default header

Cosmic War

Galactic War discussion

Moderator: JC Denton

Unread postby jeffrobot494 » 30 Jan 2015 18:53

Another way to eliminate turtling would be to copy how PA eliminates it. Put some resource nodes on the map that encourage expansion. Maybe one or two systems in the middle of the galaxy give you +1 commander per turn if you control them.
User avatar
jeffrobot494
 
Joined: 26 Sep 2012 23:59
Location: California, USA

Unread postby icycalm » 30 Jan 2015 18:54

When making the rules I was thinking of the long term, not the first mini-war between us. And in the long term there will be clans with dozens of players in them (RLM has a couple hundred), so many of your objections about factions having too few commanders will be invalid. And in the hypothetical scenario when ALL the clans join in (which, after all, is our ideal and ultimate goal), we are talking about hundreds of players, at which point if you give bonus commanders to them the war could last for months, if not close to an entire year.

Nevertheless, your objections seem correct to me, as far as the first galactic war goes. I was planning on modifying and refining the rules WHILE playing, but from your analysis it is plain that 3 commanders per side is not going to cut it.

If we adopt your suggestions, we will have to make a clear distinction between players and commanders. So a team might end up having 10 commanders at one point, but only 2 players. The commanders can be used in the turn-based portion to cover more ground, but the team, having lost one commander at some point (and hence one player), can only field a maximum of 2 commanders for any given battle, regardless of how many it has at its disposal.

And what happens to the rest of them, when there are commanders in a system that can't be used? They automatically retreat to the nearest friendly system. If there are none, they are lost. (Which gives an incentive to capturing many systems.)

How about that?

We give each team 9 commanders to start with, and for every faction they eliminate they gain 2 more, as per gmase's suggestions, which are inserted at the system where the battle took place (and in the lore we explain it as the winning side repairing and reprogramming 2 of their enemy's commanders).

As for your encryption idea, it's a life-saver. I was racking my brain to find a neutral arbitrator, and couldn't think of one. Are you sure there's no way for a team to cheat, by using multiple encryption keys, and only revealing which one they used after the other teams have posted?
Last edited by icycalm on 30 Jan 2015 19:05, edited 3 times in total.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 30 Jan 2015 18:56

Just read jeff's idea.

It's a good one too. And we can save our coolest systems for this purpose, since battles will probably take place on them.

I wonder if we can combine both ideas, while not flooding the game with too many commanders. I guess that's what the first few wars will be about: experimentation.

And when we expand the team sizes, and/or the clans join in, we adjust the rules on the fly to compensate.
Last edited by icycalm on 30 Jan 2015 19:08, edited 1 time in total.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 30 Jan 2015 18:59

By the way, just in case it's not clear, my rule that a killed commander kills also a player was put in to remove the Clan Wars problem, where, even though all these big clans are involved, only their 5-6 top players actually ever fight, making the war feel nothing like a clan war. It is an absolutely essential rule that we must retain at all costs.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby jeffrobot494 » 30 Jan 2015 20:04

icycalm wrote:If we adopt your suggestions, we will have to make a clear distinction between players and commanders. So a team might end up having 10 commanders at one point, but only 2 players. The commanders can be used in the turn-based portion to cover more ground, but the team, having lost one commander at some point (and hence one player), can only field a maximum of 2 commanders for any given battle, regardless of how many it has at its disposal.


Having commanders that can't fight seems odd to me. "I have ten commanders in my army, but because my clan only has two players, I can't use them". What's the justification of that aesthetically?

A good rule might be that you cannot have more commanders than you have players. Whenever you would gain a commander, either from eliminating a faction or controlling a special system, you need a real player queued up to take control of that commander the minute it hits the map.
User avatar
jeffrobot494
 
Joined: 26 Sep 2012 23:59
Location: California, USA

Unread postby icycalm » 30 Jan 2015 20:20

You don't seem to be getting the practical issues involved with your suggestions.

We don't have more than 3 players per team for the first war. We just don't. So we can't have both gmase's rule AND yours at the same time. If you like gmase's rule for extra commanders at the start, and the possibility of gaining commanders during each war, you have to forget the rule you just suggested, and vice versa. We are trying to start a game here, so there's no point making idealistic posts that don't move us forward. I would like to be grafted inside my commander model in perfect virtual reality, but it's simply not possible at this stage, so we'll have to make do with a less immersive system. Don't be a weeaboo.

