Jump to content
  • Sign Up

Server Relink


Talindra.4958

Recommended Posts

Fulls will get mediums or high. VH will get Highs and one VH will be combo together if it were to happen right now on NA. They can't exactly give everyone a diff link everytime so some combos see alot of each other. If you are always top 4 in hours you only got basically 4 med options

Link to comment
Share on other sites

The biggest problem is linking links; who are already quite populated or prone to bandwagoners, to already populated hosts.

Linking Deso and FR hasn't been that bad, apart from the language barrier, but if they link us with an international EU server, the more likely the link will get bandwagoned.

It may be worse than having a lower overall population than alot of the other more populated servers, who are more stacked with guilds or random players, on either the link or host, but most of the time, it's the link.

Link to comment
Share on other sites

Alliances will fix it soon. Just be patient. I mean they just announced them recently. <3

While waiting I will enjoy that great baruch bay "we suck at the game and lose every single balanced fight, so we overrun you every night at 4am with 80 players" experience this matchup gives everyone else since 2012.

Link to comment
Share on other sites

Id like to suggest ANet rethinking what does their "full server" term mean. Because cutting link from a server, seeing it drop from T1/2 to T5 and struggling even there, with 1 map que at best during primetime and keeping that server "full" is ridiculous. How many times it already happend, server theoretically full, no link, game unplayable for 2 months so ppl started transferring to other realms which ended in dead server, lol

Link to comment
Share on other sites

@"Widmo.3186" said:Id like to suggest ANet rethinking what does their "full server" term mean. Because cutting link from a server, seeing it drop from T1/2 to T5 and struggling even there, with 1 map que at best during primetime and keeping that server "full" is ridiculous. How many times it already happend, server theoretically full, no link, game unplayable for 2 months so ppl started transferring to other realms which ended in dead server, lol

They definitely dropped the population thresholds for a server to reach it's cap, since last time around. Last time around, you even had T1 servers like WSR and RoF open to transfers and Riverside dropped to High pop for a week.

It definitely isn't full if there are hardly any prime time queues, or even only two queues on reset night which turns into one queue after two hours. Or you have servers with barely enough population to have enough coverage, but you have servers like AG and RoS, who have so many guilds. On Tuesday, I also noticed 5 FSP/UW WvW streamers during prime time including Mighty Teapot, which is ridiculous.

We lost 5 guilds earlier this year and gradually our activity has dropped, but some link Servers still remain open and far more stacked right now, along with their hosts.

Link to comment
Share on other sites

@CrimsonNeonite.1048 said:

@"Widmo.3186" said:Id like to suggest ANet rethinking what does their "full server" term mean. Because cutting link from a server, seeing it drop from T1/2 to T5 and struggling even there, with 1 map que at best during primetime and keeping that server "full" is ridiculous. How many times it already happend, server theoretically full, no link, game unplayable for 2 months so ppl started transferring to other realms which ended in dead server, lol

They definitely dropped the population threshold since last time around, where you even had T1 servers likw WSR and RoF open to transfers and Riverside dropped to High pop for a week.

They likely did.

Think about if they have placed two servers as the ‘full threshold’. One on EU and the evil empire on NA. So, when those two drop population, the threshold drops.

By doing this, it would allow them to gradually push redistribution to multiple servers, increase the number of ‘full servers’ and gradually spread out the players.

Links become more valuable and with up to 12 ‘full’ servers, you are likely to have less room to move to other servers.

People have clamored for even distribution of players to ensure better population based matches.

Of course this was to be the intent of alliances, but we know how that story has played out.

What it doesn’t account for, and alliances wasn’t going to account for, is time zone play hours.

Link to comment
Share on other sites

Anyway, maybe four tiers is better for the EU, but removing some servers would be easier to balance, than 2 links for two host servers.

At least back when we last had Four tiers, it was still quite active on my Alt, but even when Gandara had two links, they weren't overpopulated.

Now generally, the population is spread out in some ways, but some servers are lacking in population, because they are locked out from transfers or cause people or guilds or random players still stack on the same couple of Servers.

Of course it will all change somewhat, next Friday.

Link to comment
Share on other sites

the next relink date would be 28 August if correct :)

Totally understand that servers linking are difficult task for dev as there are too many complex variables to manage. But as a player, I really like to stress out to Arena net that, Please do not relink the same servers consecutively. If Anet has plan to keep this two servers together permanently, please inform us or communicate with us, I believe players will not mind the idea of permanently merch two servers together providing players interest are looked after. We would be happy to be provided with a free server transfer to Gunnar server or another unpopulated minor server by choice not by force and have FoW server permanently removed.I cant represent the entire FoW community but many of us has enough fun and collaboration with Gunnar and are looking forward to play with a different major server in the next relink.

Link to comment
Share on other sites

@DKRathalos.9625 said:

@Fish.2769 said:31st July is next re-link. It's every 2 months of un-even months now since it got pushed back over Xmas.

