SGOTM 13 - One Short Straw

[shyuhe]

I'm surprised at how far off that is from my estimate... I wonder if there is something substantially different in the test game? In any event, I think it's worth a shot to gift alpha to Ragnar, unless mdy is 100% confident that we can parry Ragnar's galleys with our limited resources.

x-post with Dhoom: Yes I used epic speed beaker amounts.

 

[Dhoomstriker]

The next steps
...
2. Next, if he was Cautious, then we want to confirm or deny that "the lowest Cautious value +5 = Pleased."

I have just confirmed this fact in a test game, i.e. that "the lowest Cautious value +5 = Pleased."

So, now we really would like to know... was Ragnar Cautious with us at a "0" total Attitude? If yes, then we're in the money with being able to gift Alphabet + increments of 10 Gold being able to get him up to Pleased status.


EDIT: Hopefully, we have manual saved games of every turn (I have them for every turn that I played), since we're not supposed to load an Autosaved game.

 

[Dhoomstriker]

I'm surprised at how far off that is from my estimate... I wonder if there is something substantially different in the test game? In any event, I think it's worth a shot to gift alpha to Ragnar, unless mdy is 100% confident that we can parry Ragnar's galleys with our limited resources.

x-post with Dhoom: Yes I used epic speed beaker amounts.

Well, it looks like from the SGOTM 12 thread that a lot of factors are involved... game speed, difficulty level, map size, known # of Civs... presumably the Civs that the AI knows, the length of time that we've known the AI, and more.

The test game that I used matches everything as closely as possible, with only the "map size" possibly being wrong... but it says "map size" and not "map type," so actually I should be fine (I used a Big and Small Cylindrical Standard-sized map, but it's actually probably only the "Standard" part that matters).

With such a complicated formula, a lot of things could be easy to miss or miscalculate, so I would tend to trust the test game a lot more than any formulaic value that we could come up with.

It didn't take me long in terms of adding +10 Gold increments in order to get to +4 Fair and Forthright trading (somewhere less than 60 Gold).

Seeing as how it is very safe to Demand 50 Gold back from Ragnar (as long as he is Pleased towards us), I think that this approach is better than gifting him 2 techs... basically, as long as +4 Fair and Forthright Trading actually gets him up to Pleased status.


My saved games seem to go from +1 to -1 to -2 total Attitude with Ragnar, so I don't have data for the "0" total Attitude with Ragnar... but since we have climbed back up to +1, mdy should have that value in one of his saved games.

 

[mdy]

That locatin is NOT a blocking position. There is a Coast 1 square to the north of there. Ragnar can simply sail around you with no loss in movement points.

Actually that tile is an ocean tile, so this blocking plan will work and we don't need to worry about giving techs to Ragnar.

 

[shyuhe]

I agree with mdy here - we can block Ragnar's current 3 galleys without any gifts. Mdy was referring to 3W2S of Marble as the blocking point though. It goes like this:

T+1 IT: Rag's boat goes 3N3E
T+1: northern boat moves 2E to block marble; southern boat chases Ragnar's boats going 3N2E.

T+1 IT: Rag's boat goes 2N2E

T+2: northern boat moves 3N2W to block gold; southern boat moves to 2N3E to block marble. We declare war here.

T+2 IT: Rag's boats are stuck unless he suicides his galleys on our triremes.

We will want to send our third trireme over to help kill Ragnar's galleys in any event. We just have to hope for the best that a barb galley doesn't show up.

If Ragnar's boats during T+1 IT head straight east without declaring war, we're obviously safe (Vicky target). If they head NW, then again we're safe (western AI target). I don't *think* Ragnar will attack a trireme with an escort galley + 2 galleys filled with troops. If it's 2 escort galleys + 1 galley with troops, we'll have to rely a bit on the RNG if Ragnar decides to suicide.

Note that we can set up the T+2 position even if Ragnar declares war on us during the T+1 IT.

