Guild Wars 2

The firing speed issue with ranged weapons? I’ve been all over the profession boards posting about it. It’s really very obvious…

Rifles have a 3/4 second activation speed, and roughly a .8 recast on their #1 skills, which is about appropriate. Consequently, rifles feel pretty strong to professions that use it.

All other ranged weapons (with the possible exception of Shortbow, which is in the middle somewhere) – Pistols, Longbows, Staffs, Scepters, all either have the same activation speed but slower recast as the rifle, or have a faster activation speed but the same recast.

Because of an issue with the rate of fire of their #1 skill, the following weapon sets are all noticeably weaker than they should be; it’s either a bug with the activation or a design issue with animations or the GCD:

Thief – MH Pistol
Engineer – MH Pistol
Ranger – Longbow
Warrior – Longbow

I haven’t extensively tested Staffs and Scepters yet, but I strongly suspect they are affected as well. Anyone have any thoughts or insight on this?

There’s also the travel speed of the projectile.

The guardian scepter moves so slowly that if your target moves at all (not even dodge, just a step to the left or right), it will miss.

mulch.2586:

There’s also the travel speed of the projectile.

The guardian scepter moves so slowly that if your target moves at all (not even dodge, just a step to the left or right), it will miss.

What? Since when? It tracked back in the beta, did they break it or something?

Well Staves and Scepters are a whole other ballpark as their attacks can cause markedly different effects from profession to profession.

In regards to the basic attack speed, yes, it’s roughly the same between pistols andrifles for the primary attack. What makes pistols and other ranged weapons better is the skill diversity they bring to the table. The primary attack isn’t that important overall.

Conncept.7638:

mulch.2586:

There’s also the travel speed of the projectile.

The guardian scepter moves so slowly that if your target moves at all (not even dodge, just a step to the left or right), it will miss.

What? Since when? It tracked back in the beta, did they break it or something?

It still tracks as long as you have it targeted. Though you can move out of line of sight and lose the targeting.

Vermillion Hawk.9037:

Well Staves and Scepters are a whole other ballpark as their attacks can cause markedly different effects from profession to profession.

In regards to the basic attack speed, yes, it’s roughly the same between pistols andrifles for the primary attack. What makes pistols and other ranged weapons better is the skill diversity they bring to the table. The primary attack isn’t that important overall.

No, not really. Any weapon set’s #1 skill sets the baseline sustained damage potential of that set, with the other skills serving a tactical or supplementary role, and this is exactly how it is with melee weapons as well, and skills are no more diverse among ranged weapons than they are among melee weapons. In particular, t’s a huge problem that the Pistol fires at the same speed as the rifle, because it isn’t supposed to. If what you were saying was true, Rifles would fire more slowly than Pistols to compensate for being a lot stronger, and they don’t. This makes it clear to me that there’s some unintended design issue with the way recasting works on non-chain attacks.

MH Pistol is by far the worst weapon for thiefs simply because the DPS is horrendous, and even though it’s intended to be a condition weapon, the only way to make it halfway viable is to stack power and spam Unload, which erases any potential utility it might have due to the Initiative sink. It is very clearly not working as intended.

There’s also a dramatic difference in the recast time of Longbows and Rifles (1.2 sec. and .8 sec), leading to the latter being objectively stronger than the former, since #1s on Longbows are not any stronger to compensate for the attack speed difference. This is largely why there’s a fairly strong consensus that warriors are better ranged fighters than rangers.

But with 1h weapons you can have 2 sigil vs 2h weapons have only 1 sigil so 2h weapons may be stronger but 1h weapons have a great deal more flexibility. That alone should even out every thing about guns and ranges weapons.

Something that has been brought up on the Guardian boards a few times is that while our ranged attack has a 1/4s attack refresh, it is in actuality closer to 1 1/4 second attack refresh.

Many of our chain attacks seems to be more aligned to a 1s attack tempo regardless of what the tool tip says.

So fast attacking weapons that strike faster than 1s get tied to a 1s interval which affects the perceived effectiveness that you would think it would have via the tool tips.

CMF.5461:

Something that has been brought up on the Guardian boards a few times is that while our ranged attack has a 1/4s attack refresh, it is in actuality closer to 1 1/4 second attack refresh.

Many of our chain attacks seems to be more aligned to a 1s attack tempo regardless of what the tool tip says.

So fast attacking weapons that strike faster than 1s get tied to a 1s interval which affects the perceived effectiveness that you would think it would have via the tool tips.

