[State of the Thief] End of Dragons Thief Balance Summary and Discussion

[saerni.2584]

In light of the upcoming release of End of Dragons and the next Thief Elite Specialization (Specter) it is a good time for a holistic top-to-bottom discussion of the Thief. While this won't cover every trait and skill I'm going to be running through a significant number at the start. If you think I missed something please post it and I will endeavor to include it as best I can.

To that end, I will be (attempting) to monitor and update the thread's Original Post(s)™ to reflect the ideas posted here and elsewhere in the Thief Forum. Full disclosure, I have played Thief since roughly the launch of Guild Wars 2, and have been a contributor on these forums (older and current) for many years, and help to run the Thief Forum Guild, of which several posters here are members. My goal is not to solely advance my, or anyone else's, agenda (buff Thief!) but only to ensure the Developers are fully aware of the perspectives of Thief players and have the opportunity to consider our balance suggestions going into EoD. I encourage the posters in this thread to be specific in your feedback (helps me summarize accurately), measured in your suggestions, and to justify your reasoning (helps make a point more persuasive). I certainly do not know enough (or have the time) to discuss every single possible issue/skill so I appreciate additional/differing feedback. Please be patient as it took me a long time to write down just my own ideas for this. Summaries of everything recently posted will be a project measured in weeks and not days.
 

I will be posting proposed changes in the format:
 

“Suggestion # Justification #” where the numbers match between the suggestion being proposed and the justification. While I have made a number of suggestions under the number “1” this does not mean all # “1” ideas are mine (for example if someone else posts on something I did not include). It likewise does not mean that there is any relation between number 2, 3, 4, etc. The only relation is within that specific individual balance suggestion.

Edited November 1, 2021 by saerni.2584

[saerni.2584]

As this is the “Pre-EoD” State of the Thief, I'll start by summarizing the new Specter Elite Specialization and related feedback first. Specter represents a unique attempt by the Developers to make a single target oriented support Elite Specialization that also has the capability to be a “bruiser” duelist with Shadow Shroud, Barrier, and healing potential all making the Specter much harder to kill without as much reliance on Stealth or Evade.

The Developers mentioned during the Guild Chat that their vision for Specter is one in which the trade-off of base Initiative for Shadow Shroud means that the Specter is more reliant on Initiative regenerating utilities and traits, especially in the Shadow Arts and Acrobatics Specializations, respectively. As a result, it could be said that unless a build takes a utility oriented approach to supplement initiative regeneration, the choice for many players will be between Shadow Arts-Trickery-Specter and Acrobatics-Trickery-Specter. This dynamic makes the discussion of those two Specialization traitlines (see below) especially relevant both for considering Core Thief balance but especially for the forthcoming Specter Elite Specialization. However, because Specter lacks innate Initiative regeneration the Specter heavily benefits from both traits that maximize its Shadow Force generation (allowing the Specter a reprieve to regenerate Initiative while in Shadow Shroud) and also from Alacrity (which indirectly boosts Initiative regeneration by lower cooldowns on Siphon and Utilities that can restore Initiative). 

[WORK IN PROGRESS: BETA FEEDBACK SUMMARY]

FORMAT: Description, Issue #, Solution #.#, Explanation #.# (First number matches between issues and solutions, second number is for a particular proposal/explanation match). 

As a single target support Elite Specialization a number of testers during the beta raised the general issue that the "Target Ally" hotkeys are somewhat difficult to use properly. This is a general issue for Ally Targeting in a game which never needed this before to the same extent. While there are a number of various proposals out there to address this issue, I will suggest that the easiest and most straightforward option is to create a personal "call target on ally" targeting option (the target marker would be blue and personal rather than for the group) and a second hotkey for select allied target).

Siphon [Steal]. Siphon replaces Steal with a ranged ability that can either be used on an enemy or on an ally. In "enemy" mode ("Enemy Siphon"), Siphon applies Slow, grants substantial Shadow Force, and interacts with the traditional Thief traits that enhance Steal. In "ally" mode ("Ally Siphon"), Siphon applies barrier and "tethers" the Specter to that particular ally for the purpose of Shroud skills. 

Issues/Bugs: (1) Ally Siphon fails to trigger "on Steal" traits for their positive benefits: Thrill of the Crime (Boons), Bountiful Theft (Boon), Mug (Heal) , Improvisation (Recharge-), Hidden Thief (Stealth), Kleptomaniac (Initiative+). (2) Ally Siphon does not give any Shadow Force. (3) Siphon requires a "path to target" and therefore both the Ally Siphon and Enemy Siphon may simply fail to activate. This is notable on major map "metas" where there are single target nodes (Octovine, Dragonfall, Drizzlewood) because often those nodes are "obstructed" and cannot be targeted with Siphon (also affects Deadeye's Mark). (4) Siphon has a different cooldown in different game modes. Going between modes this means the core game play mechanic plays very differently. (5) Siphon activates Hidden Thief on a delay which makes the Specter unable to smoothly activate a Stealth Attack for 0.5-1 seconds after Siphoning a target. 

Solutions/Proposals: (1.1) Make a new F3 Ability "Shadow Tether" that provides the Ally Siphon functionality (but on an independent cooldown). (2.1) [See Solution 1.1] Shadow Tether should provide a small amount of Shadow Force, maybe 10, to provide support functionality base synergy with Shroud. (3.1) Path-to-Target and Obstructed issues should both be addressed for historical metas if at all possible. EoD metas and events should be explicitly tested for these interactions in internal testing to ensure Specters can use their core mechanics properly. (4.1) Normalize the cooldown of Siphon across game modes and adjust the amount of Shadow Force on Siphon as needed to split between game modes. (5.1) Adjust Siphon to trigger Stealth from Hidden Thief to coincide with the initial activation of other traits. 

Explanations: (1.1) The primary goal of Ally Siphon appears to be to provide a way to tether the Specter to a particular ally for the purpose of Shroud Skills. However, because this shares a slot with the F1 this creates confusion about the applicability of certain traits and also makes the Specter choose between Siphoning off an Enemy (with the inherent bonus Shadow Force) and Siphoning an ally for support. A new ability eliminates the trait interactions issue entirely and puts the focus squarely on "marking" the ally for assistance from Shroud. (2.1) Teleports can provide significant Shadow Force through Traversing Darkness but there is otherwise a lack of synergy between supporting an Ally and gaining the ability to continue that support in Shroud mode. This change may require reductions in Shadow Force gain elsewhere but would be a net positive improvement to support based gameplay. (3.1) Deadeye's Mark today still has issues in HoT metas (Octovine) through to Icebrood Saga. This issue/bug/flaw is a fairly significant frustration that should be avoided as much as possible going into EoD for Specter. (4.1) Splitting between game modes should avoid cooldowns as much as possible and focus instead on the effects of those skills to make the Specter "play" the same between modes. Otherwise it is hard for players to adjust to timings and variable gameplay flow when playing across modes. (5.1) This will fix the issues related to activation delay making the gameplay around this trait less smooth or feeling "clunky." 

Enter/Exit Shadow Shroud. This in some sense replaces Stolen Skills with a skill that allows entry into Shadow Shroud as long as there is sufficient Shadow Force available to the Specter. 

Issues/Bugs: (1) Enter Shadow Shroud has a relatively low cooldown of 8 seconds in PvE, WvW/PvP, but Shroud Skills have a longer cooldown in WvW/PvP resulting in entering into Shadow Shroud potentially 4 seconds or more before multiple skills are available to use. (2) Enter Shadow Shroud randomly chooses a nearby ally for the purposes of Shadow Shroud's supportive role. However, because the tether disappears when the ally leaves the 1200 range of Siphon the risk that the randomly tethered individual will leave the radius is high, but dependent somewhat on chance. (3) If a tethered ally leaves the 1200 range while the Specter is in Shadow Shroud it breaks the tether and no new tether target is selected. 

Solutions/Proposals: (1.1) Enter Shadow Shroud should have a cooldown of 12 seconds in WvW/PvP. (2.1) Enter Shadow Shroud should auto-tether to the nearest ally. (3.1) Tether should stay on target up to 1800 range but only be effective within 1200 range. 

Explanations: (1.1) Shadow Shroud skills already have a longer cooldown than in PvE. In particular Dawn's Repose and Eternal Night have 8 second cooldowns in PvE and 12 second cooldowns in WvW/PvP. The Enter Shadow Shroud skill should likewise be split to match the longer skill cooldowns in WvW/PvP so the cooldowns match between the Shadow Shroud skills and the gameplay flow of entering into Shadow Shroud. (2.1) This makes a faulty interaction by the ally popping out of 1200 range less of a potential issue. (3.1) Making the tether active at up to 1800 but only effective when inside 1200 prevents accidental loss of the tether on an intended target while still requiring the Specter to be aware of and maintain under 1200 distance from their tethered ally. 

 

Shadow Shroud. Conceptually, Shadow Shroud provides crowd control, Stability, Healing, Barrier, and Cleanse for a tethered ally while also being tied to successfully striking enemies with Shroud abilities, which inflict substantial condition based damage as well. Importantly, because Shroud skills do not cost Initiative one clear benefit of Shroud is a respite from the Initiative costs of weapon skills. Cooldowns are the only limitation of a series of niche skills that each fulfill a specific offensive and supportive role.

Haunt Shot [Shroud 1]. Haunt Shot provides Torment, with both self-Might and Might and Healing to a Specter's Tethered Ally. 

