They also have the same amount of hp and all defensive stats, if you’re wondering.
You have to be on the server for a full match before you start gaining pips. It kind of sounds like you get that, so perhaps your complaint is only that it activates when switching from linked worlds to host worlds. If that’s your complaint, I’m not really a fan, but discuss as you wish~
Since ruins do literally nothing outside of bloodlust, it would make sense that they don’t give participation outside of bloodlust either. It’s a silly daily since it has 0 impact and dailies should incentivize behaviours that have some sort of impact, so I wouldn’t mind seeing it removed from that rotation either. Instead, have a repair daily or something.
I imagine there will be ways to increase pip gain in future updates. They’ll probably be conditional to incentivize certain behaviors or create some sort of economic movement. Just like every other system, the base is released and later additions make it move a little faster.
As a start, I think this system is ok. I got through Diamond on the first full week but only got partially through Plat this week. This is while working 2 jobs and generally not being too gung ho about GW2 atm, but also with lots of outnumbered since I’ve been playing mostly havoc. I’m not sure how many tickets I have because I don’t particularly care about that stuff.
It’s a bit annoying when people hop onto a map and kill the outnumbered buff for the tick, but I’d be playing with or without pips so I don’t get too upset over it.
You mention an anti-zerg mechanic, but your only suggestion is to add GvG. It’s not even clear if this GvG you want to add has anything to do with WvW.
I’m not sure what to do with this post.
Also, while you provided no details, I don’t think an anti-zerg anything is going to fly. You can reward players for splitting up, but you can’t punish them for sticking together.
Oh, just wait till you see mesmers…between the blocks, evades and invulns, you’ll probably get bored and log off before there’s an opportunity to hit them.
It’s dumb, but it’s pretty far down on the list of urgent problems. More annoying than game-breaking.
The feature still exists. I’ve no idea what could have broken on your end, though.
This is clear in hindsight, but there wasn’t any way of knowing how successful the update would be. We all expected a population bump but no one knew it would be this large and this enduring. Also, it’s been a little under two weeks—population likely hasn’t settled yet.
Thus, there probably wasn’t a way to handle this much better with the information they had. Whenever population spikes this hard, some things are going to break.
I spent a solid hour last night running supply from camps to keeps to repair walls. It was critically important and should definitely remain as a participation booster.
However, it could get the ruins treatment and only give partial participation time. That is, decay sets in faster if all you’ve been doing is repairing.
It’s tough starting out, but if your server is ever not in last place or is outnumbered then it speeds up considerably. Once you rank up a bit, it gets even easier.
I’m not saying there’s absolutely no room for adjusting the threshold, but I’m ok with a game mode that requires sustained effort to succeed in also requiring sustained effort to get one set of rewards.
Perhaps a good solution would be not to reset pips for players who haven’t reached the wood chest yet. It wouldn’t be something people would farm as they’d lose a commitment pip and not get rewards for that week, but it would make things a bit easier on folks with minimal playtime.
It’s probably best to see how they handle the next set of links before transferring.
Why should a single person be able to defend against massive blob with 6 catas?
Is anyone saying that?
Even if you have 20 people versus a blob of 40, you don’t have much choice but to use defensive siege to counter their siege or you’re going to get steamrolled. If you can’t use defensive siege because it gets blown up by a few guards dry-humping the catapults, then you’re out of luck.
To be clear, even when you do this, your structure will take massive supply damage and may still fall outright. You’re not going to be able to defend indefinitely, but if the enemies did something silly like stacking all 6 catapults on top of each other, you may repel the initial assault.
I’ll choose badges if I want to pick up Supply Traps or Siege Disablers. Choosing the random Trick sometimes gives me Stealth Traps and those are basically useless.
By roaming do you mean taking small objectives, exclusively fighting other players, or some mix of the two? Guardian is a good all-rounder if you just want one class that can do everything, but Warrior is also good in zergs and is better at dueling imo.
I was referring to the SB #4 poison cloud. Calling it smoke was pretty unclear, so sorry about that.
It was 2 stacks of bleed with a very long duration. I had about 15k hp max but was probably at about half since he showed up while I was in combat. I stealthed out but was hit by one of his dodges and just kind of watched it to see how much damage it would do. It was enough that I had to blow my heal to keep watching and still almost went down. It took a long time, but not so long that any condi clear would come off cooldown. Thus, when I went back to fight him, I made sure to stand back while he ran himself out of dodges before engaging.
