Jump to content
  • Sign Up

World Restructuring


Gaile Gray.6029

Recommended Posts

@"ixl.7190" said:Let's say you are really motivated and manage to scrape together 1000 players who put in 50 hours/week on average. So our Alliance weighted average is now 50k hours a week. To balance this, ANET has to link FIVE 1000-strong 10 hour/week WvW Allainces just to get equal potential coverage, so kiss any notion of any extra bodies diluting the talent pool and activity on Green team. Congratulations, the motivated players just ensured every single WvW player on their side is using a META build, in one of two guilds, in the same voice comms, and is reliably showing up to do work. Levels of organization previously impossible are now not only attainable, but realistic, and the "penalty" for excessive hours tagged up actually helps intensify the effect. And even if the 10-hour-a week crowd on Blue Team manages to hold their own against this foe, do you honestly think Red team isn't now reduced to whomever is left over, and soon to be never logging in again? Maybe I'm wrong, but regardless of how many players there REALLY are who will play WvW going forward, a version of this is already being planned and implemented. There is no system ANET can create which motivated players cannot find a way to break and exploit to achieve their goals without hurting everyone as a whole.

This is the reason that the "new" idea isn't going to work.

Link to comment
Share on other sites

@shiri.4257 said:Instead of having players pay for transfers have guild costs for wvw designations of X number of players. I hate seeing dead guilds and bank guilds take up name space and no maintenance fee.-Every 8 weeks an alliance and guild maintenance cost along these lines

Guild Fees:-100g/50 wvw player designation slots (that's 2g/per person every 8 weeks) maximum of 500 slots (1000g/guild). This would also require Guild UI to confirm/remove wvw designations internally.

Alliance Fees:-100g/5 guilds in alliance with a predetermined max number of guilds per alliance and player cap. (20 guilds/500 players). Even a 5man roaming guild can come up with 20g/8weeks.

You can even put a Gem Fee for the max one, so guilds can pay like 1000/2000 gems every 8 weeks for maintenance

Why maintenance fees?1) determine your actual guild numbers vs inactive.2) Players that want to switch guilds/alliance can do so every 8 weeks and it will be the in the hands of the guild/alliance to accept them for a fee if they wish.3) I would actually recommend maintenance fees much higher than the proposed.

Just no. Don't monetize belonging to a guild.

Link to comment
Share on other sites

@obastable.5231 said:

@ixl.7190 said:Full Disclosure: The following wall of text is coming from the deepest salt mine on Blackgate.
snip

Please outline the part of your post that's supposed to be a strong and valid argument that
server pride
is a measurable metric to be used in providing a solution to the following:
  • coverage disparities
  • scoring
  • poor match-ups

Keep in mind this applies across
all time zones
, and that every time zone and every player is entitled to the same WvW experience as your prime time players (whatever that happens to be for your server). Your sentimentality is nice and all but for those who don't share in it there's a real lack of relevance.

I provided solutions that retain servers, but address all these issues. Regardless of the outcome of this restructure, the players on any server or Alliance ACTUALLY showing up and behaving ethically is what makes or breaks the game mode.

Link to comment
Share on other sites

@Caliburn.1845 said:From my perspective the reaction to the WvW changes have been mostly positive on these forums.

Overwhelmingly positive on the parallel Reddit thread.

And from people who don't read either, even more positive, although examples of this type of person are for most of us anecdotal.

The one unofficial poll that sort of asked the question on these forums had the changes at something like 78% support.

I would love Anet to do an official poll in-game to reach the largest amount of players, and am confident it would have more than a super majority(66%) in favor of altering WvW as discussed.

I wholeheartedly agree, as long as the the poll includes 3 options instead of just a yes/no.

Link to comment
Share on other sites

@ixl.7190 said:

@Caliburn.1845 said:From my perspective the reaction to the WvW changes have been mostly positive on these forums.

Overwhelmingly positive on the parallel Reddit thread.

And from people who don't read either, even more positive, although examples of this type of person are for most of us anecdotal.

