Rifle feels undertuned

[CallousEye.5018]

Rifle 1 must have the worst auto in the game and on a power weapon? It feels like too much value is being put into targets being pierced so the damage was gutted or something.

Rifle 2 feels like it should be an explosion so it can synergize with traits as well.

[Alcatraznc.3869]

It doesnt feel undertuned, it IS undertuned.

 

As far as performances goes, Hammer is straight up better. As far as fun factor goes, well that is subjective but it seems a lot of people arent exactly having fun as well. And dont get me started on "bUt rIfLe hAs RaNgE" because if you need range you'll be running grenade kit or mortar.

[Kuma.1503]

Overnerfed because of Mechanist. 

I'm still not agaisnt the idea of turning the auto into a proper shotgun blast if they want to discourage AFK 1200 range gameplay. 

Alternatively, return it to 3 round burst and move some of the auto damage onto Blunderbuss. So the AFK benchmark power Mech benchmark remains low. This will allow them to buff the weapon back to a decent spot. 

CMC's "solution" is just bad. Overnerfing core weapons needs to stop happening. 

[jason.1083]

It's super undertuned, comparing Rifle Burst to Sun Edge (according to the wiki) we get that Rifle AA has 133 less base damage with 49% less power coefficiency compared to sword AA

(For non- Holos, Rifle has 162 less base damage with 54% less power coefficiency)

Compared to hammer, it's 92 less base damage with 46% less power coefficiency.

Don't want to get too detailed here, but you get the point, compared to our other power options it's far worse.

And on top of that it doesn't have anything else going for it, sword gives Vulnerability, Hammer gives Might, Rifle only has a grenade which feels like a second attack rather than 2 quick attacks in succession which still has lower base damage/coefficiency  compared to the other 2.

[Arantheal.7396]

Sorry I have to say it a-net, but someone is sitting in your balance team that doesn't understand multiplicative math.
If you look across ranged power weapon AA's in the game (rifle on other classes / longbow), they are currently balanced around 420ish base damage and 1.0 power scaling, with slight variations around class mechanics and related traits. On engineer they decided to take our multiple strikes per cast into account, and therefore correctly split our damage into a weaker first projectile, and a stronger second followup.
The logical error here is that both skills should still scale with power the same.

lets do the math. I choose the warrior rifle as benchmark, to show what a properly tuned AA on rifle should look like, so that's the first we'll be doing.

Formula: [[strike dmg] * [[power scaling] * [power]] / [target dmg resistance]] / [sequence] == [n]dps (before precission, ferocity, and vulnerability, as those scale consistently across the board)

benchmark: warrior "fierce shot" with current values pulled straight from the wiki
input:
strike damage: 422
Power scaling: 1.0
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 0.96 seconds
output total: 879.16 dps

benchmark: engineer rifle "rifle burst" with current values pulled straight from the wiki
input:
projectile 1 strike damage: 190
projectile 2 strike damage: 253
projectile 1 power sclaing: 0.45
projectile 2 power sclaing: 0.6
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 1.13 seconds
output projectile 1: 153,33 dps
output projectile 2: 268,67 dps
output total: 422 dps

Do you see the issue A-net? If we'd combine the strike damage of both engineer rifle projectiles, we'd end with 443 total strike damage, roughly around, but slightly above the warrior ballpark. And if we do the same with our power scaling, we'd end with 1.05, again slightly above warrior, which would then be correctly offest by the fact that our sequence is also longer than warriors, so we get slightly more bang for our buck. Totally fair and balanced, right?
Well, that would be true, IF MULTIPLIKATIVE MATH WOULD WORK THAT WAY. But you didn't just split one multiplicator (strike damage OR power sclaing), you split both, which is why engineers rifle benchmark under the same conditions will always end with below half the damage of warrior's rifle AA, or thief rifle AA or any longbow AA (ironically, with the exception of warrior longbow, since you did the same projectile split errors there, too).

This is not just an engi problem, this is a problem you will encounter multiple times across all classes, where the decission was made to balance around multiple projectiles.
you want proof, a-net?

