Guild Wars 2

NevilleDevil.4530:

That is 24*4**100 = 9600 people. That is the amount of people that can be in WvW across all the servers at any given time. Reducing this number is not the solution.

Should only 7000 want to play at reset …
Should only 5000 want to play at a normal prime-time …
Should only 1000 want to play at a normal off-time …

Then of course it is time to reduce the 9600.
The question is when and where!

• reduce the 24, i.e. servers, e.g. till capacity of a normal prime-time matches the space
• reduce the 4, i.e. the number of maps, e.g. freeze a map in the off-time, only let it be playable in prime-time
• reduce the 100, i.e. the cap per map, e.g. reduce the cap of map in the off-time such that it is closer to the actual demand.

NevilleDevil.4530:

Before you say all I like is winning. in PTT. I hate PPT; I am in a fighting guild that has fun finding other zergs and killing them. I enjoy the 24/7 coverage because I always can find a zerg to fight, and there many more in T1 that agree with me, all of T1.

Then you should really try to understand John’s proposal and some of my interpretations of it: The population and coverage in such future matches are HIGHER than in current T1, as current T1-server will be only PARTS of the future teams.

Otokomae.9356:

Really, it’s “Don’t touch the system where people are actually enjoying WvW”.

Last week I enjoyed EU-T2 a lot! It was a totally open match till Wednesday and could still be turned at Friday.

Unfortunately our “reward for winning” was to play in T1 this week. A match clearly decided on Saturday morning. And which I did not entered afterwards, because I hope we are loosing it, to get the “reward of the looser”: go back to T2 to have another great match full of suspense next week.

That was much better till January this year, but then EotM took the queue away and season 2 enforced a lot overstacking on the expected winners. Since them I think the top servers have to be broken up to restore balance.

But other proposals that solve the imbalance as discussed here are fine as well.

Dayra.7405:

WEXP should go into 0) 1 pt only Siege Master and Repair Master 1) max Siege Bunker 2) max Defense against Guards 3) max Guard Killer and then into the sieg weapons you like most. After the first or second siege weapon skills consider Siege Might to max their damage.

Was my opinion. But I also agree that Defense against Guards and Guard Killer only have a value if you max them. And if you WEXP-level slow, it may be a good idea to leave them out till next reset.

I still think that Siege Bunker is more valuable than anything, the damage reduction is clearly noticeable in the arrow rain in front of doors.

storiessave.3807 I agree with you
The start point of the new system should be:

a todays server = a tomorrows alliance

but it may develop (on players choice) into something different over time, if you allow players to found new alliances (e.g. for a large amount of gems) and leave existing ones.

Concerning counting players for WvW I proposed (in the upper part of https://forum-en.gw2archive.eu/forum/wuv/wuv/Solution-to-fix-the-population-imbalance/4443728 where I described how I interpret alliances) to make it automatic, a player
- is counting, if (s)he was in WvW during the last 7 days
- is not counting, if (s)he was not in WvW during the last 7 days
A player that is not counting, will still be in the alliance, i.e. as passive with respect to WvW.

Even alliances consisting only of passive WvW-players could have a value for PvE, if alliances are placed together in the mega-server as todays server.

Mattargul.9235:

Dawntree.7246:

Regarding multi guild, let’s say everybody has to pick a primary guild (as proposed in guilds CDI) and they can play only for the alliance of said guild, its other side guild has to be “freelance” or still member of the same alliance.
If a player wants to leave a guild and its alliance, has to wait the new match up to join with the new guild.

This seems to get complicated because of the one player->many guilds feature.
.

How about handling anyone that hasn’t a unique (i.e. none or many) alliance as freelancer?

Maybe a “Commander Tag Block List” (similar to the chat-block list) is what we need.
If I put an account on this list I never see that (s)he tagged.

Seraph.6853:

• Outnumbered should do one thing only: Half Supplies needed to build Siege
  …

I am not sure if half-supplies is right thing, but otherwise I agree with you guerrilla-warface should be supported by outnumberedbuf.
Alternatives for “Half Supplies needed to build Siege” that might be considered
https://forum-en.gw2archive.eu/forum/wuv/wuv/Outnumbered-Make-It-Useful/4428256

EnemyCrusher.7324:

Have you considered my earlier suggestion of 2v2v2 matchups? Each server would be paired with another for each match. The highest ranked server could be paired with the lowest ranked server, etc. This system would automatically protect the lower ranked servers from falling into a state of hopelessness, and thereby reduce the urge for people to abandon low ranked servers.

.

Why stop at 2vs2vs2? If I got john right he proposed
MvsNvsO
where the
Total-pop(M) ~ Total-pop(N) ~ Total-pop(O)

Of course M,N and O could be all 2, but that’s only a special case (and likely one that has only a few or even no solutions for Total-pop(2) ~ Total-pop(2) ~ Total-pop(2)), it could also be

• 1vs1vs2
• 1vs2vs3
• 2vs2vs3
• …
• 15vs3vs6

I guess he wanted that to be unspecific in that question to stipulate variety in our thinking, discover our wishes/preferences and finally: freedom for ANet to decide later.

Match-making seems to be like in season 2.