Issues/Bugs: (1) Haunt Shot, unlike other Thief weapon skills, does not have an alternative mode when launched from Stealth. (2) Haunt Shot is not a projectile finisher so it lacks Field interactions. (3) A tester noted that Haunt Shot does not auto-attack and has to be pressed constantly to fire repeatedly.

Solutions/Proposals: (1.1) When in Stealth Haunt Shot should be replaced with "Haunting Barrage." Haunting Barrage would be a 1 second duration, 3 projectile barrage of the basic Haunt Shot effect, with a projectile finisher effect. (2.1) Haunt Shot should be a projectile finisher. (3.1) Allow Haunt Shot to auto-cast when in range.

Explanations: (1.1) Not having a Stealth Attack in Shadow Shroud feels thematically off-putting especially because Grasping Shadows is a blast finisher and can therefore grant the Specter Stealth while in Shadow Shroud, but gains no special attack for doing so. Haunt Shot is normally 0.5 seconds per shot. By making a 1 second duration for 3 shots the effect will be improved but not so much as to encourage a large spike of damage out of Stealth. (2.1) Projectile finishers make sense to include for Shadow Shroud's auto-attack. (3.1) Quality of life improvement to usability of the basic Shadow Shroud attack.

Grasping Shadows [Shroud 2]. This skill provides AoE damage, Cripple, Torment with Healing and Cleanse for the Specter's Tethered Ally, and a blast finisher on a 3 second cooldown in all game modes. However, this skill has a 180 radius and the skill has a 1 second delay before activating after being placed. 

Issues/Bugs: (1) Grasping Shadows has a low radius and a delay on activation making it harder to actually use effectively against even a single moving target. 

Solutions/Proposals: (1.1) Decrease the 1 second activation delay to 0.5 seconds. Increase the radius of the skill to 240. Lower the maximum targets to 3. Increase the cooldown in WvW/PvP to 6 seconds. 

Explanations: (1.1) The activation delay and radius make the skill less usable as intended. Increasing the radius and lowering the delay help to make the skill more user friendly. However, to offset the easier usage the maximum targets should be adjusted downward and the cooldown increased (at least in WvW/PvP). 

Dawn's Repose [Shroud 3]. Dawn's Repose is a direction based 600 range dash skill that inflicts Fear on the targets in a 120 radius around the end point. This skill Heals and grants Barrier to the Specter's Tethered Ally with additional bonus Heal/Barrier for each target Feared, up to 5 targets.

Issues/Bugs: (1) Dawn's Repose is difficult to accurately control because the Specter first needs to be at a location suitable to enter the Dawn's Repose dash and end up striking the intended target (which may also be moving). Basically, Dawn's Repose contributes heavily to the sense that Shadow Shroud is best used solely as a healing battery using Consume Shadows. 

Solutions/Proposals: (1.1) Dawn's Repose should be reimplemented instead as a targeted 600 maximum range Dash to ground target. Maximum Barrier should be reduced by 50% in WvW/PvP. Add a 0.5 second pre-Dash animation to telegraph the skill. 

Explanations: (1.1) Making it possible to control where within 600 range the Specter will end up allows the Specter to use Dawn's Repose effectively. The low radius should remain given the powerful nature of an AoE Fear effect and the associated Heal/Barrier that can result. The easier usage means the total Barrier potential should be reduced for competitive play. Adding an animation pre-Dash would telegraph the skill for competitive game-modes and still leave the skill effective (comparatively much more effective than the current version).  

Eternal Night [Shroud 4]. Eternal Night is the only melee attack on Shadow Shroud, providing strong damage, Chill, Weakness, and Poison, as well as Healing both the Specter and a Tethered Ally. This skill fits into a niche that the Shadow Shroud would otherwise lack and provides an opportunity for a follow-up strike after a successful Dawn's Repose puts the Specter into melee range. Some testers felt the name "Eternal Night" was confusing because it is similar to "Endless Night." This issue, however, isn't much of an actual problem because most players will probably just say "Endless Night" and "Shroud 4." 

Issues/Bugs: (1) Eternal Night has no obvious range indicator and at 130 range requires being very close to the target for a 1 second cast time. Testers have reported some difficulty in landing this ability and feeling like there is not a sufficient "tell" for when the skill is in-range.

Solutions/Proposals: (1.1) Eternal Night should have slightly improved range of 150.

Explanations: (1.1) Eternal Night has a longer cast time and uniquely requires melee range. This feels fine in many cases, but the lower range makes it very easy to think you are in range when you are in fact just outside of it. 150 is only slightly larger but would match better with the size of the animation and produce fewer "wasn't I in range?" moments. 

Mind Shock [Shroud 5]. Mind Shock is a slow charging AoE Stun that also provides Stability to the Specter and a Tethered Ally with longer durations based on the enemies struck. 

Issues/Bugs: (1) The ability has both a base cast time of 0.5 seconds and a charging time of 3 seconds. This makes it both interruptible and difficult to land. However, because it can be activated and continues to charge even after leaving Shadow Shroud Mind Shock is more flexible than some other skills. 

Solutions/Proposals: (1.1) Mind Shock, because it doesn't have an effect until after a 3 second charge period should be instant cast for a reduction in total cast time of 0.5 seconds. 

Explanations: (1.1) Mind Shock, which is already balanced around reduced damage in WvW/PvP has a strong effect but feels too slow to use effectively. Given the strong effect the charge time should still be significant but with that long charge time there is little sense in including a cast time on top of that. It simply feels better to hit the button and see the charge immediately start building up. 

 

Scepter.  Scepter is a dual support and condition oriented weapon depending on whether an ally or enemy is the target. While the support aspect of Specter is also split between traits and utilities, the Scepter is a primary vehicle for Specter's support ability.

Issues/Bugs: (1) None of the Scepter attacks are coded as projectiles for the purposes of Combo Finishers. (2) The "Action Camera" mode does not permit a number of attacks from triggering. (3) Only Shadowsquall will Reveal a Specter in Stealth when supporting an Ally. 

Solutions/Proposals: (1.1) Make the appropriate Scepter attacks into Projectile Finishers. (2.1) Fix the Action Camera issues by addressing the need for a target in that camera mode similar to other projectile weapon skills. (3.1) All support skills should remove Stealth in a similar manner to Shadowsquall when used in Stealth. 

Explanations: (1.1) Projectile attacks having the unique property of interactions with Fields is integral to the overall games design. Specter's Wells are Dark Fields as well and the Scepter should be able to combo with those Dark Fields, and other Fields, like other projectile weapons. (2.1) Functionality in Action Camera is important to a number of players. (3.1) This would make the interaction between support skills and Stealth consistent and ensure that a Specter is unable to permanently Stealth while supporting an Ally. This is especially important in competitive modes.

Shadow Bolt/Double Bolt/Triple Bolt [Scepter 1]. This attack chain has a smaller amount of damage and applies Torment on enemies and Barrier on allies. This attack chain has strong synergy with Dark Sentry when used to apply Barrier due to the application of Rot Wallow Venom with each Barrier applied, giving the Specter a constant ability to do damage while focused on supporting an individual ally. 

Issues/Bugs: (1) On Human (possible racial animation bug) Triple Bolt triggers twice before the chain repeats. This same bug may not occur on Asura (not verified). (2) The sound effects attached to Shadow Bolt/Double Bolt/Triple Bolt are quite loud and can becoming grating overtime.

Shadowsquall [Scepter Stealth Attack]. Shadowsquall applies significant damage and Poison to enemy targets over the course of the relatively long period of 2.5 seconds. When used on an ally this Stealth Attack Heals and applies Regeneration. This skill provides a powerful cover condition and potential source of damage in addition to the Torment application other Specter skills and traits focus on. While several testers thought it was thematically strange to apply Poison and not Torment, they did not articulate a reason applying Poison was "bad." 

Other testers noted that there are no "support" versions of the other Thief weapon "Stealth Attacks." Of course, while Deadeye changes Stealth Attacks and the Stolen Skills the Daredevil maintains the Stealth Attacks and Stolen Skills of the base Thief Profession. However, because there is room for unique styles of "Support Stealth Attacks" it was requested that unique support versions be implemented for the Stealth Attacks on Pistol, Dagger, Sword, and Short Bow. This would improve the diversity of weapons usable by Specter without changing the playstyle of being able to target Allies when out of Shroud, while limiting that to only the Stealth Attacks to retain Scepters role as the best direct support option. 

Issues/Bugs: (1) Shadowsquall is particularly slow compared to other Stealth Attacks. However, it is also unique in terms of the number of projectiles involved, and therefore should have a somewhat longer cast time. That said, the cast time per projectile ends up at an awkward 0.313 seconds. (2) When used on an Ally, Shadowsquall heals for over 500 per hit for a total Heal of over 4000 without any Healing Power. (3) Shadowsquall can fail to trigger properly when Shadow Bolt/Double Bolt/Triple Bolt are active and break Stealth before Shadowsquall queues.

Solutions/Proposals: (1.1) Shadowsquall's channel time should be reduced to 0.25 seconds per hit, for a total time of 2 seconds. (2.1) Shadowsquall should have a base Healing value reduced to 370 per hit and the Healing Power coefficient should be increased from 0.11 to 0.3. (3.1) Reduce the aftercast on Triple Bolt to allow Shadowsquall to queue faster.

