Carefully reading emails from ArenaNet and visiting NPCs that most players normally don’t interact with very often to avoid missing challenges is apparently an innovative new part of the game to keep the players engaged in the game content. It’s like having the join a party in order to get the champion box in PvE events on stacked megaservers. All exciting new content to keep players on their toes.
doesnt mean its set in stone
That’s kinda the whole point of complaining vigorously, to change those plans.
in fact, it sounds like hes testing the waters to help prevent another situation like commander tags where a couple people had to work an extra weekend to appease us.
And that’s fine, so what’s the problem with people strongly objecting like they did with the commander tags?
But of course you are moving the goalposts now. You asked, “why is everybody acting like anything john said is more than an idea at this point?” I answered your question. What he said clearly indicated that they were more than simply “ideas”.
Or, people could leave the zerg mentality and scout/roam/communicate with their team/server…
Yeah, all one of them at many times of the day on lower tier servers, where a dozen people can be a “zerg”. Maybe they could play a Mesmer so they could “split up” and scout multiple objectives at once, right?
People who don’t read the instructions, don’t read the emails sent by ANet to indicate that their rewards are now available to be picked up and must be picked up before the end of the following week likely aren’t “customers” who can work out the complex task of purchasing gems with real money.
I’ve purchased lots of gems and I missed out on a one of the weeks because, quite frankly, the apex of my guild wars experience is not visiting the Battle Historian to get a reward. Assuming that they have a great reason to use that delivery mechanism, is there any reason why the battle historian couldn’t hold a player’s cumulative reward until they go an pick it up?
why is everybody acting like anything john said is more than an idea at this point?
There were a handful of other tweaks brought up in our discussions that we are planning. These include reducing the number of people who can rally off of a kill. Removing white swords from objectives to give defenders a stronger role in alerting their team as well as giving attackers more of a chance to get through having their siege disabled. We are also going to test out awarding points per kill. PPK was part of the scoring discussion and something we wanted to try. We will be looking for your feedback on that change after it goes live.
It isn’t live yet, but that suggests a lot more than an idea. It suggests implementation plans already underway.
I despise what the megaservers have done to server identity and communities in PvE and have two friends who played daily and have over 10,000 achievement points each (one of whom got me into GW2) and plenty of gem store purchases, who have stopped playing Guild Wars 2 because of what the megaservers did to our server community in PvE. I have no desire to see the same thing forced on WvW. I’d rather my server fall back to 24th place and get stuck there again than be subjected to a forced merge.
I scout/roam loads of my wvw time like 95% and with no swords i cannot afford to get 3feet away from a structure scared i would lose it then.
Even with swords, I’ve seen objective get captured a few minutes after I checked them and refreshed the siege in them, especially if they are not upgraded yet.
Devs, I hope you reconsider before it kills bronze league wvw.
Maybe we shouldn’t assume the goal isn’t to kill Bronze League?
Well, we can do both by merging the bottom 3 and closing the top 3. Compromise. Why does everything have to be either/or?
If it wasn’t clear, I have concerns about forcibly shutting down any servers, top or bottom, because I think there are plenty of people at both ends who like their communities and are even largely happy with the play on their servers. I think what the servers at the bottom need are some handicapping or a modest amount of transfers to improve their coverage, not a forced transfer or merging into a whole other community and style of play.
Can you stop with the stupid ideas of pricing that “punish” bronze servers all day long ?
NO FRIGGING SERVER should be cheaper to transfer too, in ANY way, than a server ranked below in WvW, PERIOD.
I would argue that to get a discount, you should need to move down at least one tier, if not one league. This would prevent the servers from within a league from cannibalizing each other, too. The question, of course, is whether the goal is to even out the population across all of the servers or to stack the population on a few servers at the top.
they can kill yaks in combination with the krait/ascalonian/etc mob spawns
Actually they can’t, because they aggro on the mobs instead of aggroing on the yak. Also, wildlife doesn’t spawn on Borderland maps, so caravan just walk past sentries.
