Guild Wars 2

There’s a post on a thread somewhere from Justin saying it does hold classes back to try and form teams with variety, but when there are a lot of one class queueing, it’s impossible to create every team without duplicates of those classes. Eventually, rather than keeping the imbalance of those classes held back, once a certain waiting time is reached for them they just get put into a team regardless of that team’s composition.

So, yes.

Stun break is one option. The other is to disable them in return. Steal when traited with sleight of hand can be used defensively as well as aggressively.

jim.5380:

bountiful theft was granting u as much might as u removed
and priority was always aegis stability

Nop, if an ele has 12 stacks, then applies 1 more stack from fire attunement, your steal will remove the 13 stacks and grant you and your allies 1 stack. If the ele instead applied 3 stacks with a area might combo, your steal will remove the 15 stacks and grant you and your allies 3 stacks. It’s the last application that you get – and that last application can be more than one stack.

Also, aegis is not the highest priority because it can block steal if the target has it with stability and protection together.
- Stability
- Protection
- Aegis
- Regeneration
- Retaliaton
- Fury
- Swiftness
- Vigor
- Might

Uh... I haven’t tested it, but surely since corrupt boon corrupts all stacks of a boon it corrupts, and stability is a) not a stun breaker and b) you might not still be getting the stability from dagger storm if you were interrupted (again, haven’t tested, either way point a. holds up), that corrupt boon or dark path will still fear a dagger storming thief for the full 1-2 seconds depending on duration modifiers, and of course interrupt the cast and stopping the reflection.

Oh and don’t get greedy vs. engis. They can survive at low health for quite a bit kiting and blocking, and all it takes is one juke from them and you’re suddenly the one running or worse.

Of course you did better than others at your "level!" You took the initiative to investigate PvP concepts to improve, while most just jump in, fail, continue to fail, and blame everyone else!

There are circumstances where you might want to engage a bunker, such as to hold a neut for backup or just to delay the enemy cap, or because you know you can kill or at least pressure the bunker on his point quickly enough that you’ll either win or force them to send backup to help the bunker. If you can’t, you don’t push it alone, and may instead opt to go with another high damage team mate to relive some pressure from another point or to take it when you can’t take something else, or just to be aggressive while the enemy is respawning.

A good engi will always force you off a point, not the other way around. You can win, but alone, it usually takes more time than it’s worth investing into the endeavour. You need to dodge their grenade 2 or cleanse the bleeding asap (usually the first thing they throw when they swap to grenade kit), try to anticipate overcharge shot or keep him blinded as much as you can hoping it’ll miss (it’s an instant cast skill). When you’re knocked down, stun break and dodge the burst, dodge their immobilise, and just try to be aggressive because otherwise they’ll be able to tank you. Engineer’s are weak to conditions since their only cleanse is their turret, so a good move is to - after their block and heal turret are on cooldown - immobilise them and drop throw gunk (stolen skill) on them. If you’re panic strike this should be easy to do; if you’re S/D you’ll need to stealth and swap to shortbow for the stealth attack. Cluster bomb in the gunk (not detonated) will give you (and any allies close enough) chaos armour.

With the plot-twist that was the buff to ports, I thought maybe the problems that have plagued shadow trap for so long, making it unreliable in many positions – one step to the left can be the difference between a successful port, going nowhere, or getting stuck in a wall – and thus unreliable in practice, would be reduced or removed! Unfortunately shadow trap, a skill with so much potential, remains as it was before the patch.

A potential fix, I imagine, would be to execute the shadow step that is 10k range, 1k at a time instead of 10k at once. Those points can each be plotted at a valid location, avoiding obstructions, resulting in the player making up to 10 valid shadow steps instead of 1 which will inevitably have numerous, problematic obstructions. Were a thief to walk or infiltrator’s arrow to their destination, they would be sure to move around those obstructions anyway. It now works often flawlessly with short distances, so perhaps taking advantage of that for the longer range step is the answer to finally removing its bugs!

Yeah sorry for the length, but I figured going into yet more detail might help, so I copied, pasted and commented on some of that information from wikipedia. I didn’t write as much as it seems; at least half of it is quoted from the wikipedia page for lag. Alas, too little gets contradicted, a little more doesn’t get understood or provide sufficient clarity, and a little more again gets ignored! What more can one do =P.

I just used the -repair argument. It took a while, but when the launcher re-opened it downloaded the last file fine. Perhaps having to close and re-launch it during the first attempt caused some problems that had to be repaired before it could succeed in a folow-up attempt.

