Jump to content
  • Sign Up

Auto Attack Data (or Is Engi's Rifle auto the worst in the game?)


bethekey.8314

Recommended Posts

Prompted by a comment, I'm now curious. How do melee/ranged/aoe/projectile/aimed auto attacks compare in a controlled way? What are the outliers? It's too much work for me to go through every class and weapon combo, but maybe with some community help?

How it works:

Clear your build template so you have no traits, sigils, relics, or runes.

No outside buffs or help.

For elite weapons, equip only the required elite traitline. No traits.

Only equip Berserker amulet for power, or Carrion/Rabid for conditions. I believe Carrion is typically higher damage.

Measure time to kill (3-5x, in seconds) the Light Golem in HOTM. Start the time when you press the first attack.

For melee attacks, stand in melee.

For ranged attacks, stand 900 distance away (or max, if lower). Many attacks probably have 1200+ range, this is a more realistic, happy medium. You can measure the distance by when the little red "out of range" indicator appears under skills with 900 range.

Report your class, the weapon, the amulet (if not obvious), the times to kill, and the average time to kill.

I'll start with Engineer. Like a leaderboard, in order of fastest to slowest time to kill (TTK):

Scrapper - Hammer - 8.09, 8.22, 9.23 = 8.51 s TTK

Engineer - Bomb kit - 8.82, 7.94, 7.83, 9.63 = 8.55 s TTK

Engineer - Tool kit - 9.32, 10.5, 8.09 = 9.3 s TTK

Engineer - Grenade kit (traited, 100% hit) - 9.64, 9.98, 9.71, 8.61, 8.56 = 9.3 s TTK

Holosmith - Sword - 11.98, 11.02, 11.03 = 11.34 s TTK

Mechanist - Mace (berserker) - 10.56, 15.82, 11.57, 12.21 = 12.54 s TTK

Engineer - Grenade kit (untraited, 100% hit) - 13.97, 10.5, 11.3, 13.24, 13.71 = 12.54 s TTK

Engineer - Flamethrower kit (berserker) - 12.5, 14.04, 13.96 = 13.5 s TTK

Engineer - Flamethrower kit (carrion) - 16.51, 16.26, 16.52 = 16.43 s TTK

Engineer - Pistol (carrion) - 17.12, 17.18, 17.19 = 17.16 s TTK

Engineer - Elixir Gun kit (berserker) - 18.64, 19.18, 19.2 = 19 s TTK

Engineer - Elixir Gun kit (carrion) - 19.6, 19.5, 19.6 = 19.56 s TTK

Engineer - Elixir Gun kit (rabid) - 23.13, 22.26, 22.64 = 22.67 s TTK

Engineer - Rifle - 27.29, 28.81, 26.03 = 27.38 s TTK

I don't expect others to do full class sweeps like me. Even if you just test one weapon, I'll try and add it to the leaderboard here.

With more data, it might be interesting to see how weapons cluster by the variables like class, melee/ranged, projectile/nonprojectile, aimed etc. Weapons that do damage through Confusion or Torment will obviously be skewed by non-moving, non-attacking enemies.

Edited by bethekey.8314
  • Like 1
  • Thanks 3
  • Confused 3
  • Sad 1
Link to comment
Share on other sites

Yea, but rifle autos are explosions, meaning when traited, they provide fury and fire aim assisted rockets, thus proccing an orbital strike which does 5k dmg on crit in PvP(massive AoE). Rifle autos hit so low because they have to. It was balance with traits in mind. 

Edited by WhoWantsAHug.3186
  • Thanks 3
  • Confused 3
Link to comment
Share on other sites

12 minutes ago, WhoWantsAHug.3186 said:

Yea, but rifle autos are explosions, meaning when traited, they provide fury and fire aim assisted rockets, thus proccing an orbital strike which does 5k dmg on crit in PvP(massive AoE). Rifle autos hit so low because they have to. It was balance with traits in mind. 

Requiring a trait for the auto attack, a major part of every weapon, is not how things are balanced. Other builds want to, and should be able to use rifle without Explosives.

Other, stronger auto attacks benefit from traits as well. Rifle is not alone in this.

  • Thanks 5
  • Confused 3
Link to comment
Share on other sites

1 minute ago, bethekey.8314 said:

Requiring a trait for the auto attack, a major part of every weapon, is not how things are balanced. Other builds want to, and should be able to use rifle without Explosives.

