[GS] World Congress and Diplomacy Anticipation Thread

[Mr Jon of Cheam]

Is anyone else troubled by the fact that any one can vote on any number of possible resolutions, due to the drop down list? The votes are all scattered. There's no cohesion or sense of tension around whether the civs will vote for or against some specific proposal; instead everyone can vote on any one of 20 different possibilities. It feels off.

It's quite hard to say until we see it in action over the course of a full game but as I understand that will only be a problem for standard WC sessions. They mentioned there are also special sessions that will presumably give you the tension you're looking for.

 

[acluewithout]

@Infixo Well, I guess we'll see how it plays out come 14 Feb!

 

[Myomoto]

It also seems like resolutions does not expire. Whatever the outcome of the votes, it will stick until the end of time unless there will be another resolution that lifts or cancels the previous agreement. This I am still unsure if I am going to like it.

I believe resolutions last until the next sessions of the World Congress, per the Mali stream.

 

[MrRadar]

I believe resolutions last until the next sessions of the World Congress, per the Mali stream.

Judging by the screenshot in this post it seems that we get both sorts of resolutions: temporary (mercenary companies in the example) and permanent (urban development) ones.

 

[Republic of San Montuoso]

I don't like the randomization of resolution neither ; but if that could create some diversities in resolutions and not having 10 times in a row Theodora wanting to ban gold, I say: why not. At least we could have a world with actual global decisions and not just a world where everyone want everybody to be unhappy.

 

[Trav'ling Canuck]

The voting system is a bit complicated and has the potential to be extremely chaotic.

As I understand it, the structure is first to tally all the A votes and all the B votes. Presumably a tie means neither passes, otherwise the one with the most votes pass.

Then, within the winner of A or B, you look at the specific resolutions, to see which one of those received the most votes. And that's the game play effect.

Thoughts:

• There's no way to avoid a game play impact from a resolution, unless you can somehow finagle a tie between A and B. Something's going to happen from each vote.
  
• Will there be a way to ask an AI to vote in a particular way on a resolution? Or has this system from Civ 5 been dropped in favour of just buying that AI's votes directly and casting them yourself? I'm guessing the latter.
• Will there be a way to know which way the AI is leaning with respect to a vote? Could this be based on diplomatic access? Or has this been dropped from Civ 5, as it's an unfair human advantage? Note that it didn't need to be a procedurally unfair human advantage if the AI's votes aren't cast until after the human has voted and if the AI, at the right diplomatic visibility, got the same insight as the player does into the AI's intentions. However, in figuring out how to vote, the human will have an advantage so the more information available the easier it is for the player to get their way rather than things being purely random.
• I think people are right that the resolutions coming up randomly was likely an acknowledgment that it's too hard to teach the AI what resolution would be best for them to propose. That still leaves the question, however, of how to teach the AI how to cast it's votes in what is now a more complicated voting system than Civ 5.
• Even for the player, trying to guess how to vote may make your head hurt. Take the resolution about Diplomatic Victory points we saw early on. You can either try to add a Diplomatic Victory point (A) or subtract one (B). Assuming you don't have enough favours to just outvote the whole planet (which will be hard with the escalating favour cost per vote), you have to guess "Are they going to try to take a point away from someone or add one to themselves?" If you think they're all going to try and take a point away from someone, then you want to vote B, too, and throw enough votes into a sub-resolution where someone else loses the point and not you. On the other hand, if you think enough AI will vote to add a point to themselves that A will win, then you can weigh in with an A vote, too, and as long as you can cast one more vote for yourself than the next highest voter in A, you get the point.

 

[Infixo]

• I think people are right that the resolutions coming up randomly was likely an acknowledgment that it's too hard to teach the AI what resolution would be best for them to propose. That still leaves the question, however, of how to teach the AI how to cast it's votes in what is now a more complicated voting system than Civ 5.

It's just the opposite.
In Civ5 AI was programmed to score each resolution in terms of gains, but also in terms of loses for other players and how it was aligned with its current strategy.
In Civ6 there is no "current strategy" within the engine, nor any anticipation of others are doing. All the AI needs to do is to score gains from a particular resolution and it already does that when scoring policies, civics, buildings, etc.
More resolutions doesn't mean that the system is more complicated. Maybe for a human because of "wow, so many" effect. Computer can score 1000 resolutions in a milisecond.
I really recommend to play once with Vox Populi. AI in Word Congress is ruthless and unforgiving. You really need to make friends, otherwise you'll be outvoted each and every time.