Here's a benchmark of engi rifgle AA, if the power scaling would have been left alone. Still two projectiles, still offloaded the mayority of the dmg on the second projectile, simply set the power sclaing to the combined total of 1.05 on both projectiles:

input:
projectile 1 strike damage: 190
projectile 2 strike damage: 253
projectile 1 power sclaing: 1.05
projectile 2 power sclaing: 1.05
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 1.13 seconds
output projectile 1: 353,09 dps
output projectile 2: 470,18 dps
output total: 823,27 dps

Would you look at that...
Still slightly behind warrior in raw performance, therefore justifying all the trait interactions we get, still 2 projectiles, still the second one being the heavy hitter...
I know I might sound condescending here, but I refuse to believe that the results we see in the game from multiple projectile weapons and attacks in general, aren't the result of math errors deep seeded in the balancing of the game for a couple of years now.

Edited October 4, 2023 by Arantheal.7396

[Alcatraznc.3869]

3 minutes ago, Arantheal.7396 said:

Sorry I have to say it a-net, but someone is sitting in your balance team that doesn't understand multiplicative math.
If you look across ranged power weapon AA's in the game (rifle on other classes / longbow), they are currently balanced around 420ish base damage and 1.0 power scaling, with slight variations around class mechanics and related traits. On engineer they decided to take our multiple strikes per cast into account, and therefore correctly split our damage into a weaker first projectile, and a stronger second followup.
The logical error here is that both skills should still scale with power the same.

lets do the math. I choose the warrior rifle as benchmark, to show what a properly tuned AA on rifle should look like, so that's the first we'll be doing.

Formular: [[strike dmg] * [[power scaling] * [power]] / [target dmg resistance]] / [sequence] == [n]dps (before precission, ferocity, and vulnerability, as those scale consistently across the board)

benchmark: warrior "fierce shot" with current values pulled straight from the wiki
input:
strike damage: 422
Power scaling: 1.0
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 0.96 seconds
output total: 879.16 dps

benchmark: engineer rifle "rifle burst" with current values pulled straight from the wiki
input:
projectile 1 strike damage: 190
projectile 2 strike damage: 253
projectile 1 power sclaing: 0.45
projectile 2 power sclaing: 0.6
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 1.13 seconds
output projectile 1: 153,33 dps
output projectile 2: 268,67 dps
output total: 422 dps

Do you see the issue A-net? If we'd combine the strike damage of both engineer rifle projectiles, we'd end with 443 total strike damage, roughly around, but slightly above the warrior ballpark. And if we do the same with our power scaling, we'd end with 1.05, again slightly above warrior, which would then be correctly offest by the fact that our sequence is also longer than warriors, so we get slightly more bang for our buck. Totally fair and balanced, right?
Well, that would be true, IF MULTIPLIKATIVE MATH WOULD WORK THAT WAY. But you didn't just split one multiplicator (strike damage OR power sclaing), you split both, which is why engineers rifle benchmark under the same conditions will always end with below half the damage of warrior's rifle AA, or thief rifle AA or any longbow AA (ironically, with the exception of warrior longbow, since you did the same projectile split errors there, too).

This is not just an engi problem, this is a problem you will encounter multiple times across all classes, where the decission was made to balance around multiple projectiles.
you want proof, a-net?

Here's a benchmark of engi rifgle AA, if the power scaling would have been left alone. Still two projectiles, still offloaded the mayority of the dmg on the second projectile, simply set the power sclaing to the combined total of 1.05 on both projectiles:

input:
projectile 1 strike damage: 190
projectile 2 strike damage: 253
projectile 1 power sclaing: 1.05
projectile 2 power sclaing: 1.05
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 1.13 seconds
output projectile 1: 353,09 dps
output projectile 2: 470,18 dps
output total: 823,27 dps

Would you look at that...
Still slightly behind warrior in raw performance, therefore justifying all the trait interactions we get, still 2 projectiles, still the second one being the heavy hitter...
I know I might sound condescending here, but I refuse to believe that the results we see in the game from multiple projectile weapons and attacks in general, aren't the result of math errors deep seeded in the balancing of the game for a couple of years now.