Other, stronger auto attacks benefit from traits as well. Rifle is not alone in this.

Yea let's buff rifle so it's busted with explosives and viable without it!.....

This is why anet shouldn't take advice from the forums. 

  • Like 3
  • Thanks 2
  • Confused 2
Link to comment
Share on other sites

17 minutes ago, WhoWantsAHug.3186 said:

Yea let's buff rifle so it's busted with explosives and viable without it!.....

This is why anet shouldn't take advice from the forums. 

Rifle can be changed without having it be overpowered. A simple a change as reverting it to it's old, single shot self would improve DPS while also reducing the Aim-Assisted Rocket interaction.

Simple mindedness and thinking in extremes is why forums are bad. But I'm not the one contributing to that in this case.

  • Like 4
  • Thanks 1
  • Confused 4
Link to comment
Share on other sites

@Kuma.1503 had an opinion on this that was adjacent to simmering annoyance at the fact that rifle auto was buffed, then nerfed to be worse than it was originally. As a non-engie main, I'm inclined to agree with that for what it's worth. 

I didn't mind the quick doubleshot or the old single shot. What I -do- mind is when balancing ventures make the skill worse than it was before the venture. If you're going to rework a skill that isn't broken, your end result shouldn't feel worse or be worse to use. If that's the result you would have been better off not exerting the energy to annoy the players who are using it. 

Edited by Azure The Heartless.3261
  • Like 7
  • Thanks 1
Link to comment
Share on other sites

1 hour ago, WhoWantsAHug.3186 said:

Yea, but rifle autos are explosions, meaning when traited, they provide fury and fire aim assisted rockets, thus proccing an orbital strike which does 5k dmg on crit in PvP(massive AoE). Rifle autos hit so low because they have to. It was balance with traits in mind. 

Which is why Grenades, which do all of those things, kill in a 1/3rd of the time!

Wait...

  • Like 2
Link to comment
Share on other sites

1 hour ago, bethekey.8314 said:

Requiring a trait for the auto attack, a major part of every weapon, is not how things are balanced. Other builds want to, and should be able to use rifle without Explosives.

Other, stronger auto attacks benefit from traits as well. Rifle is not alone in this.

OK, but then rifle autos won't provide fury via explosions. So you're back to square 1 for why would you ever want to use rifle autos. Your argument is pointless and circular. I'm perfectly fine with having a traitline providing extreme optimization for DPS builds using a certain weapon. Hello pistol and Firearms, or mesmer with greatsword and domination.

Spamming weapon autos on engi is usually never optimal anyway. Engi just needs some serious reworks on its kits, so you can swap around for your short CD 2 skill. 

Shortbow is on the way and I have high hopes for it and it won't have a traitline to buff it! Hopefully. Maybe they'll be explosives and you still take explosions bahaha

Link to comment
Share on other sites

To "fix" the perceived issue, the devs need to remove the explosion skill fact from the rifle AA and buff the coefficient accordingly. In other word it mean reverting the AA to what it was before the change. However, 11 years of GW2 balance show that the balance team would rather tweak numbers around a perceived issue than work directly on the source of the perceived issue.

38 minutes ago, WhoWantsAHug.3186 said:

Shortbow is on the way and I have high hopes for it and it won't have a traitline to buff it!

Shortbow can very well carry explosion skillfacts on it's skills (just like rifle does) and be balanced around this fact. Keep in mind that the dev team manage to create catalyst which basically revolve around aura like tempest. Once the devs take a liking to a mechanism, they tend to exageratly make things gravitate around those mechanisms.

Heck! Despite repeated failures with consumable bundles droped by skills/traits/effects over the years they never fail to grasp an opportunity to try to introduce this kitten mechanism into the game (They managed to make a relic do that...). Sometime it feel like they never learn from their mistakes.

Another example of that is torment and necromancer. They've significantly abused torment on necromancer's last 2 elite specs which result to unreasonable amount of damage output once one e-spec got access to the tool of the other. Wouldn't it have been more reasonable to focus on different conditions for each spec? I mean, there is 5 damaging conditions in the game and only 3 e-specs per profession yet both Scourge and Harbinger manage to focus heavily on a single one of them.

