SGOTM 13 - One Short Straw

[mdy]

I think your ignoring the fact that the AI always takes one of the quickest free routes to it's target. Under no circumstances does this mean that the AI will end it's turn with it's galleys in the potentially dangerous position if we DOW on T+1 (if we waited till T+2 it might happen depend on how the AI evaluates it's path.)

 

[Dhoomstriker]

I think your ignoring the fact that the AI always takes one of the quickest free routes to it's target.

Ummm, maybe before the test games you could have made such a claim, but in the last few messages on the subject, what I have been doing is reporting the results obtained from playing Civ 4.

So, I guess that "Civ 4" is ignoring this fact about how an AI moves towards its war target?

Or, is the claim simply that the World Built units do not represent a situation akin to what would be seen in a "real game" of Civ 4? I would gladly welcome help if someone were able to improve upon my test game approach.

Under no circumstances does this mean that the AI will end it's turn with it's galleys in the potentially dangerous position

How about under the following circumstance:

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.

shyuhe has a point. For all that we know, Ragnar might move 1, 2, or all 3 Galleys into our territory before declaring war. Or, there might be some sort of logic that allows the AI to move into our Cultural Borders in order to declare war but also allow them to "abort" the movement if such a movement would result in terrible combat odds. Certainly, such code is justifiable, otherwise you could just park 4 Machine Guns in the Forest within your Cultural Borders that is closest to Monte and then he would be "forced" to suicide his entire stack simply because he "moved it" using "Stack Attack" "onto" that square when declaring war. I could certainly see an AI programmer putting in "abort code" that would still make the war declaration happen but would cancel the attack itself, allowing the units to walk around your Machine Guns on the following turn (or in this case, sail around the Trireme on the following turn).

Otherwise, it could be relatively easy to abuse the AIs, right? I mean, what we're trying to do by positioning the Trireme there is tantamount to abusing the AI code... if it is FORCED to take the shortest path to Paris (or even just the shortest path to our nearest Cultural Borders) and if it can only declare war by MOVING into our territory, then it would be FORCED to attack with at least one of its units. In that case, we could certainly exploit the AI's behaviour by parking a strong defensive unit where we expect it to assault us. Yet, as we can see from the test game, the AI's Galley is not forced to attack. Therefore, it is very believable that this "easy to exploit" case of the AI behaviour was altered by giving the AI the ability to abort the attack, even though war will still be declared.

And, unlike my test game, if the units will "stick together as a stack," and if the "AI behaviour is completely predictable," then given the above allowance in the code to avoid AIs from being exploited so easily, it should actually be the EXPECTED behaviour that all 3 units will end up NW + NW of the GHorse (assuming that shyuhe is right about where the units get "kicked out to being that square--I really have no idea where they would get kicked out to) and that there is no possible way that they will end up where shyuhe's screenshot demonstrates. That's going with the assumption that, as you seem to be suggesting, that the AI behaviour is going to be completely predictable.

if we DOW on T+1 (if we waited till T+2 it might happen depend on how the AI evaluates it's path.)

Can you perhaps back up your claim with results from a test game?

I have NOT tested scenarios where we declared war, so I can't say for certain one way or another at this point what might happen.

However, even if this situation were possible, I think that you can admit that it relies on a fundamental assumption, and that assumption is that Ragnar's boats will stick together in a single stack of units. So, we must minimally accept this assumption as fact. On top of that, we'd need to believe that proactively declaring war would give Ragnar only one possible path for his Galleys.

That's a lot to take on faith, so it would be nice to see test game evidence to support it.

How might we get an AI to "create a stack"? Well, I've been thinking about it and what I'm going to try is to place Cathy's units inside of a City that is located at the same position as Nidaros in the real game. She has already decided to attack us. She can't reach us from her mainland, but perhaps this approach will work in terms of getting the boats to "stick together."


If you have other ideas of how I might go about getting an AI to "stack" its units together, I'd love to hear them.