The dolyaks generally never get killed by mobs and sentries in the Borderlands but the cumulative damage from mobs in combination with sentries can kill dolyaks in EBG, which is why I’ll often run the guards before double supplies on EBG if nobody is watching the dolyaks. Places where I know this specifically happens are between Pangloss and Veloka (where the dolyak can be attacked by bats on the way out, the sentry, and additional mobs on the way to Veloka) and between Danelon and Langor (where the doyak can get attacked by bats on the to the exit, harpies upon exiting, and a sentry before the bridge), but there are other places where dolyaks can also get killed due to cumulative attacks and a few where one set of mobs, alone, can take down the dolyak (e.g., happens often between Golanta and Jerrifer’s).
All this leads to a very restricted non-tactical gameplay of forming huge blob of players around commander and overpowering everyone that comes across its way.
Sounds a bit like EotM.
I heared QueueAge has been renamed now to GrindAge .. or will this just happen at the 4th ?
Maybe they should call it LaborPointAge.
By the way, here is an actual example of an uplevel getting map chat abuse in WvW. I expect PPK to make this worse.
Omg just merge the bottom 6 servers already so NA can have 3 Leagues of 6 servers each and we can all move on to fixing the rest of WvW.
Just break up the top 6 servers already, since it’s clear that T1 and T2 don’t belong in the same league and can’t have a decent match-up with anyone else, either.
Or maybe we can all just stop making suggestions about how to ruin other people’s servers.
When my server dropped down to the bottom NA tier, we were in a similar situation. We’ve gone up six ranks in the past few months – things change. Eredon Terrace (also on NA) has remained strong (in spirit) throughout their time in the Glicko ghetto and, thanks to some hot transfer action, now seem to have steady coverage in places they were light before and are on the rise.
We’ve gone up, hit a wall, lost some people, went back down, got some more people, and now going back up again. I’m good with it all. And I have fun winning or losing most of the time because there is always something even one person can do (solo camps, upgrade and defend what you’ve got, pick off lone enemies, etc.). Make it about doing the best with what you’ve got rather than winning. That said, I do think some sort of score handicapping or incentive to show up when losing is inevitable could improve participation.
One other purpose that white swords can serve is a probe or a a distraction. A single player can knock on a door to create white swords to (A) see what the defender response is and (B) to draw defender attention to an objective to draw them away from the real objective. There is a psychological game that can be played with the things that can be seen on the map — white swords, dolyaks, sentry flags, and so on.
Further, white swords also reward disciplined groups that can take out guards, build a bunch of siege weapons, and scrape siege without tagging the gate or walls to draw the swords until the are ready to use catapults or rams en masse to quickly take the walls or gate down once the attack starts. Pick up groups rarely pull this off (someone won’t understand not to hit the gate or will do so by mistake) but disciplined groups with good communication can and do pull this off.
Having megaservers forced on players that weren’t looking for a megablob experience with no way to opt out of it or get away from it.
I’m still trying to figure out why they set the population cap for the megaservers past the point at which anyone with moderate hardware is going to have a horrible graphic experience at any large event.
Eredon Terrace in WvW. Some random megaserver packed full of people from other servers I don’t recognize for PvE. I rarely see anyone from ET in PvE anymore. Heck, I’ve had trouble getting on the same map with people I’m in a party with.
The problem I have seen is when the duel takes place at an objective or close to it. You see someone from your team and you decide to jump in to help him preserve or take the objective. Now imagine the surprise when the player, who made a move to help you thinking he would have your support, end up being alone in fighting the ennemy.
That’s why I usually just keep shooting them with siege weapons so they can’t have a clean fight to make them go away. I don’t care if they duel down by the windmill or otherwise away from objectives and I’ll respect that if they tell me it’s a duel. But if an opponent is fighting near an objective, there is no way to tell if they are scouting or otherwise helping the enemy in other ways.
If they had just left the quaggan and krait alone in their lake instead of committing genocide against them to install the ruins, none of this would be a problem. A player could do the quaggan events solo if they wanted.