So, we've already explained why we need more commanders than players, mechanically: To gain map control instead of having everyone sit at their base and wait. And the aesthetic justification is the same: Why don't we have 5v5s in the first war, even though each team has 5+ commanders? Because the leaders prefer to send some away from a fight, scattering them across the system to maintain a stronger presence.

I know that this doesn't make 100% sense when a leader is fighting his last fight and cannot leverage all his commanders, because he doesn't have enough players, but we cannot allow the aesthetics to dictate every single mechanical decision at the cost of the fun of the game. If you want a better explanation, consider that each galactic war is merely one among many, and a leader might well decide to send his subcommanders to attack or defend more strategically important systems, so that's why, in some battles, he is not leveraging his full commander numbers.

Note also that, as I explained above, these are problems that will diminish in importance the more players and factions/clans start joining the war. It's okay for the first few wars to not be perfectly congruent aesthetically. I am quite resourceful anyway, and can concoct sufficient explanations for the aesthetic part of what is happening.

I mean, for that matter, what happens when a leader is killed? How does he continue fighting in other galaxies? I am sure I will be able to concoct some bullshit explanation for this. Moreover, I will HAVE to concoct an explanation, because if I don't we can't play the frigging game!

As for this:

jeffrobot494 wrote:A good rule might be that you cannot have more commanders than you have players. Whenever you would gain a commander, either from eliminating a faction or controlling a special system, you need a real player queued up to take control of that commander the minute it hits the map.


Not only does it blatantly contradict all of gmase's proposals, which you seem to have agreed to in general terms above, but it is even utterly unenforceable. I have already explain in the Galactic Clan Wars thread on the PA forum that we cannot bond each commander to each player, because players will not always be available to play when we need them. There will ALWAYS be absent players, and therefore commanders hanging about that can't be utilized, once we get a lot of players and clans/factions participating.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 30 Jan 2015 20:33

Another idea is to involve AIs somehow. But that's a huge can of worms I am not sure I am prepared to open. Aesthetically, and even practically, it will be awesome (practically, in the sense that you could still play even if many players could not make a specific date), but you have to consider what it will do to the actual battles, and what it will probably do to them is lower their quality significantly. No one wants to see a 10v10 between 2 players per side and 16 AIs. And that's what a lot of the matches would devolve to if we allowed leaders to use all their available commanders even if they don't have enough players at hand to control them.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby gmase » 30 Jan 2015 20:42

Yep I like that if a player loses all of his Commnaders he is eliminated "forever". Also the idea of systems that grant additional Commanders is great to incentivize exploration (exploration vs exploitation).

And jeff, we can explain it like this: There are 3 Commander minds (3 players), each one installed in 3 Commanding units. If all 3 Commanding units sharing the same mind are killed that mind is lost forever (well just forever in that galaxy because they have a backup somewhere else they can use for the next galaxy war xD ). The Commander mind can only control one Commanding unit at the same time so if there are more of his Commanding units in the system the rest stay idle. So if a team reclaims a new Commander because they control a special system, they can install on it one of the remaining Commander minds so no comming back to life of the defeated.
User avatar
gmase
 
Joined: 29 Jan 2015 20:50

Unread postby icycalm » 30 Jan 2015 20:50

gmase wrote:Yep I like that if a player loses all of his Commnaders he is eliminated "forever".


Ummm... that's not exactly how it works. Players don't "have" commanders. The leader has the commanders, and moves them around the map. He only calls players in to play when he decides to. And if they lose the battle, they all die. The winning side, meanwhile, only loses commanders, not players, because it would be very difficult in practice to tell which player died on the winning side.

Just wanted to clear that up.

Guys... We are just going to have to play the first war and see how this format works in practice. Past a certain point, more speculation will not be helpful.

If we see from the very first few turn-based moves that the system is flawed, we fix it on the spot and restart the war, without posting anything in the PA forum. Only once we have a good war going will we start updating the PA forum thread and the official site. It's all just posts in a forum thread, after all, up to that point.

And gmase, are you pretty sure that no one can cheat using your encryption idea? Obviously, we would never cheat between us, because we are all friends and know each other, but we would ideally like a system that can still work when a dozen clans have joined the war at some point in the future.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 30 Jan 2015 20:53

Or maybe the winning side too loses players, if some of its commanders are killed in a battle, but their leader chooses which ones have to leave the game? Call it "winner's advantage" or something. Which would actually be less advantageous than not having any winning players be eliminated. This way the leader will remove only the least experienced players, but it will still be something, and will ensure that the deaths of the losing players were not completely in vain.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby gmase » 30 Jan 2015 21:20

Ummm... that's not exactly how it works. Players don't "have" commanders. The leader has the commanders, and moves them around the map. He only calls players in to play when he decides to. And if they lose the battle, they all die.