 

[Dhoomstriker]

Okay boys, I have gotten as close to the real game as I possibly can here:

1. Cathy's boats are all AI_ASSAULT_SEA

2. They all start in Nidaros. Okay, the City is actually 1N of where Nidaros sits in the real game, since our map is "cut off" at the south in our test game, but the movement required to reach the same square when leaving the City is the same, so it's not really functionally different

3. Cathy has 2 defensive Archers for her City

4. Cathy has 6 CITY_ATTACK land-based units that Cathy HERSELF has loaded onto Galleys

5. I even managed to get her to group and move the Galleys as a STACK! I first "land locked" her units for a turn and then she did the rest. I'm updating this message with updated screenshots now that I figured out how to do this last bit


So, there you have it: your stack of Galleys exists.



What's the most compelling thing for me, is that I started to pay attention to HOW Cathy declared war. In fact, it does look like she attempts to "move into" the closest Cultural Border of ours, "aborts" this movement due to our Trireme being there, gets "booted out" to the west (NW + W of the GHorse)--or perhaps doesn't even "move" at all, and then has 1 movement point remaining (the "booting out" didn't cost her a movement point, apparently, which makes sense if she didn't even need to "move" at all) which she uses to move 1N... putting her smack dab in the exact spot that we DO NOT want her to be in: NW + NW of the GHorse. Observe:

She then continues by moving her stack exactly to the square that we don't want her to move to. So, if you believe that with Ragnar's Galleys "moving as a stack" is "predictable," then you should be of the belief that there is only one location that the Galleys will go... exactly where we don't want them to go:

Meanwhile, if I do everything else the same except move Trireme 2 three squares to the east instead of just two squares to the east, Cathy "enters" our Cultural Borders, but, meeting no "resisting Trireme that it would allow her Galley to abort its movement," her Galley simply cruises right into our Cultural Borders:

That's pretty compelling evidence to me: if we position Trireme 2 two squares to the east, then we're doomed.

If we position Trireme 2 three squares to the east, then we have a chance if we use all 4 boats AND if Ragnar's target is not Marble City... but we'd still be relying on the random number generator to allow 3 Triremes + 1 Galley to destroy all 3 of Ragnar's Galleys in a single turn, and we'd also be relying on Ragnar to ignore Marble City.

So, even my suggested approach of utilizing all 4 boats carries assumptions and risks.


The only possible way out that would not risk our fate to the random number generator would be gifting Alphabet + Gold + Demanding as much as 50 Gold back. Of course, I don't think that we're certain that Ragnar is Cautious towards us at a "0" Attitude. If he is Annoyed at a "0" Attitude, then this gifting will not bring us up to a "Pleased" status... but, at least, on the following turn, Ragnar would not have gotten close enough to our empire to declare war, at which point we can launch our Golden Age on schedule and switch Religions to No State Religion, finally making the Demand.

We would, of course, switch Religions again to Confucianism 5 turns later, losing out on several GPP and the chance to revolt into Bureaucracy within our Golden Age. Alternatively, we'd just "pay" that turn of revolt up-front by not launching a Golden Age until a few turns later so that we could still switch into Bureaucracy at the end of our Golden Age but then we'd still be "paying" 1 turn of Anarchy to switch into No State Religion--darned if we do, darned if we don't. So, hopefully, Ragnar is Cautious with us at a total Attitude of "0," and we won't need to make this tough Anarchy decision.


Below, you can find the saved game where Cathy has loaded-up her Galleys full of units herself and MOVES HER GALLEYS AS A STACK!

 

[shyuhe]

Ok, here's the solution:

T+1: we move northern trireme 2E, souther trireme chases. WE DECLARE WAR.
T+1 IT: Cathy moves her galleys to 4W of marble as it's the only open route.
T+2: we move the northern trireme to block the path to gold. Souther trireme moves up to block the path to marble.
T+2 IT: Cathy's boat heads west around the island. Northern trireme is stationary. Souther trireme moves 1SE and completes our block.