 

[mdy]

As such, what we could do is:
1. Research Machinery until there is 1 turn left to go on it.
2. Start researching Compass until Willem has about 4 turns left to go on Machinery
3. Switch back to researching Machinery
4. Immediately trade Machinery + Code of Laws to Willem <-> Calendar

This would work, provided we are certain he would make this trade with only around 300 beakers left to go. Though it would speed up his path towards feudalism, which would be a bigger problem than Speeding Cathy up.

IF we can get our final Great Person with TWO TURNS remaining in the Golden Age, then:
1. With 2 turns to go in the Golden Age, we can Lightbulb our way up to Astronomy
2. We then set Research to Civil Service
3. With 1 turn left to go in the Golden Age, we'll be able to switch Civics, including into Bureaucracy

We can't get a GP 2 turns before the end of the golden age unless we delay it. Fortunately the final GP is being used to bulb engineering. We can get our penultimate GP in time to bulb astro, then tech CS and revolt before the end of the GA.

To Ragnar, Mysticism is likely to get him back into the AI-AI trading game. That means 5 AIs able to trade around techs to each other instead of 4 AIs.

I can't see how giving Ragnar mysticism will get him back into the tech trading game.

Do you know when we will get our first +1 open borders modifier with Cathy, and if it would make her pleased? We might be able to beg the required gold from her instead.

I think we asked for gold from Willem. We haven't been pleased with Ragnar so I doubt we begged gold from him? I may be wrong though.

That's correct.

We will want to send our third trireme over to help kill Ragnar's galleys in any event. We just have to hope for the best that a barb galley doesn't show up.

We could keep hi trapped going knowhere around those islands, and wait for galleons to destroy them. If we attacked with triemes we would have a 1 in 3 chance of losing one or more of them.

I propose using the galley to transport the workers/warrior from the iron island to the hoses island. The workers can then chop the forest there, before heading back to hook up the iron(it will be pre-roaded) after the final warrior is built. The extra production we get from the chop in iron will allow us to hire scientists there to give us the few extra beakers we need. The galley could take care. The trieme in the area will be shifted to between the capital and gold in case a barb galley shows up there.


I also think that the golden age should go ahead as planned, and that we should make the currency-calendar trade to Cathy at the last moment, and trade for Ragnar's 50 gold if we can't beg for the gold from another source by the time we need it.

 

[Dhoomstriker]

I agree with mdy here - we can block Ragnar's current 3 galleys without any gifts. Mdy was referring to 3W2S of Marble as the blocking point though. It goes like this:

Well, I agree with BBP here... this scenario does not cover all possible situations.

For example: If Ragnar declares by moving onto the square owned by our Trireme, his boats will be in a different position in your second screenshot, in a location that would require 3 of our boats to block his progress.

Since we'd only have 2 boats, Ragnar could sail past us.

Let me see if I can set up a test game to demonstrate what I'm talking about.

 

[Dhoomstriker]

This would work, provided we are certain he would make this trade with only around 300 beakers left to go. Though it would speed up his path towards feudalism, which would be a bigger problem than Speeding Cathy up.

I'm not necessarily enthusiastic about giving-away Machinery, either.

Cathy just had to be the "most hated AI" on the block, otherwise we could consider the following:
1. Sell Code of Laws <-> Cathy's 140 Gold
2. Self-tech Machinery to within 1 turn of completion
3. Self-tech Compass
4. Self-tech that last turn of Machinery
5. Trade Philosophy <-> Cathy's Calendar + whatever Gold she has made over the course of those turns

Of course, those 2 trades to Cathy will almost surely make 3 AIs give us -4 You Traded With Our Worst Enemy penalties.

Unfortunately, Vicky, Joao, and Isabella are all in conflicting Religions with Cathy and none of them have Open Borders with Cathy, so I don't really see those relationships improving over time.

What MIGHT help, though, is if one of those AIs gets angry enough at us for the Code of Laws <-> Cathy's Gold trade for us to become that AI's Worst Enemy, then we might not get the full -4 Traded With Our Worst Enemies penalty when we make the Philosophy trade.