Even if you’re polite and constructive, you still get ignored. I guess most people don’t see the point in making the effort. The forums have, unfortunately, become more of a place to vent than for constructive discussion.
I have, to date, not received a single reply to any thread asking them to clarify whether something is a feature or a bug. That’s sort of the baseline communication I hope for. I’ll keep on posting as if they care for as long as I can, then quit the forums until I can run the charade again.
edit: There are 4 Anet responses on just the first page of the GW2 Reddit.
(edited by Sviel.7493)
There was this really hilarious one on last night that would just spam dodges non-stop. 3 dodges, Signet of Agility, 2 more dodges, Channeled Vigor, 2 more dodges, then smoke cloud spam. Oh, and with Sigil’s of Energy on weapons for more dodges.
Like, I could stealth and take two steps back and just watch him go to town on nothing.
He was also wearing Dire or Trailblazer so he took forever to kill, but since he blew all his cooldowns on thin air it wasn’t a big deal. The kicker is that, if he got 2 stacks of bleed on me from one of those dodges, I’d die if I didn’t clear it. That means that if I had to clear condis once, getting hit a second time meant certain death. If the guy had any idea what he was doing, it would have been really annoying.
We don’t know if the problem was that Retaliation was too easy (in which case the current situation would not offend) or if the entire interaction was unsavory.
I suspect it’s just an unforeseen consequence like when they removed Centaurs from the borderlands but also required a drop that only came from those specific Centaurs for the guild hall. It’s not quite that cut-and-dry, though, so clarification is needed.
Is this happening around Airkeep or any other particular area?
I know the sand flying around on the east side was a problem for someone I knew. Wasn’t unplayable, but was noticeably different from the rest of the map.
Sorry, I was referring to conditions being applied to siege in a similar manner to Retaliation. That is, siege hitting players and drawing conditions in the process which leads to damage. My presumption is that Retaliation was changed because this wasn’t a good interaction, but I need some clarity so that I know if I should invest in making guild ACs.
Between the feedback damage and the doubled time-to-kill, it makes more sense to just spam siege disablers instead of building an AC. If I have surplus supply, a door treb is much better against rams. Catas are kind of GG since ballistae can’t usually hit them and take forever to kill catas if you place them aggressively enough that they can hit, so I’m relegated to just spamming siege disablers there too. 4 more disables (5 instead of sup AC) means 140s (175s) of delay. Killing the siege simply isn’t an option anymore since my siege dies or the wall is breached before the enemy siege dies.
Bump for business hours~
When WvW first launched, Retaliation could be used to reflect damage back on siege weapons. In September of 2012, this was changed.
Patch Notes17 September 2012
World-versus-WorldRetaliation: This boon no longer reflects damage received from siege weapons.
In July of 2016, almost four years later, the type of health siege weapons have changed so that they are no longer immune to conditions and critical hits.
Patch Notes26 July 2016
WORLD VS. WORLD
General
Siege weapons can now be affected by conditions and critical strikes. In addition, their base health has been doubled.
As a result, they can now be set on fire by a blocking Guardian or bled by a Necro with a chilling aura. The damage is much greater than from Retaliation. Was this intended?
In addition, since siege weapons cannot crit and do not apply significant condition damage, it now takes double the time to destroy a siege weapon with another siege weapon. Was this intended?
Gear scaling was removed from ACs like a year ago. Also, 1 Sup. AC can’t kill a Sup. Ram before a paper gate falls even with 2 disables.
On one hand, everyone’s going to be doing this now that you made a video. On the other, that might mean that Anet actually does something about it quickly.
But it probably means that everyone’s going to abuse it because fixing it probably isn’t that simple. Inc full ranger zergs.
Retaliation used to damage ACs but that was removed. This is a new thing since they made ACs susceptible to conditions and crits that was probably unintended. I’ve never seen the condis actually show up on the bar but I usually don’t have my own AC targeted.
I wouldn’t worry too much about the reaction you get here, though. The forums are chock full of internet warriors that don’t want anything in WvW that they can’t bash in with their face. It ranges from GvG to BlobvBlob but all boils down to turning WvW into a battleground deathmatch. You can tell by how often they cry foul that Anet isn’t on board, probably, but it’s hard to tell what Anet’s thinking.
Thank god. So many years of zerker or GTFO is finally over, and hopefully for good.
