[PC] Bugged Self Buffs on the Asterdahl Classes

rjc9000

The following self buffs are confirmed to be adding to the Critical Severity/Combat Advantage portion of the damage formula, rather than being a multiplicative buff in the buff category of damage.

Barbarian
-Trample the Fallen
-Raging Strikes
-Battlerage (the mechanic)
-Relentless Battlerage

Fighter:

-Vengeance above 50%
-Combat Superiority
-Determination
-Enduring Vengeance

(yes, I rechecked these today, still bugged)

Cleric

-Doomslayer
-Divine Equilibrium

The following self bugs are confirmed to be on live .

Coincidentally (or not so coincidentally?), the known instances of the bug are on all the classes that @asterdahl handled in the Mod 16 rework (GF/GWF/DC). I am gathering all the bugged self buffs I know into one thread so asterdahl doesn't need to go into separate threads to figure out what's going on. If he wants the proof, the bug reports listing the bugs are linked below.

I do note that not every self buff was mentioned on these classes. I didn't have the patience to check those other buffs, but I would bet they are bugged by being additive to CA/Severity as well.
____

Individual Threads of Proof:

https://www.arcgames.com/en/forums/neverwinter/#/discussion/1250926/barbarian-damage-bonuses-bugged

https://www.arcgames.com/en/forums/neverwinter/#/discussion/1250399/pc-fighter-bugs-self-buffs-incorrectly-calculating-heavy-slash-tooltip-incorrect

https://www.arcgames.com/en/forums/neverwinter/#/discussion/1250929/pc-doomslayer-and-divine-equilibrium-additive-instead-of-multiplicative

siggstar

This is the moment when I go "sigh".
How can those programming mistakes make it all the way to the live game? Even if they get fixed it won't lift those classes enough to be competitive in the new Trial. This is bad.

ramesh84

Making such kind of mistakes is not as bad as not placing fixing them as #1 priority after they being discovered. This bug on fighter has been reported since a while and I've not seen a single attempt fixing in preview patch notes.
I could accept a low QA engagement if devs actions would be more responsive towards players reporting.

lordaeolos

has seekers vengeance on the Ranger been tested? It may be in all the classes that have feats/ class features that add damage multipliers... possibly working as intended?

rjc9000

lordaeolos said:

has seekers vengeance on the Ranger been tested? It may be in all the classes that have feats/ class features that add damage multipliers... possibly working as intended?

Most of the Ranger stuff I tried works. I checked Seeker's, Throw Caution, Deft Strikes, and Twin Blade Storm. If skeptical, check below.

I don't feel like testing every Ranger buff in the interest of time and don't have everything unlocked on a live Ranger, so I can't test some of the higher end stuff on the roleplayer path (What's Hunter?).

Also worth noting that according to Sharpedgelord Testing™, Wizard buffs like Arcane Stacks, Controlled Momentum, amd Chilling Presence are working correctly.

With the given that all the CW stuff works and that a good portion of the Ranger stuff works correctly, we can make the logical assumption that noworries classes (HR/CW/TR/SW) have correctly programmed buffs.

Though if anyone can provide proof stating otherwise, they are feel free to disprove that notion.

_____
Conditions were naked, guild ArmorPen Boon, 84% Crit Sev, effective 9500 (10000 CA-500 dummy Awareness), 100 fixed damage weapons, and 16 STR.

Seeker's

Expected damage if bugged:

100*0.25*1.04*(1+0.84+0.0950+0.10)=52.91

Expected damage if correct:

100*0.25*1.04*(1+0.84+0.0950)*1.1=55.341

This is the result:


And I can assure you I was standing in the correct spot to get Seeker's Vengeance.



Throw Caution

I don't think anyone uses Throw Caution anymore, but it is calculated correctly. Used Seeker's for this test, in addition to Throw Caution, Crit, and companion for CA.

Expected damage if bugged:

100*0.25*1.04*(1+0.84+0.0950+0.1+0.1)=55.51

Expected damage if correct:

100*0.25*1.04*(1+0.84+0.0950)*1.1*1.1=60.8751

Result:


Deft Strikes
Used Steel Breeze for the test, but same conditions of before are held constant. I couldn't get the numbers to exactly match up, but the damage is very close to the buffed damage.