There is also the chance that Joao will switch into Hereditary Rule, as he is one of three 3 AIs that have it as his Favourite Civic. I doubt that him switching would bring him up to Cautious from Annoyed immediately, but maybe after a 10 turn interval or a 20 turn interval he'd almost certainly be up to Cautious towards Cathy...


If we DO go with this approach (angering 3 AIs, but delaying Willem's Resarch on yucky-for-him-to-get Machinery plus then delaying him a bit on his way to Feudalism), then we would want to Demand Vicky's Gold BEFORE making a trade to Cathy, since Vicky would almost certainly no longer be Pleased with us after we traded Code of Laws <-> Cathy's Gold.


The decision does not need to be made immediately, of course, but the more turns that we wait, the less Gold that Cathy COULD have if she spends or trades it... and the less Gold that we get almost means the greater the Traded With Our Worst Enemy penalties will be.

We can't get a GP 2 turns before the end of the golden age unless we delay it. Fortunately the final GP is being used to bulb engineering. We can get our penultimate GP in time to bulb astro, then tech CS and revolt before the end of the GA.

As long as we also get Calendar by the second last turn of the Golden Age, Lightbulb Astronomy by the second last turn of the Golden Age, and, as you say, plan to research Civil Service on the second last turn of the Golden Age, then you are right.

Therefore, we do need to think "short term" in terms of the Calendar trade.

But, at least we know that we can launch the Golden Age on time (Turn 167), which means that we won't need to do any re-testing of the Golden Age timing.

I can't see how giving Ragnar mysticism will get him back into the tech trading game.

Well, it is my experience that AIs like Shaka and Ragnar will avoid self-teching Mysticism, but once they get it, they can quickly shoot up the Religious tech tree. It's hard to know exactly what goes on since the game does not inform you about AI-AI tech trades, but we know certain facts:
a) Ragnar is avoiding Mysticism

b) Ragnar is the dumbest AI so AIs are avoiding making trades with him

c) I claim that if Ragnar is given Mysticism, experience shows that he will quickly get up towards Hereditary Rule. What I don't know is if he would trade for the techs or self-tech them, but he would very likely do one or both of those things

d) Once Ragnar starts shooting-up the Religious tech path, given that he also knows Construction above-and-beyond most of the other AIs, he will very quickly no longer be the dumbest AI. When multiple AIs vie for the "dumbest AI" position, then NO AI ends up being the dumbest AI... I'm not 100% certain how the mechanic decides a new dumbest AI, but it could be that a tech differential of 2 techs, instead of just 1 tech, is required, which is why it usually takes a while for AI-AI tech trading to "settle down" and for a new "dumbest AI" to be declared by the AIs

e) Even if a new "dumbest AI" gets declared after a few rounds of AI-AI tech trading, then if that dumbest AI beats the second-dumbest AI to researching their next tech, then the whole AI-AI tech trading process starts up again in haste

f) That's generally how you get a run-away tech situation, because the lower half of the AIs will ignore WFYABTA penalties... what usually keeps things from getting out-of-hand is that the "dumbest AI" gets shunned (ignored) in tech trades... but if there is contention as to which AI is the "dumbest," then "no dumbest" will actually exist and a lot of trading happens with the lower-tiered AIs. Let's just say that it gets messy really quickly


What I suggest that you do is look for some old games that you have played that have Ragnar in them (Shaka seems to also work). Find a game where one of those two AIs does not have Mysticism. Then, play the game forwards for 30 turns without trading it to them. Preferably, set Espionage to them, so you can see what they do tech-wise.

Then, play again, but trade them Mysticism on the first turn.


If the Trireme-blocking plan really were to work (which I do not think that it will in all scenarios), we could simply trade Alphabet <-> Ragnar's Gold, since then he wouldn't be grabbing Meditation and/or Polytheism anytime soon and would remain the "dumbest AI" by a margin of a couple of techs. However, I don't see that Trireme-blocking plan covering all scenarios, so I think that we're going to be forced to either:
a) Follow the Gifting Alpahbet + increments of 10 Gold plan
OR
b) Bet that Ragnar is going towards Paris and then plan things out so that his Galleys will have to "walk around" Trireme 2 for 1 square on 1 turn, which would give Trireme 1 a chance to catch up enough to be able to attack on the following turn. Triremes 1, 2, and 3, as well as our Galley, would all attack... and then we'd be at the mercy of the random number generator not to repeat the previous scenario that we saw when attacking the Barb Galley

Do you know when we will get our first +1 open borders modifier with Cathy, and if it would make her pleased? We might be able to beg the required gold from her instead.