 

Is this for PvE ? I havent done the math but it seems extremely low.  If you're telling me these value comes from PvE, then you are not ready to see the value for PvP and WvW

[Arantheal.7396]

Just now, Alcatraznc.3869 said:

 

Is this for PvE ? I havent done the math but it seems extremely low.  If you're telling me these value comes from PvE, then you are not ready to see the value for PvP and WvW

Yes, those are the PvE numbers, the supposed higher ones. It just gets worse from here on out.

[Remus Darkblight.1673]

4 minutes ago, Alcatraznc.3869 said:

 

Is this for PvE ? I havent done the math but it seems extremely low.  If you're telling me these value comes from PvE, then you are not ready to see the value for PvP and WvW

Remember these are for comparison only. The power and toughness are somewhat arbitrary and it isn't taking into account other factors such as vulnerability, so real damage would be higher for all examples in actual PvE.

The maths looks solid though and makes sense. I want to believe it's an error, and therefore has the potential to be rectified, and not that they are deliberately tanking the damage of multi-projectile auto-attacks. Because, for example, they are giving a huge amount more weight to those 20% projectile finishers or something.

Sometimes it would be nice just to understand the logic behind some of the balancing decisions.

[Arantheal.7396]

11 minutes ago, Remus Darkblight.1673 said:

Remember these are for comparison only. The power and toughness are somewhat arbitrary and it isn't taking into account other factors such as vulnerability, so real damage would be higher for all examples in actual PvE.

The maths looks solid though and makes sense. I want to believe it's an error, and therefore has the potential to be rectified, and not that they are deliberately tanking the damage of multi-projectile auto-attacks. Because, for example, they are giving a huge amount more weight to those 20% projectile finishers or something.

Sometimes it would be nice just to understand the logic behind some of the balancing decisions.

Yes, the input for power is arbitrary. 2000 is way below what you'd expect. You can set around 3600 to 3800 there instead, which would be roughly the upper end you'd get out from full asceded berserker with proper runes and food.
Also the target damage resistance is simply light armor with zero toughness. In game, that can vary as well, and usually will be higher.
I have provided the formular for calculating it in the post, too, and you can insert any values you might want to.
As long as you input the same values into all 3 benchmarks (to make sure the benchmark environment is consistent), you will get consistent results, that show the attrocious scaling as a result of spliting not one, but two multiplicators across multiple projectiles.

Edited October 4, 2023 by Arantheal.7396

[Remus Darkblight.1673]

7 minutes ago, Arantheal.7396 said:

Yes, the input for power is arbitrary. 2000 is way below what you'd expect. You can set around 3600 to 3800 there instead, which would be roughly the upper end you'd get out from full asceded berserker with proper runes and food.
Also the target damage resistance is simply medium armor with zero toughness. In game, that can vary as well, and usually will be higher.
I have provided the formular for calculating it in the post, too, and you can insert any values you might want to.
As long as you input the same values into all 3 benchmarks (to make sure the benchmark environment is consistent), you will get consistent results, that show the attrocious scaling as a result of spliting not one, but two multiplicators across multiple projectiles.

Just to clarify I completely agree with your post and understand why you used the numbers you used. I was attempting to explain why the final numbers were so low for a PvE calculation. The actual result is of course irrelevant, it's the comparison between the two, seeing that multi projectile auto attacks are parsing at 50% that of single projectile ones because of how the coefficients are split.

My concern is that it isn't an error, but intentional.

Edited October 4, 2023 by Remus Darkblight.1673

[Arantheal.7396]

The sad part is, this very error is also the same one responsible for why Flamethrowers AA is kitten. I mean, you even see it in the tooltip.
It tells you it has a 2.5 power scaling for a 2.57sec sequence AA, but that is just the actual 0.25 power coefficient added together 10 times for the 10 hits. but since power scaling is a multiplactor, it simply can't be summed up that way, turning FT's tooltip into the strongest evidence I have available, that this is in fact a erroneous calculation on a-nets side, a mistake they have made multiple times across all classes.
In reality, FT AA is 10x89 strike damage, with a power scaling of 0.25 per hit, on a 2 and a half second cast. It hits like a wet noodle, because it got hit by this error 10 times worse.
to normalize it with warriors rifle AA, Flamethrower AA would need to have a power scaling of 1.0 on all hits, and an individual strike damage of 112 per hit, if the sequence stays on 2.57 seconds.

