into one going in her direction?

7. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
Okay, tests done. I should not have overcomplicated the stuff like determining the date when Alexandria was founded...
Anyways, it really does not matter.

I hypothesized Alexandria was founded something like 20 turns prior to T93 when we met Hatshegawa. At that time, it founded two cities after Alexandria...
Between T122 and T93, there is a margin of ~29 culture turns coming from the sole culture building (Obelisk; no buddist religion). Thus for a probable of ~50 .

The artist took 5 turns to take control of the two forests. Culture control goes from 2%,11%, 18%, 40% and 50%.
But we can't have CoL for a while for possible Optics bulbing...so artist is a bit impossible for now.

Meanwhile, just growing with monument culture takes no more than 13 turns (2,6,9,11,13,15,16,17,18,19,33,42,48,53 %) to get control of the two forests. I tried some governor tweakings to see if it was possible to get some hammers in advance like getting culture pop can lead to redirect citizen in advance to hammer tiles or citizen specialist to gain an additional hammers (in hope to save one turn) before ending the turn; but nope it doesn't work as getting technology goes before hammer investment in builds...
So, if Alexandria gets around 50-60 , then it should takes 13 turns to get access to the forest.
At T13 of CornTown existence, forests are controlled and copper mined.
At T14, whip the boat, then starts to chop forests for more boats?

Sad thing, it is ~20 turns after peace treaty, she might DoW us again, right?

Artist solution doesn't work much...unfortunately because CoL should be avoided.

----

I verified, the copper next to Byblos to too close to Genghisgawa city, this impossible location for settling.

8. ### bcoolChieftain

Joined:
Jan 25, 2008
Messages:
5,569
Location:
San Francisco, CA
On t127 or T128 or whenever we think we have enough espionage we can

steal machinery
steal compass
bulb optics (or Steal it!)
steal CoL

So running an artist is definitely a possibility.

9. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
I doubt there are OCC as Lizgawa has only 7 cities, so cannot be the leader in lands.
Hatshegawa has presently ~142 tiles (approx. counted) and present leader has 161 tiles. I think it takes time for tiles to be counted in score, but in demographics, I don't think it is the case. Given a team, both lands are additional, then makes the team really a big chance to be leader in lands as much as other aspects I guess...unless I miss something.

Sorry if this was answered by someone else meanwhile. Trying to catch up in a fast way...

10. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
Okay, I just got it. I haven't viewed yet you were going to put espionage slider at max for next turns. Otherwise, I was confused: where is the necessary EP for all those stealings? Checked.

If I understood well, if we get a GProphpet, CS bulb (better capital, paper reduction)? GScientiest -> Optics (obvious reason), GE -> Machinery/Engineering (more movements) ? Read and acknowledged.
Okay, I see Optics bulb is impossible due to Theology enabling Paper. Oh. Optics is just before paper in GScientist preference. That is nice.

I'm still not sure about paper use; trade maps? Or something else? Because Toku, whatever the attitude is a skunk who never trades maps just like normal Elizabeth.
Okay I'm partially seen the purpose. Optics rocks better IMO.

Are you going to liberate LizTown and Trojan Horse while being at war (no diplo hit)? DanF's infinite while at war trick corollary... EDIT, I think so after more readings.

I guess Trojan Horses have more spies in order to capture the city asap within or next and avoid Toku getting it.

11. ### mabrahamChieftain

Joined:
Aug 9, 2007
Messages:
4,027
Location:
Stockholm, Sweden
OK thanks to Tachy's testing, we won't get these forests immediately. Shucks. We can't pre-chop while they're in Hatse's culture. So perhaps one of Eiffel or Hoover can road towards the city S of Memphis (which we may want to capture later) while the other roads Satsuma towards the corn, so they can both converge on the corn city around T130 in time to road the forest hill and be ready for chops after possible artistry. These workers will likely not survive the mission when Hatse DOWs us some time after T131.

With two chops and a whip, we could maybe squeeze out two triremes depending on timing details next turn set, so working coast is probably better than working a citizen in the corn city.

OK. Hatse will have lots of war chariots but I guess at least one of the spears will survive a defensive combat unscathed and we have a sword and a few cats in extremis.

ok

Yes.

Yes. We expect at least 35 turns of game, which I think will pay off in ability to grow on lighthouse-coast and lighthouse-lake.

Yeah. We have no evidence that we need a northern offensive army. Sending two stacks was a nice idea, but I think it has expired. We do have evidence that Hatse will DOW us and we want to get a strong punch somewhere to set up a possible capture of a coastal city.

The wheat city does not need an MP. The woodsman axe in Delhi should be sent to a front and a 0-1XP axe used as the MP.

We should put a forward unit 1NE of Kyoto and another on the copper 3S of Satsuma to get good vision of Toku's culture so we know where Hatse has units as we approach T131.

