Longturn.net

akfaew

Administrator

Posts: 622

Shared victory - max alliance size

Let's assume that currently victory can be shared among 4 players. aloril suggested we should change this setting. Please write your propositions here and we'll have a poll later.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

With a 30 player game we usually had a winning alliance of up to 6 players, I think most players agree that this is a fun and honorful way to win. With 70 players that would extrapolate to 14 players.

Another way to think about it: usually about 2/3rds of the starting players are still alive and active in the middle phase of the game, and strong enough to stand a chance to win. If those are divided among 3 alliances it creates a fun and interesting situation. That would be about 15 players per alliance.

20 players may be an upper limit, beyond that it becomes too easy to gang up and get a boring endgame. Below 10 players it may become too difficult to win and the endgame could get stretched out too much.

akfaew

Administrator

Posts: 622

Re: Shared victory - max alliance size

By the way, I made a thread about a future feature, an Alliance Manager:
http://forum.longturn.org/viewtopic.php?pid=332

jhh

Player

Posts: 79

Re: Shared victory - max alliance size

IMHO I don't think there needs to be limits for this big games except that the game wouldn't stop. It's not very likely that this big game will end because everybody joins up and starts a peace cult. I don't even think it would never be 1 vs 1 for a long while -- there might be epic world wars but I don't believe those would last long.

Also an alliance in the game is also something that might mean anything. We at freeciv.fi usually have made in-game-alliances just to share roads etc. I don't think it needs to have any special meaning outside the standard in Freeciv -- except maybe the limit to end the game, if everybody has alliance with everybody, but that has never happened for us and I'm pretty sure it would never happen.

Btw if it's not fun in the long run, I don't think people would even be part of a big alliance. There would be (real) civil wars and all would be fine in the world of warfare...

Aloril

Player

Posts: 8

Re: Shared victory - max alliance size

Some alternatives:
number_of_active_players_at_T30/2.5
number_of_active_players_at_T20/3

Alternatively number of alive players.

These might give similar numbers to what Marduk said?

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

I think the proposal is to leave players free to make any alliance they want during the game, but to prevent the game from ending until there is an alliance below the maximum size that claims victory (and all other living players accept their victory).

So far almost all LT games had a clear beginning and end, with several strong alliances battling it out until one survives and claims victory. The worry is that if lots of players gang up to win the game together the end may come too soon and they'll get victory too easily.

But it's definitally possible that with a much bigger player list the game becomes more like an ongoing cycle, which would be quite cool as well. There have been players before who play pacifistic and try to create a kind of United Nations rather than an alliance that tries to kill all others, would be very interesting if that happens in this game. But just in case the game has a more standard ending it would be good to set a maximum size.

akfaew

Administrator

Posts: 622

Re: Shared victory - max alliance size

We must also decide whether to end the game at turn T160 and then count scores, or agree on some other form of game end if no alliance is victorious.

monamipierrot

Player

Posts: 171

Re: Shared victory - max alliance size

What I understood from differnt posts and discussions of international LTurners vs. the finnish LTers, I think they're  very different game concepts and "traditions". The international LT had been (always?) bound to the overall ranking: points and victory were the only thing that counted. An aggressive gameplay was developed, and bigger players agreed in forming big gangs to destroy everyone else as soon as possible. Secrecy and intelligence were very important, a blog like Jhh's would have been impossible to think of.
The finnish instead played more "normal", everyone playing for his own pleasure and goals. Thus war was mixed with peace, UN-like organizations were forced to exist to help the weaker aganist the bad ones. Alliance were formed only on focused tasks (am I wrong?), not to win all together.

In my opinion, the finnish concept has several advantages: a more "natural" way of playing, giving a chance to the smallest. But it has the same disadvantage of international gameplay: there's not a superior goal to achieve.
What I mean is: in Int LT you can win with other 5 players (is this a VICTORY???), but in Finnish one you can't tell who is the real winner, cause everyone played his way.

In my modest opinion, there should be ONLY ONE winner. Up to him how to achieve victory, by having anyone else surrender. Alliance would be still important to win middle game wars, but you would watch your back cause you know that all your allies want to win the game against YOU. This is more or less how History worked on our Earth.

Thus, I would vote for "1" winner. If that's not possible, I would vote for not having any maximum number, cause it doesn't make sense. If there are 50 players who want to gangbang the other 20, they're free to do it, but it would be much less fun for all 70 of them. Maybe next time they will change gameplay.