Well, you’re not wrong, but I think we traded it for something worse.
Zerker at least fits the action-oriented combat system they set up. Condis do not in their current form.
Anytime a person’s main objective in playing is the rewards, they aren’t going to be contributing as much as possible to the actual competition. There’s no changing that and there’s no point breaking everything else in the effort. The goal of this change is to have people feel rewarded for playing a game mode they find at least moderately fun. It’s not to incentivize people who don’t care about the game mode at all to clog it up in hope of getting new shinies.
Also, reducing rank-based rewards due to what EotM used to be is short-sighted. Sure, some folks now get more rewards because of that, but that has nothing to do with anyone else. New players are 100% unaffected and veterans aren’t adversely affected either. It feels a little bad, but why make it worse for everyone for all time including yourself?
The Outnumbered bonus would be better served with a delay. You don’t get it the first skirmish, but you do get it all the rest. This is similar to the lock-out on pips from switching servers, but on a smaller scale. It needs to remain high to encourage people to stick around, but it should not be so easy to abuse.
On home borderland, it’s tough to hold more than the northern corner with 2 enemy zergs running. When we take Hills/Bay or southern towers, it’s more to reset upgrades. We can’t really defend them successfully, but we do burn supply as much as possible as they go down.
Also, new players tend to flock to the tag no matter what. That means that you have a 50-man zerg taking an empty northern tower. It’s a huge waste, but many don’t know enough to go and do things without a tag.
I think it’ll even out, with time. I’m seeing a whole lot of new pseudo-scouts and that makes me happy.
You can’t cap it while stealthed, but if your opponent has to run away, you don’t have to stealth.
I dislike it because it’s no fun to fight. There are counters that make it easy to beat and then there’s everything else where you’re just shooting in the dark. There’s nothing that I’m doing in the fight that is really engaging. I clear condis at certain intervals that don’t seem to matter much since they’re reapplied so easily. I try to land attacks through the stealth/block/evade but whether I land enough in time seems largely out of my hands depending on what class I’m playing. Worse, it’s not like I’m being outplayed—it’s just someone spamming skills off cd. Win or lose, it’s just a pain to deal with.
Contrast with pre-nerf Gunflame warriors. Getting nearly one-shotted (depending on class) wasn’t fun, but if I managed to dodge or reflect it then I felt pretty good. There were clear avenues to take and the trick was choosing more of the right ones than my opponent. It may not have been balanced, but it did feel like a fight. Even if I was playing a glassy character, I had things I could do to try and win.
Pirate meta was more about stability and boonstrip than power/condi. The latest stab changes were a good compromise in stepping back from that. The complaints about condi center around the binary nature of it. Either you have the means to render it useless or you get destroyed. There’s not a whole lot of in between. This is more or less the case for solo, small team and zerg fights.
@Roxanne
Everyone can run away from you, but if you’re taking an objective that’s not really an issue. Also, the class is annoying not because of it’s raw power but because of the way its power is implemented. It has the standard condi problems and, since reapplication is so easy and durations are so long and confusion/torment are the most damaging conditions, it literally hurts to deal with. Throw in stealth, blocks, evades en masse and even the bad ones feel grimy.
On the other hand, if you’re at the max rank threshold and come in third place in every skirmish, you’re guaranteed 8 pips per tick. That puts you in the ~15 hour range.
I think it’s ok that some rewards are beyond the reach of the average player. That means that people who have no life that go above and beyond get rewards that go above and beyond.
I like this approach. It may need to have fewer stages, though.
No. Please stop asking for this.
If I’m tracking several mid-to-large size groups, I may end up having my participation decay from lack of actually killing anything. Even if I hit a group with a supply trap and pressure their build sites so that they retreat, it doesn’t count towards participation. As a result, my rank is much lower than a zergling with comparable time.
But, really, who cares? I’m not going to make a fuss over a few pips. Also, I can still play when I’m outnumbered whereas a zergling isn’t getting much exp then…so the pips sort of even out.
I feel like I’m in the twilight zone…WvW players complaining about too much population, content being released too quickly and even some thread asking to buff rewards in EotM.
We’re not in Kansas anymore…
http://www.mmorpg.com/mobile/features.cfm?read=11783&game=473&ismb=1
The breakdown is +1-3 from match placement, +1-7 for Rank, +1 Commander, +1 Outnumbered, +1 Loyalty.