 

[Republic of San Montuoso]

I really recommend to play once with Vox Populi. AI in Word Congress is ruthless and unforgiving. You really need to make friends, otherwise you'll be outvoted each and every time.

Playing with Venice, building all the diplomatic Wonders, buying all my buildings (more time to build envoys and such), have crazy gold with the Rialto District, constructing embassies everywhere, gaining influence by Merchant of Venice and buying city-states allied to others, completing the Statecraft tree... I had the World Congress in my hand. Really easy, I must say.

 

[sonicmyst]

I believe resolutions last until the next sessions of the World Congress, per the Mali stream.

The only resolutions in WC that expire as far as I know are the World Fair, World Games, ISS, and special sessions like asking for aid.

But all the rest as I understood it as permanent. Hoping I am wrong.

 

[Trav'ling Canuck]

It's just the opposite.
In Civ5 AI was programmed to score each resolution in terms of gains, but also in terms of loses for other players and how it was aligned with its current strategy.
In Civ6 there is no "current strategy" within the engine, nor any anticipation of others are doing. All the AI needs to do is to score gains from a particular resolution and it already does that when scoring policies, civics, buildings, etc.
More resolutions doesn't mean that the system is more complicated. Maybe for a human because of "wow, so many" effect. Computer can score 1000 resolutions in a millisecond.

Yes, the computer can do the calculations to score each possible options very quickly. I'm talking about teaching the AI how to score those options, i.e. what formulas to program it to use. If an existing base system can be used, great. I envisioned needing to teach the AI a separate scoring system for each different resolution.

Then there's also the issue that unlike in Civ 5 the AI doesn't get a fixed number of votes to cast each session. It also needs to decide how many votes to buy with favours for each resolution, with each additional vote costing a higher and higher number of favours. Presumably this can be handled by a standard procedure that weighs how important the resolution is based on the score output per above.

The AI casting it's votes naively I expect is something that will just be a fact of life. By that I mean the AI casting it's vote for its preferred resolution, say a subset of A, even though there's only a 1% chance that this succeeds if the AI supports it, compared to the AI casting it's votes for its second most preferred resolution, say a subset of B, which might have a 50% chance to succeed if the AI supports it.

Finally, the more predictable the AI is in it's voting, the less random things will be for the player. Which will reduce the degree to which the congress seems like an RNG-fest, but also increase the degree to which the congress benefits the player versus the AI, as the player will be the only one adjusting to the diplomatic environment.

 

[Infixo]

Playing with Venice, building all the diplomatic Wonders, buying all my buildings (more time to build envoys and such), have crazy gold with the Rialto District, constructing embassies everywhere, gaining influence by Merchant of Venice and buying city-states allied to others, completing the Statecraft tree... I had the World Congress in my hand. Really easy, I must say.

I see it this way: I chose the best civ for the goal, and did all I could to win, focusing all my effort on it. Yup. That is how you win in VP.

 

[Human Crouton]

This is a huge shortcoming of the WC feature then. Imho, it contradicts the core idea of the WC as a place where you can control things on a global scale.
I really can’t believe that anyone would consider that a good idea (hey, UN, what would you say if nations were to vote on random proposals?), so it seems like a result of simplification during design and implementation. Meaning: somebody said, hey guys we hate AI, so let’s make random proposals, that way we don’t have to code anything. Because modifiers are already in the game (and resolutions are clearly implemented as modifiers) and the only thing needed to make it work, is a scoring system. The game already has several scoring subsystems, so making another one is pretty easy.
That is so very disappointing.

Yeah, it's a really terrible idea. The WC is basically just a random event now instead of an actual tool the player can use with intent.

Random proposals, random disasters. Is there any new major mechanic to this game that isn't based on random chance?

 

[Eagle Pursuit]

Yeah, it's a really terrible idea. The WC is basically just a random event now instead of an actual tool the player can use with intent.