In 9-server leagues where 1 always wins and 1 always looses the final outcome will be:

• 1) 800 tickets, won all matches
• 2) 2 server get 650 tickets each
  2nd in last top match
  1st in last middle match
• 4) 3 server get 600 tickets each
  Looser last top-match
  2nd last middle match
  Winner last bottom match
• 7) 2 server get 550 tickets
  Looser last middle-match
  2nd last bottom match
• 9) 1 server get 400 tickets, lost all matches.

applecup.3047:

My problem with the alliance thing is that it boils down to turning wvw into alliance v alliance – megaguild v megaguild. To an extent that is how wvw rolls at the moment, at least in high tier servers, but it also has the effect of shutting out roamers, newbies, small and/or unconventional guilds, etc.

I see more space for them in my imagination of Alliance-WvW, then they currently have. They could even build an own alliance and compete with all the other alliances. It depends of course on how alliances are scored, but assume for a moment it is based on current-scoring:

alliance-score = (alliance-claimed-objectives-per-tick + score-not-from-tick)/alliance-size.

A large alliance must control a large area to be successful.
A small alliance of some roamer can be successful by just kill dolyaks and enemy roamers and occasionally conquering and keeping a tower over the tick.

And concerning newbies and alliance less people, I proposed to add a temporary OpenAlliance for each Match-team. newbies and alliance less people would be automatically be assigned to the OpenAlliance of the smallst team for this week.

timmyf.1490:

I will say that Darya’s interpretation of how alliances work is not the same as my interpretation and I consider it a very troubling interpretation.

Of course it is only my imagination how it could be, neither yours nor Johns nor ANets nor …
I try to describe my imagination, such that others can see it and can like or dislike (parts of) it. And describe your own imagination maybe using parts of it.

At the end john and ANet will implement their picture of it, and it may or may not contains (or be influenced by) parts of each imagination.

Team composition replaces match making, i.e. It happens every Friday at 20:00.

You do not have to rely on the other alliances of your team, each alliance can be scored separate based on its individual performance in the match. Maybe you don’t want to cooperate, but are in a race of taking stuff from the enemies.

Otokomae.9356:

Dayra.7405:

Schtizzel.5497:

And then I can’t really grasp how an alliance system would help tackling night capping or off hour imbalances.

It allows new solutions to coverage. The root of the coverage problem is that the amount of players that want to play WvW is different all around the clock, whereas the match-capacity is equal all around the clock, i.e. the match-capacity per team is always fix around 400, while the number of players per team vary from 0-1000.

Let’s assume for a moment that we will have just ONE match in the future, where all the alliances are distributed over the 3 teams, such that overall man-power is nearly equal.

4 map is not enough for that in prime-time, we likely need 10-20 maps in primetime, but in off-time we need much less maps maybe only 4.

We have

• EB-map
• (rebalanced) EotM-map
• 3-BL-map (can be used only together due to their asymmetry)
• maybe new maps in the future

We may use them in two versions

• large (capacity is 100% of current)
• small (capacity is 50% of current)

We may have an 3-BLs-large map(s) that fits around 300 player per team, we may have a EB-medium map that fits around 50 player per team, etc..

These are the blocks ANet can setup the match-structure. ANet knows (could know) the man-power/time graphs! of all alliances from last week and can estimate the man-power/time graphs of the teams from it and can setup a match-structure that fits it well from it.

E.g. always the match consists of the match-structure of today

• EB-large
• 3BLs-large

thats 400 capacity per team.

In prime-time that’s far from sufficient to fit all player, we need more capacity for prime-time, so lets add maps that can only be played during prime-time each day

• 3-BLs-small ( + 150 capacity)
• EB-small ( + 50 capacity)
• EotM-large (+100 capacity)
• EotM-small (+ 100 capacity)
• …

till the capacity exceeds the estimated demand. The maps are saved when their demand ends and restored when they are demanded again.

• This cannot be manipulated because it is estimated from last weeks demand.
• If all alliances are stable it will fit, and you will have no queue and no coverage problem
• Only if many alliances are rather unstable, you may have queue or coverage problems, but that’s your fault, be more predictable.

Blaeys.3102:

Instead of basing the alliances around guilds and players, base it on servers and add maps to the match to accomodate the numbers.

So, it would be Servers A, C and E vs Servers B, D and F vs Servers X,Y and Z playing across 12 maps (nine borderlands and 3 EBGs). And, make it so only your server can defend your borderlands (to give you somewhere to retain server pride).

Everyone wins.

Wait… what? So you’re saying that the “Alliances” are just the same old EotM-style Green/Blue/Red teams, with everyone playing on the same teams all the time, no more Tiers or unique matchups, and Servers and guilds who never see each other on the map or even speak to one another will now have to count each other in order to win each week?

Please, keep the casual, EotM-style teams OUT of WvW.

Alliances are not blue, green, red, but blue, green, red are the teams of a match.
The initial alliances are the current servers, 1 today server = 1 future alliance.
Several alliances together are assigned to a team, such that teams have equal size.

Differences to today’s EotM

• EotM has no 24/7 match, alliance WvW will have 24/7 matches.
• alliances fight for their permant ranking in the leaderboard, whereas in EotM all score is forgotten after 4h

Imagine it as one gigantic map where all alliances are placed to fight for their score. 2/3 of the alliances are hostile, 1/3 of the alliances are friendly (you cannot fight each other, how much you actually cooperate depends on rivalry)

If an alliance controls territory proportional to its size it is as expected and stays at it’s current score, if it controls more territory than expected it gains score, if it controls less than expected it looses score.

