Guild Wars 2

Keviin Snow.5760:

a way to cheat , u can glide into tower , if you are fighting they can glade away into a tower

Away from the fact that it is not easy to glide out of a fight, especially up into a tower
it doesn’t really make a difference, if you

• glide
• stealth
• Mist Form
• …
  into a tower.

DemonSeed.3528:

Dayra.7405:

It’s a nice first step. But of course it could be made even better, till we finally have air and land battles.

I’m still waiting for underwater battles, we need submarines. All jokes aside, where are the exciting water battles? Underwater skills and weapons need a look at. Also give us an underwater map.

In fact, the alpine border lands had once a very large lake where today the ruins are. And yeah, I remember large underwater battles in this lake. I also remember very nice underwater battles on one of the first PvP maps.

But there is also a diffeence: a battle is either land or water, but it can be both land and air.

It’s a nice first step. But of course it could be made even better, till we finally have air and land battles.

Great News. Will definitely add some tactical variations to defense!

Finally Ranger and Ballista have a use

Will there be gliding and territories in EotM as well?

maybe.5086:

I thought the current www population algorithm is not based on inactive server population. So I don’t understand what is so important about purging inactives.

The point the OP is making:

a) not all inactive population is always inactive, there is a not so small percentage of “fair-weather” player.
b) “fair-weather” player come when it is running good, turning a “running-good” into an imbalanced match. And staying away when it’s not running good, turning a “not-running-good” into an imbalanced matcch as well.
c) the base-population can always join a match on the side of a server, independently how full this server is.

and last but not least:
d) I am sure there is a not so small amount of people that have accounts on several servers. Of couse only, only one is active, but all other have their reserved place on servers, and become active, when it is needed least, namely when the ser ver is already winning.

All Matters Fecal.9560:

Zergs and groups avoid PvP fights, people capturing towers/camps/zones for little to no apparent reason.

Yeah basically it’s a PvE-race (who captures more faster) in a PvP-zone.

How much PvP it is strongly depends on time (evening 8-11pm middle US for NA / CEST for EU) is much more pvp that the rest of the time) and the group you belong to. Some guild groups just fight eah other and ignore most PvE.

Chaba.5410:

Someone posting from their F2P account?

F2P cannot post into the forum
https://help.guildwars2.com/hc/en-us/articles/230165307-Account-types-Free-Core-HoT

Tournament is a bit boring when the winner is determined by ANet via full-status and server-linking.

Rather sooner than later you swim in Tome’s of Knowledge (and Birthday lvlup 20-40 lvl so far), allowing you to be 80 with as many chars you are willing to buy slots for. So use your lvl-80-boost now, before you don’t need it anymore.

PS: And nothing hinders you to explore the world and story with a lvl 80. I even like that more, as I don’t like that during lvling I need new clothes and weapons every half an hour

zealex.9410:

Dayra.7405:

Better: Give people an option (check-box in preferences) to auto-respawn on nearest unblocked Spawn.
Choosing this parameter should gurantee that you do not loose a bag.

Would help to:
* reduce the inactive, lying around waiting for the win that does not happen as they lie around
.* reduces the winner takes (revive) all.

PS: I was once lying beside my dropped Dawn in SM outer-ring, made it to get it with respawn.

Auto loot has a max range. If you die before the loot drops (op’s lazy problem) then every loot that drops wont be picked and when you isnta respawn you want be able to loot it automatically.

“Auto loot has a max range” I think this is irrelevant, the bags appear at your feet, and your feet are in auto-loot-range.

zealex.9410:

I also cant see how it wont reduce the inactves since i would just afk in keep.

Me not either, did I wrote it will affect living?

But I think it will reduce the number of people that do not respawn as are undecided, if they will be faster in action, by waiting for revive or by respawn.

Better: Give people an option (check-box in preferences) to auto-respawn on nearest unblocked Spawn.
Choosing this parameter should gurantee that you do not loose a bag.

Would help to:
* reduce the inactive, lying around waiting for the win that does not happen as they lie around
.* reduces the winner takes (revive) all.