Where it would change my behavior is that there are a lot of times where I’ll throw myself into a fight that I know I can’t win just to see how long I can stay up and to make things more interesting for my opponents. I’ll charge into a fight outnumbered or I might stay in a lord’s room when the wall goes down. Most of the time, I die from DPS so there are no points involved. If it’s going to become a serious points issue, I’ll do that less or maybe not at all because there is a penalty to my server for doing so.
I wish they’d fix the Mad King runes to be ravens attacking, as advertised. I have a Raven Norn that I’d probably go back to using them for if they fixed that.
1) Guild sets up guild catas at hills wall and takes it. Another server caps from them and if their catas aren’t destroyed the other server has an hour to go back and use the same catas without incurring any additional costs.
I don’t consider this a problem. It’s similar to driving an enemy off of the catapult spot just before the wall goes down and forgetting to fix the wall, so the next time they can knock it down in seconds with a single catapult. There are some things defenders should have to pay attention to.
2) Trebs in red keep can just keep continually taking SM w/o any added costs the second they lose it, until they (if ever) lose their keep.
You don’t have to take the red keep to stop the trebs. You simply have to counter-treb them or break into the lower keep to clean them off. I’ve seen it done to my server and I’ve seen my server do it to other servers. Yes, I think it’s obnoxious that the red keep and SMC can treb each other, but I think that’s a map issue, not a siege weapon issue.
I spent almost my whole day today running around my server’s corner of the Eternal Battlegrounds moving dolyaks, running upgrades, building siege weapons, and refreshing siege weapons. It took a lot of my time to keep all the siege weapons there and all the upgrades cost me around 5 gold. Please don’t make the defender role more expensive or time consuming than it already is. It’s bad enough having the Use [F] the siege every hour without having to also run supply to it all, competing with upgrades and so on for supply.
The problem is anet left almost everything clientside. Real MMOs try to keep everything server side for this very reason: if it’s at the server end, you can’t tamper. Until everything is serverside (which will cause lag), this game is going to be a hack fest.
I understand the problem with trusting clients and also why letting the client do the work is desirable for performance, but the client must report a great deal about the location, movement, damage, health, and so on in order for it the impact the game world and for others to see it so what I don’t understand, as a programmer, is why they can’t implement a set of assertions related to the data that client reports to the server to pick up at least the most overt examples of hacking such as flying (any character that has coordinates in the air or under the map), teleporting (if the client reports a huge location shift that doesn’t end near a waypoint), excessive movement (no character of a class should ever be able to move more than a certain distance over a certain time), characters doing damage that can’t be seen, excessive damage or healing, and so on that must be reported from the client to the servers for other players to be able to see it or be effected by it. Seriously, how can they not detect a flying character? Or a character that teleports to a chest that’s nowhere near a waypoint without reporting any points of movement nearby? That information must mass through and be available to the server or nobody else would be able to see it. Report it for investigation. Sure, there might be some false positives, but the criteria can be tuned until they are minimal.
It would also allow superior forces to coordinate to have minimal forces go in and figure out the outnumber buff. Then, after making sure they keep outnumbered, knock down the walls, open all the gates or whatever, before calling in the horde to take everything quickly without swords popping until the very last second. Imagine 5 players using catas on 5 different targets. Right as they knock down the wall to inner bay, hills and 3 towers they call in the zerg. The zerg splits into 5 groups and they take all five targets within seconds. I can only imagine seeing 2 keeps and 3 towers all changing color simultaneously.
While in practice the Outnumbered buff seems to be fairly fluid as players enter and leave the game and I think it would be hard to predict it reliably, I think this is a perfectly legitimate criticism and I can imagine something like this happening (earlier today, we set up a many-trebuchet breach of a SMC and I called in an off-map guild group as the outer walls went down). So you are right. No swords is probably never a good idea, so I retract that part of my suggestion.
Seriously. As a defender, I don’t need a “stronger role”. The commanders already take me seriously. How about a few rewards for my role? Even I can only scout for so long before I want to go for a run too, but there is no one to take my place.
It cost me about 5 gold today to spend the day running dolyaks and upgrades and building siege (not counting the cost of the blueprints) instead of running with our offensive force capturing things.