Edited by Dadnir.5038
  • Like 2
Link to comment
Share on other sites

3 hours ago, Dadnir.5038 said:

To "fix" the perceived issue, the devs need to remove the explosion skill fact from the rifle AA and buff the coefficient accordingly. In other word it mean reverting the AA to what it was before the change. However, 11 years of GW2 balance show that the balance team would rather tweak numbers around a perceived issue than work directly on the source of the perceived issue.

They generally don't revert changes, even when the previous implementation was working better. Or when X makes - among other things - Y overpowered, they nerf Y (which isn't the problem), then later nerf X, but don't undo the nerf to Y (which never needed adjustments).

I rarely wish for ANet to balance anything anymore over worrying they'll "balance" it.

  • Like 2
  • Thanks 1
  • Haha 2
Link to comment
Share on other sites

12 hours ago, bethekey.8314 said:

Prompted by a comment, I'm now curious. How do melee/ranged/aoe/projectile/aimed auto attacks compare in a controlled way? What are the outliers? It's too much work for me to go through every class and weapon combo, but maybe with some community help?

How it works:

Clear your build template so you have no traits, sigils, relics, or runes.

No outside buffs or help.

For elite weapons, equip only the required elite traitline. No traits.

Only equip Berserker amulet for power, or Carrion/Rabid for conditions. I believe Carrion is typically higher damage.

Measure time to kill (3-5x, in seconds) the Light Golem in HOTM. Start the time when you press the first attack.

For melee attacks, stand in melee.

For ranged attacks, stand 900 distance away (or max, if lower). Many attacks probably have 1200+ range, this is a more realistic, happy medium. You can measure the distance by when the little red "out of range" indicator appears under skills with 900 range.

Report your class, the weapon, the amulet (if not obvious), the times to kill, and the average time to kill.

I'll start with Engineer. Like a leaderboard, in order of fastest to slowest time to kill (TTK):

Scrapper - Hammer - 8.09, 8.22, 9.23 = 8.51 s TTK

Engineer - Bomb kit - 8.82, 7.94, 7.83, 9.63 = 8.55 s TTK

Engineer - Tool kit - 9.32, 10.5, 8.09 = 9.3 s TTK

Engineer - Grenade kit (traited, 100% hit) - 9.64, 9.98, 9.71, 8.61, 8.56 = 9.3 s TTK

Holosmith - Sword - 11.98, 11.02, 11.03 = 11.34 s TTK

Mechanist - Mace (berserker) - 10.56, 15.82, 11.57, 12.21 = 12.54 s TTK

Engineer - Grenade kit (untraited, 100% hit) - 13.97, 10.5, 11.3, 13.24, 13.71 = 12.54 s TTK

Engineer - Flamethrower kit (berserker) - 12.5, 14.04, 13.96 = 13.5 s TTK

Engineer - Flamethrower kit (carrion) - 16.51, 16.26, 16.52 = 16.43 s TTK

Engineer - Pistol (carrion) - 17.12, 17.18, 17.19 = 17.16 s TTK

Engineer - Elixir Gun kit (berserker) - 18.64, 19.18, 19.2 = 19 s TTK

Engineer - Elixir Gun kit (carrion) - 19.6, 19.5, 19.6 = 19.56 s TTK

Engineer - Elixir Gun kit (rabid) - 23.13, 22.26, 22.64 = 22.67 s TTK

Engineer - Rifle - 27.29, 28.81, 26.03 = 27.38 s TTK

I don't expect others to do full class sweeps like me. Even if you just test one weapon, I'll try and add it to the leaderboard here.

With more data, it might be interesting to see how weapons cluster by the variables like class, melee/ranged, projectile/nonprojectile, aimed etc. Weapons that do damage through Confusion or Torment will obviously be skewed by non-moving, non-attacking enemies.

I applaud your endeavor. I hope that others take up the challenge.

  • Like 1
Link to comment
Share on other sites

12 hours ago, bethekey.8314 said:

Prompted by a comment, I'm now curious. How do melee/ranged/aoe/projectile/aimed auto attacks compare in a controlled way? What are the outliers? It's too much work for me to go through every class and weapon combo, but maybe with some community help?

How it works:

Clear your build template so you have no traits, sigils, relics, or runes.

No outside buffs or help.

For elite weapons, equip only the required elite traitline. No traits.

Only equip Berserker amulet for power, or Carrion/Rabid for conditions. I believe Carrion is typically higher damage.