The one unofficial poll that sort of asked the question on these forums had the changes at something like 78% support.

I would love Anet to do an official poll in-game to reach the largest amount of players, and am confident it would have more than a super majority(66%) in favor of altering WvW as discussed.

I wholeheartedly agree, as long as the the poll includes 3 options instead of just a yes/no.

What do you have in mind for a third option? Undecided?

Because the choice as presented by Anet at the start of this thread is very binary. Move towards the proposed Alliance solution. Or, leave WvW as it is.

There is no third party candidate in this race.

Link to comment
Share on other sites

@"Caliburn.1845" said:What do you have in mind for a third option? Undecided?

Because the choice as presented by Anet at the start of this thread is very binary. Move towards the proposed Alliance solution. Or, leave WvW as it is.

There is no third party candidate in this race.

"For WvW, ANET can start by admitting this game is five years old, anyone who didn't come back when PoF launched isn't coming back, bite the 'PR humilation of server mergers' bullet, and take a hard look at what they are using to determine server capacity restrictions. Next, choose the top 9 population servers in NA, and give everyone else free transfers to wherever they want for 30 days according to your new server capacity determination. Anyone logging in for the first time after this time frame ends gets 48 hours of free transfers. Once time is up and you've chosen your home by typing your server name into the same darn box i have to verify annoying item names in i want to delete, ANET gives you 500gems as a one time apology for making you move. People who don't declare a different server than they were on before this takes effect don't get anything. Yeah, some people won't be happy, but this affects the LEAST amount of customers, and gives a token consolation prize for the inconvenienced. If WvW is still imbalanced, then we can clearly say ANET tried and the players chose to let it fail. Now the concerned Dev team can concentrate on improving the actual game mode with the next six months instead of dealing with bad player behaviors counter-measures.

Another option is to invest in some more powerful physical game servers, take the maps up to 100 per side by making them bigger and resdesigned so that winning will require three zergs, instead of one, to control a map. One 3-server matchup/week and the top three servers get to host everyone else as linked servers. EotM continues as an overflow, but contributes score based on all existing EotM instances averaged out to count as one for contribution (with tight controls on when they open/close), and gives pips. This will mean people who are only wanting pips can stick to EotM, but players waiting on Eb/BLqueues get equitable rewards and the fight on EotM matters. Meanwhile, any karma train collusion on EotM will only serve to give the same score contribution to all three sides due to win trading the map simultaneously. If the community who will consistently WvW is as small as I think it is, this would work just fine. How do you determine who the top servers are to host these matches? A combined metric of player activity in WvW, based on actual activity, like real button presses. Could this be gamed? Sure, and I'm sure it could be improved on, but it is a different angle to attack the problem using mostly existing systems and retaining servers."

If they're interested they have my phone number on file and can use it should they need things spelled out in more detail.

Link to comment
Share on other sites

@"Gaile Gray.6029" said:A message from McKenna Berdrow:Even though making a choice between the two systems might seem like too drastic a change for some people, we have been exploring other designs to deal with WvW populations for years and we believe that World Restructuring or World Linking are the only solutions that meet our requirements.

Anet has already explored other options apparently. There is no third option they feel is viable. "Bigger maps, more powerful servers" seems a tad less viable for them then "Alliances"

Link to comment
Share on other sites

@ixl.7190 said:

@ixl.7190 said:Full Disclosure: The following wall of text is coming from the deepest salt mine on Blackgate.
snip

Please outline the part of your post that's supposed to be a strong and valid argument that
server pride
is a measurable metric to be used in providing a solution to the following:
  • coverage disparities
  • scoring
  • poor match-ups

Keep in mind this applies across
all time zones
, and that every time zone and every player is entitled to the same WvW experience as your prime time players (whatever that happens to be for your server). Your sentimentality is nice and all but for those who don't share in it there's a real lack of relevance.

I provided solutions that retain servers, but address all these issues. Regardless of the outcome of this restructure, the players on any server or Alliance ACTUALLY showing up and behaving ethically is what makes or breaks the game mode.