Explanations: (1.1) Shadowsquall isn't a large enough source of damage or sustain that lowering the channel time by 0.5 seconds will present a balance issue. This change would make the skill feel better to use from the Specter's perspective and in WvW/PvP against (because the longer a skill channels damage the less there is to dodge at any given point). (2.1) At 800 Healing Power Shadowsquall would Heal for the same amount but it would lose roughly 30% of its maximum Healing potential at 0 Healing Power. Shadowsquall would gain Healing potential above 800 Healing Power. (3.1) I believe issues with other projectile attacks breaking Stealth before a Stealth Attack can queue have been previously addressed in a similar manner. 

Shadow Sap [Scepter 2]. Shadow Sap applies Weakness and self-Might on enemies and Barrier and Might to an ally. In one sense, Shadow Sap fills a similar niche to Weakening Charge by giving the Specter the ability to heavily mitigate against Strike Damage. A number of testers strongly felt that Shadow Sap is a bad skill and not workable. However, it was also pointed out that Superior Rune of Aristocracy turns Shadow Sap into a Might generation powerhouse (10 stacks of Might per use in PvE and 9 stacks in WvW/PvP on a 1 second internal cooldown for the Rune bonus effect). There is also the significant boost to sustain that permanent Weakness offers. For those two reasons further testing is likely required before writing off Shadow Sap as "useless" or in need of a rework. Note: Runes of Aristocracy may need to be nerfed as this interaction is likely not intended. 

Triple Threat [Scepter/No Offhand 3].  Triple Threat applies damage and Torment or Barrier to an ally.

Issues/Bugs: (1) Triple Threat only fires 1 projectile instead of the indicated 3. (2) Triple Threat is basically Triple Bolt with an extra second of Torment. 

Solutions/Proposals: (1.1) Fix the bug. (2.1) Triple Threat should have base Torment duration of 8 seconds in line with Twilight Combo (8 seconds, 3 stacks) and Endless Night (6 seconds, up to 7 stacks).

Explanations: (1.1) Fixing bugs is good. (2.1) Triple Thread shouldn't have only 1 second more Torment duration than an auto-attack for 4 Initiative. As a "no-offhand" weapon skill the skill should still have effects appropriate to the costs. 

Twilight Combo [Scepter/Dagger 3]. Twilight Combo provides Chill, Poison, and Torment on enemy targets and Barrier, Swiftness, and Healing on a target ally. Twilight Combo is actually a slow moving projectile and a double-projectile follow-up, which accelerates the original projectile to high speed before striking a target. 

Issues/Bugs: (1) It is possible to Stow/Interrupt the second projectile from being launched. Jumping in certain ways may also result in a failure to fire the second projectile properly. This leaves a slow moving projectile that travels the full 900 distance or until it hits a target. (2) It is not possible to fire this skill when at greater than 900 range. Measured Shot/Endless Night and Dancing Dagger/Head Shot all allow activation at greater than 900 range (but anything that cannot hit shows as missing with "out of range"). (3) Twilight Combo can provide the Specter with Swiftness if used at close range on an Ally. 

Solutions/Proposals: (1.1) Reduce the delay between the projectiles being launched to reduce the timing during which the attack can be interrupted. (2.1) Enable firing this skill at greater than the maximum range. (3.1) While this is noted as needing a bug fix, a number of Thieves have suggested allowing the Specter to also gain Swiftness from Twilight Combo when used on an Ally. 

Explanations: (1.1) Making it harder to interrupt the fast projectiles helps keep the attack behaving consistently as a fast attack rather than a painfully slow lingering projectile. (2.1) Twilight Combo should function the same as other Thief ranged weapons (and weapons across Professions). (3.1) The Specter boosting an ally with Twilight Combo can struggle to keep up with the boosted ally if the ally is moving and the Specter is trying to chase after them. Unlike with offhand Pistol, there is no teleport towards the ally to enable staying in range. Giving Swiftness to both the Ally and Specter makes sense to help the Specter fulfill the support role when wielding an offhand Dagger. 

Measured Shot [Scepter/Pistol 3]. Measured Shot teleports the Specter up to 300 distance away from enemies (up to 900 distance) and towards an ally (80 minimum distance). 

Issues/Bugs: (1) Against enemies, Measured Shot can teleport you from close to 900 range to beyond 900 range. This puts Endless Night out of range. Further, because enemies tend to move, and it is natural to use Measured Shot at variable ranges to gain access to Endless Night, Measured Shot has a tendency to move the Specter to where it is necessary to move just to get back into range. Likewise, allies in need of support tend to be in danger and the Specter currently effectively lands on top of the ally when using Measured Shot. (2) Measured Shot when used within 80 distance on an ally teleports the Specter 900 distance from the ally instead of the intended effect. (3) Measured Shot's tooltip incorrectly refers to the "Maximum" distance from Ally. (4) Some testers report that Measured Shot can hit an unintended target and that this makes the skill hard to use. 

Solutions/Proposals: (1.1) Measured Shot should have a 600 maximum distance from enemies and a 300 range for an ally. (2.1) Requires a bug fix for usage cases inside the minimum range. (3.1) Change tooltip to refer to "Minimum Distance from Ally." (4.1) 

Explanations: (1.1) If Measured Shot had a maximum distance of 600 it would not tend to put the Specter into the position of struggling to stay in range. Likewise, making the minimum distance for an ally target at 300 would make the Specter able to stay outside of dangerous AoE's while still being close enough to ensure support skills will land properly.

Endless Night [Scepter/Pistol 3 Flip].  Endless Night provides Quickness, Regeneration and Barrier in a line to up to 3 allies and applies Slow and Torment to enemies. Endless Night is made especially useful by effectively piercing and applying to up to 3 targets at once, rewarding skill and group coordination."

Issues/Bugs: (1) Endless Night prioritizes NPC pets/minions/summons to the detriment of players in the line. This makes supporting the primary targets, already somewhat difficult with a thinner line based application, needlessly more difficult.

Solutions/Proposals: (1.1) Prioritize the targets to first boost players up to the maximum targets, then boost NPCs if there are any in the path.

Explanations: (1.1) Given the limited number of targets possible this skill needs to focus on players or it will be overshadowed by support Profession builds that have AoE boon support for a larger number of targets (easier to target and less worry about pets/minions/summons). 

 

Wells. Wells act both as a Shadowstep to target as well as applying damage, boons, conditions, or crowd control depending on the Well used. Unlike other Wells these skills place the Specter directly in harms way making them into aggressive mobility skills. All Specter Wells are Dark Fields for the purposes of Combo Finisher effects and affect up to 5 targets.

Issues/Bugs: (1) Testers are reporting that the "Snap Ground to Target" does not work with any of the Wells. (2) Wells lack underwater versions. (3) Well's cast times initiate before the teleport and the "Well" is laid on the ground.

Solutions/Proposals: (1.1) Fix the bug. (2.1) Make Wells act as underwater port towards target with the AoE effect around a point. (3.1) Wells should instantly teleport and then cast the Well after the cast time expires.

Explanations: (1.1) Squish all bugs! (2.1) Although this may take some resources to implement it seems awkward to not be able to use any of the Specter's utility skills underwater. (3.1) Wells feel strange because Shadowsteps are usually instant. Separating the teleport and the "Well" portion would make the skill feel better to use without changing the functionality, while opening up options for counterplay in competitive modes. 

Well of Gloom [Heal]. Well of Gloom initially self-Heals, and pulses Heal and applies Cripple to enemies, for five pulses. Issues/Bugs: (1) This is the only Well that pre-casts the animation before the Specter teleports. The animation should only activate simultaneously with the Specter arriving at the target location. (2) Tooltip doesn't show interaction with Concealing Restoration. Solutions/Proposals: (1.1) Fix the animation. (2.1) Fix the tooltip. Explanations: (1.1) Appears to be unintended or a bug. (2.1) Oversight that is an easy fix.

Well of Bounty.  Well of Bounty provides 1 Boon per pulse (Regeneration, Protection, Resistance, Swiftness, Vigor, Quickness, Alacrity, Fury, Might, Resolution, Stability, Aegis), in a progression based on the boons the ally already has on them, for five pulses. If a recipient has all Boons already, the Well will apply an additional stack of Might per pulse. 

Issues/Bugs: (1) Some Thieves feel that the Well is underwhelming in terms of total boons applied and some Thieves similarly believe that Well of Bounty should prioritize different Boons because the Well can only natively apply 5 boons over its duration and the more valuable Boons are unlikely to ever be applied. 

Solutions/Proposals: (1.1) Well of Bounty should progressively pulse more Boons per pulse (1, 2, 2, 2, 5). The cooldown should be increased to 30 (from 20 seconds) to reflect the increased effectiveness. 

Explanations: (1.1) This would make Well of Bounty worth standing in for the full duration while also providing a tangible benefit to late comers who only catch the last pulse. The limited boons applied currently are only somewhat justified by having a short cooldown. However, in order to make the skill feel uniquely impactful it is preferable to give up some of that low cooldown (reducible with Alacrity) in exchange for a stronger effect. 

Well of Tears. Well of Tears is a short 20 second cooldown that applies no conditions but has increased Strike Damage compared to other offensive Specter wells. A number of testers indicated that they felt the pure Strike Damage had no place in the condition oriented Specter Elite Specialization. A few testers also suggested that there were several gaps in Specter Wells: chiefly a lack of stun breaks and a lack of cleanse support for Allies. Several of those testers suggested replacing Well of Tears entirely with some combination of those effects or similar. Because the Well of Tears is technically not an issue but presents the most generic profile of the Specter Wells, I am including this feedback under this Well. 

Well of Silence. Well of Silence is a Well that combines Daze with inflicting Chill, Weakness, Slow, and Cripple on foes that are interrupted with any of those conditions the enemy did not have previously.