PS: I was once lying beside my dropped Dawn in SM outer-ring, made it to get it with respawn.

Haha, a divide by nearly 0

In fact Vabbi’s start rating is not 0, but around 1k

Whow, thanks. What a cool system, random catapults you somewhere and then you trickle week by week to the place you belong.

Title says it, are we – after 1 reset try – already back to glicko-random reset?

XenesisII.1540:

When you move you are penalized by not being able to earn pips until you’ve been on the server one entire matchup, which breaks the cycle of the commitment pip requirement of 100 pips the previous week.

yes you cannot earn pips after transer, but you can make wood before you transfer on your old server (Fri-Thu), transfer friday afternoon and be ready on the new server for the new match. So no break.

Korgov.7645:

Sylvyn.4750:

Korgov.7645:

Due the population imbalances we will see lopsided biweekly repeated matchups. “Oh it’s an even week. I don’t think I will even bother to play. See you all next week.”

You’re making the assumption that server populations remain constant…any changes could push a different server up or down as a result.

There is still no migration control beyond setting a server full, transfer gem costs, and recently the loyalty pips. Stacking has happened and nothing’s been done to rebalance the server populations.

loyality pip is gone yesterday.
Now we have a Commitment pip, that requires “completing wood division the previous week”.
Unfortunately release notes are a bit lazy here, wood division on the same server or any server? I fear any server.

Yeah a long time ago (4 years?) I proposed to have the
Predicted rating evolution as on http://mos.millenium.org/eu/matchups
in the WvW menu as the main outcome of a match.

And yeah it’s easy to exlain: You gain Glicko, if you perform better than expected, you loose Glicko if you perform worser than expected.

Chinchilla.1785:

So for NA, fighting the same match ups over, and over is numerically better by population balance.

But even that is easier with the current system, just tune down the weight of the random roll. If you do to much you get total stagnation in match making.

In 1-up 1-down you always have imbalanced matcches in the even weeks.

Winner up – Looser down is clearly worser than the current matchmaking (tuned to be a bit less variant, which would be very easy).

Here an example:
1 week : 1-2-3 & 4-5-6 & 7-8-9 & 10-11-12
2 week: 1-2-4 & 3-5-7 & 6-8-10 & 9-11-12
If the ranks really represent the strength then
3 week as first
4 week as second and so on till some strength change

Disadvantages:
- 1 always plays vs 2 and 11 always plays vs 12
- Always the same matches in a bi-weekly turn
- Never a match between 3 & 4, and 6 & 7 and 9 & 10

This is without manipuation, but I see a lot of manipulation incoming in all tier: “oh no, no again vs XX, lets try to not win” … So the even rounds will be imbalanced and the uneven rounds will be manipulated. Whow, what a improvement over today.

In the seasons the swiss-tournament turned out to produce mostly Winner up – Looser down matches, it was really so bad as in my example.

Do not take it out, it needs a bit random to work correctly (do you remember before it’s introduction? most tiers where completely separated, andd the same matches were repeated for month) , but reduce it, either by switching back to 3-2-1 (this reduces randomness by increasing glicko distance) or by turning the constant, which is there, just it had to be used again.
(was adapted once after season1: https://www.guildwars2.com/de/news/wvw-matchup-variance-reduction/)

But it will always be a trade-off between staleness of matches (complained about a lot) or imbalance of matches (complained about a lot as well)

It is random!

The 5-4-3 skirmish scores moved the server close together glicko-wise, such that everything is mostly random now.

Randomness never makes any sense

You can turn BoH’s into siege-blueprints (2nd tab of siege merchant), which you can (upgrade and) send to your favoured commander or sell in the AH.

Just go there outside primetime. At 0:00 most queue will be gone till 18:00 or 19:00.

Better rewards for skirmishes, can be good to attrackt new player, but as far as ANet did so far, I fear it will likely only have the bad effect of rewarding overstacking a single server.

As for the seasons, repeatedly done by ANet in a way that encouraged overstacking, instead of being improved to reward good play with few people, e.g. https://forum-en.gw2archive.eu/forum/game/wuv/Give-Bonuses-to-Gold-Silver-League-Servers/3920804

joneirikb.7506:

@Dayra.7405

Love that old video, and it does sum up WvW pretty perfectly.