Retaining a server-based teams system isn't the practical solution to those issues, just a perpetuation of "same old, same old". Compare WvW with sPvP. Both are rooted in the same Conquest style design and we see the same "coverage, scoring, poor match-ups" issues occur in sPvP when one side has a DC or skill level between teams is off. All that is being done with WvW is making match-making more like sPvP so teams are matched up more equally every 8 weeks rather than 5 year old teams.

Link to comment
Share on other sites

@Chaba.5410 said:

@ixl.7190 said:

@ixl.7190 said:Full Disclosure: The following wall of text is coming from the deepest salt mine on Blackgate.
snip

Please outline the part of your post that's supposed to be a strong and valid argument that
server pride
is a measurable metric to be used in providing a solution to the following:
  • coverage disparities
  • scoring
  • poor match-ups

Keep in mind this applies across
all time zones
, and that every time zone and every player is entitled to the same WvW experience as your prime time players (whatever that happens to be for your server). Your sentimentality is nice and all but for those who don't share in it there's a real lack of relevance.

I provided solutions that retain servers, but address all these issues. Regardless of the outcome of this restructure, the players on any server or Alliance ACTUALLY showing up and behaving ethically is what makes or breaks the game mode.

Retaining a server-based teams system isn't the practical solution to those issues, just a perpetuation of "same old, same old".

Precisely. Retaining servers doesn't give them the means to shift populations around as they fluctuate, rather it locks us in to the same problem we currently have.

The game format itself is designed to always feel like a 2v1 scenario, because that's exactly what it is. It should feel like an uphill battle trying to control a map with 2 offensive teams coming at you from 2 different directions at the same time. This isn't the issue, and nothing should be done to change this any more than they already have (siege creep needs to be dialed back too). Changing map sizes or map caps doesn't change the population issue, or the scoring issue, or the fact that outside of a match-up with 3 servers having the same population & organization in the exact same time zones there are tens thousands of people who are suffering an awful game experience. Saying they should continue to suffer a shitty game design because of server pride (ie. not sharing your sentimentality toward your server) is disingenuous and, essentially, is trying to move the focus away from the core issues and instead implies that the problem is a lack of said sentimentality in others. It isn't, by the way, and it never was.

Link to comment
Share on other sites

@"Chaba.5410" said:Retaining a server-based teams system isn't the practical solution to those issues, just a perpetuation of "same old, same old". Compare WvW with sPvP. Both are rooted in the same Conquest style design and we see the same "coverage, scoring, poor match-ups" issues occur in sPvP when one side has a DC or skill level between teams is off. All that is being done with WvW is making match-making more like sPvP so teams are matched up more equally every 8 weeks rather than 5 year old teams.

One could also say World Versus World, and the notion of persistent battle for your server, was a key selling point of the game, and drastically changing it 5 years in, to the point where the notion of a persistent world server ceases to exist, would be a more appropriate change as part of the design of Guild Wars 3. So now we are just debating core game design and how much can be responsibly changed by a Developer. How far is too far. Where do we think the line is, in our own opinion.

Link to comment
Share on other sites

@ixl.7190 said:

@"Chaba.5410" said:Retaining a server-based teams system isn't the practical solution to those issues, just a perpetuation of "same old, same old". Compare WvW with sPvP. Both are rooted in the same Conquest style design and we see the same "coverage, scoring, poor match-ups" issues occur in sPvP when one side has a DC or skill level between teams is off. All that is being done with WvW is making match-making more like sPvP so teams are matched up more equally every 8 weeks rather than 5 year old teams.

One could also say World Versus World, and the notion of persistent battle for your server, was a key selling point of the game, and drastically changing it 5 years in, to the point where the notion of a persistent world server ceases to exist, would be a more appropriate change as part of the design of Guild Wars 3. So now we are just debating core game design and how much can be responsibly changed by a Developer. How far is too far. Where do we think the line is, in our own opinion.

And my response to that would be that the selling point was marketing speak for something that was the result of limited development resources at the time before a hard date on launch. They didn't have megaserver for PvE either at launch, but that got implemented pretty quickly, didn't it? In the meantime, the idea of battlegroups has been around for 2-3 years.