 

[Dhoomstriker]

T+2 IT: Cathy's boat heads west around the island. Northern trireme is stationary. Souther trireme moves 1SE and completes our block

Unlike the original approach of declaring war, where we could use one single location to block Ragnar and then declare war on him so that he would have had to have sailed around his entire island to reach us, this new plan would force us to use 2 permanent locations to block Ragnar.

So, it does look like he will not attack a Trireme with a stack of 3-fully-loaded Galleys, which is fortunate.

However, are we prepared to COMPLETELY COMMIT 2 Triremes to NOT MOVING, even if Barb Galleys appear, say, to the north of Gold City? I ask because that's what this plan seems to be offering: a permanent commitment to keeping the Triremes in place. Is that really what we want to do?


We would also anger Ragnar and any AI that likes him.


Also, we're not 100% certain that he can't loop around and surprise us from a northern passage to the north of Gold City, as we haven't completed the exploration up there.


So, basically, our empire would be protected from Barb Galleys by 1 Trireme and 1 Galley (although the Galley would most of the time be busy shipping units around). Is that a level of risk that we are comfortable with?


Shouldn't we at least try the gifting Alphabet + increments of 10 Gold approach to get +4 Fair and Forthright Trading, and then if it fails to get Ragnar to be Pleased with us, have a vote on whether or not we should Declare War on Ragnar as per the above plan versus Revolting into No State Religion for 5 turns in order to get him up to Pleased?

 

[shyuhe]

I'm not familiar with how many more turns of research we have to do before we're good to go on the offense. I assume it's more than 10 turns, in which case taking a 10 turn peace with Ragnar will not solve our current dilemma anyways--it merely delays the inevitable (unless Ragnar is going after someone else).

If we declare war on him as opposed to him declaring war on him, the "refuse to talk" window is much shorter as well.

Ragnar getting mad at us is more or less a non-issue.

 

[Dhoomstriker]

I'm not familiar with how many more turns of research we have to do before we're good to go on the offense. I assume it's more than 10 turns, in which case taking a 10 turn peace with Ragnar will not solve our current dilemma anyways--it merely delays the inevitable (unless Ragnar is going after someone else).

Well, coincidentally, it should be precisely 10 turns from now that we'll be Revolting back to Slavery.

Now it is Turn 166. Next turn, Turn 167, we will launch our Golden Age. The Golden Age will last for 10 turns. On the 9th turn of the Golden Age, we will be revolting into Slavery.

The plan is that on the 8th turn of the Golden Age, we will start to build some units so that they will have Hammers invested into them. That way, on the 9th turn of the Golden Age, when we switch Civics out of Representation and into Slavery (amongst other Civic swaps), we will be able to immediately whip some units in the then-Unhappy Cities that lost the bonus Representation Happiness.

Galleons could certainly be some of the unit types that get whipped.

Hence, 10 turns from now, we could be whipping Galleons, which would appear 11 turns from now. These Galleons will have 5 movement points and 4 strength.


By taking an Enforced Peace now, we could then commmit 1 Trireme to following his stack of boats to see what he does, leaving the other Trireme free to guard our Fishing Nets.

If he moves back into his capital, then we can go back to the original plan of blocking him in with a single Trireme once the enforced Peace Treaty is up, leaving our Galleons free to go wherever they wanted. Essentially, we'd reset things back to before the current situation.


If he goes elsewhere, we'll be able to keep an eye on him and plan for it, presumably by planning an ambush of Galleons that would take his army down in one fell swoop.

If we declare war on him as opposed to him declaring war on him, the "refuse to talk" window is much shorter as well.

But, to what end? We wouldn't get techs out of a war where we did no damage. In fact, we'd likely have to pay the AI some Gold to end the war.

 

[shyuhe]

Ok, then I'm in the "gift alpha then reconsider" camp, unless we can determine that it's futile (i.e., +4 fair trade won't bump him to pleased). If it gets Ragnar to pleased, then we shadow his boats. If he doesn't go to pleased, then we try demanding 1 gold on T+1. If that doesn't work, we declare war (with our boats in position as I described). We don't know what Ragnar's power is but if it's close to ours, he may give us a 1 gold tribute--worst that happens at that point is that Ragnar declares war on us.

