(edited by Stand The Wall.6987)
Lets talk population overhaul pros and cons.
one more assumption for you
https://forum-en.gw2archive.eu/forum/game/wuv/Population-Balance-1/6058497
Henge of Denravi Server
www.gw2time.com
one more assumption for you
https://forum-en.gw2archive.eu/forum/game/wuv/Population-Balance-1/6058497
Thats interesting if true. So their alliance system will be similar to the megaserver problem of newer servers (potentially but probably) being underpopulated.
inb4 qq over this.
Updated OP.
Edit
Removed. Not a trusted source.
(edited by Stand The Wall.6987)
You forgot one.
PPT and PPK scaling based on population – Scaling max PPT down will reduce extreme score curves during times of population imbalance and not allow one server to exploit the other two servers inactivity for matchup winning gains. Scaling PPK up will make players matter more during the same times.
Pros
- It doesnt change the current WvW layout.
- It fixes whats probably the biggest issue in WvW.
- Its universal across all tiers no matter the time.
Cons
- It will make nightcappers cry salty tears.
You have to start WvW redesign from even further away – from the very basics of scoring and points and match system. Basically deciding if WvW is place for players to fight each other as mass-PvP or not.
-If you scrap the silly 24 h PPT system you can design much better WvW.
-Points per Capture for example creates completely different gamemode where effects of timezones are greatly reduced.
-You can even avoid all objectives-based points and just increase score from kills.
-Or you can reduce the time WvW is available, for example if WvW will be open for 1 hour only every 6 hours and scores are increased once after the open-hour ends.
-Or you could scrap the static 4-maps system which is horrible for finding an enemy and replace it with dynamic maps based on number players currently logging into WvW.
-Or you could change the number of servers fighting in the matchup, so there wont be only 3 servers but 6 or 9.
Both of you have great and valid points, however I dont think this is the right thread to be discussing that because the scoring system supersedes population. I updated the thread title to reflect this.
(edited by Stand The Wall.6987)
You have to start WvW redesign from even further away – from the very basics of scoring and points and match system. Basically deciding if WvW is place for players to fight each other as mass-PvP or not.
-If you scrap the silly 24 h PPT system you can design much better WvW.
-Points per Capture for example creates completely different gamemode where effects of timezones are greatly reduced.
-You can even avoid all objectives-based points and just increase score from kills.
-Or you can reduce the time WvW is available, for example if WvW will be open for 1 hour only every 6 hours and scores are increased once after the open-hour ends.
-Or you could scrap the static 4-maps system which is horrible for finding an enemy and replace it with dynamic maps based on number players currently logging into WvW.
-Or you could change the number of servers fighting in the matchup, so there wont be only 3 servers but 6 or 9.
These suggestions have been circulating the boards for a very long time. Sadly, many have been countered, such as the problem with Points Per Capture being that players would (when the enemy have active capping teams) leave objectives in enemy hands to prevent them from capturing them.
I like the time-window idea, where you score fewer points overnight. Do I know it’ll work? Nope, I just think it’s better than nightcaps.
Mentioned some variations in a few other threads, if you’re interested I’ll sum them up here:
-Alliance-
Servers -> Factions
X servers -> Alliance
3 Alliances = Match-up
By combining multiple "Factions" into one alliance, this can shift every week, so population has dynamic adjustment each week, prevents band-wagon.
Pro
* Negates band-wagon and server stacking
* Keeps some server identity
* Dynamically adjust number of tiers to population
Con
* Removes low pop play style
* Reduces server identity
* Cross faction communication and cooperation might be problem
-Time Slice Server-
Each server is assigned a 6 hour slot out of 4 possible. That server is only active during that time period. Ex: NA, EU, Sea, Ocx (?).
Each user can join up to 4 different servers, from different time slices.
That way you can have match-ups with dedicated servers for each time-slice.
Pro
* Concentrates players more in time zones with less players
* Still retain a large amount of servers for the populated times
* Retains some server identity
* Allows people to explore different servers and cultures in other times
Con
* Close off while servers change from one to another server for a few minutes
* Lack of server identity for those playing in other hours for servers (*)
* Probably a whole slew of other things I just can’t think of right now.
* (example Kaineng would be considered a night server, so one guild I’m a member off that only plays early na prime would lose their server pride)
“Understanding is a three edged sword: your side, their side, and the truth.”
“The objective is to win. The goal is to have fun.”
You missed this one (it’s way back in the pages so I understand why ).
https://forum-en.gw2archive.eu/forum/game/wuv/Suggestion-Paired-tiers/first#post6030243
Pros:
- Flexibility in matches
- Retains server identity
Cons:
- Would need some work by Anet to allow us to choose map instances
- Comms between servers – would need a WvW color chat channel adding at least (allied voice comms preferred).
Feel free to add pros and cons I haven’t thought of.
If it’s anything like EOTM it won’t address any issues.
The exact same population & coverage issues are present in this map.