I’ve been stuck on 99% for almost an hour now. I restart the launcher when I get the flashing on/off connection error, and it starts downloading normally again, but it’s been trying to download this one last file for almost an hour. It keeps going between 30KB/s and 150KB/s (which is very low for me), and hasn’t moved from 0%.

Bountiful theft always removed all the stacks. It just didn’t grant all the stacks to you and your allies - only the last group applied i.e if 3 might stacks were applied, and last, you’ll rip all the stacks and share 3.

Engage from behind an obstruction with infiltrator’s strike. Then just auto attack. Don’t faceroll or do any pre-planned rotation. React to them. When you see the lines go around their longbow, use flanking strike to evade it, then comes rapid fire, which you dodge or infiltrator’s return away from. They’ll likely swap to GS now. Re-engage. If you see them holding the GS up to block, or see "block" float on your screen, dodge immediately to avoid the knockback. Next comes 2 or 3, you can dodge those with a dodge roll or flanking strike - they are very telegraphed. You might get hit by GS 5 which is the daze, in which case you can steal to daze them as well, break out of it, or just dodge until it’s gone. You’ll have to do this s couple of times because most rangers will use signet of stone. When they use it, just go back behind your obstruction, or CnD if you can’t, and wait it out. Be sure to dodge their pet’s knockdown and fear. The knockdown is when they dog crouches, and the fear is when he tilts his head back.

As for engaging with CnD+steal, if the ranger sees you start CnD, they may dodge, in which case you miss you stealth and your steal. Not good. It also means because you didn’t engage with infiltrator’s strike, that you have no return port to escape rapid fire or wait out signet of stone.

And just for clarification here, because I think you might be missing the link between latency and lag: "Latency is a time interval between the stimulation and response, or, from a more general point of view, as a time delay between the cause and the effect of some physical change in the system being observed." You see, latency is latency, and lag is lag, but latency includes lag. Latency is also one of the potential causes of lag, but regardless of its cause, adds to your latency.

The client-side section on wikipedia is actually well written, so imagine that <here> I just quoted the whole lot of it for you to read (I know, lame that you linked it to me now I’m linking it to you.. Childlike comebacks right? But seriously, if you didn’t read it yet, do, it explains in detail everything I said). Basically, as a means to combat high latency, one or two techniques are employed: extrapolation and interpolation. If you’ve read it, you’ll now understand that extrapolation is one of the causes of the desync you see when someone goes down but you can’t stomp them. That is because a delay caused the client to predict where the player was going to be, but he went down. (The other cause is when that player is lagging and rubberbands after going down. Either way, this is something that can be fixed by ANet either by a fix at server or client-side, forcing the game to update the position of the downed player - this is why it fixes itself when they move using a downed skill, for example). Now, the very fact that these measures are invoked prove that a desync has been caused by latency. The techniques rarely succeed in rectifying the problem, and instead attempt to minimize its effects. Make no mistake, just because the measures to minimize the effects of lag are on the client, does not mean the problem was with the server. It could have been at the client, at the server, or anywhere inbetween. Latency. A good example of this is when you and your buddies are not lagging, but see someone that is rubberbanding everywhere. That rubberbanding on your screen is caused by extrapolation, and yet you are not lagging, meaning there is no lag at your computer, the server, or anywhere inbetween. It’s at the client or route between the server and client of that player. This often leads to desyncs, not caused by the server, and yes caused by latency.

"Often, in order to allow smooth gameplay, the client is allowed to do soft changes to the game state. While the server may ultimately keep track of ammunition, health, position etc., the client may be allowed to predict the new server-side game state based on the player’s actions, such as allowing a player to start moving before the server has responded to the command. These changes will generally be accepted under normal conditions and make delay mostly transparent. Problems will arise only in the case of high delays or losses, when the clients predictions are very noticeably undone by the server. Sometimes, in the case of minor differences, the server may even allow "incorrect" changes to the state based on updates from the client."

This is exactly what I was talking about. And yup, even wikipedia says it’s caused by high delays or losses, A.K.A lag and thus latency.

Oh there’s something I didn’t see mentioned on wikipedia, which is a possibility (although I didn’t skim through all of it). It says this: "As clients are normally not allowed to define the main game state, but rather receive it from the server, the main task of the client-side compensation is to render the virtual world as accurately as possible. As updates come with a delay and may even be dropped, it is sometimes necessary for the client to predict the flow of the game." But doesn’t say that packet loss can also occur at client-side, completely unrelated to the client, if there are some networking flags set which your hardware is too old to recognise.