Schtizzel.5497:

And then I can’t really grasp how an alliance system would help tackling night capping or off hour imbalances.

It allows new solutions to coverage. The root of the coverage problem is that the amount of players that want to play WvW is different all around the clock, whereas the match-capacity is equal all around the clock, i.e. the match-capacity per team is always fix around 400, while the number of players per team vary from 0-1000.

Let’s assume for a moment that we will have just ONE match in the future, where all the alliances are distributed over the 3 teams, such that overall man-power is nearly equal.

4 map is not enough for that in prime-time, we likely need 10-20 maps in primetime, but in off-time we need much less maps maybe only 4.

We have

• EB-map
• (rebalanced) EotM-map
• 3-BL-map (can be used only together due to their asymmetry)
• maybe new maps in the future

We may use them in two versions

• large (capacity is 100% of current)
• small (capacity is 50% of current)

We may have an 3-BLs-large map(s) that fits around 300 player per team, we may have a EB-medium map that fits around 50 player per team, etc..

These are the blocks ANet can setup the match-structure. ANet knows (could know) the man-power/time graphs! of all alliances from last week and can estimate the man-power/time graphs of the teams from it and can setup a match-structure that fits it well from it.

E.g. always the match consists of the match-structure of today

• EB-large
• 3BLs-large

thats 400 capacity per team.

In prime-time that’s far from sufficient to fit all player, we need more capacity for prime-time, so lets add maps that can only be played during prime-time each day

• 3-BLs-small ( + 150 capacity)
• EB-small ( + 50 capacity)
• EotM-large (+100 capacity)
• EotM-small (+ 100 capacity)
• …

till the capacity exceeds the estimated demand. The maps are saved when their demand ends and restored when they are demanded again.

• This cannot be manipulated because it is estimated from last weeks demand.
• If all alliances are stable it will fit, and you will have no queue and no coverage problem
• Only if many alliances are rather unstable, you may have queue or coverage problems, but that’s your fault, be more predictable.

Torsailr.8456:

I think he was saying that if BG has 10,000 WvW players and alliances have a 10,000 member cap then all of BG could move to one alliance and be called the BG Alliance.

However if alliances have a cap of 5000 then BG would have to split into 2 alliances and they’d both need unique names.

For sure, smaller alliances make team-composition easier and more variant than large alliances.

Also for sure it does not make sense to allow alliances to grow larger than match-team-size.

But if you say, current NA-T1 teams are perfect size to fill up 24/7 4-map matches and we stick to such matches, then todays BG-size (or SFR-size, no idea who is larger) would be the max-size of an alliance.
BG-alliance would always be one team, whereas other teams would consist of several alliances.

I’ve no idea how many matches it would give if all teams are BG-size, maybe 2, maybe 3, maybe 4. But ANet for sure can determine this.

It is also possible to have matches of different sizes. Could be nice for variance.
In one week your (small) alliance is placed in a 4-map match with many others, in another week you alliance is placed in a 2-map match (assuming we have two balanced maps), with fewer others. In one week you play on maps with high (current) map caps, in another week you play on maps with low map cap (assuming you fit).

You could even say there is only one matchup of 3 teams of alliances consisting of 10 maps every week, then of course even BG becomes a small alliance that cannot play alone. And you can even vary the match-structure from week to week.

John Corpening.9847:

Alliances would provide continuity for the community.

Very interesting idea, John,

Let me explore this idea a bit to create a possible concretization of it.

An alliance is kind of an extended guild, it may have (a) alliance leader(s) direct members and additionally it has other guilds as members. It has a total size limit, far larger than a today’s guild. A guild and a player(*) can be member in only one alliance and probably it’s a good idea that a guild can declare only a subset of its members to be part of the alliance (count against the alliance limit), e.g. I am in a large PvX-guild where only half of the members are interested in WvW and it’s probably not a good idea to count the WvW-inactive members against the alliance limit, or bother them with the alliance chat. Maybe that can be automatic, if I was in WvW during the last week I count, otherwise I don’t.

Initial alliances are the today’s server. All guilds of a server and all guild-less player of a server form one alliance, e.g. the SFR-Alliance, the Elona-Reach-Alliance, …

It has an general alliance chat, that visible to all direct members and all members of sub-guilds, and a WvW-chat that is only visible to the active WvW-player of the alliance.

Some goodies I can see,

• it has a treasury and all WvW upgrades by officers of the alliance are payed from the alliance treasury.
• it has a battle scheduler/calendar
• commander visibility level: guild-only, squad-only, alliance-only, everyone-in-team

Match-making could be totally different from today.
We have a fixed set of matches, and matchmaking would only be to place alliances into the teams of the matches, such that all teams have nearly equal man power, e.g. in one week SFR-alliance is placed as one team to fight against the team of Elona+Gandara alliances as second team and Vabbi+Millers+Abaddon+Jade sea alliances as third team (assuming the total alliance sizes of the teams are nearly equal.). At the end of the match each alliance in the team receives the team-score as alliance-score and in the leaderboard alliances and it’s score are listed (not the temporary teams). Next week, team composition of alliances could be totally different, alliances that were in a team in one week, may be in enemy teams the next week.

Also in tournaments the alliances are ranked, the temporary match-teams are only the vehicle to earn the score.