Random proposals, random disasters. Is there any new major mechanic to this game that isn't based on random chance?

Random proposals are better than the AI constantly trying to ban your luxury resources.

 

[Human Crouton]

Random proposals are better than the AI constantly trying to ban your luxury resources.

Then maybe they should have the AI propose other things. Just because you like an aspect of VI's WC better than V's doesn't mean that VI's version isn't terrible.

 

[acluewithout]

Judging by the screenshot in this post it seems that we get both sorts of resolutions: temporary (mercenary companies in the example) and permanent (urban development) ones.

No, I don't think so based on that screen shot. If the negative urban projects resolution passed, no one would ever again be able to build e.g. amphitheatres, art museums etc. That can't be right.

The difference in wording is more likely just FXS having not yet getting all the text right.

 

[MrRadar]

No, I don't think so based on that screen shot. If the negative urban projects resolution passed, no one would ever again be able to build e.g. amphitheatres, art museums etc. That can't be right.

Why? If you're one of the first to get and build those buildings, wouldn't you be tempted to monopolise them?
Although yes, I was meaning "permanent" rather in terms of "until next potentially successful repeal vote", not "permanently permanent", just one without an automatic expiry date, as in the mercenary example. The presence and absence of respective wording does seem quite intentional. They did think that embargoing city states forever was a nice option to have in Civ 5 after all. Once passed, it was almost hopeless to get repealed, unless you had majority of the votes.

 

[Human Crouton]

The WC being able to happen before you meet every player moves this game away from a competition against other players, and a competition against the game. Why are people I never met able to vote and have an effect on my nation? Why are the resolutions random? It's because we're moving from a game of player vs player to a game of player vs the game.

The game is actually becoming less interactive between the players, and just because they put the faces of the other players on a screen doesn't change that.

 

[Stringer1313]

It also seems like resolutions does not expire. Whatever the outcome of the votes, it will stick until the end of time unless there will be another resolution that lifts or cancels the previous agreement. This I am still unsure if I am going to like it.

I'm pretty sure it only lasts until the next WC sessions=

Judging by the screenshot in this post it seems that we get both sorts of resolutions: temporary (mercenary companies in the example) and permanent (urban development) ones.

Good catch. That is weird.

 

[Stringer1313]

I think we have to distinguish between two issues.

1) One is the random selection of proposals instead of letting the player or anyone propose a single specific resolution for all to vote on. I actually have no problem with this piece, and I think some randomness in the game is a good thing.

2) The separate issue is with the drop down menu, and the potential for civs to vote on 20 different proposals and scattering votes everywhere. This I have a problem with. I'd even be OK with the random proposal generator focusing on one item in the drop down menu to vote on (or maybe the proposal is random, but the "leader" gets to choose which item in the drop down menu to vote on). I think I understand why they are doing this from a gameplay perspective but it just feels kind of immersion breaking. When you "vote" on something, you typically have to know what you're voting on, and this system seems to break that fundamental premise.

 

[Eagle Pursuit]

I think we have to distinguish between two issues.

1) One is the random selection of proposals instead of letting the player or anyone propose a single specific resolution for all to vote on. I actually have no problem with this piece, and I think some randomness in the game is a good thing.

2) The separate issue is with the drop down menu, and the potential for civs to vote on 20 different proposals and scattering votes everywhere. This I have a problem with. I'd even be OK with the random proposal generator focusing on one item in the drop down menu to vote on (or maybe the proposal is random, but the "leader" gets to choose which item in the drop down menu to vote on). I think I understand why they are doing this from a gameplay perspective but it just feels kind of immersion breaking. When you "vote" on something, you typically have to know what you're voting on, and this system seems to break that fundamental premise.

1) I agree this isn't that bad. Yeah, it prevents you from using the WC as part of a longterm strategy, but it also prevents the AI from abusing you with annoying resolutions. Plus, selecting resolutions and then voting on them is an additional tedium that I am happy not to deal with.

2) I feel like this is something that we have to wait and see before passing judgment. It could be that it causes a lot of unpredictability in the results. And it might also make it too easy to just use a bunch of saved up Favor to gain plurality over the AIs multitudinous small selections. It really depends on how miserly the AI is about spending Favor for a result.