 

[Dhoomstriker]

If he doesn't go to pleased, then we try demanding 1 gold on T+1.

Demanding 1 Gold will be no different from Demanding 50 Gold (confirmed by reading the code).

Before the "amount" of a Demand is considered, an AI will check:
"Am I Pleased with you?"

Yes? -> Okay, check how much you're asking for
No? -> "Is my Power greater than (your Power * 4 / 3)?"

Yes? -> I don't care how much you asked for, I'm saying "no," giving you -1 Diplo, and "remembering" the fact that you made a Demand from me
No? -> Okay, check how much you're asking for and use the same criteria as if I were Pleased with you


Our Power = 100000
Our Power * 4 / 3 = 100000 * 4 / 3 = 400000 / 3 = 133333

The highest AI's Power is 162000, and it is likely that Ragnar is at least somewhere near there.


So, we're all but guaranteed that Ragnar will not only refuse a Demand if he is not Pleased with us, will also give us -1 negative Diplo (which you might not care about), and will increment his "Demanded from me" counter, which means that when our Power IS greater than it is now, which it will be soon after we start whippipng units 10 turns from now, we won't be able to make a successful Demand out of him for a long time.


If we can't get him to Pleased, then saving a Demand for the future could be useful strategically--keeping him off of our back if another war is going on, getting some Gold out of him, etc.


But, it's all moot anyway if we are able to get him up to Pleased, so let's see if gifting (not trading--it has to be a gift where we get nothing in return) Alphabet, plus then gifting increments of +10 Gold until we get to +4 Fair and Forthright Trading, is able to get him up to a Pleased status or not.

You should be able to click on our Gold (not our Gold per Turn) in the trading window and then click on the "up arrow" once in the "Gold: How much gold would you like to offer?" pop-up window, in order to auto-populate the amount with 10 Gold.

He'll remember every Gold coin, so if you have the patience, you could do smaller increments by manually typing them in, if you really want to earn your SGOTM paycheck, by minimizing the amount of Gold that we would need to give to him.

After each gift of Gold, you can hover your mouse over his Leaderhead picture and it will tell you if we have arrived at +4 Fair and Forthright Trading, at which point we'd obviously stop gifting him Gold, regardless of whether he is Cautious or Pleased with us.

 

[Dhoomstriker]

Our Power = 100000
Our Power * 4 / 3 = 100000 * 4 / 3 = 400000 / 3 = 133333

The highest AI's Power is 162000, and it is likely that Ragnar is at least somewhere near there.

Some further numbers:
The Average Power is 131833

Let's assume that Ragnar's Power is not the top Power (we'll say that it's Willem's). The bottom one has got to be Miss Isabella "I love Archers," at 95000.

131833 * 6 = 790998
790998 - 162000 - 95000 = 533998 / 4 = 133499.5

So, the average Power of the 4 remaining AIs would give us a refusal.

From the Graphs, we can see that Vicky's Power is about 30% of the way between our Power and Willem's Power. Even if Willem doesn't have the highest Power, she's still one of the AIs that is dragging-down the Power Average.

Thus, the Average of the Power of Ragnar, Cathy, and Joao is GREATER than 133499.5.


Seeing as how Ragnar has been building an army, I find it extremely unlikely that he is going to be an AI out of the three remaining AIs (Ragnar, Cathy, and Joao) that has a Power level below 133000.


Therefore, it is very safe to conclude that Ragnar will refuse any Demand unless he is Pleased with us. Given this conclusion, I would say that we should not make such a Demand, not even "just in case," as we could make use of the "right to make a Demand" for strategic purposes later in the game.

 

[babybluepants]

Agree with trying for the 10t enforced peace. I don't buy the Mysticism argument and don't like giving away cash, but I'll live.

 

[mdy]

Ummm, maybe before the test games you could have made such a claim, but in the last few messages on the subject, what I have been doing is reporting the results obtained from playing Civ 4.