Link to comment
Share on other sites

@ixl.7190 said:

@Caliburn.1845 said:From my perspective the reaction to the WvW changes have been mostly positive on these forums.

Overwhelmingly positive on the parallel Reddit thread.

And from people who don't read either, even more positive, although examples of this type of person are for most of us anecdotal.

The one unofficial poll that sort of asked the question on these forums had the changes at something like 78% support.

I would love Anet to do an official poll in-game to reach the largest amount of players, and am confident it would have more than a super majority(66%) in favor of altering WvW as discussed.

I wholeheartedly agree, as long as the the poll includes 3 options instead of just a yes/no.

The poll on this forum with the 78% support has 5 options, the standard (strongly agree/agree/neutral/disagree/strongly disagree) you see on many surveys and evaluations, and the verbiage for each option is reasonably neutral. The strongly agree + agree is running at 75%+ versus the other 3 combined.

Link to comment
Share on other sites

@Chaba.5410 said:And my response to that would be that the selling point was marketing speak for something that was the result of limited development resources at the time before a hard date on launch. They didn't have megaserver for PvE either at launch, but that got implemented pretty quickly, didn't it?

Megaserver solutions to population decline in the most popular MMO, and other lesser ones, existed long before GW2, which is another reason why some people like me were looking for a new game to begin with. Once again, we are at impasse due to our preferences. One of us prioritizes the notion of building a community to be proud of, and then staying in it and fighting for it, the other values innovation and new experiences more (or whatever causes you to embrace the suggested change). Both preferences are completely valid.

Link to comment
Share on other sites

@"Euryon.9248" said:The poll on this forum with the 78% support has 5 options, the standard (strongly agree/agree/neutral/disagree/strongly disagree) you see on many surveys and evaluations, and the verbiage for each option is reasonably neutral. The strongly agree + agree is running at 75%+ versus the other 3 combined.

My bad, I would have been clearer if i had said "three solutions" instead of "options".

Link to comment
Share on other sites

@ixl.7190 said:

@Chaba.5410 said:And my response to that would be that the selling point was marketing speak for something that was the result of limited development resources at the time before a hard date on launch. They didn't have megaserver for PvE either at launch, but that got implemented pretty quickly, didn't it?

Megaserver solutions to population decline in the most popular MMO, and other lesser ones, existed long before GW2, which is another reason why some people like me were looking for a new game to begin with. Once again, we are at impasse due to our preferences. One of us prioritizes the notion of building a community to be proud of, and then staying in it and fighting for it, the other values innovation and new experiences more (or whatever causes you to embrace the suggested change). Both preferences are completely valid.

Creating fresh teams every 8 weeks from a total population is the megaserver version of WvW and the correct solution to population decline as you say has been done long before GW2. Both preferences are valid, but only one is the correct solution for the whole population.

Link to comment
Share on other sites

I cannot wait for this; such a great idea. Guild Wars for real!!!

Please let the first season come out at least 8 weeks before wow xpac. I would love for GW2 to absorb some of that wow population that used to play here while they are going through their down period of content. My hope is they love it enough to stay!

Link to comment
Share on other sites

@Chaba.5410 said:Creating fresh teams every 8 weeks from a total population is the megaserver version of WvW and the correct solution to population decline as you say has been done long before GW2. Both preferences are valid, but only one is the correct solution for the whole population.

And i still contend that merging the least active and populated servers affects the primary player base less, and tops off the remaining servers. I personally favor the first solution I presented (if we can assume that 9 servers is a sustainable size), because it avoids disrupting the most active player bases and established communities, and actually provides the desired reset button by allowing everyone the opportunity to step up and organize like Blackgate did if they choose to. If you guys wanna WvW, take advantage of the free xfers and get organized and show up for duty. The second part of my argument is to establish the fact that what is being proposed is easier to manipulate by motivated players than the current system, and that things will get much worse.