I meant that the leader does all the moves and that specific commanders are assigned to each player. This will gives like 3 lives for each player. If a player is eliminated the first time he loses a battle, a team can be eliminated by losing just one 3v3.

And I like winning teams to lose commanders too so it can be a pyrrhic victory.

Anyway, as you say we can refine the system on the go. Let's see how it goes.
User avatar
gmase
 
Joined: 29 Jan 2015 20:50

Unread postby icycalm » 30 Jan 2015 21:27

jeffrobot494: i wanted to add in the forum that, when it comes to other people joining, intuitiveness of the rules is very important.
jeffrobot494: and the AI seem more intuitive than commanders retreating.
icycalm: but again you are placing higher priority to lower priority issues
icycalm: you don't get that the most important part are the BATTLES
icycalm: with shit battles it doesn't matter how good everything else is
jeffrobot494: and youre afraid a 20v20 with 16 AIs is less interesting than a 2v2
jeffrobot494: but im not sure it is
jeffrobot494: err 10v10*
jeffrobot494: but, up to you
icycalm: at that stage you might as well remove all the players and let the AIs fight between them
icycalm: in which case it's not PA, it's Sid Meier's Alpha Centauri
jeffrobot494: the players would still dominate the match
jeffrobot494: the AI isnt that good
icycalm: but you wouldn't be able to see which moves they made
jeffrobot494: you would, since the AIs would have to be on separate teams. no shared armies for AI.
jeffrobot494: err, separate colors not teams
icycalm: that's even worse
icycalm: no shared armies, etc.
icycalm: i guess we could give it a try
icycalm: but i have no faith in it
jeffrobot494: well lets do what youre excited about then
jeffrobot494: ill get behind it
icycalm: no, we should find a way to test this option too. if it works, it will make a lot of things a lot easier
icycalm: i'll copy-paste this chat in the forum, and we will find a way to explore it
icycalm: gtg now
icycalm: hf with the shitstorm on the pa forum
jeffrobot494: haha oh yeah
jeffrobot494: later

This is what we are referring to in the end there: https://forums.uberent.com/threads/why- ... st-1063257
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby gmase » 30 Jan 2015 21:51

I'm against using AI and I'm pretty sure most players from other clans won't like it either.

I see 2 problems with AI:

1. In a 10v10 in a system with several planets with 18 AIs and 2 people the outcome will depend mostly on how lucky you are with your AI players. Imagine 2 AIs face each other in a distant planet; the winning one will give a huge advantage to his team (maybe sending a few nukes right to the enemy player commander) and human players won't be able to do much cause they may be fighting each other in another planet.

2. Performance: AI intelligence has to be done by the server, but if we are to play huge games, the server may not be able to handle this extra work.
User avatar
gmase
 
Joined: 29 Jan 2015 20:50

Unread postby jeffrobot494 » 30 Jan 2015 22:01

gmase wrote:In a 10v10 in a system with several planets with 18 AIs and 2 people the outcome will depend mostly on how lucky you are with your AI players. Imagine 2 AIs face each other in a distant planet; the winning one will give a huge advantage to his team (maybe sending a few nukes right to the enemy player commander) and human players won't be able to do much cause they may be fighting each other in another planet.


The vagaries of war.
User avatar
jeffrobot494
 
Joined: 26 Sep 2012 23:59
Location: California, USA

Unread postby Agentx » 31 Jan 2015 03:06

Were we looking to do this at a particular time of day? It turns out I have exam/work Monday and Tuesday mornings my time, which I've realised is Sunday/Monday evenings UTC. Which is irritating because I'm free Sunday in my timezone. So what's the best way of arranging times for something like this?
User avatar
Agentx
 
Joined: 08 Dec 2011 16:33
Location: Australia

Unread postby icycalm » 31 Jan 2015 10:17

Guys, I've thought a lot more about this, and we are going to use AIs. AIs solve simply too many problems, and they are the only way to turn the turn-based phase into something that doesn't look stupid. Because it would look stupid if you have 20 commanders in a system but could only use 3 or 4 of them in a battle. So here's how I want us to use them. Hear me out and you'll see that this can work.