So, I guess that "Civ 4" is ignoring this fact about how an AI moves towards its war target?

I can still make this claim because in artificially creating these units the AI will not have selected a single traget for them, where as he will have in the real game.

What's the most compelling thing for me, is that I started to pay attention to HOW Cathy declared war. In fact, it does look like she attempts to "move into" the closest Cultural Border of ours, "aborts" this movement due to our Trireme being there, gets "booted out" to the west (NW + W of the GHorse)--or perhaps doesn't even "move" at all, and then has 1 movement point remaining (the "booting out" didn't cost her a movement point, apparently, which makes sense if she didn't even need to "move" at all) which she uses to move 1N... putting her smack dab in the exact spot that we DO NOT want her to be in: NW + NW of the GHorse. Observe:

I think this is what is happening, and we can avoid it by declaring war on Rgnar on T+1.

However, are we prepared to COMPLETELY COMMIT 2 Triremes to NOT MOVING, even if Barb Galleys appear, say, to the north of Gold City? I ask because that's what this plan seems to be offering: a permanent commitment to keeping the Triremes in place. Is that really what we want to do?

Yes, it's not a problem because the third trieme can be moved to cover the gold, and the galley can be used to cover the iron area.

 

[mdy]

I've just tried declaring war on Cathy in the first game Dooms posted and it results in him keeping keeping his entire stack together, and moving it to a tile when we can block him as expected, so it looks like this will work, and thee is no need to resort to revolting to paganism and gifting techs+gold to Ragnar which would set us back a very long way.

Edit: I've tested it with Doom's saves No2 and 3 as well, and the blocking plan works then as well, so I think we can go ahead with the original plan, with the modification that we declare war on T+1 instead of T+2.

 

[shyuhe]

@mdy - why not try gifting alphabet to Ragnar and see how much it bumps him up? It's a tech that really won't help him and it may allow us to avoid having to declare war--which as Dhoom pointed out has the added benefit of freeing our triremes to deal with any barbs. If gifting alpha pushes Ragnar to +3, then we can consider gifting him mysticism to push him to +4 to get a 10 turn peace. The added benefit is that we'll get gold out of this transaction if Ragnar bites.

 

[babybluepants]

Yeah, I don't really see a good reason not to try the bribe.

 

[Dhoomstriker]

I've just tried declaring war on Cathy in the first game Dooms posted and it results in him keeping keeping his entire stack together, and moving it to a tile when we can block him as expected, so it looks like this will work, and thee is no need to resort to revolting to paganism and gifting techs+gold to Ragnar which would set us back a very long way.

Edit: I've tested it with Doom's saves No2 and 3 as well, and the blocking plan works then as well, so I think we can go ahead with the original plan, with the modification that we declare war on T+1 instead of T+2.

The way I see it, declaring war on Ragnar and tying up two of our Triremes is far inferior to the original plan. Our original plan was to block him using 2-stacked Triremes and then leave only a single Trireme there if he would be scared-off by it.

We now know that he would have been successfully scared off by a single Trireme, so the current war-declaration plan does force us to commit a second Trireme to the task.

To me, that fact greatly reduces the appeal. I do not think that it is appropriate for the 1 remaining Trireme to guard our empire from both Joao and Barb Galleys from all sides, nor do I consider a Galley to be a defender.

As soon as our lone Trireme defender needs to respond to a threat, it will be out of position and unable to respond to a second threat. If we move to stop a Barb Galley, we could lose a City to Joao with there being absolutely nothing that we can do about it.

Galley-Galley war is a total coin toss at best, and absolutely useless if we need to ATTACK with a Galley in order to preserve a Netted Seafood Resource. We'd likely need to be on the attack, since the Galley's priority is going to be shuttling Workers around and not standing guard in the ideal location.

I REALLY do not feel comfortable with having 2 of our Triremes stuck on guard duty.


Getting 10 turns of Peace would be an ideal situation, given that we are 10 turns away from Slavery. How much more perfect could it get?

