Feedback:Developer updates/20100128

Developer Update - January 2010 PvP Updates
As we've mentioned elsewhere, we have been working on a major playbalance update to address a large number of issues, the most significant being the extreme farming and speed-clearing builds centered on the Elite Assassin skill Shadow Form. While work on this major update continues, we've decided to release a PvP-oriented portion of the update sooner to address pressing issues in a few PvP formats. The areas we are trying to address are:
 * Ranged Spike/Damage in GvG
 * Party-Wide Healing in GvG
 * Blood Spike in PvP
 * Individual Problem Skills

We had been working on both the PvE and PvP portions of the update in tandem, but when it became clear that that PvE-oriented segments would not be ready for release this month, we decided it would be better to release the PvP-oriented part of the update to make the game play better today, rather than holding it simply in order to prevent outcry from those anxiously awaiting the PvE portion of the update. We hope to have the rest of the update, including thoroughly tested changes to Shadow Form, ready in February. Incorporating the Test Krewe into our playbalance process has slowed us down a little bit, but we have no doubt that the quality of our balance updates will greatly benefit from the level of testing and feedback the Test Krewe provides.

We do not normally like to release PvP balance changes so close to a Monthly Automated Tournament, but we saw many of these as pressing issues. We released a preliminary list of changes last week to help players to prepare for the MAT. We do not anticipate regularly releasing skill updates so close to tournament times in the future.

In addition to the skill changes, we are updating some of the GvG party restrictions to make participation in GvG a little easier as well as making a number of refinements to Codex Arena.

Ranged Spike/Damage in GvG
In GvG especially, Glass Arrows Rangers and other bow spike builds are a little too effective. Looking into these builds with the Test Krewe, we identified three issues:
 * 1) The frequency, power, and speed of the spikes. For spike builds to be balanced, it is important that there be windows between the spikes for the other team to react and recover. Additionally, the spike must be vulnerable in some way to disruption or defense so that the opposing team has an opportunity to stop the spikes.
 * 2) The sustained damage output of these builds. Teams should be making a choice between better spikes and better sustained damage. In cases where teams excel at both, something is wrong.
 * 3) Survivability of Paragons and Rangers, especially when used in large numbers. As a trade-off for their ranged attacks, these characters should not be as tough as Warriors.


 * : added a recharge of 6 seconds; reduced damage bonus to 10..20.
 * : decreased damage bonus to 5...12.
 * : increased recharge to 12 seconds; increased conditional damage bonus to 5...30; decreased unconditional damage bonus to 1..5.
 * : increased disable time to 10.
 * : increased recharge to 15 seconds; decreased conditional damage bonus to 5...15; decreased unconditional damage bonus to 5...20.

These changes were meant to address the frequency and power of the spikes. Anthem of Envy and Glass Arrows were two key damage skills. We lowered the damage on both and lowered the frequency with which Anthem of Envy could be used. The damage potential for Keen Arrow and Sloth Hunter's Shot was a major part of the power of the spikes. We've lowered the damage and, more importantly, increased the recharge on these skills to decrease the frequency of spikes. Additionally, we've shifted some of the unconditional damage bonus from Keen Arrow into the damage bonus that it provides on a critical hit. This is meant to clarify the relationship between these relatively similar skills. Keen Arrow is only superior to Sloth Hunter's Shot in situations where the Ranger has an increased chance for critical hits. Finally, we increased the drawback on Marauder's Shot to allow players the option of a powerful, fast-recharging spike skill, but at the cost of defense and versatility. Since all of these were direct nerfs and many of these skills are useful in PvE, we've split all of the skills except Marauder's Shot. The split skills remain unchanged in PvE.

In order to address both the speed of spikes and the sustained damage from these builds, we've add an aftercast to all 1-second Bow Attacks. The shots come out just as fast as they used to, but now there is an additional half-second delay before the next skill can be used. This means these attacks are still faster overall than a standard bow attack, but not so fast that they can be used to quickly chain three bow attacks together. We've also introduced 3/4 second Bow Attacks for situations where we still wanted to allow attacks to be fast.


 * : increased activation time to normal weapon attack activation time; reduced recharge to 3 seconds; functionality changed to "Power Shot deals 25..50 damage if it hits."
 * (PvE):  reduced activation time to 3/4 seconds.
 * :  added aftercast.
 * : added aftercast; reduced bonus damage to 3..10.
 * : reduced activation time to 3/4 seconds; added the following functionality: "Your other attack skills are disabled for 2 seconds."
 * : added aftercast.
 * (PvE):  reduced activation time to 3/4 seconds.
 * :  added aftercast.
 * : added aftercast.