Well of Sorrow. Well of Sorrow mirrors Well of Silence, for damaging conditions, but without the pulsing Daze. Well of Sorrow applies conditions that the enemies struck did not previous have and otherwise inflicts Torment. 

Shadowfall [Elite Well].  Shadow Fall creates a distortion effect that progressive pulls opponents towards the center of the effect with an increase in Pull distance each consecutive pulse.

Issues/Bugs: (1) Shadowfall has a 75 second cooldown in PvE and a 60 second cooldown in WvW/PvP. Usually cooldowns are higher in WvW/PvP than PvP so this is likely a mistake. (2) Shadowfall's animation is different from other Specter Wells and looks like a leap and not a Shadowstep. (3) Shadowfall can be potentially visually confusing in competitive modes where enemy Specters are using Shadowfall. 

Solutions/Proposals: (1.1) Shadowfall's cooldowns should be normalized at 60 seconds in all game modes. (2.1) Change the teleport animation to match the other Specter Wells. (3.1) Change the visual of Shadowfall when viewed by an Ally to have a Green ring effect. 

Explanations: (1.1) A 60 second cooldown is higher than comparative Elite skills in Daredevil and Deadeye. Shadowfall feels impactful and not too good at 60 seconds. Making the skills consistent in PvE and WvW/PvP should happen whenever possible, which is why a universal 60 seconds is suggested. (2.1) It looks like the animation was copied from Vindicator and there are existing animations that can be used from the other Specter Wells. (3.1) A green ring effect is already in the game and shows that the effect is not dangerous to an Ally. Anet should also consider the larger orange "danger" effect when viewed by an Enemy. 

 

Minor Traits. 

Specter. Specter's unique effect, other than granting access to Shadow Shroud and Siphon, is to reduce the maximum Initiative pool by 3 and grants "Shadow Force" at a rate of 1% per Initiative spent. 

Issues/Bugs: (1) It has been reported (unverified) that some Thief weapon skills are not granting the correct amount of Shadow Force despite having different Initiative costs. (2) Scepter skills do not grant Shadow Force when used successfully on an Ally target, and some testers were confused or felt the support side of these skills should generate Shadow Force. 

Solutions/Proposals: (1.1) Review the tables for Shadow Force granted in PvE/WvW/PvP based on the game mode specific Initiative costs. Correct any discrepancies. Ensure the functionality on a per Initiative basis is working properly. (2.1) Clarify the tooltip to state that Shadow Force is generated on successfully striking enemies with Initiative using skills. 

Explanations: (1.1) Some skills have large Initiative cost differences between PvE and WvW/PvP. This will mean that per skill Shadow Force generation is actually higher in PvE and WvW. (2.1) The design of Specter is such that giving Shadow Force while boosting an Ally may result in unintended excessive Shadow Force generation. For this reason the fix is likely a tooltip clarification rather than a change to the trait itself. 

Dark Sentry. Dark Sentry boosts outgoing Healing by 20% (10% in WvW/PvP), and applies a 10 second duration Rot Wallow Venom charge to allies receiving Barrier from the Specter. Rot Wallow Venom applies one Torment for 4 seconds per charge when striking a target (and one charge is expended when attempting to strike a target regardless of whether the skill successfully hits the target). Dark Sentry is a parallel trait to how Shadow Shroud supports an Ally when the Specter is on the offensive by building offensive damage while directly supporting an Ally. 

Issues/Bugs: (1) A number of Specter testers have noticed that Rot Wallow Venom is fairly easy/generously applied and generates very high damage ticks despite being in "support" mode rather than directly engaging with a target. Shadow Bolt/Double Bolt/Triple Bolt [Scepter 1] have a 3 second Torment duration. This means a Specter can either attack an enemy to get a small amount of damage and Torment or target an Ally hitting the same enemy and get over 500 Barrier for the Ally and 6 seconds of Torment on the opponent (using Strength of Shadows). (2) Rot Wallow Venom may not always be triggering Leeching Venoms correctly. Testing suggests that Rot Wallow Venom does in fact trigger when the source of the Rot Wallow Venom has the trait selected and the user does not. However, a number of testers have reported concerns with the interaction and believe it is not triggering properly.

Solutions/Proposals: (1.1) Rot Wallow Venom should have reduced based Torment duration from 4 seconds to 2 seconds. (2.1) Additional testing is required to ensure the correct Leeching Venom trait interactions are occurring and there are no unintended interactions.  

Explanations: (1.1) Rot Wallow Venom should not make the Scepter autoattack chain fundamentally better damage when used to support an Ally. If the base Torment duration is lowered to 2 seconds then the duration with Strength of Shadow is 3 seconds. Because Shadow Bolt/Double Bolt/Triple also comes with Strike Damage this is then slightly better damage at the expense of less support. (2.1) Leeching Venoms has the potential to improve the sustain support offered by Specter but only if the trait's interactions are properly documented and function correctly. 

Penaku's Ambition. Gives Allies Barrier when first entering Stealth granted by the Specter. Gives the Specter Barrier on using Stealth Attacks. Overall this trait is a solid addition to personal sustain that rewards using Stealth Attacks and does not promote permanent Stealth gameplay.

 

Major Adept Traits. Adept Specter traits present a range of support options from general boosts to Healing, additional single target support, and sacrificing single target support for group Healing. 

Second Opinion. Second Opinion increases Healing Power by 7% of Condition Damage, +60 Healing Power, +60 Healing Power while wielding a Scepter. This trait acts a secondary boost to Healing Power of roughly 180-220 points at maximum (depending on stat selection). This is a solid option for those looking to fit extra Healing Power into a Specter without needing to sacrifice Condition Damage or other stats.

Issues/Bugs: (1) The issue with Second Opinion is that it is overshadowed by Consume Shadows as a support skill because Consume Shadows will Heal more than any increased Healing gained from Second Opinion. There is no solution within Second Opinion itself that will address the issue, however, because there is nothing wrong with an Adept trait that increases Healing Power by around 200 points when wielding a Scepter.

Solutions/Proposals: (1.1) [See Consume Shadows]

Shallow Grave. Shallow Grave protects the Specter and a tethered Ally by sacrificing Shadow Force to Heal either if taking an otherwise lethal blow. 

Issues/Bugs: (1) Shallow Grave has a 60 second cooldown in PvE and a 300 second cooldown in WvW/PvP. This makes the trait essentially unusable in competitive gamemodes. (2) Shallow Grave is an inefficient/redundant option compared to Consume Shadows because the Specter can, using the latter, simply actively trigger the same major Healing when an Ally or the Specter gets lower on Health rather than waiting for a lethal blow (with a bonus of applying Barrier as well). 

Solutions/Proposals: (1.1) Shallow Grave's "percentage Returned to Healing" should be reduced to 33% in WvW/PvP. Shallow Grave's cooldown should be normalized to 60 seconds in all game modes. (2.1) [See Consume Shadows]

Explanations: (1.1) Rather than implementing 300 second cooldowns in competitive modes (which reduces build diversity by eliminating choices from consideration), the overall power of this trait should be reduced as necessary for WvW/PvP to allow for the same cooldown and gameplay flow. (2.1) Thematically, the issue with Shallow Grave being inferior to Consume Shadows has more to do with Consume Shadows being over tuned than a flaw with Shallow Grave. 

Consume Shadows. Consume Shadows converts all remaining Shadow Force when casting Exit Shadow Shroud into Healing, with excess Healing is converted to Barrier, for up to 5 Allies in a 600 radius. 

Issues/Bugs: (1) Consume Shadows transforms the Specter into a full fledged group support Healer. This, although thematically different from the rest of Specter, is a refreshing alternative to the single target support theme. However, because the conversion is based entirely on the "Percentage Returned as Healing" mechanic, there is little reliance on Healing Power. Effectively, Consume Shadows is better at Healing than the increase in Healing Power on Second Opinion and fills the role of preventing lethal blows on Allies better by being an AoE rather than single target, and providing Barrier (prevents one-shots on squishy Allies), and being predictable (Shallow Grave may suddenly trigger but Consume Shadows is entirely within the Specter's control). 

Solutions/Proposals: (1.1) Consume Shadows should return 50% of Shadow Force as Healing (25% in WvW/PvP) with 600 radius to Allies. Consume Shadows should apply a set amount of Barrier that scales based on Healing Power. 

Explanations: (1.1) Consume Shadows eschews single target support by encouraging the Specter to spend as little time in Shroud as possible, and focused entirely on offensive skills to generate more Shadow Force to rapidly re-Consume Shadows to support Allies. Second Opinion and Shallow Grave are single target support traits oriented towards using Shroud's tether effects and Scepter's single target support game play (the choice between the three depends on how much the Specter wants to "babysit" a tethered Ally). By reducing Consume Shadows' potential single target Healing and making the trait more reliant on Healing Power to generate large amounts of Barrier the other two traits are not overshadowed (currently: why take either Second Opinion or Shallow Grave if you will get greater Healing output and prevent more down-states by simply taking Consume Shadows). 

Major Master Traits. Master Specter traits offer a range of options for increasing Shadow Force generation: between Shadow Force from Torment, increased Shadow Force from Siphon, and Shadow Force from AoE Healing from Shadowsteps.

Larcenous Torment. Larcenous Torment increases Shadow Force by 0.5% per application of Torment on enemies, while also improving Torment damage by 20%. This trait is a strong option for Specters looking for the highest possible damage output and synergizing their damage output with Shadow Force generation.