Therefore, is there any way that you can find out from your turnset what Ragnar's Attitude was at 0? Surely, you got logs when you uploaded the game, perhaps you have some manual saved games, maybe you can check your harddrive for an autolog, maybe you took screenshots of the F4 GLANCE screen, or perhaps you can search your mind's memory for the answer--you were watching the F4 screens, right?

If we can't know for certain, then I don't see how it is a waste to at least try.

Gifting Mysticism would be "too much" value relative to Gold gifts, which is another reason to favour the Gold gifts.

If we can get our Enforced Peace by gifting Alphabet and gifting increments of 10 Gold, then we can Demand up to 50 Gold back... and, if the team decides to do so, we can always sell Mysticism to Ragnar after some period of turns, as getting an Enforced Peace would mean that tech trading would still be possible, unlike in a war situation.

If we do eventually sell Mysticism, I'd prefer to wait to do it until Isabella after learn another couple of techs, so that Ragnar will continue to hold the title of the most backwards AI without any threat of competition for that honour.



If you are unable to get us a data-point for when Ragnar's Attitude changed from Annoyed to Cautious (your logs might give us this info), then I still think it is worth gifting Alphabet plus bits of Gold... that's less than 60 Gold for Peace. WE WOULD PROBABLY HAVE TO PAY HIM GOLD TO END A WAR THAT WE DECLARED ANYWAY. So, given that the focus of declaring war is that "so that we can negotiate for Peace sooner without doing any fighting," which would cost us some Gold, where is the cost-savings with your war? There really isn't such a cost savings, or if there is, it is simply the difference between 60 Gold and the cost that we'd have to pay Ragnar. Except that having to commit 2 Triremes to guard duty is an additional cost that is far higher during our vulnerable time.

If Joao shows up with a stack of 2 or 3 Galleys, will we really win out if we're forced to switch into Slavery early just to survive? Would it have been worth saving 60 Gold at the cost of delaying our Great Person Generation, plus taking at least one additional turn of Anarchy when we'd switch out of Pacifism sometime after the end of our Golden Age? I don't see that equation being anywhere close to equal.



How is giving away Alphabet going to give away the game? Every other AI knows Alphabet. Therefore, Ragnar learning Alphabet does not actually enable any additional AI-AI tech trading. He can already trade with anyone simply because only ONE PLAYER in a tech-trading deal needs to know Alphabet.

On the other hand, giving Mysticism to Ragnar means that he can quickly pick up a couple of Religious techs using self-research, and getting those couple of extra techs is actually what WOULD bring him back into the tech-trading game.


It is NOT because he does not know Alphabet that he is the backwards AI who is not getting tech trades. That's confusing cause with effect. He is actually not getting tech trades BECAUSE he is the most-backwards AI. That's how you really need to look at the situation.

His tech priorities prevent him from self-teching Mysticism, but once he gets that tech, he becomes three tech-steps away from the tech that unlocks his Favourite Civic. So, yes, he will then be able to self-tech towards Monarchy and will do so... and on his way, will learn enough techs to vie for the "most backwards AI" crown title with Isabella. Once there is more than one "most backwards AI," then the game says that there "really isn't one" and tech trading resumes with all AIs getting involved. He still will not trade around Construction (which is unfortunate), but AI-AI tech trading will increase to all 6 AIs until a new "most backwards AI" can be crowned.


I'm sure that this info can be found on the forums if someone doubts my credibility enough to search for it.


As long as Ragnar is 4-steps away from Monarchy, then he has to get really, really, really lucky in the random-number-factor part of AI-tech-research in order to self-tech Mysticism, and until he gets a couple of more techs, no AI will trade with him. So, essentially, he is "locked out" of Monarchy until we fall for the trap of enabling him to research Monarchy by giving/trading him Mysticism.

If he becomes 3-steps-away from Monarchy (which he will be if were to gift/trade him Mysticism), then he'd be allowed to beeline Monarchy as soon as his current tech research task is completed.