Open Borders positive Diplo modifiers come after 25 turns of having Opened the Borders. When did we decide to Open Borders with Cathy? During my last turnset, if I recall correctly. We should be able to hunt for that info in one of my reports.

If we attacked with triemes we would have a 1 in 3 chance of losing one or more of them.

Well, Ragnar's moves are not 100% predictable and therefore it is possible that the proposal on the table will require us to attack with Triremes, otherwise Ragnar will sail right past us.

I propose using the galley to transport the workers/warrior from the iron island to the hoses island. The workers can then chop the forest there, before heading back to hook up the iron(it will be pre-roaded) after the final warrior is built. The extra production we get from the chop in iron will allow us to hire scientists there to give us the few extra beakers we need. The galley could take care. The trieme in the area will be shifted to between the capital and gold in case a barb galley shows up there.

This stuff is really going to depend upon how we resolve the Ragnar issue. If we get 10 turns of Peace out of him, then sure... but if we go with a war plan, then we really need every single boat going west, as we do not have a fool-proof blocking plan and may have to engage Ragnar in battle depending upon how he moves his boats.

I also think that the golden age should go ahead as planned, and that we should make the currency-calendar trade to Cathy at the last moment, and trade for Ragnar's 50 gold if we can't beg for the gold from another source by the time we need it.

So, wait, let me get this straight:
You want to trade Currency to:
a) The AI that has The Temple of Artemis
AND
b) An AI that already has 5 Cities
AND
c) To an AI that will immediately be able to trade it around to every other AI, due to that tech's low Monopoly-threshold

I mean, if you want to get Calendar from Cathy ANYWAY, then we MIGHT AS WELL not worry too much about the You Traded with our Worst Enemy penalties, since we'll be getting them ANYWAY.

In that event, I would much rather give her the useless Code of Laws (which Vicky already has by the way) and Philosophy techs, plus getting her 140 Gold as part of the bargain.

The only question becomes then... how long ago did Joao learn Monarchy (i.e. how long do you think that he'll delay his revolt--if he has already delayed for several turns, then it may be a while before he revolts) and how likely is it that +1 or +2 (+2 coming after 10 turns after he revolts) will make him stop treating Cathy as his Worst Enemy?

In other words, I'd much rather see us making the Code of Laws <-> Cathy's Gold and Philosophy <-> Cathy's Calendar trades (two useless techs for her) than giving her Currency (an extremely useful tech for her that she could also immediately trade around as a useful tech for other AIs). The only question becomes: how long can we wait for Joao to possibly stop hating Cathy due to a Shared Civic bonus for a Civic that he hasn't even revolted into yet (even though he has the tech to do so--Monarchy--I just don't know how long ago he got that tech)?

 

[mdy]

Well, Ragnar's moves are not 100% predictable and therefore it is possible that the proposal on the table will require us to attack with Triremes, otherwise Ragnar will sail right past us.

Can you post one credible scenario where this could happen, because I can't see any?

e) Even if a new "dumbest AI" gets declared after a few rounds of AI-AI tech trading, then if that dumbest AI beats the second-dumbest AI to researching their next tech, then the whole AI-AI tech trading process starts up again in haste

Are you saying that AI's will refuse to trade with the most backward Civ, even if there is a potential trade with one another that they would otherwise take? That doesn't make sense to me.

 

[Dhoomstriker]

Can you post one credible scenario where this could happen, because I can't see any?

Ask and you shall receive. I have for you THREE such scenarios.

You'll have to use your imagination, because I couldn't get Ragnar to attack us... but Cathy would. Cathy has been given a complement of 3 Nagivation-I-promoted Galleys.