The Outnumbered buff only applies to population on that map. The other three maps might be fully queued, for all the game knows. Certainly, it says nothing about overall server WvW population at any point in time.
Correct, and I think irrelevant from a defending perspective, since you can’t see white swords from another map, either. Note that I doubt this will eliminate swords on waypoints, which is what you currently can see from other maps and there is no delay on them. I’ve notified the people on a map that a waypointed keep was under attack… from another map.
Also, if you tie any substantive benefits to the Outnumbered buff, it just encourages elitist guild groups trying to harass all other players off the map, so they can get the benefits for themselves.
I was thinking of this from a defender’s perspective where I think there would still be little net benefit to being deliberately outmanned, but a later reply giving a detailed scenario of how an attacker might abuse this makes me now believe that it would never be a good idea to remove swords entirely. I still don’t see much of a problem with removing the delay for white swords when the defender is Outmanned, though. There is already no delay on waypoint swords.
(edited by Berk.8561)
There is no such guild anywhere in GW2…
In the past, there were guilds that claimed they could do it. Have they all moved on or decayed or was that never really true?
Find a way to make low ranked server transfers cheaper.
There is a person who posted on Reddit who wants to transfer to Eredon Terrace with his wife (to join some other players from his server that have already transferred and are having fun) and the main thing holding him back with that it will cost him around $45 to transfer them both. Why is it so expensive to transfer to the 23rd ranked server?
Yeah because it’ll totally “solve” anything to make things more “competitive” between the 5/6 people on each side that are the whole population 18h a day right ? RIGHT ?
Maybe EU has that problem and could shed a few servers (it has more than NA), but things don’t look that bad in NA to me, even at the bottom.
One example are the NPE changes. They said the game losing new players becasue they don’t get it, they say they lost more than 10000 players due to this. They said that the NPE changes got developed based on a survey filled by those players.
I’m fine with this argument, but I think the bigger problem is that there are ways to approach being more helpful to new players without forcing other players to play on beginner mode who don’t need it. Sure, sell the bicycle with training wheels if you want to make it beginner-friendly, but don’t make them impossible to remove and force everyone to use them, even if they don’t need them.
The extra communication is fantastic
As negative as my earlier replies were, I agree with this. This is a fantastic example of the sort of communication I’d like to see more of. It gives people the opportunity to raise objections before things are finalized and implemented, even if their objections are ultimately ignored.
Zerg busting is a beautiful thing, it really is. However it is something most people struggle with (by definition if every zerg could zerg bust then we’d be in a paradox of awesome fights)
Or really lame ones that revert back to numbers being the only thing that decides the winner.
FTFY
Do we really want a refugee guild from a T1 server that can queue a map all by themselves transferring to our server? I’m happy with those who want T1 play having their T1 play. What I’d like to see are changes that make the lower tiers more competitive, not necessarily heavily populated.
How is it that you cannot see that this destroys lower tier servers all over again?
It looked like very few of the developers went to the bottom tiers. It’s possible they just don’t have any experience with it. It’s possible that they share the preferences of T1 and T2 players for guild focus and huge armies on stacked servers. It’s possible that’s one there are so relatively few WvW players on the bottom tiers that we are an irrelevant part of their market and they don’t care. For whatever reason, they keep seeming to not get it.
The megaservers were horrible to a lot of people, too, including role-players and people who enjoyed the smaller server communities.
I can only guess from the massive population caps on the megaservers, the way they group players, the massive zergs now at all of the world boss events, and the focus of the recent CDI threads that they must assume everyone is looking for a very high population experience and want to orient their game around huge numbers on a map and a large guild experience.
I see plenty of independents, casuals, people who deliberately chose lower population servers, and people who hate the megaservers (including some who have stopped playing because of the megaservers), but maybe their numbers just aren’t enough to matter, maybe their voices aren’t loud enough or organized enough to be heard, or maybe the preferences and style of such players is simply so different from what the developers expect and enjoy that they can’t relate.
It’s not ANet’s fault, but I don’t see a lot of gray hair on their staff, doubt many developers are playing with their spouses and kids, and doubt many of them are newbies to computer games or MMOs. There are a lot of perspectives likely missing there. GW2, by the way, is the first MMO I’ve ever played (not counting text MUDs). It might help a little, though, if all of the developers spent a few weeks playing on the servers at the bottom of the rankings.