Each leader has the supreme commander and his subcommanders. The supreme commander is like the king in chess: defeat him and you kick the faction out of the system. On the galactic map the supcom is a unique piece. So everyone can always see exactly where he is positioned, and try to go for him, if they want to. So the leader is welded to his supcom; when the latter plays in a battle, so too must the former. And this is practicable too, since the leaders will be among the most dedicated players of each faction, so it should be easy to arrange for them to be present when the battle takes place.

The subcommanders, on the other hand, are interchangeable with the players, so any given player can be used to control any given subcommander at any time (within a given faction, naturally, not across factions).

Now let's say 3 subcommanders from faction A and 2 from faction B land on a system simultaneously. Together with their moves, the leaders will have had to state (in the encrypted messages) whether they are controlled by AIs or players. If all 5 subcommanders are controlled by AIs, 4 of them are destroyed, and 1 remains, since AIs are taken to be of equivalent power. Only if there is at least 1 player involved, from either side, do we actually play the battle. Full AI battles are decided instantaneously and automatically.

So, this will resolve the problem of bonus commanders flooding a system, and most of them being idle because there are not enough players to play them. All the excess commanders will be destroyed in AI battles that take place, not in the PA real-time system, but in the Cosmic War turn-based one.

Now why would a leader want to commit a PLAYER to a battle? Why not always simply use AIs?

Because players are incredibly more capable than AIs. A single player can take on 2 AIs easily, and the best players far more than that. Would these matches be interesting for the "esports" crowd to watch? Maybe not, but who gives a shit, dude? Are we making this game for the "esports" degenerates, who would rather watch games than play them, or are we making it for ourselves? Because, from the PLAYER'S perspective, these battles will be EXTREMELY exciting. Imagine that your faction is outnumbered, commander-wise, and your supcom commits you to a 1v3 fight against the AI. If you win, you can turn the tide of the war in the galaxy, but if you lose you are out of the fucking game! So maybe the viewers would not care to watch this stream, but for you and your teammates, AND your opponents, the fight would be nail-biting! So this is what we are going to do.

As for gmase's objection, that a 20v20 with 18 AIs and 2 human players would be decided mostly on luck, this may be true, but the TURN-BASED MOVES THAT LED TO THIS FIGHT HAD NOTHING TO DO WITH LUCK. So some of the responsibility for the outcome of fights has to rest with strategic, turn-based moves, otherwise why add a turn-based dimension to the game at all?

So, to start with, each faction in the first GW will get 9 commanders. Defeating a supcom (and thus wiping a faction out of the galaxy) gives you 2 extra commanders, and there will also be a couple of systems per galaxy that give you an extra commander per turn.

These numbers will be tested. So we will start with 9 initial commanders, 2 bonus commanders per supcom killed, and 1 commander per turn of holding a crucial system, but we will adjust them during the first war. We might even have to reset the war and replay it until we get the numbers right. Or make any other adjustments that seem and prove to be beneficial. And only when we have played a full war to its successful and enjoyable conclusion will we post the results of it on the official site, including streams and videos, etc., and announce recruitment for the beginning of war #2.

Also, I want to make a COSMIC map too. I want the systems to be connected, just like planets are connected in the galactic map, so that factions can choose which systems to expand to, which to defend, etc. But there is no point discussing this before the first galactic war is over, so let's not. It's too much complexity and too many possibilities to handle at once. Let's make galactic war work properly, and then we'll think about the cosmic war.

Also, gmase, this is the third time I am asking you about your encryption idea. Is there a way for someone to cheat in this? Please answer me so I can stop thinking about it.
Last edited by icycalm on 31 Jan 2015 10:37, edited 2 times in total.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 31 Jan 2015 10:19

And Robby, the matches will be played as soon as the players concerned can arrange to play them between them. There will be no regular schedule, at least until some clans decide to join us and things become more complicated.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 31 Jan 2015 10:28

To get back to that 20v20 example that gmase talked about, it would be stupid for a leader to commit a player to that match, because a single player would probably not be enough to decide the outcome of the battle. So both sides would probably use full AIs, in which case the match would be instantaneously decided with all commanders from both sides being destroyed.

However, if the clans join at some point, they could easily commit 5 or 10 players to such a battle, at which point if the opponent did not do the same, they would surely lose the battle! So the loser would lose 20 commanders, but the winner would probably not lose more than half a dozen or so. Which would be a huge win for the winning side!

As for the AIs introducing lag to the matches... this is a technical issue that we will have to work around the same way we will work around all technical issues. For example, a 10v10 between 20 players on a huge system with multiple spawnable planets. That will be very laggy too, so how will we deal with it? Or a server crash. What happens when you get one of those, and the battle is interrupted in the middle?