So, using wikipedia.

"This need to communicate causes a delay between the clients and the server, and is the fundamental cause behind lag. While there may be numerous underlying reasons for why a player experiences lag, they can be summarized as insufficient hardware in either the client or the server, or a poor connection between the client and server."

You completely left out the most common and significant cause of communication delays. Not at the client, not at the server, but the multiple server hops inbetween, and the communication lines between each of those. If a signal you sent to ANet is on its way let’s say on the 10th/12th hop, ANet’s server isn’t going to know about it - not yet.

"The frame rate (or tick rate) of the server determines how often it can process data from clients and send updates. This type of problem is difficult to predict and compensate for. Apart from enforcing minimum hardware requirements and attempting to optimize the game for better performance, there are no feasible ways to deal with it."

There is some more delay between registering input from clients. This delay, much like that of preparing data for transmission, encapsulating it, transmitting it, the time it’s spent traveling the net, received, reconstructed and error checked, and passed up to the application level, is unavoidable. Sure improving the hardware will help minimize it, as will as optimizing code, but there are essential difficulties inherent in software engineering that simply cannot be overcome - at least not any time soon (talking decades here). Regardless, if you constantly improve response times by 50%, you’ll never reach 0, and doing so is a slow process and 50% is extremely optimistic - I just used a bigger number for emphasis.

And yes, wikipedia says that too! "Perhaps the most common type of lag is caused by network performance problems."

Why are we arguing, you ask?

Mrbig.8019:

It’s no latency.

if u have high latency you’ll see a desync between your dodge and your endurance bar depleting ( basically the thing you’re saying).

I’m talking about a totally different problem, or rather dodge roll evasion frames are bugged.

Ok, it’s not latency which causes desync which is the issue. It’s something totally different.

Mrbig.8019:

Again, the phenomena you’re describing is not latency.

it’s server desyncing.

Wait, now I’m not describing latency, and you’re saying it IS desyncing?

^ That’s why. Not only have you not decided what you want to argue for yet, but you haven’t decided what I was saying yet either. So we’re kind of on hold. Also, those wikipedia pages said pretty much exactly the same thing I described, they just explained it in detail rather than using an example.

For longer stealth you can can it in black powder. Tilt your camera so you are looking as far down as possible, and heartseeker will move you less distance. That way you can easily get 3 heartseekers, and sometimes 4 inside the black powder for longer stealth.

What you do as D/P is blind the hits you really need to avoid. If they jump into the air or you see some effects around their weapon, it’ll usually be something important to dodge so use shadow shot. Shadow shot is amazing, it’s a blind, gap closer, and good damage. Generally while in a fight already, you’ll only stealth to cleanse condis or escape. Use black powder - try to anticipate when you’ll need it because it’s easy to interrupt - get some damage in, then before it dissapears or when the target moves away from the field, heartseeker. If there are more people running to kill you, get another stealth combo and leave, otherwise land a backstab immediately.

It’s nothing like shooting up in the air. That’s a problem in the code, not communication timings. I’ll happily debate this with you, but I’m not going to part-take in:
- "again, yes it is"
- "again, no it isn’t"
- "is"
- "isn’t"
- etc.

Stats in PvP are lower, and more controlled because you use a PvP amulet for all your stats, not individual gear pieces. All stats are normalised so you won’t be facing full ascended as exotic. WvWers use stat boosting consumables which you can’t use in PvP.

Of course it’s latency. Even if you’re not lagging, it still takes time to communicate with the server. That time frame is where the issue can present itself, and is precisely why it happens at the start of a dodge, and not toward the end. Effects that are triggered on dodge still trigger because you still dodged. You just got hit within the window in which the server did not know you were dodging.

Aye, normally you’d press a command, it gets sent to the server, the server validates it and says ok, then executes it, and your incoming stream gets processed and your graphical information gets redrawn showing you execute that command. For dodge, it is started at client side then the fact that you are dodging gets sent to the server for it to use to grant you evasion of attacks for the duration. So in the case of dodging, you see yourself dodging before the server does, so there is a period between the start of the dodge and when it gets to their server and is validated and recognised where you can still be hit. As mentioned for skills, the hit is determined at server side, so if the server registers that you’ve been hit just before it receives the information that you started a dodge 20-40ms ago, it has already sent the information to your computer to say you’ve been hit and took x amount of damage. So you can see yourself getting hit during a dodge within the duration it takes information to travel from your computer to the server and back, from the start of that dodge. It’s not something ANet can fix, because in the scenario your computer "knows it is executing a dodge" but the server "knows you took the hit."