What is VERY important to do is to heavily PUNISH gangbanging-for-points by heavily decrement points for shared victory, e.g. using current point sistem but adding "/FinalNumberOfWinners" at the end. This would be enough for many players to make them try to break huge alliance and start internal wars, and having much more fun.

Not to set any "superior" limit with the new >30-players-LT would be recommandable for another reason: it would be very "sociologically" interesting to see what happens: will gangbanging still exist? Will internal revolutions prevail? Will economical-driven peace still be the winning choiche?

My 2 cents.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

Some alternatives:
number_of_active_players_at_T30/2.5
number_of_active_players_at_T20/3

Alternatively number of alive players.

These might give similar numbers to what Marduk said?

That method also makes sense, though T30 may be too early too tell how the game will develop. Also some players like to create their alliances early in the game and would like to know what eventually will be the maximum size to win. Perhaps it's best to start the game with a fixed number so everyone knows what to expect.

We must also decide whether to end the game at turn T160 and then count scores, or agree on some other form of game end if no alliance is victorious.

Hard to say, this game will probably be longer than most but I think it will also stay fun for a longer time (if a 20-player game drags on beyond 160 turns it's likely that it's in a deadlock, but a 70player game could be alive and well much longer). So I'm not sure if we should put a strict limit on the game length. We can also think about a procedure (for example counting up scores as you propose) and let it take effect after a poll decided that most players think the game should end at a specific turn. Such a poll would only be proposed if the game ends up in a deadlock and people think it's getting boring, which usually doesn't happen.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

What I understood from differnt posts and discussions of international LTurners vs. the finnish LTers, I think they're  very different game concepts and "traditions". The international LT had been (always?) bound to the overall ranking: points and victory were the only thing that counted. An aggressive gameplay was developed, and bigger players agreed in forming big gangs to destroy everyone else as soon as possible. Secrecy and intelligence were very important, a blog like Jhh's would have been impossible to think of.

That's more or less correct. We've developed the habit that (almost) everyone plays in order to win the game, but with a gentleman's agreement that big alliances gangbanging the rest is dishonourful. Really big alliances almost never happened, it's nearly always 4 to 6 players. The biggest I've ever seen was like 12 out of 30 players, and most people agreed that this was not a very impressive victory. Some players care more about winning while others just want a fair and interesting game whether they win or not. The idea that the game ends with a winner and that there are rules about what constitutes an honourful victory tends to prevent cheating or sabotaging (e.g. if you join an alliance and then pass strategic information to the enemy, not because it's your strategy but just for fun). In this way I think players who don't care much about winning still benefit from the tradition since it keeps the game clean and transparent. Jhh's blog shouldn't be a problem as long as his allies don't mind it (which means players who play to win may not be willing to ally him, but that's everyone's free choice).

A more mixed game with pacifists, UN-like org's etc also sounds like a lot a fun, though I think it can co-exist with the competitive game-style. Players who prefer this style will probably band together to prevent a competitive alliance from attempting to reach victory, and then disband the alliance when the threat has been solved. For competitive players this is a nice challenge to make winning the game a bit harder, and for the UN-players it's a challenge to make their model of co-operation and co-existence work against hostile enemies. If in the end nobody succeeds to win the game in the "international LT" style then we can always decide on a different method to end the game (if we want to end it).

Having a ranking system and rules about shared victory may seem like a limitation on possible gaming styles, but if enough players decide to play in a different style then they can prevent the "race for victory" from occuring. This doesn't require any rule changes, just that the UN-style players organize themselves well. I'm all for it, though I'd probaby be one of the people trying to destroy the UN

We've played games with a single winner before, and one time a player actually succeeded in winning a game single-handedly. But it took ages and in the endgame players needed to devote more time to the game than they really wanted. The idea of having a real "5-minutes per day" LT game seems only practical if players can win together in an alliance, otherwise it's just too much work. We should definitally include an option of "1" in the poll for max alliance size, but personally I would not advice people to vote for this. In our experience people simply enjoyed games with winning alliances more than games with only 1 winner. 

What is VERY important to do is to heavily PUNISH gangbanging-for-points by heavily decrement points for shared victory, e.g. using current point sistem but adding "/FinalNumberOfWinners" at the end. This would be enough for many players to make them try to break huge alliance and start internal wars, and having much more fun.

Currently the rules include two ways for deterring gangbanging-for-points: the bigger the winning alliance (as a share of the number of starting players) the less points are redistributed from loser to winner. A game-ending with a very big winning alliance will result in very little points being distributed to anyone. Secondly those points being distributed are then divided among the winners, which means that a big number of winners will have less points per player. If these deterrences are not strong enough we can think about ways to strengthen them.