Alliances of all sizes can fairly compete with each other!

Creation of a new alliance is possible but cost a large number of gems. Transfers between alliances is possible, maybe free, maybe for a small fee. You have to be in an alliance to play WvW (you are automatically in an alliance, if your guild is in an alliance). In principle you can found your 1-man or 1-guid alliance and play WvW.
ANet could setup 1 OpenAlliance per Match-Team. If an alliance less player tries to enter WvW he automatically joins (for the rest of this week) the OpenAlliance of the smallest Match-Team.

(*) A player can be in several guilds, and the guilds may be in several different alliances. If this happens the player is considered to be in NO alliance, till a unique alliance membership exist again (the player leaves guilds, the guilds switch or leave alliances)

Manuelito.6081:

EotM is a farming ground because players made it so.
here is the solution: change server.

No it isn’t you will be blue next week and farmed by green.

EotM should of course not have queue, it should just not be connected to WvW at all.

When you join (without being in a party) you get assigned to the smallest team, is all that’s needed.

Party handling is a bit more complicated, you don’t want to separate parties, and you don’t want balance disturbing taxiing.

For EU a nice alternative could be:
- One color german speaking
- One color french+spanish speaking
- One color english speaking

Color rotates every 4hours, and either the GUI-settings or the WvW-server settings are used to determine the language.

Another idea:
Both minorities are currently spread equally over the instances created by the majority, i.e. they are a minority everywhere. If there is a strong imbalance make it
Instance1 Majority -Minority1
Instance1 Majority -Minority2
….
turning it into some 2-sided matches, but both teams are likely equally strong.

The wrong thing is measured in EotM. Who cares about possession of objectives in EotM?
The really important measures are

• average karma gained per player in the match in team X
• average WEXP gained per player in the match in team X
• average Lvlups gained per player in the match in team X

Does green win in these important measures or are they inefficient because the map is to green most of the time and they have to run to much before they reach the next cappable objective?

Does it improve your rates if you run over the other teams or does it delay it?

Which commander has the best rates?

…

Server merging alone doesn’t solve anything, it only buys some time, concentration on the winner side will just continue afterwards as it did in the past.

Server merging can solve only one issue, it relaxes the balance restrictions to be placed on all servers, e.g. capacity restrictions will be less harsh, if there are fewer server to take the mean from (that’s still an achievement, but only nice for the over-populated server not for the underpopulated merge servers., i.e. their bonus is missing.)

We need active balancing measures, without them merging is not a long-term solution.

Balancing measures that already proved their value in the past are

• queue
• rewards (even if it was so far only used as an imbalancing method)

Rewards are nicer than queue’s so I would suggest rewards as the main soft-balancing method and queues as the emergency hard-balancing factor for massive over-stacking.

Rewards can also be used to make life on a merged server more sweet.
Still I think setting up new teams from scratch is

• more fair than merging
• get’s rid of ballast from pre-mega-server times (like pve-only population included in servers, server-selection before you even know what WvW is, …)
• makes it easier to add some new goodies, like a team-chat for all WvW-players of a team currently online

Sakura No Seirei.6120:

There is a way to stop blob manipulation, and that’s to introduce cooldown timers for re-entering WvW after a deliberate exit from a WvW map to a non-WvW map (which quite handily gets rid toon hotswapping).

If my client crashed or if I send him a kill-signal is not distinguishable and
if my internet DCs on purpose (switched WLAN off) or not (critical packets got lost/delayed somewhere) is also not distinguishable.

And I would be really kittened if a (currently quite frequent) client-crash/DC is additionally punished by a cool down timer.

Ansau.7326:

And if there is no winner, there is no competition. If there is no competition, people who search that won’t play WvW. If those people don’t play WvW, all it is left are pve farmer. You’re just discribing EoTM: A place where there’s no competition and wvw dedicated players don’t go there, because you only find farming blobs.

Didn’t you read the 2nd part of the original post?

WvW is not competitive, was never competitive and if you read the thread a lot do not want it do be competitive.

You do not need a match to compare manpower and coverage, the match is completely superfluous to determine the winner, just compare the man-power per time graphs of the 3 server, and give the one that “wins” (the one with the largest lonely coverage advantage) a reward. After 2 years WvW everyone should have noticed that.
The only surprise is: You don’t know the man-power per time graphs of your opponents in detail ahead of time.

Now I am confused.

I am aware that it is not allowed to critic Forum-Moderation in the Forum. That’s the reason why I mailed the following to the Support:

Just to be told be the Support that I should post it in the forum.

What’s the correct way?

Ansau.7326:

*A lot of people like 1 week matchups. There are a lot of coverage differencies between week and weekends. Servers play with that. Also upgrading things becomes worthless. Why to upgrade a keep if the matchup will end before it reaches T3?

Also a lot of people say, one-sided matches would be ok, if they would be shorter.
And 2 days are enough to upgrade your keep a dozen times

Ansau.7326:

*If people are complaining because matchups this week are outrageously unbalanced, what would happen if SFR fought 17th or below servers? They’re already stomping Baruch and Elona.

SFR becomes bored and spread out instead of Elona stops playing?
(As Baruch already do between reasons.)

Vabbi can test if their guerrilla tactics work against SFR as well?

Ansau.7326:

*Never ever reward WvW with pve rewards, but WvW rewards.

I really mean: never ever determine a winner in a playground.
If there is no winner there is no stomping.