Since the rank reward is tiered, it shouldn’t be too big of a disadvantage. We also don’t know what the maxiumum rank needed for +7 is, yet. Also, if a roamer is doing things besides just ganking players, they should still have a decent WvW rank. It really comes down to what the threshold is.
I wonder if you were one of those terribad condi mesmers I’ve met this week? The build is certainly very strong, but, somehow, some people manage to be mind-blowingly awful with it. I’m not sure what they’re doing wrong.
You’re right about the burst, though a condi build isn’t about burst damage anyway. You should be stealthed, evading or blocking almost the entire time while your condis go to town.
A bit more info now~
Speedy Yaks is definitely a 100% speed increase.
Superspeed from Speedy Yaks is actual Superspeed and can’t be boosted by swiftness.
With that in mind, and ignoring any chance of assassination, Packed Yaks will be slightly better than Speedy Yaks for upgrading structures. If the Yaks are escorted, the scales tip further.
Speedy Yaks will be much better for gaining points and is probably best employed for that purpose. It will probably still see widespread use for upgrading since it can be unlocked much earlier than Packed Yaks.
As for assassinations, I personally find Speedy Yaks more of a hassle. If they aren’t being escorted, it’s no big deal as long as they aren’t running in opposite directions (see: North Camp). If they do have an escort, there’s just not enough time to kill the Yak in many instances or I can’t survive long enough while the escort attempts to murder me. That said, if you have that level of escorting, it doesn’t really matter which kind of Yak you use.
They got an indirect 50% damage nerf that I’m not entirely sure Anet noticed. They weren’t all that great before, either, tbh. However, the inner cannon being able to hit through the gate is too powerful anyway.
WvW players like to think they’re better than PvE for a variety of reasons. Even if they’re trying to play nice, the numerous, common expressions that denigrate PvE players will probably win out. You’re right to ask for effort, but don’t expect it to change much.
The people that were going to play nice, will. The people that were not going to play nice, won’t.
-After being on the server for one match, you just earn pips at a slower rate. We don’t know how much slower.
-You’re completely locked out until you have been on the server for 1 complete match. That means if you transferred before reset on May 26th, you won’t have any down time. If you transferred after reset on the 26th but before reset on June 2nd, you’ll be locked out until after reset on June 9th. If you just transferred after reset on June 2nd, you’re locked out until after reset on June 16th.
Why make it harder to communicate in the name of historical parallels? Thematically, it’s cool, but why wreak havoc on current gameplay for that?
So Anet is telling us if we bought a tag for 300 gold we can’t use our tags unless we have five people to group with us?
Or did I misread that little tidbit?
300 gold… I can imagine refunds will be the next thing players are going to ask for since that would be incredibly unfair on so many empty maps – and I mean empty.
How are people just logging in going to find anyone if a person who CAN tag up must spam the map or team continually to find four more players…?
You’ve misread it~
You can still tag up just like you do now and nothing changes. The only update is that if you’re tagged up and you have 4 other players in your squad, you’ll earn the new rewards slightly faster (just you, not your whole squad).
So if you’re tagged up solo, everything is the same as it is now.
What about all the rest?
The most notable omission is Scouts, but you’re also missing Havoc and a variety of usually un-named roles. If I may…
4a. Stationary Scouts. They remain in a single objective and relay news of enemy attacks to the rest of their team. They build and refresh siege, manage the supply in the structure and lead repairs following an assault.
4b. Havoc Scouts. They patrol the map or a section of it and relay news of enemy havoc teams or zergs. If possible, they’ll also engage said havoc teams. They’ll check on un-scouted objectives in their area if they believe there is an attack and will relay that information as well. They flip camps and sentries and generally keep supply lines running.
5. Havoc. This can be one person or a small group. They disrupt enemy supply lines and assault enemy structures that are not guarded by large garrisons. Their goal is to drag the enemy zerg around by forcing them to respond for defense and therefore leaving opportunities for other Havoc teams or zergs to assault other structures.
If the buff is too small, it won’t be worth protecting. If it’s big enough to help against small teams but not large, it is unlikely to spur people to hold objectives but will push them toward zerging over small-team play. I don’t think there’s a sweet spot where more defensive buffs work.
Paring it down to just a trader would eliminate that problem, but people would be able to finish their business with said trader after seeing him once or twice. After that, the incentive would be gone. In addition, they can just hold an easy structure like their own Garrison/Rampart and not worry about trying to get traders anywhere else.