Thanks to sminkiottone.6972 for this discovery 2 years ago, here the link again for those who don’t know it
“WvW – always the best gaming experience for the majority of our players”
http://www.youtube.com/watch?v=gKgkYxWoZ-M

If speed is everything that you want from a mount:

https://wiki.guildwars2.com/wiki/Speed_Booster

ask for a 30% one to be added to the gem-shop (not usable in the Mists)

^^ You have mounts

The main-mounts are called glider in GW2.

And there are some other in the gem-shop from time to time:
- broom-stick
- flying carpet
- …

But I must say that I am not a fan of wasting developer resources in riding-animations of different animals.

Xenesis.6389:

Events pfft wvw doesn’t need events, we don’t even need a wvw team, we only need to fix lag and balance. Priorities people get with it.

We are probably talking about quite different times of the day. I never had laags when I played, but usually found an enemy mostly once per hour..

Currently there are 2 threads asking for new content for WvW, but what could that really be?

ANet made some attempts to add new content to WvW, but most of them failed, some even misserably.

New map with a new game-mode: EotM, a misserable fail, not recognized as additional content, but as alternative content drawing people away.

New map in the current game mode: DBL, a clear fail, the map isn’t taken positive, but mostly negative.

Tournaments: Miserably fail, it started the large-scale bandwaggoning to the winning server.

New NPCs and capabilities? Nah, that’s not PVP.

New Doors sarcasm on Would be a clear enrichmenrt of the 20/7 PvD sarcasm off b ut most likely not really be acepted as new content.

So what could new content be?

0) All PvE-stuff can also be acquired in WvW. I would support that, but thats hardly new content.

Currently I see only 2 possibilities.

1) Gliding in WvW with air combat.

2) An event system, that from time to time (especially in off-time) replaces the skirmish scoring to focus PvP action, e.g. winner of the skirmish is: who controls objective X (e.g. SN, or a keep or a tower) for the longest time in the 2h of the skirmish.
This would give reason for real battles in off-time, instead of the endless PvD trains, especially when combined with: Total dead cannot be revived, but are immediately ported to the nearest RP (revive point), add blueprints for RPs, the can be freely build and destroyed, whereever wanted.

What new content do you see? What new content do you like?

Vermillion.4061:

Also look how close everyone is in rating now.

Yeah a side effect of the 5-4-3 Skirmish scores.

Ah you mean he informed us about all the WvW-improvements from the patch?

Unfortunately, WvW not even got so much white space in the notes

Btw.: The formula for Matchrank (MR) is

MR(Server) = Glicko(Server) + runif([-1,+1]) * min((0.45*deviation + 10),100)

runif([-1,+1]) a uniform distributed random number between -1 and +1

https://www.guildwars2.com/de/news/wvw-matchup-variance-reduction/

As all servers roll, the table from Serkit.7836 has even more possibilities,
if a low ranked server roll up and a high ranked one rolls down a match between first and last is easy possible.

It is random to a large extend.

With 200 pts distance between 1st and last server and random influence of +/- 80 per server everything is possible (even first vs last), just not everything equally likely.

- The problem started with mega-servers
https://forum-en.gw2archive.eu/forum/game/wuv/Only-a-WvW-player/4006290
- Got worser with guilds/people jumping around
- Got out of hands when people started to have accounts on several worlds (1st the 10$ special actions, then the f2p accounts)
- Admittedly (re-)linking did not made it better, but I can hardly say it made it worser (ok, I am at home on a server that never got linked)

The only reasonable solution I see is still:
Drop the server-concept and make it such that it can be played as guild vs guild, vs guild e.g.:
https://forum-en.gw2archive.eu/forum/game/wuv/Proposal-for-a-new-competive-WvW

Good point I forgot about relinking

Dayra.7405:

…
I get 300 as the expected new distance between 1st and last server in the 5-4-3 system.

So range shrinks to 3/4 resulting in “wild matches” are 4/3 or 33% more likely.