To me, declaring war on Ragnar is a last-resort option (now that it requires us to commit 2 Triremes to the task) and I would vote to spend 5 turns in No State Religion above the choice to declare war. That cost seems to be a lot smaller than risking our empire's defenses to a lone Trireme.

Minimally, I'd like to see a majority out-vote me before we declare war on Ragnar.

 

[Mitchum]

I agree that we should try gifting Alphabet first to see where it gets us. We also need to know when Ragnar went from Annoyed to Cautious.

This post shows the thresholds for moving between the different relationship levels assuming no hidden modifier. If we know when Ragnar switched from Annoyed to Cautious, we can calculate the hidden modifier and then tell exactly what we need to do to get Ragnar to Pleased.

 

[mdy]

The way I see it, declaring war on Ragnar and tying up two of our Triremes is far inferior to the original plan. Our original plan was to block him using 2-stacked Triremes and then leave only a single Trireme there if he would be scared-off by it.

The "if" in this plan means you were willing to accept the possibility of two triemes being tied up before now. If this wasn't a problem then, why is it one now. In my plans I always assumed we would commit 2 triemes down there.

As soon as our lone Trireme defender needs to respond to a threat, it will be out of position and unable to respond to a second threat. If we move to stop a Barb Galley, we could lose a City to Joao with there being absolutely nothing that we can do about it.

Galley-Galley war is a total coin toss at best, and absolutely useless if we need to ATTACK with a Galley in order to preserve a Netted Seafood Resource. We'd likely need to be on the attack, since the Galley's priority is going to be shuttling Workers around and not standing guard in the ideal location.

This isn't going to happen. Jao looks like he is so far away the chances of him getting a force to us anytime soon look low, and even if he did it wouldn't matter because we would have plenty of early warning from the warrior/workers which are going to be dropped of on the horses island soon, and a single galley will be able to block Jao's entire stack.

I can shift the Stone warrior to the north which would give us warning of a barb galley

By positioning our trieme on the pigs fish we will be able to intercept any barb galley before it can reach any of our fish with a trieme (except crabs if we net it), regardless of the direction it comes from, and without taking any risks with Jao.

Galley-Galley war is a total coin toss at best, and absolutely useless if we need to ATTACK with a Galley in order to preserve a Netted Seafood Resource. We'd likely need to be on the attack, since the Galley's priority is going to be shuttling Workers around and not standing guard in the ideal location.

As explained above, we should never need to fight a battle with that galley again.

If we can get our Enforced Peace by gifting Alphabet and gifting increments of 10 Gold, then we can Demand up to 50 Gold back... and, if the team decides to do so, we can always sell Mysticism to Ragnar after some period of turns, as getting an Enforced Peace would mean that tech trading would still be possible, unlike in a war situation.

Ragnar won't have anything to trade with us, and WE NEED that gold to keep our research going, we can't afford to give away several hundred gold.

Bottom Line: I think we are fine with the trieme block plan, and can't see any reason to needlessly strengthen Ragnar and jepodise our research position.

 

[mdy]

More bad news for the gift plan:

Looking at past saves Ragnar is cautious with -1 visible, and cautious at 0 modifiers, indicating he has a hidden modifier of -2 overall, so we would need the maximum +4 fair trade bonus for the gift plan to work.

 

[Mitchum]

Ragnar won't have anything to trade with us, and WE NEED that gold to keep our research going, we can't afford to give away several hundred gold.

Bottom Line: I think we are fine with the trieme block plan, and can't see any reason to needlessly strengthen Ragnar and jepodise our research position.

At this point, I think bbp, Dhoom, shyuhe and myself are all on board with gifting Alphabet to Ragnar to see where it gets us. We have NOTHING to lose by doing this. If doing so gets us to +3 Fair Trade, we gift in increments of 10 gold (or less) until we get to +4. Don't foget that we plan to ask for 50 gold back once he is Pleased with us, so we should get it all back and then some. There is also other gold on the table that we can get too if we really need it.