The easiest rebuttal is that if Megaservers were so successful, we'd still be playing those games instead of this one. Any fact can be used to support any argument with the right phrasing.

So once again, impasse.

Link to comment
Share on other sites

From my point of view this is a really welcome change. It will make match-ups more balanced and varied! And with alliance's of over 500 players there will still be much space for communities to form. Another big plus is that it opens up the system for guilds to recruit new players.

A flaw I see conceptually however lies in capping alliances around player count solely (and possibly guild number, although having that be very restrictive would be a real bad idea imo). An alliance formed by only the most dedicated WvW guilds would achieve a much higher playtime over the week than a big community guild with averagely active players would. And I believe the difference would be quite huge.If that is the case the developers would need to decide what player limit could be detrimental to world balancing in the long run and what would still be fine. Under the presumption that an alliance forms filled with only very active players. They'd need to set the player limit rather low and it'd be much more limiting for the already existing WvW communities than it should.A solution for that would be to set an activity cap for an alliance. This would be the limiting factor for alliances formed by the most active guilds and therefore allow for a less restrictive max player count. The old WvW communities wouldn't need to split up as much.

So here is my idea on how to help with that:1) Next to the player count (indicating e.g. 632/1000 players) there would be an activity bar (indicating e.g. 90% of maximum alliance activity reached). For example the devs could set 30% of a world's total playtime as the alliance's activity cap.2) Whenever a new player follows an alliance, by setting their WvW guild, his personal playtime/activity value is added to the alliance's activity level.3) Once the alliance's activity level reaches 100% it locks, preventing any more players from following it.4) If it were that simple, players could easily exploit the system by playing less for some time / only playing on EotM etc. This way whole servers managed to open for transfers in the past, and for alliances it would be far easier. Players certainly'd do it to overstack their alliance.5) To prevent this from happening, players automatically unfollow an alliance when it hits 110% of its max activity (=e.g. 33% of a world's total playtime). As they will need to refollow now the activity cap at 100% will be reached early, forcing at least one guild to leave the overstacked alliance.A possible problem with this is that player activity may be fluctuating way too much overall for this to be viable.

Link to comment
Share on other sites

@"ixl.7190" said:I personally favor the first solution I presented (if we can assume that 9 servers is a sustainable size)Since the assumption is wrong, it falls flat.

One of the core balance problems with WvW has always been the fixed size of a server. We had no real solutions - or at least, no real solutions Anet was interested with. This restructure allow dynamic "server" and matchup size to fix that.

Its like... imagine if there are 20 players in sPvP. Thats 2 matchups right? 5/5 + 5/5 people. Now, 4 people leave and they wont come back. Suddenly you have an imbalance. Say its 5/2 + 4/5. How are you going to fix that? The full teams dont want to move, they are good with that. The good enough matchup says nothing is wrong. If you shuffle the teams and make a 4/2 +5/5 thats still a roflstomp on one matchup. The WvW restructure offer a solution. We have 16 players. That is 4/4 + 4/4 matchup. Done. Is that too crowded for many people? Then we can have 2/2 + 2/2 + 4/4 matchup. So 2 new people enter the sides randomly. Cool. next 8 weeks we can have 3/3 + 6/6. Or 4/4 + 5/5. It doesnt really matter cause we can make any sizes and any matchups while trying to achieve some balance based on people both leaving and coming in.

For some servers its not needed "now" but its needed for the future of WvW. Player counts shift all the time. The honor of players to do it themselves is not enough.

Link to comment
Share on other sites

@Gaile Gray.6029 said:We also want to give titles related to the worlds on which players currently are playing when World Restructuring goes live.

Meh. I don't feel any dedication towards the world I'm currently on, only transfered there because of my guild. However, I do still feel connected to the server that got me into WvW and where I spent most of my days. Would it maybe be possible to choose which title we want to receive?

Link to comment
Share on other sites

@ixl.7190 said:

@"Chaba.5410" said:And my response to that would be that the selling point was marketing speak for something that was the result of limited development resources at the time before a hard date on launch. They didn't have megaserver for PvE either at launch, but that got implemented pretty quickly, didn't it?