Yes, and to be honest I think this even duped the designers. I am pretty sure that when skills were being mathematically balanced early on, they were balanced around their activation speeds in a vacuum based on a single cast, not taking into account any recovery/recast delays that slow down the rate of fire artificially for non-chain attacks.

What makes the least sense to me is that Rifle attacks are fairly well synchronized with their expected activation speed, but nothing else is.

Rifles have a 3/4 second activation and a .8 recast
Pistols have 1/2 second activation but a .8 recast
Longbows have a 3/4 sec activation with a 1.2 recast
Scepters and Staffs vary across professions but tend to take a ful .5 – 1 second longer than their activation timers would suggest to recast.

I would stake my life that it was a design oversight with the way animations and skill responsiveness work, and that is the singular reason those weapon sets are so often complained about on the profession boards.

In my opinion, the ranger longbow would be fine if they increased the arrows flight speed a bit and changed/removed that horrendous distance bonus; make it a cripple at certain distances and from the sides/behind or something at least… or something that makes a little more sense.

I just did some rough testing of guardian scepter, thief pistol, and warrior rifle slot 1 attacks. It’s not 100% accurate, but counting the number of attacks in x time frame (I used 30 seconds) gives enough information to determine how the three relate to each other.
Scepter and pistol were close to equal, each getting in 34-35 attacks before time ran out. The scepter listed activation time is 1/4, while the pistol is 1/2 second. There is a clear difference in the cooldowns of these two weapons. Given their activation times, if the cooldowns were the same the scepter would be hitting many more times in the 30 second window.

Rifle was a bit slower, only getting 30-31 attacks in 30 seconds. With it’s 3/4 second activation, the cooldown must be significantly shorter to retain this fire rate.

As for the damage comparisons, I went for the most simple option of taking the stats listed on wiki and multiplying damage+condition damage by the number of shots. For rifle this was approx. 12,200; pistol was 10,336; scepter was 6720. There are some huge differences here, but as someone mentioned above, you cannot compare weapons based on a single skill. The other skills have widely varying DPS that has to be taken into account. Scepter #2 skill deals over 1600 damage to multiple targets with a 6 second cooldown, while rifle slot 2 does 135 damage (+cripple) with a 10 second cooldown. That damage difference actually makes the combined DPS of slots 1 and 2 rather close to equal for scepter (guardian) and rifle (warrior). The thief’s lack of cooldown and reliance on a regenerating resource makes this type of comparison rather difficult, so I left it out of this portion.

I do still think that the skill descriptions vs the actual timing need to be looked at, if only to make the descriptions more accurately reflect actual speeds. Accurate listed times and a game-wide standard cooldown would make on-paper comparions much easier to do, and with greater accuracy versus having to measure and determine varying cooldowns.

On the most recent AMA post on reddit, someone posed this question to a developer and they asked the poster to contact him (the developer) privately to share data/info about this issue and to discuss it further.

I found this to be extremely out of character with the rest of the fairly-useful dialogue where the dev was responding to questions openly.

Would like to know what came of the conversation, if one took place…

mrstealth.6701:

I just did some rough testing of guardian scepter, thief pistol, and warrior rifle slot 1 attacks. It’s not 100% accurate, but counting the number of attacks in x time frame (I used 30 seconds) gives enough information to determine how the three relate to each other.
Scepter and pistol were close to equal, each getting in 34-35 attacks before time ran out. The scepter listed activation time is 1/4, while the pistol is 1/2 second. There is a clear difference in the cooldowns of these two weapons. Given their activation times, if the cooldowns were the same the scepter would be hitting many more times in the 30 second window.

Rifle was a bit slower, only getting 30-31 attacks in 30 seconds. With it’s 3/4 second activation, the cooldown must be significantly shorter to retain this fire rate.

As for the damage comparisons, I went for the most simple option of taking the stats listed on wiki and multiplying damage+condition damage by the number of shots. For rifle this was approx. 12,200; pistol was 10,336; scepter was 6720. There are some huge differences here, but as someone mentioned above, you cannot compare weapons based on a single skill. The other skills have widely varying DPS that has to be taken into account. Scepter #2 skill deals over 1600 damage to multiple targets with a 6 second cooldown, while rifle slot 2 does 135 damage (+cripple) with a 10 second cooldown. That damage difference actually makes the combined DPS of slots 1 and 2 rather close to equal for scepter (guardian) and rifle (warrior). The thief’s lack of cooldown and reliance on a regenerating resource makes this type of comparison rather difficult, so I left it out of this portion.