Provide people statistics they may like to see.
Vabbi may be happy to have a higher killed dolyaks per player rate that SFR.
SFR may be happy to control the map 100% of the time.

*About competition, if you eliminate some sort of ranking and competitions, WvW becomes EoTM.

EotM is as it is due to a bunch of reasons

• random teams, quite often different for every join
• imbalanced reward structure, objective+lord farming with green beats nearly everything in GW2 by far.
• the only statistics is a meaningless score
• no voice-com
• …

Interest for a score that only measure manpower (we know it ahead), coverage (we know it ahead) and PvD-Speedrace (Elona is quite good in that, and this is all that matters in EotM as well) is not what makes play style in WvW different from EotM. A lot of people in WvW have lost interest in score and it’s still partially different. (But only partially, it’s much more similar than most WvW-elite allows itself to admit)

Jezynka.2651:

Yes, zerg gets bigger but main problem is hard-resing during fight.

No the problem is resing of the winner after the fight. That way the zerg keeps size and never has looses due to several small group that are only able to produce causalities, but aren’t able to win.

If instead every complete dead would be send to spawn, the zerg shrinks as a result of every fight. A group of 10 (with a short way) that can force a 40 zerg (with a long way) to fail by repeated attacks: 5 kills on 1st try leaving 35 left, on 2nd attack they have better odds 35:10 and they are able to kill 8, leaving 27, 3rd attack odds are 27:10 they kill 12, leaving 15, …

roxybudgy.8205:

The way things are now, if you don’t enjoy how things are on a server, you can transfer elsewhere. My brother started on Sea of Sorrows, then moved to Blackgate, then to Jade Quarry, then to Blackgate again, if he wants to hop from place to place, go for it! I’ve been on the same server (Ehmry Bay) since day one and have experienced various fluctuations in tiers, and I have a great time, whether it’s the chaos of being heavily outnumbered, or effortlessly steam rolling through maps devoid of enemies.

That’s a possibility, however in this case:

• don’t make the matches (same world opponents) 1 week, 4h to 2 days is enough
• don’t compute a score (and no ranking), but instead provide several statistics about the match, such that everyone can look on what interests him most.
• make matches completely random SFR-Deso-Kodash should have the same probability as SFR-Vabbi-Millers
• NEVER EVER make a competition
• NEVER EVER reward winning, in fact never ever determine a winner.

I personally think it would be nice to have both a

• a playground, EotM style (not necessarily EotM map, EB map and BL-triple maps may be used as well) may be best for that (EotM has a mayor problem, only attack is rewarding defense is not, turning it into a karma-train, but that’s a different problem) (maybe in EU with English-speaking vs French+Spanish- speaking vs German-speaking would be a better idea than the current WvW-color based system)
• a competition, but for a real competition it must be possible for every team to fill the map(s) 100% of the time of the match, all participants in the competition must be equal in fielded manpower and coverage. They may have 10 times as many people on the reserve bank, but on the field, the power must be always equal. I also believe that for a competition an guild/alliance based system is better, as you can control whom you want in your guild/alliance, e.g. disabling trolling and grieving.

The problem of today WvW is that it sits between those chairs. It’s not competitive but used in one. It’s not a playground as it has a score and a rewarded winner.

How does this help FA to avoid being crushed by TC, Elona, Gandara, … to avoid being dominated by SFR, …? That would not be enough merge.
Maybe merge 1st with 24th, 2nd with 23th, … (add 27 to 10+15, 26 to 11+14, 27 to 12+13 in EU) would help, but that doesn’t respect languages.
And still most maps would be mostly empty most of the time.

Time-Sliced grouping of servers.

While servers can play (partially) competitive against each other in prime-time (i.e. evening from an Eu perspective). This is much less the case in off-time (night and day from an Eu perspective).
Also the overcapacity of match-slots is worst in off-time.

How about time-sliced server-mergers?

In the night 4 server build up a team and we have a total of 2 matches:
Match1: 1+24+12+13 vs 2+23+11+14 vs 3+22+10+15
Match2: 4+21+9+16 vs 5+20+8+17 vs 6+19+7+18

During the day teams of 2 server, i.e. each night 4-team splits into 2 day 2-teams
Match1.1: 1+24 vs 2+23 vs 3+22
Match1.2: 12+13 vs 11+14 vs 10+15
Match2.1: 4+21 vs 5+20 vs 6+19
Match2.2: 9+16 vs 8+17 vs 7+18

During the evening each server plays on it’s own
Match 1.1.1: 1 vs 2 vs 3
Match 1.1.2: 24 vs 23 vs 22
Match 1.2.1: 12 vs 11 vs 10
Match 1.2.2: 13 vs 14 vs 15
Match 2.1.1: 4 vs 5 vs 6
Match 2.1.2: 21 vs 20 vs 19
Match 2.2.1: 9 vs 8 vs 7
Match 2.2.2: 16 vs 17 vs 18

The score of a server is the sum of the scores of all it’s matches, e.g.
server 1 sums up it’s score in Match 1.1.1 + Match1.1 + Match1

