Capitulation threshold?

[jkp1187]

Just wanted to continue the discussion from the original thread about capitulation threshold. Solver had mentioned that he didn't like the current level, as in a game he was recently playing, the AI capitulated after losing only one or two cities.

On the other hand, I am currently playing a game (prince level, 9 civs, normal speed, standard map size, tectonics map, 60% water), and I actually thought that AI capitulation was more or less occurring at the right levels. I took three cities from Louis in a war, after which he offered to capitulate (which I accepted). Montezuma capitulated (to Wang Kong) after losing several cities in a war against Egypt and Korea. At the same time, I ended up fighting a protracted war against Saladin, taking several cities from him (mostly ones from the rotting carcass of Egypt,) and he did not capitulate. Similarly, a war against Korea in which I took two cities from Korea has not resulted in a premature capitulation (in fact, he is demanding the return of one of those cities as the price of peace.)

In terms of the power graph, France was way down when they offered to capitulate to me (ditto Monty to Wang Kon). But Saladin has continually been high in the power graph, and Wang and I are neck-and-neck in terms of power. Basically, I'm just not seeing much anamolous behavior....but of course, I only have one or two 3.17 games under my belt right now.

It's also my view that capitulation should happen a little more often than it was under 3.13. There isn't much point in having the option available if you pretty much have to crush an enemy down to one or two cities in order to get them to surrender -- might as well just finish them off completely. So I don't see anything wrong with a civ offering to surrender even after only losing two cities, say, if the power imbalance is great enough. (Depending on leader flavor at least. Stalin should be much more reluctant to surrender than, say, Gandhi.)

The only anomalous AI behavior I've seen so far is that of Egypt. Hattie had capitulated to Saladin, and so I ended up at war with her, too, in my war against the Arabs. As a result, I ended up taking several of her cities along the way. When she was down to one city, she broke off from Saladin....then, in the same turn, agreed to become a voluntary vassal of Saladin. Odd.

Interested in hearing others' thoughts.

 

[Woody1]

I like the capitulation levels in the 3.17 patch. Especially the voluntary acceptace of being a vassal. The computer seems much smarter about becoming someone else's puppet as a matter of convenience when he's weak. Then, he'll break away when he no longer needs big brother.

It may be annoying, but I think it's a smarter way for the AI to play, rather than waiting to capitulate (or become a vassal) when he's almost killed-off. It's too late to make a come-back after that.

I really think the unofficial patch should only address things that are broken. Capitulation isn't broken, it just occurs more easily. Design-intent.

 

[Solver]

Thanks for starting the thread, this is one I wanted up.

Two notes, Woody. Design intent isn't as simple as intended or not intended. What is intended is to make capitulation easier than 3.13. But the exact threshold is tricky, you can't be assured that the threshold that 3.17 ended up with is as intended... from my experience, similar thresholds take a lot of tweaking to get right. And secondly, becoming a vassal voluntarily is a different issue. Voluntary vassalage and capitulation have different logic, and it's only capitulation I have some concerns about.

 

[Woody1]

Fair enough. I seldom accept capitulation, so it doesn't really affect the human player directly.

How is it a problem with AI interaction, though? Is the concern that the AI will form a large empire of capitulated vassals too easily? I haven't really noticed that in 3.17, yet.

 

[Solver]

Pretty much yes, it might enable an AI to gain a couple of capitulated vassals too easily. The worst I've seen so far is a 12-city civ capitulating after losing some initial ground. Not great for humans, either, if the AI can tempt them into capitulation very quickly.

 

[Willem]

I seldom accept capitulation, so it doesn't really affect the human player directly.

You mean it doesn't affect you too much. Please don't speak for all of us on this issue. I purposely try to get a civ to capitulate to me when going to war, it saves from having to wage a long and protracted campaign.

As for my opinion on the subject, I was seeing civs capitulate in 3.13 after only losing a couple of cities so I don't really understand why Firaxis felt the need to make it even easier. Maybe a few of the civs could use a bit of tweaking, the more stubborn ones, but on the whole I thought it was just fine the way it was.