The goal is still good—just the implementation is tricky.
I like the base concept of an Economy System, but I’m not sure how it would actually work. Perhaps Camps that benefit from it send out Yaks with bonus supply? That wouldn’t change anything for an empty map as the structures there all have full supply anyway. For towers and keeps, I’ll need more time to think. Or perhaps they can just function as increasing the bonus for camps?
You have to be on a server for a full match before you begin earning rewards. So if you transfer on a friday after reset, it’ll be 2 weeks.
I’ve been trying since that post and haven’t been able to document any speedy Yaks. That is, I’ve seen enemy speedy Yaks but I’ve had to kill them and flip the camp rather than taking notes on how fast they upgrade objectives on my BL. No allied speedy Yaks have completed trips while I was watching. This is likely the result of my playing in a low-pop timezone this week.
I’ll have the hard numbers for you asap. I’ve ranked my guild up to lvl 12.9 and am just waiting on aetherium while I farm the obscene amount of leather for future upgrades.
What I can say is that Superspeed is a 100% increase. Thus, it should be 2x speed increase rather than 1.5. You state that the maximum move speed for a Yak is 150% but I do not see this documented anywhere. When I have the numbers on hand, I can settle it. I have not applied Superspeed to a speedy Yak but I have applied swiftness and seen an increase in speed. That’s where my statement comes from but, again, I’ll wait until I have numbers to insist on it. My statement that speedy Yak Superspeed is a different thing derives from the observation that it stacks with swiftness as opposed to normal Superspeed which does not.
There’s a lot to break down here, which is fine, but forgive me if it’s a bit out of order.
I want to talk about your Merchant idea first. It is essentially an array of defense-oriented buffs in addition to existing systems. While it will increase the value of holding objectives, it does so by layering more things on top of existing things. The result is that defense becomes much more powerful in a way that cannot be undone through gameplay. Offense can prevent these things from happening in the first place but, once achieved, progress cannot be reset without fighting directly against them.
In addition, these changes favor the larger, dominating server over the smaller servers. They increase the disparity in power between the two and, thus, exacerbate temporary population differences.
As such, I cannot support this idea as is. I believe you have correctly recognized that mechanics are what need to be tweaked, but this is not the way to do it.
Second, I want to contest your claim that mechanics to incentivize holding objectives do not exist. I argue that guild tactivators, the current upgrade system and the related PPT system encourage holding objectives. The purely defensive advantages of higher tier walls are plenty of incentive. The higher levels of tactivators and more than double PPT are even more incentive. The problem is not a lack of incentive but that the incentive is not as effective as it should be. If players don’t care about PPT, why should they care about being able to defend? What’s the point in higher tier walls, tactivators and more PPT if you don’t care about the score in the first place?
This new update that, ideally, makes the score matter a bit more will perhaps improve our situation. If players want a high tick to get the extra pips, they’ll be more inclined to defend a 20-point keep and more inclined to keep supply camps running to get their structures upgraded. Again, the root of the problem is that players do not currently care about the score.
Third, I want to return to mechanics. Your idea to tweak them to incentivize holding objectives has merit. However, it must be done in a way that doesn’t confer semi-passive bonuses as those overwhelmingly benefit dominating servers. The best starting point is simply making sure the current bonuses apply consistently.
For example, T3 walls are a great boon in defense because they buy you more time. However, unless you can do something productive with that time, it is meaningless. Therefore, mechanics that cannot be countered no matter how much time passes render the defensive boon largely moot. For example, catapults placed in positions where they cannot be hit from inside the keep no matter what essentially nullify the defensive advantage. The only recourse is to fight the stacked enemies directly which, of course, aggravates the power disparity between large and small servers. Now, if those catapults could be hit from inside the keep but were protected by shield gens and such, a sacrificial thief with a disabler could create an opening to treb or mortar the catas down. With enough time, a smaller team could withstand an assault from a larger team. The larger server would retain the advantage as they have more people to hold supply lines, carry supply, man siege and fight, but the extra time from the higher tier defenses would be a meaningful factor. As a result, defending upgrading objectives would have more meaning. They wouldn’t just be auto-flipped when the current guild ends its raid.
edit: I should note that a smaller team assaulting a larger team cannot make use of uncounterable offensive siege as they will simply be run over. Thus, the argument that a boon for small servers is a larger boon for large servers cannot be applied here.
(edited by Sviel.7493)