There is another social convention involved in this: back in the early days I once broke up an alliance which I thought was too big, and killed a former ally in an internal war. This player then went on to sabotage me in several later games just to get vengeance. In my eyes it's perfectly fine to break big alliances, have internal wars, play evil diplomatic strategies etc. It's part of the game and adds a lot of suspense and uncertainty. But it only works if other players share this social convention. It should be not done to have personal vendettas that cross from one game to the next, it spoils the fun.

But maybe we're getting into a different topic here

akfaew

Administrator

Posts: 622

Re: Shared victory - max alliance size

monamipierrot, I think you're talking nonsense by suggesting only 1 victor - this is simply impossible. You claim to not care about points but it seems it's your most favourite topic to talk about. Chill man, and let the rest of us compete with each other. Marduk has always maintained the ranking and I strongly believe he knows what he's doing.

What about: at some point make a special poll: "end game?". If 75% live players vote "yes", then the top x (5?) players, as dictated by score, claim victory. The downside is: a sudden 7 day race for points.

akfaew

Administrator

Posts: 622

Re: Shared victory - max alliance size

We should definitally include an option of "1" in the poll for max alliance size

As I said - impossible. With 200 cities the client would have to compute way too much. And now there are 220 cities already.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

What about: at some point make a special poll: "end game?". If 75% live players vote "yes", then the top x (5?) players, as dictated by score, claim victory. The downside is: a sudden 7 day race for points.

Sure, I think that would work.

We can let the number of turns before endturn open for the one who proposes the poll, and only keep the 75% as a strict rule. If more than 25% of the players disagree with a quick end or a short race for points they can vote "no" until an endturn is proposed that is far enough in the future to give them a fair chance to win.

Perhaps let the top X be decided as the top 25% of players, rounded down? So if there are 20 left the top 5 in points would be the winners.

Since 75% support is needed to set this process in motion I think most games will still end with a winning alliance rather than be decided on points. But if 75% of the players play a "UN-style" game then they can use this rule to have a game ending compatible with their playing style (you can keep peace during the entire game and then agree that the top players in points are the winners). That way the two playing styles both have a chance to dominate the game, and the most popular one will win. Sounds fair to me.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

We should definitally include an option of "1" in the poll for max alliance size

As I said - impossible. With 200 cities the client would have to compute way too much. And now there are 220 cities already.

Ok that's a good point, if it's not technically feasible for the client then we should not include the option as a possibility (we'd be voting "shall we crash the game yes or no?").

monamipierrot

Player

Posts: 171

Re: Shared victory - max alliance size

There is another social convention involved in this: back in the early days I once broke up an alliance which I thought was too big, and killed a former ally in an internal war. This player then went on to sabotage me in several later games just to get vengeance. In my eyes it's perfectly fine to break big alliances, have internal wars, play evil diplomatic strategies etc. It's part of the game and adds a lot of suspense and uncertainty. But it only works if other players share this social convention. It should be not done to have personal vendettas that cross from one game to the next, it spoils the fun.

That's interesting, is more or less what I thought of, even before joining my 1st LT (LT29) game, only by reading some of your old posts.
A Game should be unique. Sometimes you want to be some "Alexander", sometimes you want to be the UN headquarters, sometimes you just want to be the Hitler or the Star Wars Emperor of the Game.
Even worse than personal vendetta are undestroyable frendships that cross through games.
These prevent you from betraying (cause in next "life" your ally will remember this) and thus from adding some spice to the Game. But a social convention should be: forget about ranking, real life, last or next Game, and so on.
This time I heard there are some real life fathers & sons playing at the same time. I hope they end up slaughtering each other, or even better the father being the weak - or servant - allied of the son. A Game is fun because is different from real life....

But maybe we're getting into a different topic here

Ooops!!!

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: Shared victory - max alliance size

There is another social convention involved in this: back in the early days I once broke up an alliance which I thought was too big, and killed a former ally in an internal war. This player then went on to sabotage me in several later games just to get vengeance. In my eyes it's perfectly fine to break big alliances, have internal wars, play evil diplomatic strategies etc. It's part of the game and adds a lot of suspense and uncertainty. But it only works if other players share this social convention. It should be not done to have personal vendettas that cross from one game to the next, it spoils the fun.