I do still think that the skill descriptions vs the actual timing need to be looked at, if only to make the descriptions more accurately reflect actual speeds. Accurate listed times and a game-wide standard cooldown would make on-paper comparions much easier to do, and with greater accuracy versus having to measure and determine varying cooldowns.

Thanks for the data, interesting stuff. I would still bet almost everything that I own that the activation speeds are meant to follow the recast more closely than they do (if not identically), and the difference in the number of shots per 30 seconds should be greater between the three weapons. I also think it’s the primary reason for the DPS difference between the three weapons you tested.

As for the #2 skill on the scepter, it is not as spectacular as you implied (the damage is spread out and it will never do that much to a single target). The fact is that what you provided is, proportionally speaking, a rough estimate of the type of sustained damage these weapons will do in a protracted fight, especially fighting from a distance like intended by the weapon. So the rifle has significantly higher DPS than the pistol, and both have dramatically higher than the scepter.

As for the Thief’s pistol, here’s the real problem with it. The lack of cooldowns and reliance on initiative mans that you can push the Pistol’s DPS up to get close to the Rifle’s, ONLY by stacking power (even though it’s a condition weapon) and continually dumping all of your Initiative to spam Unload, which removes any utility potentially provided by the set. The warrior does not need to do that to get maximal amount of damage and utility from their rifle, and it correctly benefits decently from both condition and power.

In short, it is clear to me that the pistol and the scepter are both weaker than they are intended to be, and it’s because some issue is significantly impairing their firing speed. The Pistol should be noticeably faster than the rifle and the scepter should be noticeably faster than the pistol, and that isn’t the case.

I totally agree that this cooldown/animation issue needs to be seriously investigated and at the very least some light needs to be shed on what’s up with the misleading activation speeds, etc.

I did a bit more digging and playing around with skills. As you brought up, guardian’s slot 2 scepter skill doesn’t give all 15 hits to a single target. You’ll hit a single stationary target 8-10 times. This leaves the scepter a bit short in comparison with the warrior’s first two rifle slots, if your target actually stands still and takes the hits.. Not being satisfied with the partial data, I decided that I wanted to see how the rest of the slots stacked up.

Using the same damage per 30 seconds data as before, I figured the damage for the rifle’s 5 slots and scepter’s 3 combined with both a focus and torch offhand. The rifle’s average at base attributes should be about 16,337 damage. With a focus, the damage falls well short of rifle, at 13,048. However it should be noted that the focus damage is also relianted on slot 4 bouncing and hitting the target twice (which it will in a close range 1v1 fight) and slot 5 being allowed to detonate. Slot 5 is also a defensive skill, so the utility of the focus can be expected to lower it’s damage output for the sake of balance. The torch, however, gave some surprising results at dealing 19,603 damage with slot 4 being used only for burning, ignoring its chain skill. Unfortunately, both torch skills require much shorter range (150 and 400) to be of any use. In comparison, only rifle #5 is melee range, but that skill knocks back the target, quickly returning the range gap.

The scepter has the potential to be the higher damage set, but that is going to require getting into melee range for a large portion of the damage, which comes from the torch and its burning. Using both weapons as ranged-only, scepter/focus ignoring skill 5, and rifle ignoring skill 5, the rifle still wins out at 15,602 vs 12,373 damage every 30 seconds.

Out of these two, I have to agree that the warrior’s rifle is a better ranged weapon (damage-wise) than the guardian’s scepter, largely due to the offhand options lacking ranged damage. In order to close that gap, you need to get the guardian into melee range, at which point you are likely better off simply swapping to a melee weapon.

All factors being considered (utility, virtues, etc), I think it’s still rather balanced. However, in a situation such as a big event/wvw assault where defense/utility can easily go unused if you’re never targeted, the warrior still has the advantage of more damage and an easier time getting credit for kills.

I’d be interested if you could test with Elementalist scepter and staff too if possible, the projectile speeds and cast rate are ridiculously slow on staff, things get interesting when you compare scepter auto attack in Air/Earth as opposed to Fire/Water.

Shlamorel.8714:

On the most recent AMA post on reddit, someone posed this question to a developer and they asked the poster to contact him (the developer) privately to share data/info about this issue and to discuss it further.

I found this to be extremely out of character with the rest of the fairly-useful dialogue where the dev was responding to questions openly.