Sundering Attack (PvP), Penetrating Attack (PvP), Precision Shot, Quick Shot, and Melandru's Shot were all given the additional half-second of aftercast. Quick Shot is essentially unchanged, since its 1-second recharge already forced an identical delay between successive uses of the skill. Additionally, Precision Shot's damage bonus was reduced in order to bring it in line with Penetrating Attack (PvP). The PvE versions of Penetrating Attack and Sundering Attack were given the faster 3/4-second activation time in order to preserve their usage in fast-attacking damage builds in PvE. Power Shot had its fast activation time removed and was reworked as a skill that does fixed, armor-ignoring damage. It should be effective in the early game, since the low-end damage on it is fairly high, and again late game, where the armor-ignoring nature of its damage should prove valuable. To make it more viable in this role, we reduced its recharge. The other 1-second Bow Attack that did not simply get an aftercast was Needling Shot. In order to preserve its rapid-fire feel without working in a spike chain, Needling Shot was given a faster 3/4-second activation time but now disables other attack skills for 2 seconds when used.


 * : functionality changed to: "Shout. For 5...17...20 seconds, all party members within earshot receive +24 armor when not moving. Ends on a party member if that party member attacks."
 * : increased recharge to 45 seconds.
 * : added the following functionality: "Whenever a shout or chant ends on you, you suffer from Cracked Armor (10 seconds)."

"Stand Your Ground!" gives a very strong armor bonus to your entire party. Because it only affect stationary characters, it was ideal for teams with a high number of midline ranged attackers. This synergized well with the fact that two Paragons (midline ranged attackers themselves) could maintain this shout 100% of the time. Builds making use of this skill tended to blur the line between what frontline characters and midline characters should be able to do. This change is meant to prevent tougher offensive characters from taking advantage of the bonus while still allowing it to help casters. The changes affect PvP only. This skill is unchanged in PvE.

Since its drop in recharge time a while back, Lightning Reflexes has benefited from the rise of Expertise/Marksmanship builds. With effective rangers running higher specs in these attributes than before, Lightning Reflexes has become an extremely powerful and versatile skill, providing both defense and increased attack speed. We've increased its recharge time to force more careful, tactical use of the skill. This change is only for PvP.

Looking at the survivability of Paragons in PvP, we decided that the Paragon should have to make a choice between survivability and offensive potential. With this in mind, Soldier's Fury was given the same drawback as Aggressive Refrain, the one other Paragon skill that increases attack speed: it now inflicts the Cracked Armor condition on the character using it. We do not believe this change will have any major impact on Paragons in PvE, so this change affects both PvP and PvE.


 * : added the following functionality: "This arrow moves twice as fast as normal."
 * : decreased activation time to 1 second and added the associated aftercast.

Because we were removing some damage-dealing options for Rangers, we decided to buff a couple of their condition-spreading skills. We gave the 1-second activation time to Hunter's Shot to make it quicker at spreading bleeding. As with all bow attacks with 1-second activation times, this skill also has a slightly longer aftercast. The arrows from Crippling Shot now move faster, making these unblockable projectiles harder to dodge. We chose this change rather than a 1-second activation time so that Crippling Shot would still be vulnerable to disruption.

Party-Wide Healing in GvG
We believe that spiking, disruption/control, and pressure damage should all have a role to play in GW PvP. Currently, pressure damage tends to be outshined by party healing in GvG. There are a couple of skills that we wanted to alter to bring this into greater balance.


 * : functionality changed to: "Binding Ritual. Create a level 1...8...10 Spirit. Non-spirit allies within its range under 75% maximum health gain +1...3...3 Health regeneration. This spirit dies after 15...39...45 seconds."
 * : changed functionality to: "Echo. For 15 seconds, you have +2...3...3 Health regeneration. This Echo is reapplied every time a Chant or Shout ends on you."

Recuperation (PvP) no longer provides Health regeneration to characters who are above 75% health. This lets it remove some pressure damage but not counteract it entirely. This change only affects PvP. Mending Refrain has been split into PvP and PvE versions. The PvP version now only targets the caster, effectively removing this skill as a party-healing option in PvP. The PvE version is unchanged.

Blood Spike in PvP
We decided to make two sets of changes to address current issues with Blood Spike.

First, and most dramatically, we've changed several defensive skills to work against life stealing. They still work against damage as well. The effects from Vampiric weapon upgrades are now termed "Life Draining" and are not affected by these changes.