Expected Damage if bugged:

100*1.5*1.04*(1+0.84+0.0950+0.1+0.1)=333.06

Expected Damage if Correct:
100*1.5*1.04*(1+.84+0.095)*1.1*1.1=365.2506

Result:


Twin Blade Storm:
Used Clear the Ground. Since it's Lightning damage, need to use magic damage and account for 12 INT rather than my STR. Used Seeker's, got a Crit, had CA, and had companion CA wall.

Expected damage if bugged:
100*0.20*1.03*(1+0.84+0.095+0.1+0.08)=43.569

Expected damage if correct:
100*0.20*1.03*1.08*1.1*(1+0.84+0.095)=47.3548

Result:

noworries

While there is a bug here, no one on this thread is going to like that it is the reverse of what is being called out.

Multiplicative buff stacking was a big problem for Neverwinter and is what would cause a normal geared group to do a boss over minutes, and a group that knew how to min max all of the stacking buffs to complete a boss in seconds.

We had made the decision in M16 to switch the buffs over to additive instead of multiplicative to keep power creep in a better place to allow content to be designed better for both elite and standard groups.

Throughout the handover of classes as staff changes occurred, this adjustment was missed for the Ranger, Warlock, Wizard, and Rogue. They are actually the classes in error here and the ones that will need to have their buffs adjusted.




I realize that the players in this thread understand the significant difference between the two methods (which is exactly why you did tests and brought this up), but for those who have not delved as far into the math:

If a player has 10 buffs that each increase the damage 10%, there are two main ways to do this. The first is to add up all of the buffs together and then multiply by the base. If the player's base damage was 1000, then the end result would be 2000 damage as it would be a total of 10 x 10% = 100%.


The second method is to multiply each bonus on top of the previous bonus. In the past this was how most damage boosts worked, and it was also the one of the biggest source of problems of having so many stacking sources that players could even potentially 1 or 2 shot endgame bosses. With the same number of bonuses as the above example, instead of ending at 2000 damage you'd end up at ~2600 damage, or in this particular case 30% more damage than the other method.


Now that might not look like much at a small scale, but let's compare bigger effects. Let's look at if in the examples above instead each boost was 20%.

Additive = 3000 damage
Multiplicative = ~6200 damage - so now the multiplicative approach is over double the value of the additive.


The bigger the value of the bonuses and the more of them that there are, the bigger the gap between the two methods. Additive provides a much more linear approach that works better for the long term health and balance of the game.

thefabricant

noworries#8859 said:

While there is a bug here, no one on this thread is going to like that it is the reverse of what is being called out.

Multiplicative buff stacking was a big problem for Neverwinter and is what would cause a normal geared group to do a boss over minutes, and a group that knew how to min max all of the stacking buffs to complete a boss in seconds.

We had made the decision in M16 to switch the buffs over to additive instead of multiplicative to keep power creep in a better place to allow content to be designed better for both elite and standard groups.

Throughout the handover of classes as staff changes occurred, this adjustment was missed for the Ranger, Warlock, Wizard, and Rogue. They are actually the classes in error here and the ones that will need to have their buffs adjusted.




I realize that the players in this thread understand the significant difference between the two methods (which is exactly why you did tests and brought this up), but for those who have not delved as far into the math:

If a player has 10 buffs that each increase the damage 10%, there are two main ways to do this. The first is to add up all of the buffs together and then multiply by the base. If the player's base damage was 1000, then the end result would be 2000 damage as it would be a total of 10 x 10% = 100%.


The second method is to multiply each bonus on top of the previous bonus. In the past this was how most damage boosts worked, and it was also the one of the biggest source of problems of having so many stacking sources that players could even potentially 1 or 2 shot endgame bosses. With the same number of bonuses as the above example, instead of ending at 2000 damage you'd end up at ~2600 damage, or in this particular case 30% more damage than the other method.