Would like to know what came of the conversation, if one took place…

I would never suggest publicly posting anything too statistical for fear of the community just nit-picking over everything in an attempt to find relevance. It just invites every person and their dog the chance to say that they know the best for the game and ’Here’s what the devs should do …’ and subsequent upset when their suggestions for the mechanical systems aren’t implemented, likely due to all manner of things. I’m surprised that privately talking about stats and such and the reasoning behind them was even suggested.

Saying that, I’m even more surprised that the mathematical reasoning behind the firing speed of weapons hasn’t already been figured out by some intrepid analyst considering the wealth of similar information on GW2’s systems out there.

Einlanzer.1627:

The firing speed issue with ranged weapons? I’ve been all over the profession boards posting about it. It’s really very obvious…

Rifles have a 3/4 second activation speed, and roughly a .8 recast on their #1 skills, which is about appropriate. Consequently, rifles feel pretty strong to professions that use it.

All other ranged weapons (with the possible exception of Shortbow, which is in the middle somewhere) – Pistols, Longbows, Staffs, Scepters, all either have the same activation speed but slower recast as the rifle, or have a faster activation speed but the same recast.

Because of an issue with the rate of fire of their #1 skill, the following weapon sets are all noticeably weaker than they should be; it’s either a bug with the activation or a design issue with animations or the GCD:

Thief – MH Pistol
Engineer – MH Pistol
Ranger – Longbow
Warrior – Longbow

I haven’t extensively tested Staffs and Scepters yet, but I strongly suspect they are affected as well. Anyone have any thoughts or insight on this?

Yep we’ve definitely noticed and commented on it multiple times in the discussion, thief, engineer, ranger forums.

At least Warriors do decent damage with theirs and can fire off two arrows in a row…..

The supposedly more agile ranger can’t even do that.

The supposedly more agile thief can’t ricochet without putting a huge number of nerfing points into a tier on the traits screen that has nothing to do with bullets or damage just to be able to hit multiples with pistols, so then the only other option for any kind of AOE is the terribly designed shortbow. Why would anyone design a shortbow attack to be even slower then the pistol on a class that survives in pve on speed alone?

And to those who theorycraft, the math all points to the same thing, it severely lowers dps for the adventurer classes by making their main attacks slower. It needs fixing. My pistol on my engineer and thief should not be as slow as my shotgun.

The wiki already has what appears to be some accurate attack timings listed for ele skills. I’ll post my own rough estimates, which are consistent with the data on the skill pages.
As before, results are based on a 30 second clock. The number of attacks and amount of damage dealt in this time.

Scepter
Fire skill 1 – 19-20 attacks – 14383 damage
Earth skill 1 – 15-16 attacks – 13590 damage

Staff
Fire skill 1 – 21-22 attacks – 5966 damage
Earth skill 1 – 21-22 attacks – 3885

The gap between staff and scepter auto-attack damage is huge. The majority of scepter damage is from burning and bleeding, while the staff attacks are relying on direct damage only. I don’t really know how to compare condition damge to direct damage. If the target remains alive and suffers the full duration of the condition, it’s pretty straight-forward. This is likely to happen with the fire scepter’s burning, as its duration is short (2 seconds).

What I wasn’t expecting to see was how slow the earth scepter is, and how fast the staff attacks are for both elements. I was expecting staff to be the slower weapon in both cases. The animations make it feel slower, but that’s certainly not the reality of the numbers.

mrstealth.6701:

The wiki already has what appears to be some accurate attack timings listed for ele skills. I’ll post my own rough estimates, which are consistent with the data on the skill pages.
As before, results are based on a 30 second clock. The number of attacks and amount of damage dealt in this time.

Scepter
Fire skill 1 – 19-20 attacks – 14383 damage
Earth skill 1 – 15-16 attacks – 13590 damage

Staff
Fire skill 1 – 21-22 attacks – 5966 damage
Earth skill 1 – 21-22 attacks – 3885

The gap between staff and scepter auto-attack damage is huge. The majority of scepter damage is from burning and bleeding, while the staff attacks are relying on direct damage only. I don’t really know how to compare condition damge to direct damage. If the target remains alive and suffers the full duration of the condition, it’s pretty straight-forward. This is likely to happen with the fire scepter’s burning, as its duration is short (2 seconds).

What I wasn’t expecting to see was how slow the earth scepter is, and how fast the staff attacks are for both elements. I was expecting staff to be the slower weapon in both cases. The animations make it feel slower, but that’s certainly not the reality of the numbers.