Edit: Actually what am I saying? Of course it can be fixed. It would just be extremely fiddly and difficult, and create other problems. For example, let’s say your client has the ability to say, "hang on, I’m dodging, I’m not taking that hit." By which time the server now knows you dodged, and needs to amend the values involved, such as your health, your endurance, and the other player’s facts like if it was a thief that used flanking strike, it would have turned over to larcenous strike. What’s more, the other player, and all other players in the area, would have seen you take the hit. You’re now out of sync, and re-syncing would be confusing, and people would then be complaining not that they got hit mid dodge, but their hit was evaded when their target wasn’t dodging! Imagine seeing you hit land, then suddenly the target dodges, their health from that hit is restored, you get a seemingly random "evaded!" floating notice, and any skills you had unlocked resulting from that hit get turned back. It gets worse. What if you activated that turned over skill you were not meant to have before the resync? What if the player went down, and you see them suddenly get back up with a dodge instead? What if you had started stomping and tanked some damage to do so, only to find the player didn’t actually go down and you went from winning to almost dead? Getting hit mid dodge is the simpler and less chaotic problem of the two to pick.

Yuo dno’t ntoice typos bcuase the brain auto-cerrocts them. In a sentence full of them, like that one, you would notice. Granted, you’re less likely to not notice if your first/main language is not the one being used, but I did not say 100% of people. What I said was that one does not notice because the brain auto corrects. Such a cause of a typo would not prevent it from being identified. Once again, you’re twisting and mixing-and-matching what was said so that you can respond to it argumentatively. If you want to play that game, you should at least make it so that it cannot be so easily done to you. We could easily say "Umm, no. Stealth does not interrupt stomps. It prevents the stomp from finishing the target but it does not interrupt anything." We could once again repeat that the build was being reviewed for the benefit of the OP, which you didn’t seem to be able to twist to your liking and so claimed to not care. We could then point out that you repetitively claimed to not care, and yet are still here and are now going on about what you do in fact use. Ironically, we don’t care what you use, but you didn’t seem to get that because it wasn’t so bluntly said. We reviewed the build posted due to the nature of the thread. No one is reviewing you; for that, make another thread. I could go back further and point out that sure, no one said "this Jana chap is good," but no one said you’re bad or could never be good. You arrived at that all on your own, and told us that you went and told your friends that it’s been said of you. From a psychological perspective, I could assert that you’re seeking attention and approval. Again, you should make another thread for that. I also point out that I in fact said the shadow arts build is good when learning a new role, which is something positive about it, when all you’ve done is invent negativity for it. Do note too, that everything was regarding PvP, and not WvW, as PvP was what the OP mentioned.

If you can’t see that by now, you’re either misunderstanding a lot or twisting people’s words so much that you believe what you’ve invented, or trolling. I did think you were just looking for conflict, but it’s one of those believable cases so I gave you the benefit of the doubt and replied to everything politely.

You’ll get mixed answers for questions like that because everyone has different experiences - they have different skill levels and encounter other classes used by players of different skill levels.

To my knowledge and experience, necro is strong in 1v1 against any class, with the possible exception of a good warrior since they have a lot of condi clear, passive heals and condition immunity. That said though, most warriors run shouts now which shouldn’t be a problem. A necro should win a fight if they cross another player off point, and should be able to pressure a point enough to force reinforcement, but they can easily be overwhelmed and are not bunkers, so 1v2 would be problematic. That said, it can be done since a lot of compositions opt for less damage and more sustain, so if you use your extra health pool and plague form really well you could sustain a while - even on point long enough for back up. An organised team that runs with a condi necro will want to protect their necro because their necro can pretty much allow them to dominate the team fight, hence why necros are usually the primary target.

Edit: To clarify, the necro build I have in mind when saying all this is the standard/meta terror build of 0/6/4/0/4; carrion amulet, not rabid.

Thief receives hate because:
- At a low level, thief counters every berserker build for every other class easily, and I mean pop in, swing your weapon about a bit and they drop - literally, easy.
- Thieves are hard to kill because they escape. Even if chasing someone off means you won, people like to get kills, so they complain.
- In tPvP, thief is only really countered by itself. It may have difficult matchups, but it’s role as being quick and sneaky to get decaps can only be countered by another mobile, aware, player.
- Thieves survive by blinding and evading, and having your hits miss often is annoying, so people complain.
- Thieves are designed to be able to adapt. People don’t like to adapt, they just want to be able to win "their way," so they complain.
- Thieves can dissapear. People don’t like having their target vanish, so they complain.