I sometimes misstiming the relink, I thought it supposed to be 24th if counting 8 weeks from 29th May. How to calculate? is it 8 weeks or 9 weeks?

It's 8 weeks but usually the last week of the month so the 31st is a Friday so the last week technically so maybe 9 weeks this time it seems

Link to comment
Share on other sites

@"KrHome.1920" said:Alliances will fix it soon. Just be patient. I mean they just announced them recently. <3

While waiting I will enjoy that great baruch bay "we suck at the game and lose every single balanced fight, so we overrun you every night at 4am with 80 players" experience this matchup gives everyone else since 2012.

your kidding right?

Link to comment
Share on other sites

@"CrimsonNeonite.1048" said:The biggest problem is linking links; who are already quite populated or prone to bandwagoners, to already populated hosts.What you are describing there, though, basically is the system.

The servers prone to bandwagoners are the underpopulated servers and they are hitched to a quite populated server. The transfers occur once the linking is done so it is people transfering to the respective host server that they want to play with's link. That's why the link servers that have been pushed up to host size have shifted over the past few years. The same goes for the host servers that are generally regarded as the stacks. They were originally the first generations of links. Since they became hosts some have lingered and some have fallen back to link status, mostly depending on the activity of their corresponding "fight" guilds (as the GvG guilds tend to transfer back and forth anyway; with a well established server mainly serving as an attraction for them between active matchups).

@"Strider Pj.2193" said:What it doesn’t account for, and alliances wasn’t going to account for, is time zone play hours.One of the saddest bits in all of this is that if ArenaNet had reacted in anything that would qualify for anywhere near of a timely fashion they might have been able to save everything with just a proper GvG Arena and a Scoring system based on live and tuned "outnumbered". That is so much easier to build. All they had to do was do that in any time that wasn't completely unacceptable. They even have the damn tech.

Now they are basically falling behind on repairing things that has broken since they first started buying time, and like you say, still need to fix A after they have fixed B.

Link to comment
Share on other sites

When Anet introduced this Linking system, they explained that the idea isn't to give every linked pair/triple/quad/whatever a fighting chance to compete for top spot. The goal was simply to provide more even sides in the tiers.

Immediately, the system was abused by server transfers who wanted to "game" the system in their favour. Arenanet was disincentivised from changing anything because they benefit from gem sales for these transfers. So far, so obvious. But this gaming of the system means that the intended purpose never came about.

What isn't discussed much is the alternative. A system where the developers want each linked pair/triple/quad etc to have a fighting chance to compete for the top spot. Sure, it means work, it means changing things which benefit Anet such that they don't, it means remaking the game-mode in such a way that, heaven forbid, people might want to win it!

But no, #lazy is the order of the day, month and year-on-year instead. Lest we forget; WvW was abandoned.

Link to comment
Share on other sites

@Svarty.8019 said:What isn't discussed much is the alternative. A system where the developers want each linked pair/triple/quad etc to have a fighting chance to compete for the top spot. Sure, it means work, it means changing things which benefit Anet such that they don't, it means remaking the game-mode in such a way that, heaven forbid, people might want to win it!I don't disagree with the spirit or message of your post.

I just want to reiterate what was said above.

If they fixed scoring to not be about numbers/coverage (ie., simply not handing out score when outnumbered is flagged) they wouldn't need to balance out populations for the chance to compete for a top spot or win by definition of the matchups and ladder.

Population imbalance and not having a fighting chance may still be something experienced when actually fighting. However, a system where scoring and laddering is more reflective of when-populated performance will also match servers up more accurately according to the content they can create and reciprocate. Such a system is very easy to design and implement. They already have the tech for it. They have simply chosen not to do it even though we have pretty much been talking about it here since they developed the tech for the July 9, 2013 patch. That, is 7 years.

They have all these systems, all of this tech, they just aren't using them to effect. The Wiki even says that outnumbered affects Skirmish points since 2017, so then it is simply a question of changing a couple of values in a database to solve the problem. However, their systems are half finished, not iterated upon, they reinvent the wheel a hundred times over and their design/systems design comes off as weak and uninsightful as a result.

Link to comment
Share on other sites

@subversiontwo.7501 said:

@Svarty.8019 said:What isn't discussed much is the alternative. A system where the developers want each linked pair/triple/quad etc to have a fighting chance to compete for the top spot. Sure, it means work, it means changing things which benefit Anet such that they don't, it means remaking the game-mode in such a way that, heaven forbid, people might want to win it!I don't disagree with the spirit or message of your post.

I just want to reiterate what was said above.

If they fixed scoring to not be about numbers/coverage (ie., simply not handing out score when outnumbered is flagged) they wouldn't need to balance out populations for the chance to compete for a top spot or win by definition of the matchups and ladder.

Population imbalance and not having a fighting chance may still be something experienced when actually fighting. However, a system where scoring and laddering is more reflective of when-populated performance will also match servers up more accurately according to the content they can create and reciprocate.