Disadvantage:
- EU-servers 25-27 have to be disbanded, player of them have to choose a different team
Advantages:
- Prime-time is as it is today
- servers are as today (beside EU25-27, sorry)
- play is 24/7 for your team
- whenever you play for your server you score for your server.
- maps for everyone are much better filled at every time, i.e. always more friends and more enemies, without more danger for queue
- you will always have your primetime enemies, e.g. 1 vs 2 vs 3 happens at all times, just during day and night the teams are larger
- match-capacity is removed at times where it is not used, i.e. it is really over-capacity reduced, no one sit more often nor longer in queue.
- each single match runs for the full 7 days
- you have better changes to pickup in a state where you left, e.g. playing full 1.1.1 may be possible.
- you see more different enemies, if you play off-time
- you see more different friends, if you play off-time
- GvG’s between e.g. 24 and 2 or … are possible (when 1.1 or 1 are active) without transfer

Each slice can be 8h (e.g. 0-8,8-16,16-24), but maybe a different slice times are more adequate to get a better filled matches, the attached figure illustrates a possible more complex flow of game and player for reset till sunday morning.

Jong.5937:

who would win in a Seafarer’s-Eredon/Blackgate-Vabbi matchup?!

I don’t know who will win, but will it be an interesting match?
In EU-prime (NA-Day) Seafarer’s-Eredon will dominate the maps 4:1, 400 bored Seafahrer will crush 80 Blackgates+30 Vabbi and tick 595:100
In NA-prime (EU-Night) Blackgate-Vabbi will dominate the maps 4:1, 400 bored Blackgates will crush 80 SFR + 30 ETs will tick 595:100
EU-Day/NA-Night/OCX-prime will probably decide the match, I don’t know who is stronger OCX at Blackgate or Day at SFR.

I think matches are only interesting if the opponents have a similar 24h manpower profile, and not if they just have their peaks at different times, i.e. if they just crush each other at different times.

ANet should be able to produce real number here is an estimate of match-capacity usage on my EU-Gold (EU #4-7 have approx similar numbers) server.
We have always a few people, at reset-evening we are (sometimes) able to generate moderate queue on all 4 maps, on most (not reset) evenings we are only able to generate queue on 2-3 maps.

An 24h player (red) compared to capacity (blue) graph is as attached.
At just 2 of 24hours we are able to fill all maps, at total our man-power is around 3200 play-hours per day and the match capacity is 9600 play-hours per day. We are able to use around 1/3 of the offered capacity.

The few servers above us may have a better capacity usage (that’s why they are above us) , maybe up to 50%, the majority of server have a worser capacity usage (that’s why they are below us).

Observation No 1: Player are able to fill less than 25% of offered match-capacity.

Observation No 2: If players would be evenly distributed this would also be true for prime-time (19:00-23:00). As they aren’t EU-Gold reaches capacity-limit for 1-2 hours a day. EU-Bronze, probably never does.

Observation No 3: During off-time (which is over 75% of the time) only 5-20% of match-capacity is used.

Observation No 4: Any game or sport that is competitive fills available capacity 100% at any time.

Observation No 5: Times with the lowest match-capacity usage are most decisive for the match-result. (See pic 2, scoring per hour of the current EU-Gold T1 from http://www.gw2score.com/server/Elona-Reach)

Implication No1: WvW is far away from being competitive, if a game is not competitive tournaments harm the game, and even scoring should not be done.

Alternative No 1: We can keep WvW as open playground WITHOUT any tournaments and score, but the 24/7 matches are a bad structure for that, some player you meet to often some you never meet.

Alternative No 2: We modify WvW to make it competitive, for that we need to restrict the capacity of the game such that EVERY team is able to 100% fill the capacity at ANY time.

Implikation for alternative No 2: To make WvW competitive we need to remove all the over-capacity (i.e. around 75% of it’s total capacity) that is never been used and we need to make sure that all server have the same capacity and the same chance of filling it.

Cut must not be equal all around the time, as demand isn’t equal all around the time.

BooDuh.2841:

Theftwind.8976:

Tie any rewards for WvW, including Karma, coin and loot drops to the server population balance at any given moment, a modifier to multiply your chances. To maximize reward you will want to play during non-peak times or against a force larger then yours. Once the large, highly populated servers realize that they get more bang for the buck on the lower populated servers they will move. Over time it will all balance out. No need to change population caps.

how would that differ from this? Asside from the fact that I love charzookas n all …
http://wiki.guildwars2.com/wiki/Outnumbered

Outnumbered is based on the odds of one map in one match.

The proposal above should be based on comparing all servers population at a time.
NA 1 has currently 200% of the mean population, it gets everything halved, NA 24 has currently only 10% of the mean population it gets everything multiplied by 10.

Magi Pendragon.7042:

John Corpening..
For tournaments, two or three RANDOM servers from NA would battle together under one NA flag, against three RANDOM EU servers who also battle together under their EU flag.

You are aware that EU-sleeps when NA has evening and NA is at work/school when EU has evening, aren’t you? Have a look here https://forum-en.gw2archive.eu/forum/wuv/wuv/World-Versus-World-suggestion-1/4086126 for a slightly more detailed discussion.

IF NA and EU can be combined merge 1 EU and 1 NA server together to build a new league of 24 servers with overall better coverage.

EU 1 with NA 24
EU 2 with NA 23
…
EU 24 with NA 1

EU 25-27 drop, let the people choose one of the 24 server.

Virtute.8251:

Dayra.7405:

Virtute.8251:

First, rank your servers by the population count you use internally to set the world population status on World Selection screen.