Stats in WvW are much higher than in PvP. You can steal put out nice damage in WvW while using shadow arts. Trickery is more for utility and survivability than damage though.

Oh I see, Jana. You took a typo that was meant to be "to," and based your responses around that instead of what it meant in order to continue deflecting. Only, it didn’t work out very well because small typos are read as they were intended 99% of the time - that is to say that the brain "auto-corrects" it while it’s reading. Using it to declare that I need to work on my English is actually a low blow considering I’m currently suffering from repetitive strain injury in my left hand and wrist, which makes typing quite difficult, but then you wouldn’t know that.

Anyway, all your responses pertaining to our discussion since you started this little conflict have been based around the straw person fallacy - that is, deliberately displaying a misinterpretation of what was said to avoid addressing it.

It’s my first and only language, and that sentence is both correctly phrased and true. If you’d like to explain to me what you’re not understanding, I might be able to help you.

The problem with every class having an energy resource to use their skills instead of cool downs is that they’ll have to be rebalanced for that. Imagine a warrior just spamming GS 3 accross a point, or a ranger doing nothing but rapid fire, or an elementalist laying down half a dozen simultaneous meteor showers.

I despair for you Jana. I did not say I don’t enjoy having the last word - anyone who knows me knows that I do. What I said was that you clearly care enough to want it. If you want to make this about me to escape giving a straight response to that fact, you’ll need to do a better job than that. Alas, the simple fact that you returned to try, confirms my assertion.

Nephrite, he was going for damaging traits, but missed executioner. For damaging traits, fury on steal trumps shadow’s embrace. At least, that’s my take on his thinking.

You can go straight into heart of the mists for PvP via the portal in Lion’s Arch, or right clicking a friend who is in a hot join or custom arena match and clicking "join friend in PvP." In HotM you are level 80 with access to most of your skills and traits.

You care enough the want the last word though =P.

Well it was amusing until I had to clarify that I was referring to you wanting the last word. I’ll edit it for you v_v.

Jana.6831:

Alright, let’s try again: my build link wasn’t in so you can all judge it and tell me just how far I’m off of the meta and that I will never become a good thief (which made my friend lol when I told him) – but to give an example as to how link builds – I couldn’t care less about what you think about my build =)

Perhaps not, but you posted it on a thread where someone is testing out builds. Whatever the reason, you must have anticipated it getting reviewed for the OP.

Off topic but +1 for saying “couldn’t care less.” So many say “could,” which is annoying =P.

The critical strikes version is 2/6/0/0/6. Faster at killing but more vulnerable to conditions and has an arguably weaker heal because it needs hide in shadows for the condi clear. That being said, both are still used, they just have different play styles.

I didn’t look at the the build link from Jana, but I have now. That’s a variation of a couple of very old builds you used to see, before the sigil changes which might be why you’re not using fire and air. One was D/D or D/P 10/30/30/0/0, and the other D/P 0/30/30/10/0. The former is a nice build for new players running thief in PvP; it lets you work on the play style and mind set a thief needs for its role while not dying all the time, but as you start meeting better players the build and play style need to adjust - you might not see that now but you’ll realise it when it happens, and why. The latter used to be used competitively as a far-point assaulter build, using shadow trap for decaps and 3 signets for might stacking (malice, assassin’s and either agility or infiltrator’s [I forget]). It was alright, but rather restricting in versatility/adaptability, both for you and your team.

Edit: Got my "latter" and "former" mixed!

Sounds like an attempted variation of the old "carry" build. Back when you’d have a glass thief carrying a bunker to a point, killing anyone who tried to stop them before they could join the fight to support their allies. The old build was 25/30/0/0/15. Your build trades the key traits for finishing targets off quickly, for a bit of extra survivability, but less than the panic strike D/P which opts for condi cleanse over a bit of swiftness and vigor (6/0/2/0/6). The former build also has vigor from bountiful theft, and the ability to rip and steal boons which can give it swiftness, stability, vigor, fury etc. and as mentioned already, sleight of hand may sound like a near worthless trait at first, but after using it you realise just how much more you can do. The 6/0/2/0/6 also gets the damage modifier from 5 in trickery, while your build gave up its two damage modifiers which are 5 and 6 in critical strikes, meaning it probably has more damage than your build as well as more survivability.

A nice build craft though. It may be far from as effective as the meta builds but don’t let that disuade you. After all, someone made those at some point too.