 * : functionality changed to: "Enchantment Spell. For 8 seconds, the next time target ally would take damage or life steal, that ally gains that amount of Health instead, maximum 15...67...80."
 * : functionality changed to: "Elite Enchantment Spell. Remove 0...2...2 conditions from target ally. For 8 seconds, the next time target ally would take damage or life steal, that ally gains that amount of Health instead (maximum 20...84...100)."
 * : functionality changed to: "Enchantment Spell. For 8 seconds, damage and life steal received by target ally is reduced by 3...15...18. When Shielding Hands ends, that ally is healed for 5...41...50 Health."
 * : functionality changed to: "Binding Ritual. Create a level 1...10...12 Spirit. Whenever a non-spirit ally in its range takes damage or life steal, it is reduced by 15 and the spirit takes 15 damage. This spirit dies after 30...54...60 seconds."
 * : functionality changed to: "Weapon Spell. For 8 seconds, the next time target ally takes damage or life steal from a foe, that ally steals up to 15...51...60 Health from that foe."
 * : functionality changed to: "Elite Weapon Spell. For 8 seconds, the next time target ally takes damage or life steal from a foe, that ally steals up to 15...63...75 Health from that foe and loses 1 Condition."

Second, we've adjusted the lead-in Blood Spike skills, toning down both the speed and frequency of spikes.


 * : increased recharge to 15 seconds.
 * : increased recharge to 15 seconds.
 * : functionality changed to: "Spell. Steal up to 15...51...60 Health from target foe and up to two other foes in the area."

Although they were not featured in every Blood Spike build, Lifebane Strike and Shadow Strike provided a potent combination of damage and life steal, making them strong skills for starting a spike. We see this use as in-bounds, but we've increased their recharges in order to decrease the frequency of spikes. Vampiric Swarm served as a lead-in spike skill for entirely different reasons. The life steal on the initial target was delayed (to simulate it being a projectile), which gave the spiking character time to cast a second skill that would steal life at nearly the same instant as the delayed life steal from Vampiric Swarm. The result was an unreasonably small window to react to the spike. While updating the skill, we also corrected the description to reflect the fact that the skill is not (and never has been) a projectile.

Individual Problem Skills

 * : functionality changed to: "For 5...29...35 seconds, if you are wielding an earth weapon, your melee attack Skills [sic] also cause Blindness for 4...9...10 seconds."

Previously, Ebon Dust Aura worked on both ranged and melee attack skills, enabling characters that kept enemies Blinded from range without the vulnerability of casting spells. We're changing Ebon Dust Aura to only trigger from melee attacks. This skill will no longer be useful for ranged Blindness spamming. This change affects both PvE and PvP.


 * : increased casting time to 2 seconds; lowered duration to 8 seconds; lowered recharge to 5 seconds; decreased healing to 36...103...120.
 * : increased casting time to 2 seconds; increased Energy cost to 15.

These numerical adjustments are meant to make Illusion of Pain more vulnerable to disruption and more demanding to use effectively. Similarly, the increased casting time and Energy cost for Mirror of Ice makes it harder to maintain against opponents determined to stop it. These changes affect both PvE and PvP.


 * : increased Energy cost to 10; increased recharge to 15 seconds.

Although Winter's Embrace has the worst recharge-to-duration of any of the common Water Magic movement-speed hexes, its fast casting time and additional damage have been more than enough to make the skill worthwhile. We did not want to fundamentally change this skill or to remove it from play. In order to add a little pressure to combining this skill with the new, more expensive Mirror of Ice and the other 10-Energy hexes, we've increased the Energy cost from 5 to 10. We've also increased the recharge time from 12 to 15 to ensure that it does not outshine the Elite skill Icy Shackles, which had the same recharge time but a slower casting time. These changes affect both PvE and PvP.


 * : decreased Energy gain to 1...5...6; increased recharge to 25 seconds; changed duration to "5 seconds plus 3 seconds for every rank of Soul Reaping."

While examining Blood Spike, we took a hard look at Cultist's Fervor. After some consideration, we determined that this skill was not the key problem with Blood Spike, but it was too powerful. We've made a number of adjustments, and we are confident that players will still find this to be a powerful skill. First, we lowered the Energy gain by 1 point. This allows a little bit more Energy attrition over the course of a long battle. Additionally, we've increased the recharge time, making the skill more vulnerable to removal and disruption. Finally, we've associated its duration with Soul Reaping. This effectively restricts its use to primary Necromancers and it requires a slightly greater spread of attribute points when used in a Curses build. These changes affect both PvE and PvP.


 * : added standard teleport aftercast.

Unlike other teleport skills, Ride the Lightning did not have a long aftercast. Favored in some Air Elementalist spike builds, the short aftercast also made it a little too powerful in the hands of melee attackers, who were able to attack almost immediately after teleporting in. These changes affect both PvE and PvP.