Nades have the same issue, and it's not just AA's, this error affects all weapons and utility skills that are balanced around multiple hits.

Edited October 4, 2023 by Arantheal.7396

[Jzaku.9765]

53 minutes ago, Arantheal.7396 said:

Sorry I have to say it a-net, but someone is sitting in your balance team that doesn't understand multiplicative math.
If you look across ranged power weapon AA's in the game (rifle on other classes / longbow), they are currently balanced around 420ish base damage and 1.0 power scaling, with slight variations around class mechanics and related traits. On engineer they decided to take our multiple strikes per cast into account, and therefore correctly split our damage into a weaker first projectile, and a stronger second followup.
The logical error here is that both skills should still scale with power the same.

lets do the math. I choose the warrior rifle as benchmark, to show what a properly tuned AA on rifle should look like, so that's the first we'll be doing.

Formula: [[strike dmg] * [[power scaling] * [power]] / [target dmg resistance]] / [sequence] == [n]dps (before precission, ferocity, and vulnerability, as those scale consistently across the board)

benchmark: warrior "fierce shot" with current values pulled straight from the wiki
input:
strike damage: 422
Power scaling: 1.0
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 0.96 seconds
output total: 879.16 dps

benchmark: engineer rifle "rifle burst" with current values pulled straight from the wiki
input:
projectile 1 strike damage: 190
projectile 2 strike damage: 253
projectile 1 power sclaing: 0.45
projectile 2 power sclaing: 0.6
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 1.13 seconds
output projectile 1: 153,33 dps
output projectile 2: 268,67 dps
output total: 422 dps

Do you see the issue A-net? If we'd combine the strike damage of both engineer rifle projectiles, we'd end with 443 total strike damage, roughly around, but slightly above the warrior ballpark. And if we do the same with our power scaling, we'd end with 1.05, again slightly above warrior, which would then be correctly offest by the fact that our sequence is also longer than warriors, so we get slightly more bang for our buck. Totally fair and balanced, right?
Well, that would be true, IF MULTIPLIKATIVE MATH WOULD WORK THAT WAY. But you didn't just split one multiplicator (strike damage OR power sclaing), you split both, which is why engineers rifle benchmark under the same conditions will always end with below half the damage of warrior's rifle AA, or thief rifle AA or any longbow AA (ironically, with the exception of warrior longbow, since you did the same projectile split errors there, too).

This is not just an engi problem, this is a problem you will encounter multiple times across all classes, where the decission was made to balance around multiple projectiles.
you want proof, a-net?

Here's a benchmark of engi rifgle AA, if the power scaling would have been left alone. Still two projectiles, still offloaded the mayority of the dmg on the second projectile, simply set the power sclaing to the combined total of 1.05 on both projectiles:

input:
projectile 1 strike damage: 190
projectile 2 strike damage: 253
projectile 1 power sclaing: 1.05
projectile 2 power sclaing: 1.05
power: 2000
target dmg resistance (armor + toughness): 1000
sequence: 1.13 seconds
output projectile 1: 353,09 dps
output projectile 2: 470,18 dps
output total: 823,27 dps

Would you look at that...
Still slightly behind warrior in raw performance, therefore justifying all the trait interactions we get, still 2 projectiles, still the second one being the heavy hitter...
I know I might sound condescending here, but I refuse to believe that the results we see in the game from multiple projectile weapons and attacks in general, aren't the result of math errors deep seeded in the balancing of the game for a couple of years now.

I don't where you get off pulling completely imaginary fake calculations from - the wiki has a very clear page on exactly how to calculate damage in this game. The first massive glaring issue in your post is that base damage does not exist in GW2. Skills strength is effectively purely based on the coefficient listed on the wiki. You might be misinterpreting "Weapon Strength" in the formula - please click on the wiki link for it to see what it's actually referring to.

Plugging your assumed own values (including the sequence duration which is a big assumption that you have it right including aftercasts) in the actual formula the game uses gives us:

Quote

Damage done = (Weapon strength) * Power * (skill-specific coefficient)/(target's Armor)

Warrior Rifle:
(~1150 * 2000 * (1.0) / 1000) / 0.96 = 2395 dps

Engi Rifle(both hits calculating damage then dividing by the sequence duration):
[ (~1150 * 2000 * (0.45) / 1000)  + (~1150 * 2000 * (0.6) / 1000) ]  / 1.13 = 2137 dps

 

So the auto is only ~10% worse as opposed to the ~50% your unhinged imaginary formula is giving you.