It sounds to me like you want to play necro but are afraid you’ll be dissapointed and find it doesn’t have what you want. Necro can roam, but it’s slow. Condi necro is a threat to any class in 1v1 and can counter the celestial builds of the current meta. It’s usually the first target to be focused down in PvP, and I imagine it will have a high skill cap to be good at in WvW because every other class can out run it. Condi necro does its damage from range.

Engineer has what you want your class to have. It is a great roamer in PvP and WvW, and can be very good at applying conditions or be a jack of all trades with fair burst, condi, and support. It prefers to throw grenades from range but has no shortage of utility to deal with close combat situations, has lots of different skills, strong area heals, and damage mitigation. Ironically for you though, condi necro counters engineer because engineers don’t have much condi cleanse, and necro’s both deal great damage from conditions and transferring conditions they get affected with back to their target.

Thieves, mesmers and rangers are more of a survive through skillful play and escape to survive rather than out last like an engineer would. However, for WvW, you can’t rely on being able to out last your opponents as much as you can in PvP with allies always nearby, so mobility is great for WvW.

I’d say if you want to do mostly PvP, try your necro or an engineer. If you want to do mainly WvW, look at engineer, ranger, or thief. That’s not to say ranger and thief are bad in PvP, mind. I’m recommending based on your comments.

I just realised. With the changes to these ports, preventing us from chaining ports and stopping vertical ports, warrior will be the most mobile class in the game.

It all makes sense now. Well played ANet, well played.

witcher.3197:

Yes, thief’s only meta in Dungeons, PvP and WvW roaming. Anet pls buff this poor profession.

Broaden your mind before you start trying to be witty, even if it is with sarcasm. There is a huge nerf incomming, and it’s to the only thing that makes this class viable: mobility. Every class is designed to excel at something, and each class gets buffed in those areas. Thief gets nerfed in every other area with its area of specialty mostly left alone. Now they are nerfing that area massively, and as far as we know, there will be absolutely no compensation.

Although I’m hoping it won’t be as bad as it seems. If it is, I’ll be beyond angry. Most PvP teams have 2 eles. Eles and engineers are all extremely popular. Mobility is a good way to try and work against these celestial compositions, and yet now it’s getting nerfed. Aside from the fact that my favourite class - actually the only one I enjoy playing because of the fast mobile style - is getting nerfed yet again, I’m losing interest with the way this meta is, and seems to be going further into.

Arikyali.5804:

Alright. Next question:

I’ve been going straight for mid right at roundstart, for the sake of ease and better ganks. But I know some thieves will choose to go far at the start, and can turn the points to their team’s favor. What is the deciding factor?

Your team’s composition and the enemy team’s composition. Anticipate where they will go, and go to where you will be most effective. Typically with regard to the far push, it’ll either be two high damage builds like thief and mesmer, or a sustain build that can 1v1 anything like a D/D elementalist. In some cases you might send a thief with a sustain class to land a burst and leave to ensure the point is won or to force someone from another point to have to run far and help their team mate. It’s all situational and composition dependent.

Often you’ll see a thief and another stealth up and wait somewhere, for the enemy to make their moves first before deciding what to do. On a map like forest, you might even stealth out the side and go through the middle to spike down a squishy enemy player before they even make it to mid, or delay the enemy’s far push allowing your close point capturer to get the full cap before it gets contested.

It depends on the type of thief you are.

Generally, thieves decap the enemy captured points to force them to have to run back and re-capture it, or lose out on those points. When doing this, there’s not really any solid way to explain when you should and shouldn’t pull off. It’s extremely situational, but in general you want to pull off for a decap when you know your team can at least sustain themselves on the other point(s), and stay when you are needed more to take down or finish an enemy player such as a berserker, lich, or condi necro who would otherwise be left to free-cast, stay to cleave a downed enemy to ensure they don’t get back up, stay when you know the enemy has a respawner who is likely to anticipate a decap and stop you before you can complete it, and sometimes stay just to win a fight that much faster in order to secure an important cap or cause the enemy respawns to stagger.

The same mobility that lets you decap can also work against the enemy decap attempts. This means if you see them going for a decap you can defend your point, at least long enough for a slower team mate to arrive to bunker it.

Again, generally, thieves act as +1s to fights because of their mobility. If you’re not traveling the map a lot, you’re either not doing it right or are being forced to compensate for a lacking by someone else on your team. Typically though, it’s acro+trickery S/D that is able to compensate, and critical strikes D/P that would -try- to carry.