GvG Restrictions
Based on feedback from the GvG community, we've eased the restrictions for participating in Guild Battles. The new rules are as follows:
 * Ladder Play:
 * Requires 2 guild members.
 * A guild member must be the party leader, and at least one guild officer must be in the party.
 * Henchmen are restricted to a maximum of 4.
 * Daily Automated Tournaments:
 * Requires 2 guild members who have been in your guild for at least 7 days.
 * A guild member must be the party leader, and at least one guild officer must be in the party.
 * Henchmen are restricted to a maximum of 4.
 * All party members must be registered with Tolkano.
 * Monthly Automated Tournaments:
 * Requires 4 guild members who have been in your guild for at least 7 days.
 * All players in the party must be members of your guild, and at least one guild officer must be in the party.
 * Henchmen are restricted to a maximum of 2.
 * All party members must be registered with Tolkano.
 * Removed Balthazar faction penalty for new guild members.

Codex Arena Changes
Looking at player response to the Codex Arena, we believe there are a number of changes necessary to improve and polish this format. These changes are intended to address the following issues:
 * Polish—Codex Rotation Time and Codex Skill List
 * General Balance Issues—Priest Maps, Protector's Defense, and Ebon Dust Aura
 * Balance between Professions—Physical Attackers vs. Spellcasters
 * Core Skill Combinations within Single Professions—Minions, Spirits, and Assassin Attack Chains

Polish

 * The Codex now changes every 6 hours.
 * Added a new sorting option in the Skill Panel (K). This option allows you to toggle between viewing all available Codex skills and only skills available to your profession combination.

Previously, the Codex changed once per day. This meant that players in certain time zones never got to enjoy the experience of playing right as a new Codex was introduced. We considered even shorter time intervals than every 6 hours but decided against that because whenever the Codex changes, further matches under the old Codex are disallowed, thus cutting off any win streaks for active players.

Since Codex Arena debuted, players have been clamoring for a way to see all Codex skills currently available. The new sorting option provides just that. We thought this was especially critical with the change to a 6 hour interval between Codex changes. With the 24 hour interval, our wiki community was valiantly maintaining an up-to-date list of the Codex skills, but we did not want to ask them to do so four times a day.

General Balance Issues
Due to limited offensive options and inconsistent access to utility skills, we keep a close eye on defensive skills in Codex Arena. We have identified two that were causing problems:


 * : changed skill's functionality (see above).
 * : restricted from Codex Arena.

For Ebon Dust Aura, the problem applies to any 4v4 play, not just Codex Arena. With that in mind, we made changes to the skill directly. For Protector's Defense, the problem is specific to Codex Arena. We made no changes to the skill itself, but instead added it to our restricted list. This skill will never appear in any Codex.

Limited defensive and healing options in Codex Arena caused another balance issue in maps with priests. Players often camp near their own priest to take advantage of his Heal Area ability. This leads to overly defensive battles. We made two changes to the priest to address this:


 * Removed Heal Area from the priest.
 * Increased priest Health to 800.

These changes allow us to maintain this game type in Codex Arena without encouraging players to use overly defensive strategies.

Balance Between Professions
Having observed the Codex in action for a while now, we believe we should increase the number of skills available to some professions and decrease the number of Elites available to others. The new totals are as follows:
 * 25 Codex skills (including 5 Elites): Elementalist, Necromancer, Mesmer
 * 23 Codex skills (including 4 Elites): Ranger
 * 20 Codex skills (including 4 Elites): Monk, Ritualist
 * 20 Codex skills (including 3 Elites): Assassin, Dervish, Warrior, Paragon

We chose these numbers on a profession-by-profession basis, using the following factors:
 * Compared to physical attackers, spellcasters need greater synergy between their skills in order to be effective. Thus, spellcasters need more skills in the Codex than physical attackers do.
 * Because Factions and Nightfall professions have smaller skill pools to draw from, they should have fewer skills, especially Elite skills, in the Codex in order to keep builds from being too stagnant.
 * Since offensive options are limited, it is important that defensive characters not be too powerful. Professions capable of healing and providing defensive buffs should have fewer skills available.
 * A slight increase in the number of available skills overall should make designing builds more interesting.

Core Skill Combinations
A few professions have a large number of skills meant to play off whole categories of skills. For example, Necromancers have skills that require having a minion. When no minion-summoning skills are present in the Codex, these other skills are effectively useless.

Previously, we addressed this problem for Assassins (but not Necromancers or Ritualists) by always providing Desperate Strike and Fox Fangs in every Codex. We have replaced this with a system that allows us to address the problem in a better way:


 * Codex Assassin skills will now be guaranteed to include one randomly selected Lead Attack and one randomly selected Off-Hand Attack. (Neither of these will be Elite, and the Off-Hand attack will be one that requires a Lead Attack.)
 * Codex Necromancer skills will now be guaranteed to include one randomly selected minion-summoning skill.
 * Codex Ritualist skills will now be guaranteed to include one randomly selected Binding Ritual.

Additional Lead Attacks, Off-Hand Attacks, minion-summoning skills, and Binding Rituals may appear in the Codex as before. These changes merely guarantee there will be at least one of each.