(edited by Berk.8561)
These include reducing the number of people who can rally off of a kill.
The correct number is probably zero, but experimentation is a good way to test things out.
Removing white swords from objectives to give defenders a stronger role in alerting their team as well as giving attackers more of a chance to get through having their siege disabled.
As others have already said, this would be a disaster for low population periods in the lower tiers where there might be only a couple of people on who can’t possibly scout everything. Elsewhere, I’ve recommended:
- Defender Outnumbered = No Delay Before White Swords
- Attacker Outnumbered = No White Swords.
- No Outnumbered Buffs = White Swords with existing delay
This would allow ninja attacks against superior forces that have the numbers to scout but also help outnumbered servers defend their stuff from much larger forces.
We are also going to test out awarding points per kill. PPK was part of the scoring discussion and something we wanted to try.
I’m skeptical this will be a good thing but I’m willing to give it a try.
My concern about the last two ideas is that they seem to hurt the weaker or outnumbered opponent who is going to die more, unless they avoid most fights, and has fewer resources to scout their map for attacks. I don’t think fixing WvW involves making things worse for an outnumbered server with coverage holes. If your goal is to help the underdog, then giving the more powerful server even more points for rolling over a smaller force and giving them no warning before they lose objectives is not the way to do it.
Also bear in mind that rallying and points per kill make inexperienced players, casual players, and uplevels a liability to their server and, on the case of rallying, any hardcore or experienced group they try to travel with. That runs counter to what I thought was an important design goal for GW2, which was to always welcome more help. It also makes me more incentive not to stand and fight if I can’t win, and fewer fights isn’t going to make WvW better.
I’m pleased that the developers are listening, pleased you are talking to the community and each other, and pleased that you are giving us some insight into the direction you are thinking about going in, but I do have some serious concerns from this that my server and how I enjoy playing WvW are going to be casualties of future changes, just as they were in PvE with the megaservers.
(edited by Berk.8561)
My main issue with loot is that i am getting fed up with all them Blues and Greens i get at lvl80. I guess a rare item is really rare, cos i rarely get a rare, unless a rare material also counts as a rare (i get those alot more).
And an Exotic, cant remember the last time i got one.
Try improving your Magic Find. Do you break down your blues and greens and consume the luck, or do you sell them?
I think that maybe the Bronze League situation is different in NA and EU. In our bottom match-up, our 23rd and 20th level servers have been fighting hard for first place and even the though the third place server (24th) has been trailing quite a bit in in points, they’ve gotten waypointed keeps and have managed some nice captures from the other two servers. While each server has quiet periods with only a few people on, it’s not dead down here in NA and we have more people playing on ET than we’ve had in a while.
I’m sort of writing to think, but I’d like to understand what truly reaches the point of “I’d like more info, but I’m satisfied with what you’ve been able to tell us right now.”
Gaile, if you haven’t read it before, I recommend reading The Web Is a Customer Service Medium by Paul Ford. In particular, this paragraph is key here:
“Why wasn’t I consulted,” which I abbreviate as WWIC, is the fundamental question of the web. It is the rule from which other rules are derived. Humans have a fundamental need to be consulted, engaged, to exercise their knowledge (and thus power), and no other medium that came before has been able to tap into that as effectively.
As I said earlier in the thread, I’ve seen recent evidence that ArenaNet is listening to what people are saying on the forums but I also said I had concerns about the direction the game is going in and am concerned that ArenaNet doesn’t always interpret the message from the players correctly. I’m also concerned that ArenaNet will listen to the complaint of some players, which may be perfectly legitimate, and then implement a fix that absolutely destroys the game for other players. Without any back and forth dialog or consultation with the entire player community, the odds of such problems happening are fairly huge.