So now it happen, distance between first and last shrink to 225 pts in NA (350 in EU) and matches get wilder as well (2 vs 2nd last in NA)

I think 3-2-1 scoring was better than this. Wilder matches are to single sided.

PS: As the 5-4-3 scoring is still in use I would not call this necroing a thread, but as resuming a discussion now that reality has even outperformed it’s prediction.

McKenna Berdrow.2759:

Like others have mentioned in this thread, we don’t think this would be a healthy feature for WvW. Many players will not participate in WvW if they don’t see a commander tag, and we try to avoid adding features that can be used to exclude players.

Maybe a tag can only be private, while there is already at least one public tag on the map?

This way no one would be excluded, but directed to a place where is is wanted and needed.

Haha no I wasn’t the one who diggred it out just the post I cited doesn’t exist anymore

And no, it wasn’t meant as lame joke, people are gone now, it would need a mayor release to get more than a handful back.

Diku.2546:

Please keep it constructive to help provide feedback that ANet can use to improve WvW.

Sorry, I think it is much to late to improve GW2 WvW, maybe the only contructive way can be: make proposals how GW3 WvW should be

Not in 1 match, but over time, i.e. several matches, and yes, to lazy for an exact computation (only Pi * Daumen). Deviation and Votality are on the Leaderboard.

Korgov.7645:

Dayra.7405:

If the server come closer in Glicko, the matches are more likely to contain server far away from each other.

So if the 5-4-3 runs longer, I guess matches like e.g. 1 vs 12 become more more frequent than they are today.

Can you show some mathematics to back up those statements?

JTGuevara.9018 did a good job to explain the basics.

lets assume 3 server played in 2-1-1 and the result was
A won all 84 skirmishes
B and C where 2 and 3 each 50% of the time
Thats 168:84:84
now the same 3 server play in 5-4-3 and the skirmish result is the same, the final result will be 420:294.294

As Server A did not got a 2 : 1 as in the week before but only a 1.42 : 1 it looses Glicko (did not matched it’s Glicko expectations), while Server B and C where better than their expectation so they gain Glicko.
Currently the NA-Glicko-Range is 420 points in 2:1:1 system, if I divide that by 2 and multiply it by 1.42, I get 300 as the expected new distance between 1st and last server in the 5-4-3 system.

So range shrinks to 3/4 resulting in “wild matches” are 4/3 or 33% more likely.

Deviation is around 160 (http://mos.millenium.org/na/matchups), so random roll maximum is approx. 80. So matches with servers up to 160 points away are likely, matches with servers 320 points away are possible (but not very likely) due to match allignment (assume first and last are 300 away, 1st rolls -80, everyone above 150 rolls +80, everyone below 150 rolls -80, except the last that rolls +80, afterwards, there is no server left between 1st and last, so they may end up in the same match)

I wonder about the effect 5-4-3 will have on Glicko and Match-Making.

I think the looks closer is something Glicko will take over, as (if I remember correctly) Glicko-Score is based on the Quotient.
So in 5-4-3 the winner has only 5/3 = 1.66 times as many points than the looser,
while in 3-2-1 the winner had 3/1 = 3 times as many points than the looser.

I guess this let the server come closer in Glicko points.

If the server come closer in Glicko, the matches are more likely to contain server far away from each other.

So if the 5-4-3 runs longer, I guess matches like e.g. 1 vs 12 become more more frequent than they are today.

Kimi Chaika.9862:

Anet needs to stop with this artificial manual balancing junk, open ALL the full servers up and let the players naturally balance out. Hasn’t their economist heard of trickle down theory? Naturally players will trickle down and fall into lower servers.

Good example. natural balancing works as well as economist heard of trickle down theory. The top server get more and more people and the bottom server die out, very similar to all money trickle only up (richer get richer and richer) and nothing down (the poor get poorer).

There is a not assigned keyboard commando, called “select next friend”, probably it works on tactivators as well.

I always like to have a pool of downed in front of me especially now as I need to stomp my halloween achievement.

Concerning a poll: I like downstate.

Concerning revival of totally dead: I would like a poll to teleport them to spawn immediately on dead.