Her Galleys are in the same position as Ragnar's Galleys are.

Our boats are in the same positions as in the real game and they have used/not used their movement points appropriately.


In the first scenario, all 3 of Cathy's boats move to the position that was indicated in shyuhe's screenshot, so I moved the 2 Triremes as indicated in shyuhe's screenshot:

So far, so good, right?

Well, what happened when we ended the turn was not what we saw in shyuhe's screenshot. Instead, Cathy seems to have blocked our Trireme 2 from being able to get into position:

So, I tried again. Version 2 has the same scenario for T + 1, but T + 2 gives us the following situation. From here, Trireme 1 could attack that southern boat, and if it wins, then Trireme 2 could block Gold City, at which point Cathy would probably move her 2 remaining boats SE + SE + SE. We could then block her by putting both Triremes to the east of her (S and S + S of Marble City's western Clam) and then, on the following turn, send one of those Triremes towards Gold City to block and then trap her... BUT, all of that scenario relies on a successful battle when we attack with Trireme 1, PLUS having Trireme 1 be strong enough not to be attacked on the one of the following turns by one of the remaining Galleys. So, it's not much of a "safe bet," either:

Version 3 has a different scenario for T + 1, where Cathy has already decided to split up her boats. I don't have a screenshot for that scenario, as I am running low on web server space again, but if that scenario happens, we have zero chance of blocking all of her boats using only 2 Triremes given the Triremes' current positions.


Each of the three scenarios has a corresponding saved game that you can check out below.

Are you saying that AI's will refuse to trade with the most backward Civ, even if there is a potential trade with one another that they would otherwise take?

Yes.

That doesn't make sense to me.

Does everything that's coded in this game make sense?

 

[shyuhe]

@Dhoom - your latter scenario should not happen unless the bottom boat has no troops in it, in which case we can ignore that boat anyways. The AI should not send unescorted ships filled with troops. So assuming 2 out of the three galleys have troops in them, Ragnar should not split up that stack. If he does split it up, we just go after the stack with 2 boats.

As for your former situation, did Cathy declare war on the inter-turn? The only scenario I am worried about is if Ragnar declares war during our inter-turn and then attempts to move its boats north because he doesn't want to attack our trireme (due to bad odds). If this happens, he will be able to land next to gold.

From an old thread:

^If the galleys are w/ AI_ATTACK_SEA they will attack, however since they are AI_ASSAULT_SEA (i.e transport).
The spawned barbs are not transport but produced ones are.

If they are in a city and loaded they will skip the turn, if the city is in danger (i.e. enemy units nearby). That results into unloading the troops.
Next time the troops are loaded again and the check about danger is performed and the same.

How about we declare war on Ragnar on T+1? I think this should keep him from splitting up his boats, in which case we can block him using the placement I suggested earlier. Basically if we can make Ragnar keep his boats together, we'll be good.

@Dhoom - do you mind re-trying your test with two boats filled with troops and AI_ASSAULT_SEA script and one boat without troops with AI_ATTACK_SEA? I'm pretty sure the AI won't split that stack but I'm not 100% sure.

 

[mdy]

I think we can safely ignore all there of these possibilities, as all three of Ragnars galleys are obviously part of the same stack which Ragnar will move as one, and he has already decided his target.

The AI would not have decided to treat these units as a stack when you created these units in worldbuilder, this decision would not have been made, hence the different behaviour.

 

[Dhoomstriker]

So, back to the Calendar trading issue... if you really think that we should get Calendar from Cathy, then I'd much rather give her Code of Laws + Philosophy than give her Currency.

Thus, on that note: do you know when Joao learned Monarchy? If he learned it several turns ago, then he might not continue to revolt for a while. If he learned it last turn, then he could revolt at any time. If he revolts soon, then we may be able to see him get +2 with Cathy within a reasonable timeframe (10 turns after he revolts), which might or might not be enough to shake off the Worst Enemy status. So, we may or may not benefit from waiting some turns before making the Code of Laws <-> Cathy's Gold trade.