When +1ing, the thief usually goes for their targets (mobile/squishy players) first to relieve pressure on their allies. After that, they go for the target that is closest to dying - this can mean low health or the squishiest of the bunch - either way, if the target you know you can finish fastest is not the current team target, call the new target. In an ideal scenario, you’d be coordinating your burst with another high damage team mate such as a shatter mesmer, pewpew ranger, offguard, or fresh air elementalist. D/P thieves tend to look for an opening, land some damage, then get out to lose the target on their head - getting out can just be stealth, kitten, then get back in, or pulling off. S/D tends to just dive in, and keep pressuring relentlessly until they need to pull off or recover some initiative or endurance.

How to know when and what to +1? Any point that needs help. Help doesn’t mean "going to lose," it can be a stalemate or a situation that your team mate can win, but can be sped up by you landing a quick burst and getting back to where you were before. Abuse that mobility for such things.

Moving around a lot also makes you the eyes for your team. Call every enemy movement you see. If you can’t see any, go and look at their respawn area to see where they go next.

When does what type of theif you are come into play? All the time. It determines just how much mobility you have and thus whether you’re fast enough to make some plays. It determines how useful you’ll be in a particular fight. For example, if you’re S/D and you don’t need to avoid much damage, you wouldn’t be as effective as D/P because you have less offense and more defense. On the other hand, if you’re needing to avoid damage while pressuring, D/P will spend more time out of the fight than S/D, and D/P’s mobility can be reduced by other players obstructing your shadow step projectile.

Assuming it happened exactly as you say, then the ele was using dagger offhand, which means we can already determine he was not using the standard fresh air build. If that is the case, then we can accept the possibility that he is not using the same traits as that build. A few tweaks can easily squeeze out an extra 1k from a lightning strike.

With ride the lightning, arcane blast, arcane wave, lightning strike (skill), lightning strike (sigil), flame blast (sigil), and lightning flash, all of which (excluding RTL) are instant casts, it is not hard to imagine 17-18k burst in a few seconds. I’d estimate those skills putting out 12-14k using 0/6/0/4/4. With say, scholar runes, stone splinters, and maybe some points into fire, you’re looking at 16-19k.

http://lmgtfy.com/?q=gw2+meta+builds

In a duel? As in where the only purpose is to fight 1v1 and win? Against a good D/D ele it’s a matter of whoever messes up first, but your mobility with infil return should allow you to reset the fight if you need to. Against an offguard you just have to get the first engage, anticipate and dodge their condi cleanse, land every steal, and ideally your mace skull cracks. Keep count on their teleport meditation too; you’ll want withdraw, a infiltrator’s return, or steal ready. Steal isn’t reliable for it though because they’ll often blind when they teleport. To land steals, be very wary of their blinds. Try to use it to interrupt a hammer/gs skill. If you miss a steal - especially right at the start of the fight - reset by kiting or using refuge.

It’s not "bad," but it’s certainly a noticeable difference, and being further to travel the possibility of getting lag is broader. You’ll likely get hit mid dodge a lot. If you have say 150ms ping, you’ll have a 300ms shorter window to dodge stuff.

I’ve often wondered if removing hot join would make queue times shorter and result in not meeting the same few people over and over as if you are the only ones queueing.

+1

It’s not like anyone PvPs in hot join anyway. The first team to 10 points wins, because after that it becomes a match to join the winning side via auto-balance rather than to reach 500 points first.

What. The. What.

Shout heal war isn’t overpowered. It’s good if you want to support as a warrior, but ele, engi and guardian are all better in that role. The only reasons for it are a) you want to support as a warrior, b) you want to have support potential but find ele/engi too difficult, c) your team needs much less support than a guardian can give and a bit more damage than a guardian can give. It’s basically the celestial guardian. It’s easier to kill, offers no blocks but retains condi cleanse and low easy-mode supportive healing, hits a bit harder but those skills are easy to dodge, has a bit more mobility with a leap, can stack a bit more might, can tank more hits but mitigate less damage, and it has a highly interruptible rally banner. Eles and engis still deal more damage, offer stronger support, have more mobility, have more utility, and are more likely to win a 1v1.

MithranArkanere.8957:

Yeah. I know. That’s why I say “more things should be turned into that kind of ‘trap’”.

Ha. A combination of your use of “should” and my skim-reading. Gg. Apologies =P.

The issue is that each class has something they excel at, and for thief it’s mobility. First, take away or reduce that mobility, and compensations are needed such as increased shadow step range, reduced initiative costs, or buffs to other areas such as sustain, offense and support. Based on previous changes, when it comes to thief, there are never compensations. Second, the reason players love their classes are because of those areas they excel at, so if you enjoy thief because of the speed and mobility, you’re going to be unhappy with the changes no matter what – if any – improvements are made in return.