Just tried to level up some hero points tonight on an Engineer and spent the whole night getting wiped by zergs 3-4 times our size. Must have gone 2 hours without a successful tower cap – they were everywhere.
80 Ranger (3), 80 Warrior (3), 80 Thief (3)
80 Ele (2), 80 Engi (3), 80 Rev (2)
If it’s anything like EOTM it won’t address any issues.
The exact same population & coverage issues are present in this map.
Just tried to level up some hero points tonight on an Engineer and spent the whole night getting wiped by zergs 3-4 times our size. Must have gone 2 hours without a successful tower cap – they were everywhere.
Ah but tbh, other games have already solved this “problem” – transparency in map population and transfering between instances.
In theory the system would work like this: on instance 12 the population is 60/12/6. Badly imbalanced, right? So then there is a call to arms. Either the other two sides reinforce instance 12 with troops so that the battle can continue – they get a boon for joining of course – or the instance shut down, handing a victory and points to the outmanning force. The closing of the instance will ensure that the leftover forcs are spread out and that the victor cannot exploit inactivity (they still get points, but not like ticking 600 for 8 hours).
Server identity here would change into something… else. I call them Anet guilds. Servers literally become ingame guilds. So in the case above, there would just be a note in the main chat… “Dont worry, Vabbi got this”.
(edited by Dawdler.8521)
The alliance system is already mentioned here and the time slice system creates more problems then it solves.
Thats an interesting idea, but its too similar to the current alliance system mentioned on reddit. There are unique things about it but we could go for days on all of the ways to variate the main overhauls.
Mentioned some variations in a few other threads, if you’re interested I’ll sum them up here:
-Alliance-
Servers -> Factions
X servers -> Alliance
3 Alliances = Match-upBy combining multiple “Factions” into one alliance, this can shift every week, so population has dynamic adjustment each week, prevents band-wagon.
Pro
- Negates band-wagon and server stacking
- Keeps some server identity
- Dynamically adjust number of tiers to population
Con
- Removes low pop play style
- Reduces server identity
- Cross faction communication and cooperation might be problem
More Cons
- Too complicated
- Faction pride evaporates overnight
- Gameplay seems vapid and unsubstantial
- People will quit
- Gametype will die
Doesn’t seem such a hot idea. UNLESS, the alliances are somehow permanent.
The alliance system is already mentioned here and the time slice system creates more problems then it solves.
Since the Alliance system you listed above was about 3 guilds, I listed another one using X servers, with variation. Do as you please, just listing options.
More Cons
- Too complicated
- Faction pride evaporates overnight
- Gameplay seems vapid and unsubstantial
- People will quit
- Gametype will die
Doesn’t seem such a hot idea. UNLESS, the alliances are somehow permanent.
- Not really complicated at all, since all will be run by a machine in the background. Really not much harder than the current match-up/glicko checks each week.
- Could be, don’t know. To fickle a thing to predict.
- It will be more EotM’ish, but I’m almost certain whatever we do get will be more EotM’ish anyway. So at this point it’s mostly about getting the lesser evil. And as I’ve said before about this specific idea, it is “a little bit of everything, and a little bit of nothing. The ultimate compromise.” Quite aware it isn’t perfect.
- Sad fact here is that they will do no matter what gets done (or not). So it isn’t really a good point.
- Sad fact here is that people will say this no matter what gets done (or not). So it isn’t really a good point.
- If you make the alliances static, you retain most of the existing problems we have, and lose most of the advantages with moving over to such a system as this. AKA you hate it, I get that. I wouldn’t be to overly fond of it myself either to be honest. Just posting ideas to spark discussion.
“Understanding is a three edged sword: your side, their side, and the truth.”
“The objective is to win. The goal is to have fun.”
Keep it the way it is
no problem solved
Server alliances
EoTM style but with an option to where to go to
Server merge
another EoTM style but definitely delete some servers
Dynamic server capacity/ megaserver
if WvW is defined as World vs World, is this a GvG Galaxy vs Galaxy or MvM, wait M2M i love thier songs lol
PPT and PPK scaling based on population
we all want to be fair here dont we?
will this spawn a P2Win item for fair scoring?
Gate of Madness
1. Server merges, maybe have only 2-3 tiers total.
2. Make PPK have equal value to today’s PPT for total scoring.
3. Impliment PPC (and PPD points per defense) and replace the otehr half of PPT with it.
4. Tie best/maximum personal rewards to killing players or capturing objectives while in a group of 25 or less. The bigger the blob, the less bonus it gets.
5. Watch and profit.
Communities preserved with extra people joining in, no ETOM style crap, and people will overstack / blob less to actually get stuff that is worth getting. Win win win on all counts.
I hope they do SOMETHING at least.
YB wins T1 other servers quit…
TC stacks in T2 for a glicko push others servers quit… (thanks T1 quitters)
Tried a little T3 morning sesh but meh guess mornings are all PVD and/or empty maps
Still no actual word on when, or even if this is going to happen?