 

1 hour ago, Alcatraznc.3869 said:

 

Is this for PvE ? I havent done the math but it seems extremely low.  If you're telling me these value comes from PvE, then you are not ready to see the value for PvP and WvW

 

38 minutes ago, Remus Darkblight.1673 said:

Just to clarify I completely agree with your post and understand why you used the numbers you used. I was attempting to explain why the final numbers were so low for a PvE calculation. The actual result is of course irrelevant, it's the comparison between the two, seeing that multi projectile auto attacks are parsing at 50% that of single projectile ones because of how the coefficients are split.

My concern is that it isn't an error, but intentional.

^ Just FYI. 

[Remus Darkblight.1673]

8 minutes ago, Jzaku.9765 said:

Warrior Rifle:
(~1150 * 2000 * (1.0) / 1000) / 0.96 = 2395 dps

Engi Rifle(both hits calculating damage then dividing by the sequence duration):
[ (~1150 * 2000 * (0.45) / 1000)  + (~1150 * 2000 * (0.6) / 1000) ]  / 1.13 = 2137 dps

 

I think I see where you are coming from now. It's been a while since I have actually looked at exactly how the damage calculations worked and it looked sound at first glance.

I assume what you're saying here, and correct me if I am wrong, is that the tooltip skill damage for Fierce Shot and Rifle Burst/Rifle Burst Grenade are actually the results of this calculation and not the input. Which would make sense.

I suppose really the sentiment here though still is that it feels under tuned, and come to think of it warrior riffle doesn't feel great either outside of gunflame cheesery.

[Jzaku.9765]

10 minutes ago, Remus Darkblight.1673 said:

I assume what you're saying here, and correct me if I am wrong, is that the tooltip skill damage for Fierce Shot and Rifle Burst/Rifle Burst Grenade are actually the results of this calculation and not the input. Which would make sense.

Yes that's correct, the wiki assumes a specific set of values, and the ingame tooltip actually does the calculation on the fly - you can see it go up and down as you gain/lose Might. For most intents and purposes, the coeff alone is an accurate representation of the strength of the skill.

(Note that this is only true for power damage - condi damage and healing do actually have base damage/healing values)

Edited October 4, 2023 by Jzaku.9765

[Arantheal.7396]

In that case I concede my points. I was indeed putting in the tooltip damage as the actual damage, since I didn't factor in the weapons own base strength.
And I'd like to thank you for pointing out my mistake and correcting me, if you didn't went full keyboard warrior and threw terms like "unhinged" around.

[Jzaku.9765]

Just now, Arantheal.7396 said:

In that case I concede my points. I was indeed putting in the tooltip damage as the actual damage, since I didn't factor in the weapons own base strength.
And I'd like to thank you for pointing out my mistake and correcting me, if you didn't went full keyboard warrior and threw terms like "unhinged" around.

I take serious issue with disinformation stated as fact, especially when it presents a situation wildly deviating from reality. 

[Arantheal.7396]

3 minutes ago, Jzaku.9765 said:

I take serious issue with disinformation stated as fact, especially when it presents a situation wildly deviating from reality. 

Which are your personal issues, I'd prefer to not be carried out on me. I made a mistake, made erroneous asumptions based on the mistake I made, and now stand corrected. If you feel this is grounds for getting personal, I feel this is abusive.

Edited October 4, 2023 by Arantheal.7396

[Gendalfs.7521]

10 hours ago, CallousEye.5018 said:

Rifle 1 must have the worst auto in the game and on a power weapon? It feels like too much value is being put into targets being pierced so the damage was gutted or something.

Rifle 2 feels like it should be an explosion so it can synergize with traits as well.

Rifle suffers because of Mech. Heres guys calculating AA, but what they need to take on is added Mech AA values to the whole damage.

Rifle AA consists of 2 attacks where 2nd is grenade that triggers explosion traits, but it wouldnt hurt Rifle 2 be also finisher or trigger explosion.