Generally speaking and based on the above, I’m not surprised that damage caused by the sceptre is more direct and the staff less so, or for the sceptre to provide additional utility and support options. That duality is fairly common route to follow in order to clearly give alternative options for both choices, but the more subjective defensive/options make weapon vs weapon comparison kinda tricky – a ‘weaker’ weapon may be worthwhile to a player if given a particular defensive option that is useless to another. And really, attack speed is all about damage over time. My main is an Engineer and the difference between a two-handed weapon (rifle) and dual-wielding weapons (pistols) is very similar – one clearly focuses on direct damage, while the other on conditional damage over time that is actually made more considerable because its base damage is less and kills things by itself, without relying on conditional damage, less often.

That the staff is also two-handed, while the sceptre allows the user to hold a dagger or focus also gives the sceptre a little more variability, adding options to split and differentiate stats rather than lump all bonuses into fewer areas.

mrstealth.6701:

Staff
Fire skill 1 – 21-22 attacks – 5966 damage
Earth skill 1 – 21-22 attacks – 3885

I have no idea how this wiki information can even be accurate.

I’m looking at my auto-attack tooltips right now as I type this and the fire auto-attack is listed as having a 1 second cast time, while the Earth (and air and water) auto-attacks all have a 3/4 of a second cast time.

I don’t see how the fire and earth attunements can have the same amount of auto-attacks in a 30 second window when the tooltips themselves have a 25% difference in cast time.

If I’m reading anything wrong here please fill me in.

P.S. I realize you’re copy and pasting data from wiki – I’m not shooting you, the messenger

Shlamorel.8714:

mrstealth.6701:

Staff
Fire skill 1 – 21-22 attacks – 5966 damage
Earth skill 1 – 21-22 attacks – 3885

I have no idea how this wiki information can even be accurate.

I’m looking at my auto-attack tooltips right now as I type this and the fire auto-attack is listed as having a 1 second cast time, while the Earth (and air and water) auto-attacks all have a 3/4 of a second cast time.

I don’t see how the fire and earth attunements can have the same amount of auto-attacks in a 30 second window when the tooltips themselves have a 25% difference in cast time.

If I’m reading anything wrong here please fill me in.

P.S. I realize you’re copy and pasting data from wiki – I’m not shooting you, the messenger

The wiki shows the difference you are talking about. This is exactly the problem we are highlighting, the activation speeds of the weapons represent functionally useless data and there’s some design issue making the firing speed on ranged weapons really erratic in comparison to the supposed activation speeds. This is why a lot of ranged weapons in the game feel weaker than they should be.

So, basically, Earth, Air and Water autoattacks are supposed to be faster than fire autoattacks with the staff, but they aren’t, which is probably a major component of why you see people running fire way more often than any of the others.

There’s also a problem with the camera affecting whether abilities fire off.

It’s felt the most by engineers using FT/EG/p/p because any cone attack will cause this problematic to either miss every shot or completely not fire at all and one has to wait 4 seconds before attempting it again. It’s really not helpful in the slightest especially against mobs like those in Orr that run all over the place.

I experience this on my thief all the time when i try to use #3 skill unload. It just simply doesn’t fire. whether I’m manually using the mouse or using the keyboard button command it just doesn’t fire.

Shlamorel.8714:

mrstealth.6701:

Staff
Fire skill 1 – 21-22 attacks – 5966 damage
Earth skill 1 – 21-22 attacks – 3885

I have no idea how this wiki information can even be accurate.

I’m looking at my auto-attack tooltips right now as I type this and the fire auto-attack is listed as having a 1 second cast time, while the Earth (and air and water) auto-attacks all have a 3/4 of a second cast time.

I don’t see how the fire and earth attunements can have the same amount of auto-attacks in a 30 second window when the tooltips themselves have a 25% difference in cast time.

If I’m reading anything wrong here please fill me in.

P.S. I realize you’re copy and pasting data from wiki – I’m not shooting you, the messenger

Actually, I’m not copying anything from the wiki, but I did use it to get the base attack values with no +stats in order to make fair comparisons. I noticed the listed attack rates on the wiki pages after I had done my own research into it. The two happen to be pretty close, though. I could have used more accurate methods, so the wiki data is likely more accurate than my own.

In any case, the actual attack speeds and the activation time in the skill tooltips are very different. The activation itself might be the same, but the cooldown between attacks is different for each skill.

Anything new going on here? I’m going to try to spend some more time looking at the numbers this weekend and seeing what all I can get to the bottom of.