(edited by displayname.8315)
First, let us deal with the mechanics that really have chased players away.
1. Broken and OP arrow carts – REMOVE
2. Shield Gens – REMOVE
3. Tactivators – REMOVE
4. Automatic updating of keeps and too high HP walls and doors resulting – lower to amount before HoT came out
5. Too much PvE in the DBL’s – REMOVE
6. Scoring – redo or remove GLICKO
7. Rotating maps
The above points will help to bring back players then we can deal with what to do with the population after it comes back up a little.
1. Server merges, maybe have only 2-3 tiers total.
2. Make PPK have equal value to today’s PPT for total scoring.
3. Impliment PPC (and PPD points per defense) and replace the otehr half of PPT with it.
4. Tie best/maximum personal rewards to killing players or capturing objectives while in a group of 25 or less. The bigger the blob, the less bonus it gets.
5. Watch and profit.Communities preserved with extra people joining in, no ETOM style crap, and people will overstack / blob less to actually get stuff that is worth getting. Win win win on all counts.
Some Cons:
1: This will remove any low pop play style, and arguably most existing communities.
2: Like PPK in principle but from the stats I’ve seen on it it tends to even out a lot. Might end up making every match "close" to the end, to the stage where people might no longer care about close scores.
3: Depending on the implementation of PPC/PPD, worst case could end up as above that you never get away from each others in points. With a good implementation would be more interesting.
4: Do like.
Think it has some interesting parts, though I personally am very much against getting stuck on a high pop server. 2+3 sounds interesting, but I wonder if it might completely negate winning/losing by more than say 10%, and that would be hell with the current Glicko system.
First, let us deal with the mechanics that really have chased players away.
1. Broken and OP arrow carts - REMOVE
2. Shield Gens - REMOVE
3. Tactivators - REMOVE
4. Automatic updating of keeps and too high HP walls and doors resulting - lower to amount before HoT came out
5. Too much PvE in the DBL’s - REMOVE
6. Scoring - redo or remove GLICKO
7. Rotating mapsThe above points will help to bring back players then we can deal with what to do with the population after it comes back up a little.
Agree on some (1,2,3)
Mostly wanted to comment on #4:
Just an idea, if the auto upgrade only went to the first level. And the 3rd level could be activated by the tactivators. NB: requires fortified walls and gate as 2 separate upgrades. That should be fairly cheap and easy to get (while in WVW!!! not pve). And takes tactivator spots, so less other silly stuff.
Still would like to see workers (very) slowly repair damaged walls and gates as long as they gain supply and are alive. So boring looking for 1% damage to walls to top them off.
(Not going to keep beating the dead horse about rotating maps, think everyone knows my stand on that by now)
“Understanding is a three edged sword: your side, their side, and the truth.”
“The objective is to win. The goal is to have fun.”
Sad it before say it again, easyest solution remove PPT then only points your server will gain is from kills, upgradeds and capping/defending buildings. So if there is no enemy the only Points you can gain your server is Points that the enemy can get back when you leave.
What do tou think about factions?
3 Servers → 1Faction.
Low +mid +High population → 1Faction.
The server pride is the same, you are playing with the same people. Maybe, the lenguage….like what happen in Baruch Bay, that is the only Spanish server.
What do tou think about factions?
3 Servers -> 1Faction.
Low +mid +High population -> 1Faction.
The server pride is the same, you are playing with the same people. Maybe, the lenguage….like what happen in Baruch Bay, that is the only Spanish server.
I would prefer all servers having high population. If they don’t fix population problems there won’t be any point in WvW.
the issue is each servers not holding more than 80 players on each BL, if they resolve this, we could have much less servers that would bahave liek factions and something like a continent each server their spawn.
I think this is the real issue, that does not allow WvW to get improved, do they cover that with gimmicks after gimmicks.
the issue is each servers not holding more than 80 players on each BL, if they resolve this, we could have much less servers that would bahave liek factions and something like a continent each server their spawn.
I think this is the real issue, that does not allow WvW to get improved, do they cover that with gimmicks after gimmicks.
If you want larger numbers of players than 80+80+80 on a borderland map, then the lag becomes unbearable.
Personally would prefer a lower limit (60 perhaps), and rather open more copies of maps. And deal with over-population in that way. (Con: stacking would mess this up bad).
Dynamic Maps would solve this.
“Understanding is a three edged sword: your side, their side, and the truth.”
“The objective is to win. The goal is to have fun.”
the issue is each servers not holding more than 80 players on each BL, if they resolve this, we could have much less servers that would bahave liek factions and something like a continent each server their spawn.
I think this is the real issue, that does not allow WvW to get improved, do they cover that with gimmicks after gimmicks.
this was actually discussed during launch and anet respond with technical difficulty. it is the same for why you can only hit 5 targets with AOE, it is because of technical difficulty.
the server is simply incapable of handling that large amount of data. though personally i have seen way larger number of people at one spot fighting in other mmorpgs
Henge of Denravi Server
www.gw2time.com