Then, there's the issue of wanting to Demand from Vicky BEFORE trading Code of Laws <-> Cathy's Gold, and if we want to wait to see if Vicky will accumulate more Gold, then we may also want to wait several turns to see if she gets more Gold for us to Demand.


There isn't an immediate rush to grab Cathy's Gold, as it doesn't look like she has been spending it and with so many AIs that she cannot trade with (the three AIs that treat her as a Worst Enemy plus the "world's dummy" Ragnar), she isn't likely to trade away her Gold to anyone else anytime soon. So, we can probably afford to wait a few turns, but if we wait too long, Vicky might go down to only having +2 Fair and Forthright Trading with us, at which point we would not be able to make the Demand out of her.

I'm not sure how many turns we can afford to wait, but probably 5 turns would be the most that I'd feel comfortable waiting, after which I would suggest that we:
1. Demand all of Vicky's Gold (under a reasonable amount--I said 60 Gold or less earlier but the maximum amount will have gone up by now--if Vicky ends up with more than 60 Gold, let me know and I'll do a calculation as per that exact turn)
2. Trade Code of Laws <-> Cathy's Gold
3. Later, trade Philosophy <-> Cathy's Calendar + any Gold that she might have collected in the meantime

 

[mdy]

1. Demand all of Vicky's Gold (under a reasonable amount--I said 60 Gold or less earlier but the maximum amount will have gone up by now--if Vicky ends up with more than 60 Gold, let me know and I'll do a calculation as per that exact turn)
2. Trade Code of Laws <-> Cathy's Gold
3. Later, trade Philosophy <-> Cathy's Calendar + any Gold that she might have collected in the meantime
May 24, 2011 09:00 PM

I reckon we can safely postpone the trade for calendar, and any gold demand till T172 without closing down any options. As no one else has currency they won't be trading there gold anytime soon.


Jao did not have monarchy in 155BC, so he must have learnt it in the last 3 turns. My autosaves have been overwritten now, so I can't be more precise than that.

 

[shyuhe]

How much gold are we short?

 

[Dhoomstriker]

@Dhoom - your latter scenario should not happen unless the bottom boat has no troops in it, in which case we can ignore that boat anyways.

I'm not sure which scenario you are talking about, since I had 3 scenarios, called "Version 1," "Version 2," and "Version 3."

However, in every Version, all 3 boats were fully-loaded.

The AI should not send unescorted ships filled with troops. So assuming 2 out of the three galleys have troops in them, Ragnar should not split up that stack.

I do not think that we can make this assumption, as the facts that were presented tell a different story.

The 2 Archers in Nidaros were clearly City Defending units, as they had been sitting in that City for a long time, and thus those Archers would not be given a command to get loaded onto a boat.

However, Ragnar was in "hands full" mode and had 2 Galleys with a Catapult in the same City. The Catapult would have immediately "disappeared" into one of the boats had he been in "hands full" mode and had either of the Galleys had room in them.

Therefore, we must conclude that both Galleys were full.

However, Ragnar did not attack with two full Galleys.

So, he waited for a third Galley to arrive/be built in Nidaros.

But, he was also "amassing" further ground troops (as evidenced by the Catapult in Nidaros).

Therefore, we really need to believe that all three Galleys were collecting troops.

Since an AI will not send out a partially-full Galley, we know that instead of having 4 troops in the first 2 Galleys and a single Catapult in the 3rd Galley, he must have 6 troops--2 in each Galley.

As for your former situation, did Cathy declare war on the inter-turn?

She declared on the inter-turn in each of the three Versions.

The only scenario I am worried about is if Ragnar declares war during our inter-turn and then attempts to move its boats north because he doesn't want to attack our trireme (due to bad odds)...

How about we declare war on Ragnar on T+1? I think this should keep him from splitting up his boats, in which case we can block him using the placement I suggested earlier. Basically if we can make Ragnar keep his boats together, we'll be good.

I'm worried about far more than that.