What happens in any complex game like this is that improving the game for one segment of the player community can harm the game for another segment of the community. For example, if you listen to hardcover players who want challenging content that can only be done through skill and coordination, you’ll alienate casual players who can’t do it. If you make all your content casual-friendly, you’ll bore your hardcore players. And players who are unhappy often don’t make much noise, so what can happen is that a change will be made to appease unhappy noisy players that destroys something that other players loved that maybe ArenaNet wasn’t even aware of, thus destroying the game for them. I think that happened with the megaservers.
The megaservers solved one set of problems that players were complaining about (empty maps and events that never or almost never got completed on low-population servers) and may have been very useful for ArenaNet by allowing the consolidation of servers, but it harmed or even ruined the game for other players, and I am not exaggerating with the word “ruined”.
Two of my friends (who played daily, have over 10,000 achievement points, spent money in the gem store, and got me to play Guild Wars 2 as my first MMO) have entirely stopped playing Guild Wars 2 because of what the megaservers did to the PvE part of the game and world events. They didn’t leave to go play another game. They just stopped playing Guild Wars 2 because they don’t enjoy it anymore. The megaservers wrecked the small town community feel that our server had, when a handful of familiar people would do events.
It also destroyed the community feel for people who liked to hang out in the cities like Lion’s Arch and chat. It made it hard for the role-players to find each other and play together. It made it nearly impossible to recruit for WvW in the game using map chat because people are no longer grouped by server. It ruined events for players who wanted the challenge of soloing them or doing them in small groups.
Sure, the megaservers fixed one problem but made a huge mess in a lot of other places. And we’re stuck with it without having been consulted and left wondering just how much of this was considered by ArenaNet before it was implemented. As a result, I’m very concerned about how ArenaNet might make other fixes to things like WvW (e.g., forced server merges) without considering what they’re going to destroy in the process.
In other words, people are concerned that if we don’t find out about the changes until they are a done deal that we’re going to be surprised in an unpleasant way by what we get because that’s already happened a few times.
(edited by Berk.8561)
wvw has been stale for so long. No major changes, and no sign of future big changes.
I think the single simplest thing ANet could do for WvW to make it feel less stale and bring former players back, at least for a while, is new maps and then rotate the maps around. Give players some new WvW maps to learn how to play (not necessarily with more gimmicky terrain and mechanics like EotM) and that will go a long way toward making WvW feel less stale and giving former players a reason to come back to experience something fresh.
Changes that make absolutely no sense in the context of the discussion with the player base.
Ranger CDI is a great example.
Actually, I think several of the Ranger changes in the last patch reflected things people asked for in the Ranger CDI thread — specifically, the removal of Signet of the Beastmaster and having signets apply to the Ranger as well as their pet by default and improvements to longbow power damage to make power and burst damage longbow builds more viable. Did people get everything they asked for in that thread? No. But what did get changed suggests that they actually did pay attention to that thread.
Note that I do have serious concerns over the direction the game is going in, but I also see evidence that they are also paying attention to what players are saying, though I don’t always think they get the right message when they interpret the comments.
(edited by Berk.8561)
The problem is that a lot of stuff was thrown around to generate hype. Stuff like : Precursor Crafting, Fractal Leaderboards, new legendary weapons and other legendary types etc.
These are things that players care a lot about and as a result when nothing else was heard on this front and all these were kinda left to disappear into thin air with no news players got a bit doubtful.
And that’s probably exactly why they’ve stopped talking about anything that’s not almost ready to be deployed, because anything they say that they are working on or intend on implementing with generate type and players will care about it a lot and they’ll get angry if it never happens or doesn’t happen for a long time.
Yes, you can enjoy playing on a bottom server, but your experience will suffer because the game mode was designed for large-scale battles supporting hundreds of players at the same time.
People keep saying this but:
- While they are interesting when they happen occasionally, the truly large scale battles are some of the worst play experiences I have in WvW because they stop being about individual movement and targeting and at best are an exercise in sticking to a commander and following orders and at worse are a chaotic mess of people running all over the place and that’s far from ideal, in my opinion, even if it didn’t come with low frame rates, mass skill effects, and skill lag. When massive groups attack a target, it can not even matter if you disable or destroy their siege weapons because they can DPS the gates down. I don’t consider any of that the ideal I’m looking for in WvW.