We will just have to come up with a "random event" like the ones used in board games like Monopoly or whatever, that postpones the fight or moves it to another, more appropriate system. A solar flare or something. We'll figure it out.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby gmase » 31 Jan 2015 12:22

Sorry icy, forgot about it. I think the encryption algorithm they claim to use (AES) is completely secure for this purpose.

I'm not strongly against AIs but I think it can discourage many players from other clans to join. Anyway I think it's a pretty good solution for this first war with few players involved.
User avatar
gmase
 
Joined: 29 Jan 2015 20:50

Unread postby icycalm » 31 Jan 2015 13:19

Yeah, I think we will try to scale the game up slowly. So instead of allowing a FULL clan to enter, and upset everything, we will be allowing INDIVIDUAL players to sign up.

Let's say 4 Promethean guys decide to sign up. Then we throw them into the war as the newly-formed Promethean faction. They don't even need their clan leader's approval to join, so there will not be such huge pressure on the leaders to commit their ENTIRE clans immediately. In this way, I am sure we will be able to attract small groups of players into the game, and scale it up slowly, in small steps.

If, as the game is scaled up, AIs prove to be a problem, we will slowly phase them out as real players join and take their place.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby icycalm » 01 Feb 2015 03:56

Guys, here are the rules for the first GW.

Each team gets 3 commanders to start with. Exactly the same as their number of players.

There are no homeworlds to defend. It doesn't make sense that they would have homeworlds on systems they just arrived at. We will figure out a way to use homeworlds once the factions have a number of galaxies under their control. I merely borrowed the idea from Clan Wars, but I haven't figured out how to adapt it to our needs properly. The point is, we don't need homeworlds to introduce strategy to the turn-based movements. You've seen the map in the artwork thread. The 4 stars around the central square will give +1 commander/turn, and all the factions will spawn one star away from them. So right away we'll probably get 1 or 2 battles, on one of these valuable systems, and then a chaotic rush for the rest of them. The whole thing will probably be over in under 5 or 6 turns, and there won't be much time for these systems to generate many bonus commanders and flood our battles with AIs. The peripheral systems will mostly not be used, but they make the galaxy look good, so we'll keep them.

So all factions will want to grab, and hold on to, as many of the 4 valuable systems as they can, and smash the other factions, and gain the galaxy. So there is plenty of incentive to start moving right away. Or a team could choose to not move, and see how things play out with the others in the meantime, but then that will give an advantage to whatever team DOES move, and thereby gains a valuable system unopposed, and an extra commander for the second turn. Or a team could divide its forces, and try to go for 2 valuable systems. So even though the galaxy is so small, and everything is packed densely together, there are enough options to make sure there is some strategy involved at the turn-based level, but not so many as to take a lot of turns and complicate things too much and get difficult to handle, and perhaps even boring. So I think it is a perfect starting galaxy and a perfect starting set of rules, for our purposes.

There'll be some tense fights, some lucky turn-based decisions involved, and a little strategy, and within a few moves one faction will have gained a galaxy. And then we can evaluate this experience and think about how to proceed with the second war. We'll add a few players, and maybe some rules, and take it from there.

If you have any important objections that can't wait until the end of the first war, post them here now and I'll consider them.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Unread postby gmase » 02 Feb 2015 21:42

I think we have almost everything to start the first galactic war. I'm not gonna make any more suggestions until we play and see how the game develops.

However 2 points:

1. We have to clarify what happens if enemy comms intersec while moving aroud the galaxy. For example: Comms from system A going to B, comms from B going to A and comms from C going to A.

A---> <---B
^
|
C

We have to set an arbitrary rule to decide where the battle takes place.

2. I propose this notation to describe the moves: I move N comms from system X to system Y (and repeat). I think that's all you have to say in a galactic turn. That message encrypted will be posted by each leader in a thread. After all the leaders hadveposted, they reveal the uncrypted message and password. If the movement is illegal the comms of that faction remain still.
User avatar
gmase
 
Joined: 29 Jan 2015 20:50

Unread postby icycalm » 04 Feb 2015 00:55

We can just say that the move of whoever posts his encrypted move in the thread first takes precedence. This should also help make people post their moves faster, and keep the turn-based game moving along at a nice pace.
User avatar
icycalm
Hyperborean
 
Joined: 28 Mar 2006 00:08
Location: Tenerife, Canary Islands

Previous

Return to Cosmic War