Amplified Siphoning. This trait improves Siphon to grant 10 additional Shadow Force (over the 25% base value) and improves direct barrier applications by 100%. This trait is a strong solo-support trait built around sustaining Allies. 

Traversing Darkness. This trait Heals Allies and grants Shadow Force to the Specter when Shadowsteping (5% base per Shadowstep and 1% per Ally Healed), while also making all Specter Wells apply Alacrity on the first pulse. This trait is a strong group support option by potentially granting permanent Alacrity while permitting the Specter to repeatedly generate large amounts of Shadow Force from Wells and synergizing with Consume Shadows for AoE Healing and Barrier. 

 

Major Grandmaster Traits. Grandmaster Specter traits offers a choice of improved sustain and damage options. 

Strength of Shadows. Strength of Shadows improves the Torment duration of Rot Wallow Venom from Dark Sentry by 50% and also grants 13% of the Specter's Vitality as Expertise as well as reduces incoming damage by 2% per Condition on the opponent. This trait broadly improves the Specter's condition damage output and potentially reduces incoming strike damage. 

Hungering Darkness. Hungering Darkness transfers Conditions from a Specter's tethered Ally when in Shadow Shroud while cleansing 1 Condition per second from the Specter, while Healing the Specter and giving the Tethered Ally Barrier per condition cleansed. This trait presents a strong mix of personal and single target support sustain against all condition effects.

Issues/Bugs: (1) Cleansing 1 condition per second when in Shadow Shroud seems to be over tuned compared to other "Cleanse" traits, in combination with Heal/Barrier.

Solutions/Proposals: (1.1) Hungering Darkness should have a 3 second internal cooldown on pulling and cleansing conditions. 

Explanations: (1.1) Shadow's Embrace is a comparable example of a cleanse trait on a 3 second interval. The difference between an Adept and Grandmaster in this case is the support Hungering Darkness provides both to an Ally and the Specter.

Shadestep. Shadestep grants self-Barrier on Siphon and revives Allies by 33%, while also causing Shadowsteps to transfer any Barrier on the Specter to nearby Allies. 

Issue/Solution/Explanations: (1) If Ally Siphon is replaced by a new F3 [See Siphon] then the self-Barrier should remain on Siphon and the Revive aspect should be placed on the F3.

Edited November 8, 2021 by saerni.2584

[saerni.2584]

 

Core Thief remains in a relatively inferior place because both Daredevil and Deadeye have superior damage application and sustain thanks to superior mechanics, and Specter seems destined to follow by providing unique support gameplay. Core Thief retains the 1200 range Steal and the advantage of being able to take three Core Specializations at once. Steal has a much longer base cool down, however. And the real advantage of being able to take three Core Specializations at once is limited by the degree to which an Elite Specialization is basically better at “role” or more flexible at that role than a Core Specialization. Daredevil versus Acrobatics is the often cited example of an Elite Specialization taking the same “role” and doing it with more flexibility. Deadeye has this to a lesser degree but can also somewhat can replace the Critical Strikes line (access to Boons and bonus critical chance) or the Deadly Arts line (access to bonus damage modifiers).

– – –

The “Daredevil” is likely our most popular Elite Specialization. Daredevil has seen a few changes over the years, from putting Exhaustion onto Unhindered Combatant to the change from Steal to Swipe. However, because Daredevil utilities offer a mix of ranged and melee Crowd Control options, a channeled block, and a heal with good synergy for the evade-dodge play style, Daredevil is in a good place overall and a reason people say “Thief is in a good place.” Daredevil isn't locked into playing either condition or power oriented builds, although Staff is a power weapon, due to its unique on-dodge traits and the sustain boosts Daredevil has compared to Core Thief make it the preferred choice.

Bandit's Defense. @Graven X.6091  raised the issue that Bandit's Defense does not interrupt channel or casts despite activation of a block being a timing based imperative. 

Suggestions: (1) Bandit's Defense should be allowed to interrupt channels when activated.

Justifications/Explanations: (1) A defensive ability should be able to be triggered on demand and interrupt anything else that is currently channeling. This would be a quality of life improvement. 

– – –

The Deadeye, although somewhat less popular, still occupies a positive unique niche for ranged Thief builds and playstyles. In large part because Deadeye's Mark operates at up to 1500 range, ranged builds can take more advantage of the distance between themselves and a target. Deadeye, however, suffers from a few “dead” (no pun intended) trait choices and utilities. Further, there are a few traits that are under-performing in comparison to alternative options, in ways that negatively impact the diversity of build options and play-styles within the Elite Specialization as a whole.

Collateral Damage. The main issue with this trait is that this will often trigger on an isolated enemy and therefore not do any damage. This applies equally to PvE and PvP/WvW making it an unpopular to dead trait selection.

Suggestions: (1) Collateral Damage now triggers an AoE damage pulse on Stealth Attack around the target. Justifications/Explanations: (1) AoE cleave and tagging potential are always welcome in both PvE and WvW. Thematically it makes more sense to damage around a target you are currently attacking (actual collateral damage) than to trigger an explosion only after the target dies.

Fire for Effect. The main issue with this trait is that it feels subpar by comparison to alternative support options. However, because Core Thief lacks adequate support traits/utilities, Fire for Effect has to effectively carry the entire burden of making Deadeye into support which makes it hard to balance without making the trait either overpowered or ineffective. There is overlap with Maleficent Seven as well, because the Thief will likely increase their own damage taking Maleficent Seven more than the group's overall damage with Fire for Effect.

Suggestions: (1) Replace Fire for Effect effect with: Stolen Skills now apply Stability (5 seconds), Resistance (5 seconds), Resolution (5 seconds), Regeneration (5 seconds), and Protection (5 seconds) when used at 5 malice stacks. Justifications/Explanations: (1) Deadeye Grandmaster traits all follow the general theme of using Boons to enhance the Deadeye. However, despite Deadeye being the least mobile Thief Elite Specialization, it has limited/no access to Resistance, Resolution, or Stability to overcome this mobility deficit. This change would provide the Deadeye with those enhancements to sustain while still requiring them to build Malice sufficient to get access to the additional boons. It also makes the Grandmaster traits a clear tradeoff between Maleficent Seven (superior sustained damage, enhanced mobility), Be Quick or Be Killed (superior burst damage) and Fire for Effect (superior personal sustain).

Maleficent Seven. The main issue with this trait is that it does many things at once at the expense of other Grandmaster options. Might and Fury combined with Initiative regeneration and two more points of Malice (for Stealth Attack related bonuses) make the trait a major damage enhancement. Protection, Swiftness, Regeneration, and Vigor, however, are significant boosts to sustain. This makes the trait out of balance with the other Grandmaster traits in terms of purity of purpose.

Suggestions: (1) Remove Protection and Regeneration from Maleficent Seven. Justifications/Explanations: (1) Protection and Regeneration should be integrated with Fire for Effect for a defensive versus offensive trade-off. Maleficent Seven should retain Swiftness (because Swiftness helps to secure melee range damage) and Vigor (because Vigor contributes to more Silent Scope access) as Boons with an Offense oriented benefit.

Be Quick or Be Killed. The main issue with this damage oriented trait is that it is out shined by Maleficent Seven. Especially in PvE, building and cycling Malice with Maleficent Seven for additional Initiative and Might stacks is always going to be better damage output even in “burst” time frames. Deadeye's options for access to Quickness make it hard to actually use the trait to quickly do damage and secure a kill (again for a burst damage niche). Deadeye's Mark applies Quickness but the animation of marking a target covers part of the duration of the Boon. This is worst in sPvP/WvW where the duration of the base Quickness applied is significantly reduced. “Steal Time” is the only Stolen Sill that offers the Quickness boon, requiring fighting a Mesmer or getting a lucky “One in the Chamber” roll.

Suggestions: (1) Make Be Quick or Be Killed now also apply 2 seconds of quickness when Stolen Skills are used at 5 malice. Reduce the duration of Quickness on Steal Time from 3 to 2 seconds. Justifications/Explanations: (1) This mirrors the proposed change to Fire for Effect but focuses solely on the Quickness boon. This would give a player more control over when Quickness is applied for more consistent ability to take advantage of the Be Quick or Be Killed power/precision bonuses. Quickness on Deadeye's Mark helps the Deadeye quickly generate Malice and then Quickness on Stolen Skill (at 5 Malice) gives the Deadeye synergy between using Stolen Skills for Stealth + Stealth Attacks.

Shadow Flare. The main issue with this utility is that the projectile that applies the flare is rather slow and can be blocked/obstructed. Suggestions: (1) Increase the speed of the Shadow Flare orb projectile and make it unblockable (projectile, not the AoE damage pulse). Justifications/Explanations: (1) This significantly improves the usability of the skill by making sure the projectile itself lands where it is intended to.

Edited November 1, 2021 by saerni.2584

[saerni.2584]

Utilities:

Preparations. Preparations added a nice variety of options to Thief for damage, crowd control, mobility and utility. However, all the Preparations universally suffer from the built in “arming time” that makes them clunky to use in PvE and more difficult to successfully set up in sPvP and WvW.

Suggestions: (1) Remove the arming time from all Preparations. Justifications/Explanations: (1) Arming time makes the Preparations difficult to use in most situations, making an alternative utility preferred in almost all cases. This would be a quality of life improvement to these skills without needing to adjust damage/effect numbers individually to make the skills more appealing to players.