However, with a new class and new specialisations which of course means new traits, skills, weapons etc., a balance change is necessary. Perhaps such changes are to make way for something new and better, that would otherwise be overpowered, and to set a clear difference between something a class has and what its specialisation has.

Still, it’s hard to be optimistic when it’s a class that is constantly nerfed, and receives “improvements” that are so good that no one uses them (even after those “improvements” have been buffed a few times since their addition).

MithranArkanere.8957:

So they should not ignore anything that can’t be evaded or blocked like Static Field (conversely, static field should not affect someone who exists it with a mesmer portal).

Also, more things should be turned into that kind of ‘trap’. For example, Ring of Warding should also knock down someone who shadow steps in an out of it.

Shadow stepping in, out or through static field or ring of warding DO cause the cc affect to be applied to you.

Know where he is before either of you attack.
Use line of sight while he is invulnerable or you had to recover from something.
Dodge the pet’s immobilise, knockdown and fear.
Try and get the engage, and force him to use signet of stone early.
Dodge their melee skills and longbow 4.

It’s just something you need to be determined to do, and stick to it. I used to click my utilities, and just changing from that took time and practice. You will feel the urge to fall back into the old style when pressured or wanting to be sure you’ll succeed in what you’re trying to do, but you shouldn’t, because that’s the best learning experience. Muscle memory plays a significant part in reacting quickly when pinned down, and they won’t remember if they don’t first do. You will also find yourself checking your hand’s position on the keyboard perhaps frequently. That’s normal, just like you look at your skill bar when you’re playing a new class, and eventually you’ll just realise that you stopped doing it.

The key to good key binds is having everything within comfortable reachable distance of wherever your hand’s normal position is, and can be pressed while moving in any direction. For this reason, moving your movement keys down is not a good idea - especially if you’re keeping the default profession skill keys. Ideally, the most commonly used keys will be over 3 rows and your mouse. Also if you try other people’s key binds, remember not all keyboards have the same key sizes and spacing. For example, some have F1 reachable right above 1, while others have F1 above 2 with a gap in-between.

Here are mine:
Forward: W
Left strafe: Q
Right strafe: E
Backward: S
Jump: Space
Dodge: Mouse button right next to the scroll wheel
Auto run: G
Stow weapon: C
Swap weapon: `
Weapon 1: 1
Weapon 2: 2
Weapon 3: 3
Weapon 4: 4
Weapon 5: D
Heal: A
Utility 1: Left side mouse button
Utility 2: X
Utility 3: Right side mouse button
Elite: V
Profession skill 1: R
Interact: F
Target: T
Call target: Caps lock
Next target: Middle mouse button (scroll wheel)
About face: Tab
Look behind: LCtrl

Hit them with a projectile and go to where it landed. Sometimes it’s close enough to the position you see them on your screen and you can interact to stomp, other times you just have to cleave.

Seems I’ve traded up my ping spikes for packet loss! Yay!

Location: England
Connecting to: EU (/ip 206.127.159.198)
ISP: Virginmedia

Ping test to Frankfurt, Germany:
http://www.pingtest.net/result/117222682.png

Ping plotter to 206.127.159.198 for ~30mins:
http://s15.postimg.org/nk8u6kwdn/Capture.png

The packet loss is new, and even though it is showing on other servers as well as NCSoft, when I plot to another IP I get no packet loss, hence the ping test result to Frankfurt using pingtest.net. Either way, significant loss is still occurring on NCSoft servers.

Admiral Stukov.8791:

I am now contemplating whether I should transfer back to NA servers. The lag wasn’t this bad when I first transfer to EU 6 months ago.

I’m in the same boat. I transferred to EU (I live in England) from NA 7 months ago, and everything was great until the end of August, for 2-3 months, and since then it’s been on/off. From what I can tell, there is major packet loss occurring on the last 3 hops (NCSoft servers). It’s risky to transfer, because not only might there be the same problem on NA servers (some people in EU are fine, others get these spikes), but there’s also the issue of routing – longer distance for data to travel means more opportunities for something to go wrong along the way.

A friend had the same problem as me, starting at the same time. We both live in the UK. Then, he would lag and I would be fine, and now I’m lagging and he’s fine. When this happens, even when we route through the same server, one has lag spikes and the other does not. We were trying to form a competitive PvP team, but it’s near impossible when someone is always unable to play. The past 5 days it’s been me. It’s maddening.