- What GM designed the game mode for has nothing to do with what individual players find fun or not fun. ArenaNet designed WvW to have a score and a winner but plenty of players don’t care or worry about that. ArenaNet may have designed the game for massive battles but plenty of players like solo and small group roaming or even simply open field fights with their guilds. And in some of those cases, the large scale battles only get in the way of what they really want.
And don’t tell me everyone is happy with the current state on WvW. I talked to and still talk to a lot of people who are disappointed about WvW in Bronze, and even though some have transfered so they could have a better WvW experience, some decided to stay. Not because they love playing outnumbered, being zerged down or regularly joining EB with nobody in it, but because of the friends they made in their guild and in their server.
And don’t tell me it’s all rainbows and unicorns at the higher tiers, because we also have people who transfer down to Bronze from higher tiers and they enjoy what they find at the bottom. It’s different. That’s good because people enjoy different things. Assuming that everyone plays any part of GW2 for one reason only and enhancing that one reason to play at the expensive of all others is a recipe for disaster unless you’ve got a very large audience looking for that one specialized reason to play.
Its time for you to do the same and come back to your family in bronze…
I think we’ve had a pretty good week at the bottom, and I’m not saying that just because ET is leading in points. Both FC and AR have had their moments when they were doing well, too. Bronze is not dead.
How many people knows the truth that it doesn’t take skills to be good in WvW. It’s the numbers game. You bring 1, I bring 10, you bring 2 and I’ll bring 20. I’m sure my 20 will kill your 2 after 18 of them falls.
While numbers always matter to some degree, skill really becomes less important as the numbers increase because of the AoE attack cap and overlapping AoE boons. At smaller numbers, you actually have to get out of the red circles and move around a bit on your own to avoid dying. At higher numbers, you just stick to the center of the blob.
Important: Remove white swords (perhaps only when you outnumber the opponent). These are the bane of all small-groups and are one of the main reasons servers get away with blobbing so much.
This means that worlds who dominate and capture everything then risk losing their most precious objectives to ninja teams.
The problem with removing them entirely is that seriously undermanned servers that lack the resources to cover their map with scouts rely on the white swords to k ow where they are under attack and a dominant server is in a better position to have dedicated scouts.
What might work and achieve what you want is to remove them from a defender on a Borderland if the attacker is Outnumbered and remove the 30 second delay if the defender is Outnumbered. The potential problem with that is that there is currently no way to know when your opponent is Outnumbered, so the loss of swords could be a surprise, and it would create a tipping point where one more attacker could cause the sword delay to disappear for their server, though getting the Outmaned buff is already a big warning that a large attacking force is on your Borderland.
- Diminishing returns.
The more you hold, the less points for each thing you get. Obviously your total always goes up when you hold more stuff but the difference between holding all-but-one and all of the objectives should be miniscule. Disincentivise holding everything, give no champ bags or loot when you have significantly more people or when your score is significantly higher than the place below you – why do you deserve those things for easy-(blob)-mode?
Earlier in the discussion, I proposed a diminishing returns system based on overall points, but it could be done with PPT, too. The nice thing about looking at the points is that it avoids the problem of how to count population for handicapping and diminishing returns, if they never diminish to 0, never fully erase the benefit of wining or showing up to fight for either side. It’s never a better idea for a defender to abandon a map a and sit out the fights and it’s never a better idea for an attacker to stop attacking.
I currently think some sort of diminishing returns strategy based on overall points or maybe PPT is probably the best way to address runaway scores and give underdogs a chance without warping the incentives to play or being easily gamed by the players.
3: maximalize the number of sieges in an objective. 9ac 2 trebs, 2 catas and 3 ballistas in a normal tower is insane and boring. No good fights.
Only boring if you expect to be able to capture things in 5 minutes or less. I can build 5 or more trebuchets in a single tower, not because I normally expect all 5 will be used during an attack but because a smart and patient enemy can quickly destroy several of them and the extras provide backups. There are some towers and keeps that are very vulnerable to siege scraping with a ballista or trebuchet. The existing limits are fine.