Scorpion Wire. Scorpion Wire is an effective Pull but only when the projectile properly connects with the target. 
Suggestions: (1) Increase projectile speed on Scorpion Wire to improve reliability of the skill but add an animation precast to keep the telegraph the same. Justifications/Explanations: (1) Increasing the speed will improve reliability and the added animation precast will not make the faster projectile unfair in competitive play. 

 

Core Specializations:

Trickery. Trickery provides flexibility oriented traits designed to improve basic Thief gameplay in several ways. Primarily, because Preparedness increases Initiative by 3 points and because Kleptomaniac regenerates 2 Initiative on Steal, this Specialization is often considered mandatory across a wide variety of builds and play-styles. While the Thief community has often and emphatically asked the Developers to make Preparedness baseline, the Developers have consistently refused to do so, and with the reduction in baseline Initiative for Specter appear to have further committed to this design. For that reason, I will not be including the suggestion in this review other than by mentioning why I am not including it.

Shadow Arts. Shadow Arts serves as the primary Stealth oriented specialization for Thief. Thematically, the traits in this specialization over the years have trended between passive bonuses for being in Stealth and active bonuses upon either entering Stealth or using a Stealth Attack. Although the specialization was traditionally more passive in nature several iterations of balance patches moved the traits more towards active bonuses as a way of reducing the benefit for a Thief to build a permanent Stealth uptime build, at least in the PvP or WvW contexts.

Merciful Ambush: This trait fills the niche of a support revive trait for Thief. However, given that the choice was made to neuter similar traits for other Professions to reduce revive speed overall, this trait has fallen behind. The other traits available in the Adept tier are far superior when considering the risks associated with merely disappearing briefly from view while reviving an ally.

Suggestions: (1) Make Merciful Ambush Stealth yourself and your target when reviving an ally. Grants swiftness, regeneration, and stealth to yourself and your ally upon successfully reviving an ally. Justifications/Explanations: (1) Reviving is risky, and the risk should be rewarded when you are successful. For a Thief, and to an extent their revived ally, having a boost to healing, movement speed, and going into stealth to reposition is exactly the post-revive bonus a so-called “ninja nurse” should be applying.

Meld with Shadows: This trait provides enhanced Stealth duration and passively boosts the Thief to 50% movement speed while in Stealth. This trait is a legacy from before Superspeed was added to the game.

Suggestions: (1) Make Meld with Shadows increase Stealth duration and apply 3 seconds of Superspeed on entering Stealth. Justifications/Explanations: (1) Now that Superspeed is available, as an alternative Superspeed on Stealth would be preferable for short term reliability/usability and preventing permanent Stealth from having a permanent speed boost.

Shadow Savior: This trait provides healing to a Thief and nearby allies upon completion of a Shadowstep. As a support and sustain trait it is oriented around active play and nonetheless suffers from being slightly under performing.

Suggestions: (1) Increase the healing power contribution from 0.2 to 0.3 and the ally healing coefficient to 0.6. Justifications/Explanations: (1) Increasing the coefficients should provide some opportunity for expanded use without making the trait either a must-take or otherwise overpowered.

Shadow's Rejuvenation: This trait is responsible for Thief being able to maintain permanent Stealth by regenerating Initiative passively while in Stealth. The healing from this trait is also not insubstantial, offering a major boost to sustain on-par with Warrior's “Healing Signet” skill as long as the Thief is invisible. The most direct method to limit or disincentivize permanent Stealth game-play without significant game design changes is to modify how it grants Initiative regeneration.

Suggestions: (1) Decouple the healing and Initiative gain in Shadow's Rejuvenation. Make the healing pulse once per second as it is currently. For Initiative, make Shadow's Rejuvenation regenerate 1 Initiative upon entering Stealth and another 1 Initiative upon hitting with a Stealth Attack. Shadow's Rejuvenation should have a 3 second internal cooldown for the Initiative gain on entering Stealth. Justifications/Explanations: (1) The passive healing is useful, but only becomes a balance problem with near infinite Stealth. By adding an “on hit with a Stealth Attack” effect and also pairing that with gaining Initiative upon entering Stealth, Shadow's Rejuvenation would have a similar effect on Initiative regeneration without creating permanent Stealth game-play. The cooldown prevents potentially unhealthy interactions with the Deadeye Silent Scope trait and using back-to-back Stealth Attacks via Shadow Meld.

Acrobatics. Acrobatics provided much of the evade based support traits for Thief prior to the advent of Daredevil and the comparably superior traits in that Elite Specialization. In addition, multiple traits within this Specialization have been nerfed in competitive modes, infamously Instant Reflexes and Hard to Catch both with an unusable 300 second cooldown. Without a doubt, as Acrobatics does contain useful traits for certain play-styles, and potentially for the Specter for Initiative regeneration, there is a benefit to a fresh look at the Specialization.

Instant Reflexes. This trait currently provides 2 seconds of evade when struck below the health threshold of 50%. While Evade is not immunity to all skills, and conditions continue to apply damage, Instant Reflexes was nerfed in large part to avoid rewarding someone passively with significant additional evade time. In the split between PvE's cooldown at 40 seconds and PvP/WvW's cooldown at 300 seconds is the clearest indication that this trait is under performing when compared to the other two Major Adept traits (Vigorous Recovery and Pain Response).

Suggestions: (1) Instant Reflexes should be re-imagined as a bonus for successful evades in the vein as Swindler's Equilibrium and Upper Hand. Instant Reflexes should grant 2 seconds of Superspeed on a 1 second internal cooldown. Justifications/Explanations: (1) The Major Adept tier of Acrobatics does not have an “on evade trait.” Changing this trait in this manner allows for consistency across the tiers and also acts to help someone who dodged to evade reposition to maximize their ability to make the next move on their terms. This combines with Expeditious Dodger to make an Acrobatics inclusive build truly mobile in a unique manner from Daredevil.

Hard to Catch. Hard to Catch acts as a passive stun-break that also refills 100 Endurance. In PvE the cooldown remains at 30 seconds and for PvP/WvW the cooldown is 300 seconds. This was split so heavily in PvP/WvW because the trait again acted as a passive reward for failing to avoid an incoming disable. Acrobatics' design follows a pattern of skill upgrade-evade bonus-cleanse/condi reduction. However, Hard to Catch doesn't act as an upgrade so much as mistake mitigation.

Suggestions: (1) Hard to Catch should instead reward players who built a stun-break into their build by making them “hard to catch” with a follow-up stun by applying 1 Stability for 2 seconds on stun-break. Justifications/Explanations: (1) Acrobatics has plenty of other ways to regenerate Endurance between Vigorous Recovery, Feline Grace, and Endless Stamina. What a Thief being hit with a disable needs isn't a passive bailout for that disable but an upgrade they can use to briefly prevent themselves from being disabled again.

 

Weapon Skills:

Shadow Strike [Pistol/Dagger 3]. Shadow Strike has gone through a number of balance iterations around the question of how much Torment, exactly, this skill should apply. In 2019, Shadow Strike was also updated to include the flip skill Repeater, reducing the ability to use Shadow Strike to once every 4 seconds in PvE and every 3 seconds in PvP/WvW, but enhancing its follow up damage potential through a channeled pistol barrage. It should be noted that Shadow Strike has faced its most recent modification in part due to the 2021 modifications to the Torment Condition that increased the value of Torment application in PvE.

Suggestions: (1) Shadow Strike should apply 3 stacks of Torment in all game modes. Repeater flip skill duration should be set at 3 seconds in all game modes. Justifications/Explanations: (1) Due to the Repeater flip skill Shadow Strike does not have the ability to stack excess amounts of Torment and therefore should regain at least one of the stacks of Torment that were previously on this skill. As the modifications to Malicious Sneak Attack showed, the balance issues caused by the ability to stack large amounts of Torment came from the combination of Maleficent Seven's Initiative regeneration bonus and Malicious Sneak Attack applying 5 stacks of Torment with the bonus duration at seven Malice, and not Shadow Strike. Further, with Repeater having a variable flip skill access duration it creates problems for players moving between the modes who have to adjust to having 1 second less/more before they can use Shadow Strike again for its teleport. Normalizing this skill to 3 stacks of Torment and 3 second flip duration in all modes will be a consistency/quality-of-life improvement.

Disabling Shot [Short Bow 3]. Disabling Shot provides a ½ second backwards Evade, 2 second Cripple, and small amount of damage for 4 Initiative in PvE and 5 Initiative in PvP/WvW. A comparison must be made to Debilitating Arc (Daredevil's Staff 3 skill) which also includes a ½ second backwards Evade, 6 second Cripple, and damage, and costs 4 Initiative in PvE/5 Initiative in WvW/PvP. Debilitating Arc, however, also removes Immobilize and hits up to 3 targets, albeit at a shorter 400 range compared to 900 range for Disabling Shot. Disabling Shot's comparative advantage of being able to apply a much shorter duration Cripple at 900 range is outweighed by the far superior version on Daredevil Staff. Further, Disabling Shot doesn't as seamlessly integrate into the Short Bow's other skills as Debilitating Arc does for Daredevil Staff.

Suggestions: (1) Disabling Shot should no longer Evade. Instead, Disabling Shot should apply 3 seconds of Cripple (900 range) and Push the target if the target was within 180 units for a distance of 300 units. Justifications/Explanations: (1) Disabling Shot as currently implemented is a large investment of Initiative and doesn't integrate well with the other skills on Short Bow because the payoff of a single short evade, short term Cripple, and minimal damage isn't enough to justify the Initiative costs when other high cost skills like Choking Gas or Infiltrator's Arrow usually monopolize the priority for spare Initiative. What Short Bow needs more than anything is the ability to pair something higher cost like Infiltrator's Arrow, a potential gap closer most often used to run away, with a targeted Push for crowd control. Paired with Cluster Bomb, a targeted Push could help to safely set up a Detonate Cluster or enable the Thief to reposition against an advancing melee opponent.