[Remus Darkblight.1673]

22 minutes ago, Arantheal.7396 said:

Which are your personal issues, I'd prefer to not be carried out on me. I made a mistake, made erroneous asumptions based on the mistake I made, and now stand corrected. If you feel this is grounds for getting personal, I feel this is abusive.

Yeah I don't think using wildly unhinged was entirely necessary where a "I think you may be mistaking the tooltip damage for the base damage, you actually need to be plugging weapon strength in here".

25 minutes ago, Jzaku.9765 said:

I take serious issue with disinformation stated as fact, especially when it presents a situation wildly deviating from reality. 

It's clear from the post that the intention wasn't to spread disinformation, the intent was in the right place, it was just an honest mistake. I think your phrasing might have implied malice  at worst or gross incompetence at best. They were using the correct formula, but had just made a mistake on the initial input.

 

 

[jason.1083]

4 hours ago, Gendalfs.7521 said:

Rifle suffers because of Mech. Heres guys calculating AA, but what they need to take on is added Mech AA values to the whole damage.

Rifle AA consists of 2 attacks where 2nd is grenade that triggers explosion traits, but it wouldnt hurt Rifle 2 be also finisher or trigger explosion.

Mechanist afk playstyle is a reason for the nerf, but that's not an excuse for what we got now.

We don't count the Mech AA because it shouldn't be in the calculation in the first place, firstly because it's not just Mechanists that might want to use the weapon, but it's also unfair for the weapon itself, you could literally say the same about all the other weapons available.

And yes it does consist of 2 attacks where the second one triggers explosion traits, which as I've said multiple times before in these forums, it looks to be the A-net excuse to nerf whatever they feel like nerfing. Just slap an explosion tag in there, nerf the base skill to the ground and call it a day. And if you complain the answer is always as simple as "Just use explosives". 

What if you don't want to use explosives? Well then I suppose you'll have to come to terms with the nerf and move on.

This would be fine if the number of skills/traits with that tag was reasonable, but we're at a point where you pretty much can't create an Engi build without them, and then you just stare at the explosion tag thinking about all that lost potential cause you don't use explosives in your build. 

So no, I would much prefer for rifle to not gave an explosion on it's AA, get the triple shot back (this time with 3 bullets instead) and have it's base damage/power coefficiency match our other power options.

You don't realize it but explosions will be the death of Engineer variety and utility, I'm almost waiting for the uptade where they'll just say,

"All Engineer skills/trails will now be explosions, removed all traitlines and replaced with variations of the explosives traits." followed by a massive list of nerfs to "compensate".

[Gendalfs.7521]

They (Anet) already shot themselves in foot with weapon mastery. Imagine to balance weapons taking care of Mech dmg. Its impossible 😞 

Maybe someone has any ideas?

[Alcatraznc.3869]

15 hours ago, Gendalfs.7521 said:

Rifle suffers because of Mech. Heres guys calculating AA, but what they need to take on is added Mech AA values to the whole damage.

You're aware with that logic, pretty much all Engineer weapon deserve a nerf

[Yaki.9563]

Mechanist traits, because they grant a mech, far outstrip Scrapper traits. Scrapper was propped up (and barely, given Scrapper's lackluster dps) by the overtuned hammer. If it wasn't, you'd see rifle Scrapper being a thing. Since all specs can use the same weapons now, there's no reason to play Scrapper for dps over Mechanist. They'll need to massively buff Scrapper traits because they'll be nerfing hammer to rifle levels thanks to Mechanist.

Engineer isn't the only class with this issue. Greatsword was propping up Vindicator. Now it got nerfed thanks to Herald, so they had to buff Vindicator traits.

Edited October 5, 2023 by Yaki.9563

[Matoro.9708]

Turrets have been undertuned for like 8 years and I've seen no sign of anet ever caring at all

[jason.1083]

6 hours ago, Matoro.9708 said:

Turrets have been undertuned for like 8 years and I've seen no sign of anet ever caring at all

I swear they're doing it on purpose, it's just a meme for the dev team to laugh at. Because it's impossible to seriously ignore it at this point.

The only patch I'm experiencing to see them in is the April Fools one.

Edited October 5, 2023 by jason.1083