 

[jkp1187]

Pretty much yes, it might enable an AI to gain a couple of capitulated vassals too easily. The worst I've seen so far is a 12-city civ capitulating after losing some initial ground. Not great for humans, either, if the AI can tempt them into capitulation very quickly.

Hi Solver -- just curious, what was the power graph comparison between the capitulating power and the master power? And what size map? (12 cities is impressive on a standard map, but less so on higher maps.)

 

[Woody1]

You mean it doesn't affect you too much. Please don't speak for all of us on this issue. I purposely try to get a civ to capitulate to me when going to war, it saves from having to wage a long and protracted campaign.

I meant exactly what I said. A lower capitulation threshold does not affect the human player. You are never forced to accept a capitulation offer. If you want to take advantage of the lower threshold, then it's completely optional for you to do so. (If you like the way it was in 3.13, then don't accept capitulate until you've mostly wiped them out.)

I'm not very interested in changes that are optional to the human player. I'm more interested in how the capitulation threshold affects the other AI players, because that is something the human player has no control over.

 

[The Last Conformist]

I meant exactly what I said. A lower capitulation threshold does not affect the human player. You are never forced to accept a capitulation offer. If you want to take advantage of the lower threshold, then it's completely optional for you to do so. (If you like the way it was in 3.13, then don't accept capitulate until you've mostly wiped them out.)

Don't be absurd. By the same logic, Corporations, say, don't affect the human player, because you don't have to build them.

 

[Solver]

Hi Solver -- just curious, what was the power graph comparison between the capitulating power and the master power? And what size map? (12 cities is impressive on a standard map, but less so on higher maps.)

The master had considerably higher power. Which may be part of the problem. The master in this case was a Qin Shihuangdi AI that already had two vassals. The victim was Hannibal. I attacked Hannibal, took a couple of small island outposts and had just landed on his main continent when Qin joined the war. After losing two cities on the main landmass, which left Hannibal with 12, he capitulated to Qin. My concern here is the snowball effect - Qin needed remarkably little effort to gain a fairly powerful vassal, he already had a couple of vassals, which scared Hannibal.

 

[jpboia]

I meant exactly what I said. A lower capitulation threshold does not affect the human player. You are never forced to accept a capitulation offer. If you want to take advantage of the lower threshold, then it's completely optional for you to do so. (If you like the way it was in 3.13, then don't accept capitulate until you've mostly wiped them out.)

This thread isn't about the option, it's about the capitulation threshold.

And yes, the capitulation threshold affects the human player, because it affects gameplay.

 

[Woody1]

The master had considerably higher power. Which may be part of the problem. The master in this case was a Qin Shihuangdi AI that already had two vassals. The victim was Hannibal. I attacked Hannibal, took a couple of small island outposts and had just landed on his main continent when Qin joined the war. After losing two cities on the main landmass, which left Hannibal with 12, he capitulated to Qin. My concern here is the snowball effect - Qin needed remarkably little effort to gain a fairly powerful vassal, he already had a couple of vassals, which scared Hannibal.

Isn't that exactly what should happen, if the AI behaved like a human player in a multi-player game? He's weak, he's suddenly ganged up by two strong opponents. If he fights on, he's going to be wiped out. Much better to capitulate to the strongest player, and hope to regain independence later in the game if things change.

IMO, a lower capitulation threshold seems to result in a civilization being able to survive. Isn't that a good thing?

 

[mjs0]

A lower capitulation threshold does not affect the human player. You are never forced to accept a capitulation offer. If you want to take advantage of the lower threshold, then it's completely optional for you to do so.

Unfortunately if you don't accept a capitulation offer, someone else may. This changes the dynamic of the game completely as you may not want to go to war with an ally or the power block represented by the new master and its allies...so the early offer of capitulation is often a use it or lose it proposition.

 

[Woody1]

Don't be absurd. By the same logic, Corporations, say, don't affect the human player, because you don't have to build them.

Your logic is non-sequitur.