Such a system is very easy to design and implement. They already have the tech for it.

Twiddling with the scoring mechanism was a cheap and dirty way to fix a problem that should be about the mechanics of objectives, capturing and battling over them. Instead it seems that we get the easy-way-out mentality every single time. Some might call it efficiency - I don't - and it's coming back to bite Anet on the bottom now that people are becoming weary of the same old tripe.

Link to comment
Share on other sites

@Svarty.8019 said:

@Svarty.8019 said:What isn't discussed much is the alternative. A system where the developers want each linked pair/triple/quad etc to have a fighting chance to compete for the top spot. Sure, it means work, it means changing things which benefit Anet such that they don't, it means remaking the game-mode in such a way that, heaven forbid, people might want to win it!I don't disagree with the spirit or message of your post.

I just want to reiterate what was said above.

If they fixed scoring to not be about numbers/coverage (ie., simply not handing out score when outnumbered is flagged) they wouldn't need to balance out populations for the chance to compete for a top spot or win by definition of the matchups and ladder.

Population imbalance and not having a fighting chance may still be something experienced when actually fighting. However, a system where scoring and laddering is more reflective of when-populated performance will also match servers up more accurately according to the content they can create and reciprocate.

Such a system is very easy to design and implement. They already have the tech for it.

Twiddling with the scoring mechanism was a cheap and dirty way to fix a problem that should be about the mechanics of objectives, capturing and battling over them. Instead it seems that we get the easy-way-out mentality every single time. Some might call it efficiency - I don't - and it's coming back to bite Anet on the bottom now that people are becoming weary of the same old tripe.

It's not cheap and dirty, it's really simple and effective given its investment in time and resources. They could simply inverse the trigger from 25% (or w/e it is) to 75%. They could raise the impact of outnumbered on scoring. They could raise it to 100% if need be. The end result would be that only numerically balanced fights and captures provides (relevant) score.

That is what the system already does in terms of mechanics, it is just screwed down so low that it triggers on an irrelevant value and it adds an irrelevant impact or factor to the overall scoring. Had they actually iterated on the system and understood it's relevance in the mode to actually design for it, it could have been effective regardless of comparative time and resources spent. It is a good mechanical design on a concept level. Making things unecessarily costly or heavy-resourced is not a good, clean or luxurious way of doing things. That is time and resources wasted that could have been put onto other things that would have made the mode better. If anything, wasting time and resources is the muddled and dirty way to do things.

Again, it's not a programmer issue, it is a game designer issue. The solution is as simple as entering a database and changing figures. It is "balance" on a systems level. It is the bread and butter of that profession. It is their daily routine. It's a mode/systems level equivalent of what CmC does to classes and much as with CmC's impact on class balance, it is simply a question of understanding the systems that creates the world/life.

Also, the simplicity of it means that it could easily be tried. That is iterative design. They could change the numbers for 8 weeks and see the results. If the results are negative they can simply change it back or change it to something else. It only requires the bare minimum of attention and competence.

Link to comment
Share on other sites

@subversiontwo.7501 said:

@Svarty.8019 said:What isn't discussed much is the alternative. A system where the developers want each linked pair/triple/quad etc to have a fighting chance to compete for the top spot. Sure, it means work, it means changing things which benefit Anet such that they don't, it means remaking the game-mode in such a way that, heaven forbid, people might want to win it!I don't disagree with the spirit or message of your post.

I just want to reiterate what was said above.

If they fixed scoring to not be about numbers/coverage (ie., simply not handing out score when outnumbered is flagged) they wouldn't need to balance out populations for the chance to compete for a top spot or win by definition of the matchups and ladder.

Population imbalance and not having a fighting chance may still be something experienced when actually fighting. However, a system where scoring and laddering is more reflective of when-populated performance will also match servers up more accurately according to the content they can create and reciprocate.

Such a system is very easy to design and implement. They already have the tech for it.

Twiddling with the scoring mechanism was a cheap and dirty way to fix a problem that should be about the mechanics of objectives, capturing and battling over them. Instead it seems that we get the easy-way-out mentality every single time. Some might call it efficiency - I don't - and it's coming back to bite Anet on the bottom now that people are becoming weary of the same old tripe.

It's not cheap and dirty, it's really simple and effective given its investment in time and resources. They could simply inverse the trigger from 25% (or w/e it is) to 75%. They could raise the impact of outnumbered on scoring. They could raise it to 100% if need be. The end result would be that only numerically balanced fights and captures provides (relevant) score.

[..]

Also, the simplicity of it means that it could easily be tried. That is iterative design. They could change the numbers for 8 weeks and see the results. If the results are negative they can simply change it back or change it to something else. It only requires the bare minimum of attention and competence.

Look my lad, I know cheap and dirty when I see it, and I'm looking at it right now! (not your post, it's a parrot sketch reference).

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...