Measure time to kill (3-5x, in seconds) the Light Golem in HOTM. Start the time when you press the first attack.

For melee attacks, stand in melee.

For ranged attacks, stand 900 distance away (or max, if lower). Many attacks probably have 1200+ range, this is a more realistic, happy medium. You can measure the distance by when the little red "out of range" indicator appears under skills with 900 range.

Report your class, the weapon, the amulet (if not obvious), the times to kill, and the average time to kill.

I'll start with Engineer. Like a leaderboard, in order of fastest to slowest time to kill (TTK):

Scrapper - Hammer - 8.09, 8.22, 9.23 = 8.51 s TTK

Engineer - Bomb kit - 8.82, 7.94, 7.83, 9.63 = 8.55 s TTK

Engineer - Tool kit - 9.32, 10.5, 8.09 = 9.3 s TTK

Engineer - Grenade kit (traited, 100% hit) - 9.64, 9.98, 9.71, 8.61, 8.56 = 9.3 s TTK

Holosmith - Sword - 11.98, 11.02, 11.03 = 11.34 s TTK

Mechanist - Mace (berserker) - 10.56, 15.82, 11.57, 12.21 = 12.54 s TTK

Engineer - Grenade kit (untraited, 100% hit) - 13.97, 10.5, 11.3, 13.24, 13.71 = 12.54 s TTK

Engineer - Flamethrower kit (berserker) - 12.5, 14.04, 13.96 = 13.5 s TTK

Engineer - Flamethrower kit (carrion) - 16.51, 16.26, 16.52 = 16.43 s TTK

Engineer - Pistol (carrion) - 17.12, 17.18, 17.19 = 17.16 s TTK

Engineer - Elixir Gun kit (berserker) - 18.64, 19.18, 19.2 = 19 s TTK

Engineer - Elixir Gun kit (carrion) - 19.6, 19.5, 19.6 = 19.56 s TTK

Engineer - Elixir Gun kit (rabid) - 23.13, 22.26, 22.64 = 22.67 s TTK

Engineer - Rifle - 27.29, 28.81, 26.03 = 27.38 s TTK

I don't expect others to do full class sweeps like me. Even if you just test one weapon, I'll try and add it to the leaderboard here.

With more data, it might be interesting to see how weapons cluster by the variables like class, melee/ranged, projectile/nonprojectile, aimed etc. Weapons that do damage through Confusion or Torment will obviously be skewed by non-moving, non-attacking enemies.

No mortar!

Link to comment
Share on other sites

7 hours ago, Dadnir.5038 said:

To "fix" the perceived issue, the devs need to remove the explosion skill fact from the rifle AA and buff the coefficient accordingly. In other word it mean reverting the AA to what it was before the change. However, 11 years of GW2 balance show that the balance team would rather tweak numbers around a perceived issue than work directly on the source of the perceived issue.

Shortbow can very well carry explosion skillfacts on it's skills (just like rifle does) and be balanced around this fact. Keep in mind that the dev team manage to create catalyst which basically revolve around aura like tempest. Once the devs take a liking to a mechanism, they tend to exageratly make things gravitate around those mechanisms.

Not everyone perceives rifle as having an issue. Maybe it's AA could do slightly more dmg, but the explosion component provides trait synergy and tradeoffs. Any tier 2 and 3 trait in explosives can effectively be taken with rifle. The whole idea of balance is based around tradeoffs. 

Guessing shortbows functionality is just speculation, but yes I'm sure there will be some type of synergy with both explosives and Firearms. I'm leaning on it being a condi weapon as engi has little ranged AoE condi dmg options 

Link to comment
Share on other sites

15 hours ago, WhoWantsAHug.3186 said:

Yea, but rifle autos are explosions, meaning when traited, they provide fury and fire aim assisted rockets, thus proccing an orbital strike which does 5k dmg on crit in PvP(massive AoE). Rifle autos hit so low because they have to. It was balance with traits in mind. 

Rifles and Explosions have nothing to do with Aim-Assisted Rocket or the Orbital Strike it unleashes. Aim-Assisted rockets work with pistol just fine as well.