Korgov.7645:

Dayra.7405:

Every Friend in your AoE Area count against your AoE-Limit

Interesting idea. But…

Scenario: 5 enemies are capturing the Wildcreek tower. The lord is at 50% health and I rush in for the rescue. I manage to create 3 illusions while the enemies do not pay attention to me. I move into middle of them and hit Mind Wrack to shatter the clones. Since there are now 5 friendly targets in the area Mind Wrack does no damage. (1 tower lord, 3 clones and myself)

I guess you would not count NPCs as friendly targets which consume AoE charges.

Scenario: 4 melee jarheads are attacking a downed mesmer in the Wildcreek lord room. They are spamming #1 skills which are AoE cleave attacks. They are standing right on top of the mesmer so everyone is within each other’s the cleave area. Since there are 3 friendly targets in the area, the cleaves do no damage.

Would you exclude cleave attacks? If so then you are on a slippery slope to apply this feature on specific skills only.

Yes, in scenario 5 inclusion of minions and NPC’s isn’t indended.

In scenario 4 it works as intended I would say, if none is bright enough to step out of melee aoe they earned the result. But one could also debate, that the target is always prio 1, i.e. AoE degenerates to a single target skill in the presence of freinds.

A very weak form of friendly fire would alrady do it, and cannot be abused:
https://forum-en.gw2archive.eu/forum/game/wuv/How-can-we-changed-WvW-to-prevent-the-blob/4821283

Dayra.7405:

How about a weaker form of friendly fire (which has to many negative side-effects)

Every Friend in your AoE Area count against your AoE-Limit

So if you put AoE onto an area with 5 friends you do no damage. I guess this will take out all teeth of the Blob-Stacking Meta.

This will make single-target skills more valuable as single target skills will be the only way to hit enemies in your blob.

It will make small groups valuable that can “hide” in the enemy blob.

Both could need an upgrade for large-scale fights.

Generally I would propose the following priority in AoE target selection:
Invisible Friends > Invisible Enemies > Visible Friends > Visible Enemies
I.e. you only hit a visible enemy with AoE, if there there are less than 3/5 friends or invisible enemies in your target area, you only hit invisible enemies if there are less than 3/5 invisible friends in your target area. The higher priority for invisibles is needed as you cannot target them with directed skills.

Who comes earlier (to prepare the map for a good start score) is on it. Beside reset there is no battle royale, but first come first serve.

But yes, an overcrowded server with bad community has probably a worser mix of people on the map that a smaller server with better community . That’s part of the idea

But, the usual achievements (to determine participation) in the season should only be made during the skirmishes/maps that do not count for the tournament-score, such that you do not have to crowd the impportant skirmishes with (participation) acchievement hunting.

In the end it’s to everyones favour to not rush the map uncontrolled but to populate it with a community shared plan, but of course not everyone is able to see it favor.

Dawdler.8521:

According to Anet the seasons burned people out and led to a decline in player activity.

This is the case, if the seasons are just the normal 24/7 coverage war won by bandwaggoners overstacking a selected server as all seasons were so far.

But what if the seasons are something special especial are not coverage war¨

So here is an idea for a season:

Emdebeded into the usual weekly match (but match-making via season rules, not glicko):

10 skirmishes of the week are quasi-randomly (maybe weigthed by pop and if that slot was already selected in that time-slot) selected to be the season-skirmishes
I would propose:
- reset skirmish is always included
- 2 skirmishes randomly from Saturday
- 2 skirmishes randomly from Sunday
- 1 each from mo-fr (before reset)
For each of these skirmishes one map is selected such that in total, each side BL is include twice and EB 4 times (results in 4Eb, 4 ABL, 2 DBL)

Only these predetermined (and pre-announced) maps and skirmishes count for the season-match winner.

This way we have much less coverage war, nearly every server should be able to fill one map for 20h a week.

McKenna Berdrow.2759:

The 6-5-4 score is because everyone started with 3 Victory Points. We will be fixing it so that everyone starts with 0 Victory Points but this change won’t be in for a few weeks.

Looks like it’s done now. See a 0-0-0 score now.