That's interesting, is more or less what I thought of, even before joining my 1st LT (LT29) game, only by reading some of your old posts.
A Game should be unique. Sometimes you want to be some "Alexander", sometimes you want to be the UN headquarters, sometimes you just want to be the Hitler or the Star Wars Emperor of the Game.
Even worse than personal vendetta are undestroyable frendships that cross through games.
These prevent you from betraying (cause in next "life" your ally will remember this) and thus from adding some spice to the Game. But a social convention should be: forget about ranking, real life, last or next Game, and so on.
This time I heard there are some real life fathers & sons playing at the same time. I hope they end up slaughtering each other, or even better the father being the weak - or servant - allied of the son. A Game is fun because is different from real life....

Exactly my thought as well! I've been the "evil diplomat" in several games but also the loyal ally in other games. It's fun as long as people don't take things personally and keep an open mind.

I also agree with you on the undestroyable friendships across games. They didn't happen often, but there have been times in which one game after another was won by the same group of players who didn't admit anyone else into their little club. There's no way to make a rule against this (instead of randomizing team games), but I definitally hope we'll have social conventions against this.

monamipierrot

Player

Posts: 171

Re: Shared victory - max alliance size

Perhaps let the top X be decided as the top 25% of players, rounded down? So if there are 20 left the top 5 in points would be the winners.

Why not set NumberOfWinners = sqr(AlivePlayersAtTheTimeOfThePoll), rounded up?
If there are still 70 players, it would be 9, with 20 it would be 5, with 10 it would be 4, and so on.

just my 2 nonsensical cents.

Aloril

Player

Posts: 8

Re: Shared victory - max alliance size

We must also decide whether to end the game at turn T160 and then count scores, or agree on some other form of game end if no alliance is victorious.

Hard to say, this game will probably be longer than most but I think it will also stay fun for a longer time (if a 20-player game drags on beyond 160 turns it's likely that it's in a deadlock, but a 70player game could be alive and well much longer). So I'm not sure if we should put a strict limit on the game length. We can also think about a procedure (for example counting up scores as you propose) and let it take effect after a poll decided that most players think the game should end at a specific turn. Such a poll would only be proposed if the game ends up in a deadlock and people think it's getting boring, which usually doesn't happen.

Looking at server settings it seems we have space race enabled and that it ends the game. Given this many players and space for cities I think it will end in space race before T160 in any case unless it ends otherwise earlier.

Aloril

Player

Posts: 8

Re: Shared victory - max alliance size

We should definitally include an option of "1" in the poll for max alliance size

As I said - impossible. With 200 cities the client would have to compute way too much. And now there are 220 cities already.

I think 200 cities is possible, but I think it would be more like 600-1000 cities in the end.

IllvilJa

Player

Posts: 51

Re: Shared victory - max alliance size

(oh... long, probably incoherent, post... don't have time and probably neither the needed brains and intellect to get it structured either...)

Hm... I recall this kind of discussion from the BattleField sites in Sweden, where some people were so hardly bent on the competetive e-sport way of playing the game that they seriously believed that their way of developing as a gamer was the only possible way and thus the only one worth even keeping when trying to get more out of the game.  They completely missed out the gamers which were not bent on developing as players based on competetive e-sport but based on other ways to seriously play the game, in my case those who play for the experience itself.  I used the word "wargamer" as the name for us who played the Battlefield games without trying to measure 'who is the bes' but instead focus of the made-believe combat experience, I'm not sure that really would work here but I'll go for it for the conext of this post.

So, the "Finnish LT-game tradition" seem to be more "wargaming" or perhaps "nation-ruler-gaming" where one enjoys acting as a ruler of a country who's fate is in one's hands.  I recall stumbling over some cathegorization of gamers which would sort the Finnish tradition under the two categories "Simulationists" and "Role-players"  Keeping points and ranking as is seem to be done in LT seem, on the other hand, to be more like e-sport, at least in my eyes.

Personally, I prefer the 'simulationist/role-playing/wargaming' way of playing these games with all it's associated drama and atmosphere and I prefer a gaming system which provides room for those things.  At least in the Battlefield games I have played (BF1942 and BF2, AGES ago), this meant I quite liked certain elements in the games which, in my eyes, contributed to the gaming experience and the war-like atmosphere, but which made it harder to determine whether a certain player were better or not than another certain player. Example of such features are the presence of ground vehicles in the game, artillery barrages (that made you fly ca 50 meters up in the air if you were hit), VERY deadly attack planes wiping out a lot of infantry and ground vehicles during each strike and so on.

