There's multiple things about multiplayer that just seem off. Take this battle as an example: http://zero-k.info/Battles/Detail/125870If you are in team games room, !balance will tell people what the matchup would be if the game started right now. If there is an odd amount of players, the faction with more players can start the game without the agreement of the outnumbered team. So if i don't pay 100% attention i might be caught up in an unfair matchup. Looking at the above, 1600 and 1300 don't match up with 1450 elo. The 1600 guy should be playing alone with 2 coms, if at all. And the fact that the game can start without the agreement of the solo player makes this even more of a pain. And what is the result in this case? The solo players lost 20 elo at once, in a game he likely didn't want to play and had basically no chances of winning (except for if he can play for 2, which the highest points player would've been more likely to). This just doesn't seem right to me. If one team is outnumbered, the other team shouldn't get massive elo for winning. On the other hand, look at this battle: http://zero-k.info/Battles/Detail/125622We managed to win 2v3 (needless to say, neither of us wanted to start the game). The 1800 elo guy lost 3.7 elo, the 1400 guy 18. The 1330 guy also only lost 3.7 elo. What the hell is going on here? So guys, what's your take on this?
+0 / -0
|
As far as I know the 1-man-team can just exit/resign before the game starts and it won't count. Problem solved! :) Fish3r lost most elo because their elo was the most unreliable. Teammates got a high level, which means many games played and their elo is pretty accurate, thus doesn't need a lot of adjustment.
+0 / -0
|
Are you sure? I'd assume that's just an instant elo loss, but i'd like to be proven wrong. And for 2v3 (which is hardly ever balanced) i don't expect this to work either. Still doesn't fix the horribly weird elo wins/losses from those games. Edit: Well i can't say that i agree with this method... It's not like the newb should be "punished" for losing like this. To elaborate what i mean with "punished": Elo should describe skill in some way. The one who played worse than what his elo suggests is not the 1400 elo newb but the 1800 veteran. Also, looking at the first battle, the high elo guy got just as much as the low-elo one (10!) for winning 2v1 against somebody 200 elo below him. Wtf?
+0 / -0
|
@2v1: Like I said, elo loss is based on reliability. Both players have a pretty high level, so their elo is already quite accurate, thus both get the same change.
+0 / -0
|
If their combined elo is really close to the solo guy's elo, and they have the advantage... Why do they get so much? I mean, the solo guy lost 20 (!) elo in a match where the other team was favored...
+0 / -0
|
the one man cant exit on his own. dont know if it was possible before the exit rules were changed, but exactly because if situations like that i was against the change - one team gets outnumbered and forced into unfair game.
+0 / -0
|
Having someone who is worse than him draining half his resources makes the match much more difficult. A 1600 player with two comms is simply not balanced vs a 1300 and a 1400 player. It is not very much 'harder' to play with 2 comms, if they're right next to eachother and you're the only player on the team. If you don't like the way the game is balanced, don't vote to start, signal to the other players you are unhappy with the team comps, vote exit if the game starts prematurely, and resign if you are 'forced' to play.
+0 / -0
|
Keep in mind that elo changes are higher in games with low player numbers. So 10 and 20 are not really that much.
+0 / -0
|
What     Saktoth said.It is more balanced to have an 1600+1300 vs 1450 then a 1600vs1300 +1450. In the first situation the 1600 ELO player has a resource leech which is a disadvatnage. In the second example the player can spent all of his resources better since he is 1600 ELO. Take a loot at this game for example http://zero-k.info/Battles/Detail/125786It was me 1650ish ELO and a 1400 ELO player vs two 1500ELO players. Now notice that as soon as my 1400ELO team mate resigned, my resources doubled and I could steamroll those two players with ease.
+0 / -0
|
Right now, the system acts as if the solo guy were actually two players, which is obviously not the case. Which one will perform worse: 1 player with twice the ressources, or 2 same level players with equal ressources? It's not like you're sitting there chilling half the time playing RTS games. You're usually fully busy with managing one economy, and having twice the load to handle just doesn't make things better. Btw, will elo be distributed if i resign my "unfair" game? If it is, then there's no point in resigning.
+0 / -0
|
@MauranKilom , because of Zero-K's unique controls it's actually a quite "relax" game. Ofcourse you can micromanage all your units to infinite detail but Zero-K's fight command doesn't do that bad. The ease of using longer command structures and options like "area" mex allow you to scale quite easily. You just have more freedom in spending your resources I feel. For example if you have two seperate players with two factories, like a Jumpjetfac and a cloakybotfac. Now what if Felon shieldball approached your fortress? You don't really have an effect counter to that in Cloakybot fac so if you were two seperate players one player would be wasting his resources in relatively useless Cloakybot units while the other made a Skuttle. However if you were just one player you could just pause that other factory and krank out the Skuttle in 0.5 the time.
+0 / -0
|
"Right now, the system acts as if the solo guy were actually two players, which is obviously not the case. Which one will perform worse: 1 player with twice the ressources, or 2 same level players with equal ressources? " this actually depends very much on the overall skill level of involved people, like saktoth already worte above. a good player with double ressources might have trouble against 2 average players, while 1 average guy against 2 newbies will just trash them. thats why the larger team gets the higher rated player, in order to not give him an advantage. ofc this often fails but its probably still the best way to handle it.
+0 / -0
|
@Pinguinpanic I heard ticks rock shieldballs. Especially with erasers. But your point still stands, and i'm aware that having twice the income will allow you to build units (faster) that you couldn't afford alone. Also, I never said that there is a good way to balance 2v1 (or similar uneven player distributions). At same player levels, 2v1 will never be fair, be it newbs vs newb, normals vs normal or pros vs pro. So what do you think of the outnumbered team having at least 1 player agree in order to start the game?
+0 / -0
|
Ah yes Tick + Eraser might work. But I agree that having a majority of one would be desirable. Current equation is: winCount = cnt /2 + 1; But that's an integer so in this case the ammount of votes needed to win would be 3/2=1 (integer division) +1 = 2. I think we should make that division round up instead of down, like: winCount = (cnt+ 1)/2 +1; In that case 3 players would result in needing 3 votes. For even numbers it will still result in the same as winCount = cnt /2 + 1 due to integer division. EDIT: Fixed it. I don't think it will work immediatly but it will be in next version. Fix is here: http://code.google.com/p/zero-k/source/detail?r=8544
+0 / -0
|
|
|
imo with 2v1 the 2 players have advantage for the first 5-10 minutes while solo is expanding. after 5-10 minutes the solo has expanded and got production to the advantage swings masivly in their favour, because they have more focused resource use and more cohesive gloabal plan. at high levels 2v1 is actualy quite balanced imo...
+0 / -0
|
1v2 has been discussed already. For larger games, like 5v6, I think balance should add weight=2 to the ELO of the highest player of smallest team. ie, it should count like if there were 2 players with same ELO.
+0 / -0
|
why? having 2 comst dose not inherently make you a stronger player, 2 half decent players can beat a 2 com player easily with micro and agressive strategy the 2 com player dose not have same micro capacity as the 2, 1 com players
+0 / -0
|
[V]Sheep, even if the 2-comm player has an advantage at the beginning, metal is still distributed equally among all the players in the team. For your suggestion to be balanced, the 2-comm player should also get a double share of metal. I wouldn't oppose it if this was added, though!
+0 / -0
|
quote:
winCount = (cnt+ 1)/2 +1;
|
so 1 person will need 2 votes?? 5 people will need 4 votes?
+0 / -0
|