Once again, we are at impasse due to our preferences. One of us prioritizes the notion of building a community to be proud of, and then staying in it and fighting for it, the other values innovation and new experiences more (or whatever causes you to embrace the suggested change). Both preferences are completely valid.

I think this is the crux of the matter, although you didn't mention the primary motive for those advocating the new system, which is getting rid of the imbalance created by people entrenched permanently on a dominant server. By removing the "permanent" unit from that of an entire world (server) to that of only a fraction of an entire world (alliance), Anet is looking to removing the disproportionate share of elite/hardcore players on just one or a couple of permanent worlds, and spreading those people to multiple alliances which will be matched up against each other instead of being permanently able to roflstomp almost every other world.

For many of us returning competitive balance to a game -- and games should be about healthy competition -- is far more important than maintaining permanent entire-world-sized "communities", especially when any given player will still be able to stay in a community of 500-1000 people. I seriously doubt there are that many people on any current server who can even name individually more than 499 people who are more than semi-familiar names in a chat and whom they frequently actively play with in wvw.

Link to comment
Share on other sites

So basicly this will make every guild stack in one server for domination and the other matches will be manipulated results to keep ktrain alive, ort they cant find decent and equal timezone groups to fight....

So far this changes feel like a placebo is on the work.... i would not expect otherwise from LamerNet.

Link to comment
Share on other sites

@Norbe.7630 said:500-1000 alliance limit.....alliance per world limit...?1 BG is how many alliance?

new question, will the new system sustain all of BG in 1 world or not :s

....the answer will decide the future of WvW

According to everyone on the forums, no.

BG has 24/7 queues on 5 maps (yes, people think BG queues OS) for 6/7 days of the week. Because that is the only way BG wins.

So... unless the alliance cap is 50000, BG can't fit on one.

Link to comment
Share on other sites

@"Norbe.7630" said:500-1000 alliance limit.....alliance per world limit...?1 BG is how many alliance?

new question, will the new system sustain all of BG in 1 world or not :s

....the answer will decide the future of WvW

The new worlds will likely have fewer players available then BG currently does. It looks to be that worlds will be around 2000-2500 players.Guilds are currently limited to 500 players. It looks like they may also make the alliance limit 500 players (with no guild limit set, just a player limit).The 'number' of players in an alliance isn't the problem (although, the larger the allowed alliance size, the greater the imbalance gets).

The first problem comes down to time played. If one alliance can get 500 dedicated players who all play 40 hours a week (which seems probable), than that one alliance will have 20000 hours of playtime a week. Organized, meta build, voice comm, playtime. It is going to be really hard to come up with two more worlds with alliances that will also be on equal footing. Anet will match them with two more worlds that have a similar total playtime, but those three worlds are NOT going to be equal. I imagine that two worlds will be able to create that level of organized coverage and playtime.

The second problem is that the map spawns are not equal. So, the most competitive matchup will likely consist of a BG-level organized alliance (possibly 2) and a third world that is not quite as well organized. That world will be red in the matchup and is handicapped by map design before players are even alotted. Desert borderland is harder to defend than the alpine borderlands and the layout in EB puts red at a further disadvantage.Green has the easiest third to defend and is the strongest world in a matchup.Blue has the next easiest third to defend. If they are as strong as green in a matchup, being blue is not really a disadvantage.Red has the hardest third to defend, and their keep can be trebbed from the castle (almost always owned by the strongest world).Tower design, location, and distances make red's third the worst spawn to get, and it is given to the losing team in a matchup.

It's not a question of IF someone is going to stack an alliance, but how strong that stack will be and whether or not at least one other stack can match it.That said, if Anet doesn't switch the spawns there is no chance in balancing these matchups. If the goal is to create some semblence of balance, than the weakest world in each matchup has to be green and the "strongest" world in a matchup has to be red. At a minimum, the red keep HAS to be moved back so that it can NOT treb (or be trebbed from) SMC. They can work on the rest of the redesign of red's third later, but that HAS to happen.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...