OK.

With the units from the western stack, 2 spears is probably tolerable initially. I'd like to follow up with 1-2 more, though, as well as 1-2 crossbows.

Toku has some units wandering in our culture. That will annoy us when we gift cities back to Liz. We might want to bribe him to peace with Liz with the first tech we steal.

12. ### bcoolChieftain

Joined:
Jan 25, 2008
Messages:
5,569
Location:
San Francisco, CA
I think we have to ceasefire and gift the cities to Liz. (I'm not sure but they are not close enough to liberate so I don't think DanF's infinite liberation trick is what we are using).

Same turn we gift cities, we steal from her, then DoW again. Diplo hits are possible from some AI but not Hammuragawa.

13. ### mabrahamChieftain

Joined:
Aug 9, 2007
Messages:
4,027
Location:
Stockholm, Sweden
Oh, how does this work? I've forgotten... I thought our plan was to cease fire, gift and re-DOW immediately. Edit cross-post with bc: seems since liberation is not possible, this is not a valid approach

It just had several there already when we committed (late) to more spies and it was inconvenient to move them and get fortified fast.

14. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
@Post 2731

Oh and I forgot to add: I think no library will come up out of the blue because of the Obelisk.
I don't know when the two cottages were laid down, but supposing built quite a time, the 5 pop Alexandria is most probably working clam and rice, while hiring three specialists.
Perhaps the AI is coded to avoid whipping when set up for GPeople generation...but AI is stoopid and have a quantum mind, so I can't be sure.

Indeed, we have to hope no library and popping, otherwise we lose the corn location.

15. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
Ah yes! I forgot those cities are much much closer to our capital than Lizagawa ones. I suspected you were about to make ceasefire while reading deeper in past posts.

Indeed, not a valid approach, but would have been nice.
I just remember DanF talking about some buttom to bring up the option to liberate some cities with civ while at war with said one, but in fact, we never had precise modus operandi.

Probably the only fact I knew before catching up was Hammuragawa is our voter once he gets +10 modifier according to your tests. I think it is T136? second diplo modifier for resource for a total of +9, then share wars ends close the goals. Indeed, he might ask techs, so it can be easier.

But as regards AP rival, I'm still unsure. I've heard of Toku of Japan. If Lizagawa could be infected in its principal cities later (the pig tile near old Gateway City seems to connect to she-Toku), then is it possible to make it our rival while putting everyone at war, should do the work...

Now, Toku is well-liked and share wars make me nervous.

16. ### mabrahamChieftain

Joined:
Aug 9, 2007
Messages:
4,027
Location:
Stockholm, Sweden
How should we plan to cope with the possibility that Shaka might be planning to DOW us? I doubt we'll have enough espionage spending to see Hammu's city vision. Build and run a scout back and forth looking for trouble?

17. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
Indeed he is gunning for us. Oh well...

Putting a unit on the hilly forest 2N2W of GhostTown?Idiotic fact.

I don't know if he will try to find another path to get at us while being under war prep...

If I understood right, a AP vote for holy war doesn't stop the AI to make their own private wars?
Is this different than bribing someone on his tail, then he forget the war prep against us?

18. ### bcoolChieftain

Joined:
Jan 25, 2008
Messages:
5,569
Location:
San Francisco, CA
yes an AP vote for holy war doesn't stop an AI from stopping an AI from continuing their own war preparations. I'm uncertain if an AI already at war will initiate a new war target while at war with someone else due to an AP vote. I don't know if there is a way to tell when an AI enters war prep while already at war with someone else.

Bribing someone else to declare war is different I believe than the AP war vote. Although I don't know the exact war mechanics of bribing someone into a war.

19. ### WastinTimeChieftainSupporter

Joined:
Jan 27, 2006
Messages:
15,221
Gender:
Male
Location:
California
There is a window of time where, if we closed Toku-Hats borders, Hatty would have no access to our cities. Would that make her drop plans to attack us? and go after toku? or someone else? Then maybe when we found the corn/coastal city, it won't be a target?

(yes, I'm suggesting the embargo...again

20. ### Manco CapacFriday,13 June,I Collapse

Joined:
Mar 1, 2010
Messages:
8,051
^And CornTown, isn't this a city Hatshegawa might accede quite easily? The boat isn't built enough fast to avoid possible war dec after the 10 turns of cool down. I suppose, like I saw in the test game, her SoD got back. The thing is does s(he) retarget or is there fixed target that cannot be changed?

Wow, did I write that? Surely, I wasn't really bright at that time...haha.

Anyways, in three turns pig city and Delhi will have two happiness. Can we trade it away a gem or gold for faster diplo with Genghisgawa or is it a big no-no?
Sorry for asking this again, but where can I find the rules about diplo thanks to resources?