Same goes for how I view this freeciv campaign.  If there are design choices that improve both the competetive e-sport experience (things that simplifies measuring who's better than who)  as well as the 'wargamer/nation-ruler/simulationist/role-player' style of viewing the game, then I'm all for them.  However, if there are design choices that lessen the value of the non-competetive 'wargamer/...' aspect of the game, I wouldn't really support them with a vote (even if I would still could accept them if they ended up being implemented).  I have to admit, I'm not fully clear if max alliance size have ANY impact for non-competetive wargamers and if so, what value would benefit us the mos ;-).

(Random thought: I really like the style of gaming that monamipierrot describes above, BTW)

So, I really don't care how points are calculated or how people are allowed to win or lose or so as long as this criteria is met: There is room for people to play the game in a non-competitive 'wargamer/nation-ruler/simulationist/role-playing' way.

(Actually, I would love to make a separate agreement with Marduk that he always gives me zero points on that ranking table he keeps, so I can let my nation behave any goddamn way I like without anyone complaining that it resulted in me getting points I did not deserve due to not following some subjective convention of how to play the game they have happened to make up.  Probably such an agreement would be going too far but I'm so seriously tempted.)

But the game have some logic in it which says 'Game over', doesn't it?  We got space race enabled, and as far as I can tell, the server setting 'End Year' is 5000.  So decide some value for max number of players to get points for an alliance victory (a value I'm quite indifferent of) and then run the game until the game logic claims the game is over.  THEN, if too many players are part of the winning/surviving alliance, then none of them get any points in the ranking system (and if that situation occurs, I suspect none of the winning players CARE about getting any points in the ranking system).

Another way to put it: anything that adds epic drama to the game is a good thing!  That includes offending some more powerful superpower in the game by sinking their nosy destroyer and get ones own entire country nuked into oblivion in return!

Just my 0.02 cents of incoherent thoughts :-).

/Illvilja/Sweden who's currently enjoy gardening a piece of land in LT30.

Last edited by IllvilJa (2012-01-08 23:45:32)

munk

Player

Posts: 36

Re: Shared victory - max alliance size

Looking at server settings it seems we have space race enabled and that it ends the game. Given this many players and space for cities I think it will end in space race before T160 in any case unless it ends otherwise earlier.

if this is the case, would it allow for an alliance winning the game via space race if one of its members gets a ship to AC?

akfaew

Administrator

Posts: 622

Re: Shared victory - max alliance size

Damn, how the hell did space race get enabled? That's not right.

It's clearly disabled in the settings:
set spacerace=0

IllvilJa

Player

Posts: 51

Re: Shared victory - max alliance size

Damn, how the hell did space race get enabled? That's not right.

It's clearly disabled in the settings:
set spacerace=0

It could be my client being buggy when reporting the server settings, but as far as I can tell from it, spacerace is enabled.

But it is good to know whether or not it will be something that ends the game :-).  Sure, it's just turn 5, so I don't expect anyone to really have adaped their country development to any assumed absence/presence of the space race.

monamipierrot

Player

Posts: 171

Re: Shared victory - max alliance size

Another way to put it: anything that adds epic drama to the game is a good thing!  That includes offending some more powerful superpower in the game by sinking their nosy destroyer and get ones own entire country nuked into oblivion in return!

Great. This time we went VERY off topic, so I will continue the talk here: http://forum.longturn.org/viewtopic.php?id=77

Kryon

Administrator

Posts: 370

Re: Shared victory - max alliance size

I agree it is important to put a limit on winning alliance size. We can not easily enforce max alliance size unless a patch is written in freeciv code but we can discourage big alliances by time limit + max number of winners. I suggested doing that in LT28. The limit was set to N/5 players (N being # of players).  To enforce the rule in LT28, we set the endgame to T180. If more than N/5 players claimed victory at T180, game would end without any winners. The game ended up with 3 strong players vs the rest (~10 cooperating players, 6  of which were formally allied). The small alliance won the game but it would have been interesting to see what would happen if the big alliance beat the small one and then had to split to win the game.

As for LT30, T180 would be a very early end. I believe this game will either be won by space race or, if spacerace is disabled, by nuclear warfare. Although it had originally been my idea, I suggest we don't put any limit to alliances in LT30 and see what happens and then try to come up with a solution for LT31 only if LT30 had some unfairly large alliances. On the other hand, I believe Marduk's ranking system is a really good way to discourage large alliances as small winning alliances get much more points than larger ones.

Last edited by Kryon (2012-01-09 12:07:06)