No please, do not use this population count at all for WvW!

Absolutely yes. It has to be that way.

The reason: Server status (Full or not) is irrelevant to WvW. It’s only reason for existing is a matter of logistics for provisioning computational resources on the network.

The only reason we are being asked this question inside the WvW sub-forum is because if this is to be done, ANet would like to keep WvW-focused players happy in the process. You can not, however, have a WvW-centric server merger.

Sorry.

Your argument was partially valid (but no map ever reserved capacity for all 70000 people on a server) before mega-servers, but now with mega-servers it is complete non-sense. Servers only concern WvW anymore, and whoever does not play WvW, doesn’t need capacity in WvW, before he (re-)starts playing WvW. Or which WvW-capacity do you want to reserve for the 70000 people register on any server?

I was convinced of ensigns proposal till February 2014, but the concentration afterwards (enabled by eotm’s queue reduction and enforced by seasons 2 skins rewards for only the better servers) cannot be handled by that any longer.

Look at current NA T2, just 3 ranks between TC and FA and a 3.5:1 score imbalance, that’s out of the scope of any reasonable handicap. I am also in doubt that anyone has fun in this match.

Virtute.8251:

First, rank your servers by the population count you use internally to set the world population status on World Selection screen.

No please, do not use this population count at all for WvW!

I don’t think it changed, and pre-mega-server it was:
Everyone that has it’s current domicile on this server, maybe inactive since month or even years (likely the majority) maybe totally uninterested in WvW (a large part), the people that play WvW (a small minority in the whole number).

With over 3.5mio copies of GW2 sold in the last 2 years it is probably around 70’000 per server. The few transfers (compared to 70’000) that already made WvW imbalanced slipped unnoticed due to the inadequacy of this number,

The usage of this number (to tax or restrict transfers for WvW) is one of the main failures in the past that lead to he current imbalance.

Better numbers would be

• player join the last match
• played hours in the last match

Apokriphos.7042:

In World vs World, Balance is nonexistent.

Some players may like it that way. Perhaps because they enjoy greatly outnumbering their foe(s) and dancing on their poor, sword-ridden corpse. Or perhaps because they like the feeling of decimating, with certainty, whomever they come across that is of less number then they.

However, I don’t feel this is the way it should be. Players stack servers during “tournaments” so they can be on the most populated, guaranteed a high position and reward. Is that how you want your game to work?

ArenaNet, why not auto-balance WvW? You already do it to some extent with Hotjoin.

The idea will work like this:

…

What do you guys think?

I completely agree with the preface.

The idea is a capacity reduction based on the smallest team in the match, and as such it is very susceptible to exploitation. Let me make an example server A is in danger to loose it’s keep because a strong force of B is at it’s doors, and A’s force is to far away (or in Battle with C or …) to reach it in time. A moves all but the 5 scouts in the keep away from the map enforcing a reduction of B to 15, a now very comfortable odd to defend an upgraded keep with preinstalled defensive siege. A could even go back to the map with now with 20 people and crush B, before they are able to sort out, where their people are now. (And even if B manages to bring the people back fast, they are now at spawn and not at the keep)

If you want to make the capacity reduction match-based, you need at least base it on the mean of all three teams. A has 40, B and C have 10 that want to play, sum is 60, mean 20, half of A’s for is able to play. If one server leaves the map, the capacity of the other teams is only reduced to 2/3. Better (because it helps to balance teams of different matches, and because it is even more imune to manipulation) is a capacity reduction based on all teams demand.

Crius.5487:

Overgrowth is Green.
Green is generally the 1st place server of each tier.

This week it is unusual worse: Due to season, green is the top tier of every (9 server) league without any randomization

Boris Losdindawoods.3098:

~ ~ As someone else said, this also looks cool, but seems like it would be complicated for Anet to implement. http://www.youtube.com/watch?v=qHcO6Xo8eJ8&feature=youtu.be

Guesting isn’t a solution, it’s the opposite, it cements population imbalances. People over-stack even more, knowing they can freely guest somewhere else at will, when they aren’t needed at the “home-front”.

Balanced Alliances are a good solution in my view as well (also to grief play/trolling), but I left them out as a common response is “We don’t want elitist groups”.

And to all: Keepping it as it is (especially all freedom that broke it) and repairing it at the same time doesn’t work

And yes, if we don’t want to end as shown in sminkiottone’s video discovery https://forum-en.gw2archive.eu/forum/wuv/wuv/Matchup-Feedback-merged/4435605 T1 over-stacking has to be ended, if T1 likes it or not.

ketchup.2783:

How about having the tournament, or just daily rankings takes place only on Fridays for a designated 3 hours, for one month? Get the reward at end of each month.

Or mix servers for each world (red, green, or blue) each week.

It’s a bit off-topic, but have a look at: https://forum-en.gw2archive.eu/forum/wuv/wuv/Proposal-combine-the-best-of-WvW-and-EotM

Good question. Mostly I tend to:

Red is dead.

i.e. I don’t care who is where and why, a red on the way is just a possible score point/baggy to get, if I have nothing more important to do.

But you may be right, it’s a lack of empathy (but Anet supports that view, by anonymizing them) to handle enemy people the same like enemy NPCs.