If you want to compare my example to corporations, then the appropriate analogy would be if the tech-requirements for corporations was lowered to something like civil-service. Thus, you could build them much earlier. If you didn't like that, you could simply refuse to build them until later in the game. It doesn't affect the human player unless you choose to take advantage of the lower limit. (And before you start talking about the AI doing it, I'm only talking about the human player.)

So, my concern is not how capitulation limits affect the human player, because you can always refuse. Don't like it? Then don't accept. I don't need a patch to be my nanny.

I'm far more concerned with how the changes affect the AI, as already discussed in the posts above. There, it really could affect gameplay, because the human player can't control the interactions between the AIs.

 

[jkp1187]

The master had considerably higher power. Which may be part of the problem. The master in this case was a Qin Shihuangdi AI that already had two vassals. The victim was Hannibal. I attacked Hannibal, took a couple of small island outposts and had just landed on his main continent when Qin joined the war. After losing two cities on the main landmass, which left Hannibal with 12, he capitulated to Qin. My concern here is the snowball effect - Qin needed remarkably little effort to gain a fairly powerful vassal, he already had a couple of vassals, which scared Hannibal.

I confess, this doesn't trouble me that much. It sounds like Hannibal was weak and cut the best deal he could under the circumstances. I speculate that Qin may also have torn the heart out of Hannibal's army, too, since he was attacking directly into Hannibal's core cities, and not the fringe like your troops were. It wasn't like Hannibal lost only two cities -- he lost several cities to you AND Qin, and then surrendered to Qin. (Plus, you're right, Qin probably looked more frightening...but then again, that only makes Hannibal's decision to capitulate more sensible, no?)

Just curious, do you have a save game? It would be interesting to see how it played out.

 

[Ace of Spades]

So, my concern is not how capitulation limits affect the human player, because you can always refuse.

I'm not quite sure, but I sometimes have the situation that I am waging war on a civilization. I am intending to capture their cities, therefore I do not go for capitulation... and then the civ I invaded capitulates to a third civilization which declared war on it as well, preventing me from capturing their cities.

However, I cannot say if at a given time a civ would offer to capitulate to two different civs, or if it can only offer to capitulate to one civ at a time. I know there should be some code to encourage it to capitulate to the civ who damaged it most, but I'm not sure if it's just a tendency or a necessity.

To me, this would be one example where a lower capitulation threshold might affect the human player, as enemies might be more willing to capitulate to a third party - so it's really not just your choice as a player.

Bes Regards,
Ace

 

[Solver]

I don't have a save from that game anymore, but you guys have a good point, looking at it from the "best deal to survive" perspective. Hmm. I guess my problem is that it does sometimes result in really short wars - capitulation occurs when you think the action has just started.

Interestingly, in my case, I captured Hannibal's cities. Qin killed many of his units but I'm pretty sure I did more damage overall, and the AI is supposed to prefer capitulating to whoever did the most damage, which I think is a good feature.

 

[Willem]

(If you like the way it was in 3.13, then don't accept capitulate until you've mostly wiped them out.)

As I mentioned, I was already able to capitulate a civ after capturing only two cities in 3.13. So it must now be ridiculously easy to make them cave in.

 

[Willem]

So, my concern is not how capitulation limits affect the human player, because you can always refuse. Don't like it? Then don't accept. I don't need a patch to be my nanny.

You can't always refuse a capitulation offer because if you do, your enemy is going to turn to some other civ in order to gain protection, especially now with a lower threshold. The thing that will prevent that from happening is there's no civ that's willing to join in the war against you, which generally doesn't happen very often.

 

[Woody1]

I'm not quite sure, but I sometimes have the situation that I am waging war on a civilization. I am intending to capture their cities, therefore I do not go for capitulation... and then the civ I invaded capitulates to a third civilization which declared war on it as well, preventing me from capturing their cities.

Are you certain you can't keep attacking a civ that capitulates to a 3rd party? Why can't you just declare war on the 3rd party, and keep on rolling? Or are you forced into 10 turns of peace?