Okay, let's assume that all 3 boats will "stick together." Well, as you can see from Version 1, those 3 boats could end up being where 2 of the boats finished their turn--NW + NW of the Grassland Horse. In Ragnar's 3 boats end their turn there after having declared war, then 2 Triremes, given the Triremes' current positioning cannot block him. Thus, not even every scenario where all 3 boats "stick together" is stoppable.

That single possibility alone is enough for us to reject the plan. I don't need to come up with multiple counter-examples... I just need to come up with one feasible... and there it is.

Although I don't have a scenario where THREE boats end their turn on that square, I have one where TWO boats end their turn on that square (see Version 1). Let's assume that ALL THREE boats ended their turn on that same square (the square that is NW + NW of the GHorse).

Please convince me, if you think that it is possible, that if the three boats "stick together" but end their turn NW + NW of the GHorse that you can effectively block them with 2 Triremes. I don't see a way to block the Galleys. Do you?

@Dhoom - do you mind re-trying your test with two boats filled with troops and AI_ASSAULT_SEA script and one boat without troops with AI_ATTACK_SEA? I'm pretty sure the AI won't split that stack but I'm not 100% sure.

Okay, I tried exactly that. I even named the Galleys according to their scripts.

The results were really weird... I had to open the World Builder just to take a screenshot of T + 2. Cathy still declared war on us, but 1 turn later. The boat to the east is one of the 2 fully-loaded Galleys with the AI_ASSAULT_SEA script, while the two boats to the west are the other one of the fully-loaded AI_ASSAULT_SEA script boats plus the empty AI_ATTACK_SEA script boat.

 

[shyuhe]

Do you mind testing version 4 a few times with a jiggled RNG? You can just throw in a few warriors somewhere next to a few barb animals and have them fight to see if the RNG is affecting the AI behavior. I'd test this myself but can't do it until later tonight.

edit: What I think is happening is that Cathy moved the "transport" galley into our border and declared war, bumping the boat out. Then the other two boats headed westward around the island to avoid our trireme. Just a guess though...

 

[Dhoomstriker]

As no one else has currency they won't be trading there gold anytime soon.

An excellent point!

There's yet another reason not to give away Currency... AI-AI tech-trading is limited to "relatively equally-valued trades," that is, until they learn Currency, at which point they can use Gold to make up the difference and thereby enable far more AI-AI trades than were previously possible.


So, sure, we can wait a few turns to see if Vicky gets more Gold.

Please watch Vicky's relationship towards us on each turn, though... if she goes down to +2 Fair and Forthright Trading, we might still be able to have success... we could simply gift her increments of 10 Gold until she is back at +3 Fair and Forthright Trading and then Demand it all back (60 Gold you could safely Demand at Pleased from Vicky since we've known her for a long time, more than that and I'd want to run a calculation based on the exact turn that you plan to make the Demand).

If we wait more than 1 turn after she goes down to +2 Fair and Forthright Trading, though, then it might cost us too much in Gold to get her back up to +3.


Also, if she makes a Demand herself between now and then, and we refuse it, then we won't be able to make a Demand from her, as she should dip down to Cautious, so if she makes a Demand, we should probably weigh our options instead of automatically refusing her.

 

[Dhoomstriker]

Do you mind testing version 4 a few times with a jiggled RNG? You can just throw in a few warriors somewhere next to a few barb animals and have them fight to see if the RNG is affecting the AI behavior. I'd test this myself but can't do it until later tonight.

It will have to be later as I have to run to an appointment now, but to what end would be running this scenario?

For one, I don't believe that there's an empty Galley. That point aside, assuming that one of the Galleys IS empty, what are we looking to prove?

Are we just looking to see if your screenshotted-scenario is one of many possible paths for Cathy's boats to take? Sure, I'll grant that it is possible.

But we already have a counter-example, one where all three boats end their turn NW + NW of the GHorse, and that possibility seems to be well-supported by what we saw in Version 1, where Version 1 is actually a slightly "even worse" scenario.

 

[shyuhe]

I don't think version 1 can happen given the scripts on the boats. I'll try futzing with it tonight.