In Obsidian Sanctum (wasn’t there for while) I stayed neutral, as long as they behaved neutral. Motto: If we start a fight now, we both hinder us to reach our goal.

timmyf.1490:

It’s rather disappointing that the most commonly proposed solution to “there aren’t enough people on some servers” is “limit the number of players who can play the game mode.”

No it’s trying to find a good way to adapt the provided capacity to the requested demand.
Currently we have much to much overcapacity (nearly empty maps most of the time on most servers).

And the other thing is: try to reach more balance, and there I prefer adapted rewards (more rewards if you play with fewer players) over capacity restrictions.

There is a third issue:

3) in total not enough player to fill all maps in all matches.

PS: don’t know about the others, but I read most posts, otherwise I wouldn’t have noted the opposition. And a player brainstorming and opinion overview, saturated by ANet is probably not bad.

Hm, we have to few people to fill the current capacity of matches, but there is a strong opposition against sever reduction and against global capacity limits.

How about differentiating the capacity per permanent leagues?

• a league of 6 server play 24/7 matches on the 4 maps as currently all matches are.
• a league 6 server play matches where only EB is 24/7 and the borderlands are 6/7 (every day only in EU/NA prime time)
• a league of 6-9 server playing only EB 24/7
• a league of 6-9 server playing only borderlands 6/7 (every day only in EU/NA prime time)

(Or something like that, only ANet can get all the numbers to calculate the real demand)

It would enable all servers to fill their maps, player can choose their preferred match style, it probably is very close to the current capabilities of the server.

Transfers between leagues should be very cheap and easy, transfer between servers of the same league should be severely restricted. (E.g. By remembering the last-server in every league for every player)

Balancing methods can be much more relaxed and would fit current servers much better. But it’s probably a good thing to add some, before things run out of control again.

Concerning EU only: per league 3-4 English, 1-2 German, 1-2 French, and 1 league with a Spanish server (1st or 2nd?) would probably best. Currently I would say for league 1: 3 EU, 1-2 German, 1 French, 0-1 Spanish.

munkiman.3068:

Dayra.7405:

munkiman.3068:

Without being free, there is already incentive to transfer to lower pop servers, it’s a lot cheaper

That’s not an incentive to move down. Staying on an overstacked server is cheaper.

Yes, they’ve done free transfers in the past though, didn’t really impact balance all that much. If anything it caused the same group to shift off the high tier into ONE (a single server) in the lower tiers. Basically swapping the positions of the two servers.

If you want to turn the transfer prices into an balance incentive you need some additional things, as taxing (or not) transfers is only a function to reduce (or not) the amount of (specific) transfers, but not a way to increase specific transfers:

• accurate WvW-population counts, that are updated on every transfer/activation of former inactive WvW-player (such that underpopulated can be adequately assigned and removed whenever its isn’t true anymore)
• free/cheap transfer to any low-pop server (to test it out)
• a payment for playing WvW (better increased income from participating in WvW) on a (still) underpopulated world (it stops when it’s no longer underpopulated).

munkiman.3068:

Without being free, there is already incentive to transfer to lower pop servers, it’s a lot cheaper

That’s not an incentive to move down. Staying on an overstacked server is cheaper.

titanlectro.5029:

And then you just destroyed every last server community for your solution.

Are the top server really the best communities we have?

Didn’t all server implosions happened in the top 6, showing the instability of those communities in face of losses?

Isn’t a community that stays together even while crushed more worthy than a community that runs over everyone without challenge?

Isn’t a community that has fewer closed and cleaned threads in the good old days of match-threads more worthy than a large collection of forum trolls?

Isn’t a community with a lower percentage of new joiner more worthy than a community with a high percentage of people moving around?

My expectation is that the top-server have a small good core community with a large periphery of unsteady people. Without doubt the core-community will be able to move together to build a new community somewhere else. And the periphery that disturbed the balance has to spread out.

titanlectro.5029:

That said, server mergers will not solve the population balance issues. People will still stack to the winning servers, because they have incentive to do so.

That’s why it isn’t a good idea to merge the low-pop servers. Either

• start from scratch with fewer teams
• delete the nearly empty and the Top-Pop-server to force the overstacked people to spread out. (they have the people that are experienced in moving and that made the problems)

In both cases the balance-limits have to be implemented first (you are not allowed to overstack again)

munkiman.3068:

Dayra.7405:

munkiman.3068:

especially since objectives can go unmanned for days and still earn that server points.

That’s exactly why I proposed to that “the value of an objective decays over time, if it has no action”

What about neutralizing it entirely?

Yes, decay to zero. (maybe best a logarithmic decay: For every tick without action, the value is halved (or 75% or …), an action involving at least X player (treb-shooting, and the thief hitting the guards, better do not count) double (or add 25% or …) the value (up to some max value))

And did you noticed my edit about ruins that happened after you cited it?

Nightshade.2570:

Merging Servers vs. Mega Servers.

One is is solid the other is flowing. Either way there are actual ways to maintain a servers sense of identity and pride. I prefer the Mega servers idea as it provides a long term solution vs. a temporal one of Merging servers.

Do not do the following now, the servers are to imbalanced.

But if we face a situation in the future, where map-caps shrink to low as all teams lost players: Make a league where the teams fight for survival (last 3 are closed).
Should we ever face a situation of increasing population, that increase queue to much, add 3 new teams and give players that move there a time-limited “colonization”-bonus.

This way, server-reduction solutions can get a very long time perspective as well.