 

Link to comment
Share on other sites

Just so we're clear, I made this same test a very long time ago. Things I noticed

1) Ranged AA are weaker than melee AA, unless some weird shenanigans happen (eg: bouncy projectiles on thief sb)
2) Thief dagger has the single strongest AA in the game
3) Mesmer sword, even when striking boonless foes, is very very average

I didn't test engineer specifically back then, just a quick rundown among some classses (guardian, mesmer, thief, elementalist).

Link to comment
Share on other sites

1 hour ago, Lan Deathrider.5910 said:

FWIW, on the wiki on the first hit in the chain they have the average execution time of the chain. You can then total the tooltip damage from the chain and divide by the chain cast time for a dps/s of the chain based on the tooltips for comparison. You would add in the base condi damage to that for condis.

It gets complicated when there are crit probabilities, aoe bounces, crit from passive elite traitlines etc. With all the variables and swapping amulets it felt easier to just take some time/dps measurements in game.

 

Link to comment
Share on other sites

42 minutes ago, bethekey.8314 said:

It gets complicated when there are crit probabilities, aoe bounces, crit from passive elite traitlines etc. With all the variables and swapping amulets it felt easier to just take some time/dps measurements in game.

 

Sure, but the melee ones also cleave. AA in a vacuum is a dangerous comparison. Better to compare the bar itself to its aa for an opportunity cost breakdown.

Link to comment
Share on other sites

The analysis for most things except Ele (I'm not doing all the Ele combinations lol): https://imgur.com/a/MDQjF84

Results:

- The average auto attack time to kill was ~15 seconds using the methodology above. Range was 7.2 seconds (Guardian Hammer) to 29.7 seconds (Necromancer Staff).

- Melee (avg TTK 11.4 sec) typically does higher damage than Ranged (avg TTK 18.4 sec).

- Among Ranged options, non-projectiles performed above average (avg TTK 17.4 sec) despite having less counters.

- Engineer's Rifle (avg TTK 27.38 sec) and Necromancer's Staff (29.7 sec) / Pistol (27.9 sec) are clear DPS outliers on their own. Traits / side effects (e.g. Life Force gain) must be large to justify such poor performance.

- Whether a Ranged attack has AoE (TTK 18.15 sec) or not (18.6 sec) does not seem to affect average DPS.

- Warrior and Guardian have the lowest average TTK auto attack options. Necromancer had the highest average TTK by far. Revenant was most consistent, with nearly every weapon option having lower TTK than average.

- Berserker amulet generally yielded lower or comparable auto attack TTK to Carrion amulet, even on condition weapons. Clearly, the other weapon skills/utilties would ultimately dictate which amulet you choose.

- Infamous Grenade kit is the only aimed auto attack in the pool I tested, or possibly, the game. Traited, it has the best Ranged auto TTK so long as 100% of the grenades hit over many throws. This is unrealistic. Overall, Grenades are likely the best Ranged AoE option but Revenant shortbow is best single target.

There may be a couple label errors, but overall I think this is fairly accurate.

This data isn't meant to call for nerfs/buffs directly. Clearly other factors (traits, equipment, situation) influence how much damage a weapon does. Or put another way, theoretical damage isn't practical damage.

Still, auto attacks are powerful and ever-present tools in builds. Their lack of cooldown makes it so that a strong auto attack can define a whole build (Old Flamethrower Engi, Old Rifle Mech, OWP Staff Druid etc.). I believe outliers are meaningful and worth looking into.

Edited by bethekey.8314
  • Like 3
Link to comment
Share on other sites

I think condi autos are much, much worse than your testing shows, given how often players cleanse. As such, pistol is probably worse than rifle in practice. Likewise, melee autos are much worse than ranged autos in practice, since you'll be in range way less often.

  • Like 2
Link to comment
Share on other sites

6 minutes ago, coro.3176 said:

I think condi autos are much, much worse than your testing shows, given how often players cleanse. As such, pistol is probably worse than rifle in practice. Likewise, melee autos are much worse than ranged autos in practice, since you'll be in range way less often.

Yes, this thought came up while I was doing the testing. Here, all of the conditions are allowed to tick for as long as possible. Just like with aimed Grenades, 100% uptime is unrealistic for conditions as well.

You could probably estimate how the condition weapons perform in practice by multiplying these TTKs by a factor reflecting the % of total damage you expect to stick. The power component will remain the same of course. Weapons that apply multiply conditions (Soulbeast Dagger was surprisingly good) would be more resistant to full cleanse too.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...