Infiltrator's Arrow [Short Bow 5]. Infiltrator's Arrow is the consummate Thief mobility skill and one reason why Thieves have traditionally sought to use Short Bow more than any other offhand weapon. However, Infiltrator's Arrow has received significant nerfs in terms of the Initiative costs of the skill, for a total of 6 Initiative in PvE and 8 Initiative in PvP/WvW. Because Initiative is a shared resource pool between both weapons this high cost represents half of Maximum Base Initiative in PvE and two thirds of Maximum Base Initiative in PvP/WvW respectively. Even taking Trickery to gain Preparedness's increase in Maximum Base Initiative to 15 means that in PvP/WvW Infiltrator's Arrow represents more than half of the Thief's weapon cooldowns across all weapon sets any time it is used.

Suggestions: (1) Reduce the Initiative cost on Infiltrator's Arrow to 7 in PvP/WvW. Justifications/Explanations: (1) No Thief weapon skill should, with the increases in base Initiative that are possible through taking the Trickery Specialization, be impossible to use twice in a row at full Initiative. Thief Initiative is designed to give the Thief the flexibility to use their weapon skills at any time without regard for individual cooldowns. Any time a skill cannot be used more than once in a row while at maximum Initiative effectively imposes an individual cooldown in violation of the basic design premise behind Initiative. Reducing the cost to 7 from 8 in PvP/WvW represents the maximum cost that should ever be imposed on an individual skill from that basic design perspective. Any further nerf should come to the skill effect itself and not come on the cost side.

Edited October 25, 2021 by saerni.2584

[Lithril Ashwalker.6230]

Scorpion Wire: currently throws a projectile 1200 range at a target and pulls them to you with a 2 ammo charge. while its useful in mostly wvw or tricky mobs in pve its not very reliable on moving targets as the velocity is like tossing a wet noodle.

Suggestion: increase the velocity to match that of Spear of Justice and add a red glow telegraph on the user to help telegraph.

Preperation(Seal Area): creates an (supposed to be) impenetrable bubble that foes cannot enter or exit for about 8 seconds and can double up as a dark field for Dark Aura. the issue with this is that some leap finishers and skills will allow you to "jiggle" through or out of the bubble during its animation and uptime...Heartseeker spam is a good example of this.

Suggestion: make the barrier from the Seal Area bubble thicker to prevent these skills from getting through without proper CC protection. It would make it more reliable.

 

 

[saerni.2584]

14 hours ago, Lithril Ashwalker.6230 said:

Scorpion Wire: currently throws a projectile 1200 range at a target and pulls them to you with a 2 ammo charge. while its useful in mostly wvw or tricky mobs in pve its not very reliable on moving targets as the velocity is like tossing a wet noodle.

Suggestion: increase the velocity to match that of Spear of Justice and add a red glow telegraph on the user to help telegraph.

Preperation(Seal Area): creates an (supposed to be) impenetrable bubble that foes cannot enter or exit for about 8 seconds and can double up as a dark field for Dark Aura. the issue with this is that some leap finishers and skills will allow you to "jiggle" through or out of the bubble during its animation and uptime...Heartseeker spam is a good example of this.

Suggestion: make the barrier from the Seal Area bubble thicker to prevent these skills from getting through without proper CC protection. It would make it more reliable.

 

 

 

I integrated the Scorpion Wire suggestions above. I don't know enough about how Seal Area functions on a technical backend side so for now I'm not going to include that. We should probably have a separate thread on that (although only a Developer could probably answer the technical questions necessary to understand whether your suggestion is workable or not). 

[Trianox.3486]

I know venom have their uses, but they're not as good as most of the other utilities... I know, technically they can provide decent DPS when used with a group of allies, but solo they're lack luster. I'd rather wished we have something that works similar to  Soulbeast vulture stance with maybe a 0.5 sec internal cooldown... But then again... It's not like they'd be overpowered when you start comparing to other classes utilities...

 

Acrobatics: really like the ideas your have @saerni.2584 

[Felices Bladewing.3914]

Now that Scepter is cleary condibased i would finaly like to see a rework on D/D dualskill to be more focused around power

[saerni.2584]

I'm beginning the process of adding in Specter Beta feedback (post 2). I've been testing myself and reading some of the feedback. Lots of mixed feelings. Some people seem to have not gotten used to the potential play styles (only 1 day of testing so far). I will be attempting to sort through things carefully and include a representative sample of the feedback from across the sub-forum. 

[Trianox.3486]

I'll support and reiterate what had been said for preparations:

Alll preparation should really loose their activation delay in PvE so that we can spawn them, and trigger as soon as we see fit, and not wait a few seconds...why? In the reality of PvE, you tend to move quite quickly between targets to the point where when your preparation is "armed", the foes are dead or you're long gone from that spot, especially given the thief mobility. 

 

Yet, there those preparations have so much interesting potential... Just never good enough in PvE simply because of the delay.

[Zacchary.6183]

So after messing around on Specter, I have put together a list of observations for each part of the new spec and I will place them in the context of skill in to power out and personal suggestions to perhaps enhance it in this regard. Starting with Scepter:

Scepter Skills

Overall: The scepter is no doubt a support weapon as every skill provides some sort of buff. Endless Night is perhaps the biggest example as the amount of quickness and regen you can apply with it and there is a clear purpose between sc/d and sc/p. Otherwise, damage is not that great. The biggest issue I have seen with scepter is ally-targeted auto-attack being incredibly annoying in cities. It would be a good idea if the noise was toned down greatly and/or if attacking an ally did not activate auto unless in combat.

Shadow/Double/Triple Bolt: 

Plenty of attacks done with this skill, which is good for procing on-hit effects. Strike damage is low and Condition application potential, while better, is also low as the stack of short duration torment will be gone by the time you get to applying more. And this means that the Specter will have to load up on Expertise if they want to do any noticeable damage with scepter. Allied-targeting, the amount of barrier applied is good enough that you could probably switch between party members as you attack to spread the barrier application out. However as mentioned before, the noise is annoying.

Shadow Sap:

The might application vs cost is decent considering might is gained very easily in this game. This means that Shadow Sap on an ally will be best used as a sort of long lasting top up. Of course, this skill can't be used for damage more so than weakness application. The amount of barrier applied when ally-targeting is comparable to that of the ones applied with the auto-attack, so its really there to keep barrier application consistent.

Twilight Combo (Sc/D 3):

This two-part attack is where a good amount damage can be done with this skill, but it barely compares to core weapons. The conditions applied by this attack last long enough that it can help keep a nice stack up and the poison helps it synergize with Deadly Arts more. Chilled can be extended to keep targets in place. For allied-targeting, it provides a great chunk of barrier and a little bit of healing and swiftness. This skill would be best used on a hybrid build.

Measured Shot/Endless Night (Sc/P 3):

Reminiscent of Shadow Strike, this skill is good for keeping enemies away. The follow up attack, Endless Night, is very good for applying torment/slow and will do very well with maximum expertise. It also provides a good burst heal with a follow up buff attack that can provide near max quickness to the targeted ally with firebrand runes and some concentration, as well as buff anyone else in the beam. When playing around with this, I noticed that the beam seems to extend slightly past the target, like about melee distance. As an idea for improvement, it could be possible to keep the beam a static 900 units so that a specter that is close can sweep behind the target and get other allies. It feels clunky, otherwise.

Shadowsquall:

Decent stealth attack if you are supporting your allies as the amount of HP healed hovers around 10k with extra healing coming from the long duration regen. The enemy-targeted version also does a fair amount of damage, however the poison is short-lived and could use a slightly longer duration. But overall, Shadowsquall is good as a stealth attack. Combined with Hidden Thief and Sleight of Hand, a specter could use this every 12.5 seconds. So this stealth attack could see plenty of use in a rotation.

Siphon/Shadow Shroud:

Siphon has a very short cooldown with Sleight of Hand which is good news for every steal trait thief has access to. This also makes it easier to build up shadow force quickly. Not only that, it is fairly essential to coordinate within a group as allied-targeted use provides a good chunk of barrier out of the box. Shadow Shroud is a good concept, however I have not done a lot of work on its skills so I will have to come back to that another time.

 

Traits

Second Opinion:

This can provide a good boost to healing power for condition builds and a decent choice for those wanting a balance of conditions and support. However, it is obviously not for power builds and it competes with other traits.

Shallow Grave:

This can actually be pretty good if you want to provide a passive way to prevent a death since Siphon provides a nice chunk of force already to allow the effects to proc. 60 seconds for a cooldown also seems fair for an effect like this. While it is generally better than Second Opinion, it competes with Consuming Shadows.

Consuming Shadows:

A personal favorite, especially since this will become mandatory in power support builds because of it's scaling with Health rather than healing power and its ability to overheal. This will also incentivize thieves to load up on vitality to get the most out of it, automatically boosting their survivability. The only downside is that in order to get the insane overheals, the thief will need to generate force fast and expend it all. This can be done by keeping Siphon on cooldown and maybe using traited wells on stacked allies.

Dark Sentry:

While the torment duration is short, Specter's barrier application is so abundant that this trait seems to make up for the lack of condition damage on scepter so long as they are within a group. The outgoing healing boost is also very good for supporting in general.

Larcenous Torment:

Considering the amount of torment scepter can pump out, this is a direct boost to scepter auto and condition builds in general. Would go well with hybrid p/d.

Amplified Siphoning:

This is better for solo-focused builds or if you want to maximize Siphon effects. Otherwise, its group support capabilities are minimal.

Traversing Dusk:

This is necessary if you plan on providing long lasting alacrity to a group and great for shadowstep-heavy builds like s/x wells. A stacked group will semi-passively help you build up shadow force as you support them with the wells. You can also provide permanent alacrity with a full set of wells and some concentration since most of them have 20 second cooldowns.

Panaku's Ambition:

The amount of barrier this applies is around the level of baseline scepter, providing more synergy with Shadow Arts in terms of support.

Strength of Shadows:

This is a much better stat booster than Second Opinion, hands down. For builds that have added vitality for Consuming Shadows, this is a good trait to take if you want a boost in condition durations. Even power builds can benefit since they tend to have plenty of non-damaging conditions available. It also helps Rot Wallow torment stack, so good for barrier bots.

Hungering Darkness:

If you are playing with tether against heavy conditions or are a dedicated Tag healer in a zerg, this might be something you could take to keep conditions off them. The amount of condition cleanse you can deal in shroud is potent, but otherwise the other two traits are better options.

Shadestep:

This is definitely a must if you plan on running medic. Siphon revives allies by 33% which is a massive chunk of healing out of the way and you can use your barrier to supplement Consuming Shadows.

 

Wells

Overall: Shadowstep wells is a pretty good concept and synergizes with Traversing Dusk, Shadow Savior and Sword. You don't need condition builds to use them effectively and on such a short cooldown, your mobility and support capabilities can skyrocket. The dark field they provide also adds to support with access to dark auras, life siphon and blinds.

Well of Gloom:

This is a decent heal, especially when traited. It heals for around 8k total and provides basic support on a short 20 second cooldown.

Well of Bounty:

Definitely a must take for support and benefits greatly from concentration and other buffers. Traited, it provides 6 boons  due to this well only providing boons the affected players do not have an simply placing it will have already applied the boon.

Well of Tears:

This well does chip damage over its duration. Good in a power build for general mobility. While it doesn't do that much damage, its 20 second cooldown and synergy with many traits makes this good to slot.

Well of Sorrow:

This would have to be the condition equivalent of Well of Tears with the mechanics of Bounty. This would do really well with expertise and Deadly Arts. However, overall, its effects seem a bit underwhelming. The short cooldown and shadowstepping utility combined with Traversing Dusk makes up for this a bit.

Well of Silence:

The only gripe I have with this is that the tooltip could use a little more clarification as to how it exactly functions, specifically that it pulses daze. Good at eating away at a break bar and would do well paired with x/p for interrupts.

Shadowfall:

This well is also decent for crowd control and would work in a stacked group. It does a bit more damage than the others but the pulse interval is longer. For solo play, it would be better to use Thieves Guild since they can be buffed by your attacks and wells.

Edited October 28, 2021 by Zacchary.6183
some proofreading

[ZeroTheRuler.7415]

On 10/25/2021 at 8:13 PM, Highlie.7641 said:

Viewpoint: mainly Spvp. (currently not active)

 

Remove [Gain Vitality Based on a Percentage of Power: 7%] from

https://wiki-en.guildwars2.com/images/thumb/5/5c/Marauder%27s_Resilience.png/25px-Marauder%27s_Resilience.png Marauder's Resilience

 Add it to 

https://wiki.guildwars2.com/images/thumb/a/ab/Vigorous_Recovery.png/25px-Vigorous_Recovery.png Vigorous Recovery

 

Then we are left with  Bloating https://wiki.guildwars2.com/images/thumb/4/4f/Swindler%27s_Equilibrium.png/25px-Swindler%27s_Equilibrium.png Swindler's Equilibrium more.  Do we want to?:

 

Add Another [Gain Vitality Based on a Percentage of Power: 7%] while wielding a sword..

(This would force Swords to still use Zerker amulet, but maintaining 15ki'sh life)

 

Or Do we increase The damage section of that trait again

(your now using Maru with the power of zerker).

 

That would be a massive nerf to most builds, especially for PvE content. A DD zerker's build already has low HP as it is. To get it one would have to spec into acrobatics which isn't exactly a damage spec.

[Jugglemonkey.8741]

1 hour ago, Highlie.7641 said:

No not at all.

 

D/p would now be able to take this trait while still gaining condition on dodge (which you should be using in both Spvp and Wvw anyway) Yes you lose damage, but that's the choice your suppose to have. Damage or more sustain. not both for no cost.

 

 

 

You shouldn't really be using acro on D/P daredevil, and the cost of taking the sustain through extra health is the lack of condi cleanse on dodge. If you want D/P to take both traits then you're basically saying that your WvW build shouldn't be forced to choose between those traits when every other build should, which kinda invalidates your last comment. 

[SoulGuardian.6203]

I think that sword skill 2 should be altered.

That infiltrator's return should be removed or moved to another weapon.

It confuses the heck out of me. 

Edited November 1, 2021 by SoulGuardian.6203

[saerni.2584]

I've updated the Specter feedback post to include most of some initial feedback based on my testing and some comments I've read. Still need to finish Wells and Traits and then review the feedback/bug threads for anything I've missed. 

Keep on posting! It helps motivate me to finish what has turned into probably the largest posts I've ever written on these forums.

Edit: I'm also fixing the formatting along the way. 

Edited October 31, 2021 by saerni.2584

[Lithril Ashwalker.6230]

revert Choking Gas back to the initial at-threshold interrupt per pulse when someone has more than X amount of poison...considering the elite for specter now pulses interupts. its hardly choking when it just IMPACTS...raise the initiative cost to it by 1 if people are worried about the spam

[saerni.2584]

On 10/31/2021 at 6:53 PM, Lithril Ashwalker.6230 said:

revert Choking Gas back to the initial at-threshold interrupt per pulse when someone has more than X amount of poison...considering the elite for specter now pulses interupts. its hardly choking when it just IMPACTS...raise the initiative cost to it by 1 if people are worried about the spam

 

Can you elaborate more on why you think the current implementation of Choking Gas is bad and how the pulsing interrupt at X Poison stacks would resolve that issue? 

[Lithril Ashwalker.6230]

15 minutes ago, saerni.2584 said:

 

Can you elaborate more on why you think the current implementation of Choking Gas is bad and how the pulsing interrupt at X Poison stacks would resolve that issue? 

risk vs reward. the more u use the less initiative you will have as specter but would have extra conditions and group dps support/cc support for grouped enemies. threshold was enough of a cost and gave the skill pizzazz. having it as a single on-impact interupt that just pulses poison a few ticks isnt exactly "choking gas" its noxious cloud. that original variant had more synchronization with pressure striking than it does right now and considering we are going to have an elite spec revolving around torment it would be fitting to have that pulsing interrupt effect even if the initiative were high cost to work with the Specter. right now the skill isnt really worth using unless you just choose to autoattack with shortbow through it...adding specter is a nice point in the right direction but pidgeon holing it into a few good weapons doesnt seem to really stick to the original build craft vision i saw the game start out with...

[Drakonhammer.2148]

The main problem I have with Acrobatics from playing around in the BWE is that it is a very selfish traitline with little payoff - in solo play it's weaker than Daredevil and on Specter why take it when I can take Shadow Arts which offers initiative/sustain + group support? My only complaint with Shadow Arts was Leeching Venoms not seeming to work with Rot Wallow Venom, which is a missed opportunity as it forces us into Shadow Saviour, but even so Shadow Arts remains flat out better than Acrobatics for group support.

Acrobatics offers just initiative/self sustain which is counterproductive if you're going with a group support build and can take Shadow Arts which offers initiative + group support. If Acrobatics offered alternative group support to Shadow Arts e.g. if the boons and maybe a few extra traits like Guarded Initiation, Endless Stamina etc perhaps (at least partially) were provided to the group I could see myself taking it on Specter, as it stands now there's no point imo. It'd be great if Shadow Arts, Acrobatics and Trickery were all more or less on par.

On condi Specter I suspect I'll be taking the standard condi Deadly Arts + Trickery for the most part as Specter can spam out poison very well this way and Lotus Poison + expertise means very high weakness uptime for solo content too. Maybe Shadow Arts would be good for solo sustain due to life siphon from venoms, but I don't see myself taking Acrobatics on condi Specter at all.

Trickery: Honestly, as much as Preparedness helps with opening bursts and phases, I feel Kleptomaniac does most of the leg work on Thief - it is a huge source of initiative especially combined with Trickery's lowered steal CDs. Having either Preparedness or Kleptomaniac in an already good traitline means it is even more compulsory, and this applies for most builds even outside Specter.

[saerni.2584]

Round 1 of Specter feedback is complete! Whew, that is a lot of words.

 

Round 2 will be a review of the feedback thread and bug threads posted by Anet. Round 3 will be every other potential source, including this thread/reddit/other threads on this subforum/other threads in other subforums. 

 

Thank you all for your patience and kind contributions.

 

EDIT: Round 2 Status

I have completed a review of 8/15 pages of the primary feedback thread. There's a lot more to do so stay tuned for additional notes. 

In the meantime, please feel free to post additional feedback in this thread to respond to the Specter/other meta-posts. 

Edited November 8, 2021 by saerni.2584
Added status update