Now that might not look like much at a small scale, but let's compare bigger effects. Let's look at if in the examples above instead each boost was 20%.

Additive = 3000 damage
Multiplicative = ~6200 damage - so now the multiplicative approach is over double the value of the additive.


The bigger the value of the bonuses and the more of them that there are, the bigger the gap between the two methods. Additive provides a much more linear approach that works better for the long term health and balance of the game.

Please show me a group finishing the dps check at the end of phase 1 in tomm if you make this change. Forget about the rest of the raid. Also throw class balance out of the window and pretty much everything else. This is also completely ignoring the issue that the buffs are adding to critical severity and combat advantage.

micky1p00

noworries#8859 said:

While there is a bug here, no one on this thread is going to like that it is the reverse of what is being called out.

Multiplicative buff stacking was a big problem for Neverwinter and is what would cause a normal geared group to do a boss over minutes, and a group that knew how to min max all of the stacking buffs to complete a boss in seconds.

We had made the decision in M16 to switch the buffs over to additive instead of multiplicative to keep power creep in a better place to allow content to be designed better for both elite and standard groups.

Throughout the handover of classes as staff changes occurred, this adjustment was missed for the Ranger, Warlock, Wizard, and Rogue. They are actually the classes in error here and the ones that will need to have their buffs adjusted.




I realize that the players in this thread understand the significant difference between the two methods (which is exactly why you did tests and brought this up), but for those who have not delved as far into the math:

If a player has 10 buffs that each increase the damage 10%, there are two main ways to do this. The first is to add up all of the buffs together and then multiply by the base. If the player's base damage was 1000, then the end result would be 2000 damage as it would be a total of 10 x 10% = 100%.


The second method is to multiply each bonus on top of the previous bonus. In the past this was how most damage boosts worked, and it was also the one of the biggest source of problems of having so many stacking sources that players could even potentially 1 or 2 shot endgame bosses. With the same number of bonuses as the above example, instead of ending at 2000 damage you'd end up at ~2600 damage, or in this particular case 30% more damage than the other method.


Now that might not look like much at a small scale, but let's compare bigger effects. Let's look at if in the examples above instead each boost was 20%.

Additive = 3000 damage
Multiplicative = ~6200 damage - so now the multiplicative approach is over double the value of the additive.


The bigger the value of the bonuses and the more of them that there are, the bigger the gap between the two methods. Additive provides a much more linear approach that works better for the long term health and balance of the game.

That's not the point. The issue is not that the buffs are additive, but that they are additive to a stat based multipliers, the critical severity and combat advantage.
There isn't a valid reason that all those 3 should be added together. Buffs should be added to buffs, debuffs to debuffs, CA has it's counter stat and critical severity is on crit.

CA and Severity, and in this case Buffs should be seperated.
This is similar issue with outgoing healing, where it is added to critical severity...

What next, we should added power multiplier into the bucket too????

weapon * magnitude * ( 1 + arp_defense_reduction + severity * crit + CA + buffs + power/100000 + debuffs + everything else in existence). ??

Additive should be confined to its proper groups.

weapons * magnitude * (1 + power / 100000) * ( 1 + crit * severity ) * (1 + ca) * ( arp_resistance) * (1- targetdeflectChance * deflectSeverity) * ( 1 + SUM OF ALL DAMAGE BUFFS) * (1 + SUM OF ALL DEBUFFS)

This is how additive buffs should work.

If when things where they belong too over-perform, then the can be adjusted, if it's lower CA ratio, severity, or adjusting the buffs or magnitudes.
But the entire things should have at least some consistency and logic first.

katamaster81899

If these adjustments are made, will TOMM difficulty be adjusted to account for it?

thestia

noworries#8859 said:

While there is a bug here, no one on this thread is going to like that it is the reverse of what is being called out.

Multiplicative buff stacking was a big problem for Neverwinter and is what would cause a normal geared group to do a boss over minutes, and a group that knew how to min max all of the stacking buffs to complete a boss in seconds.

We had made the decision in M16 to switch the buffs over to additive instead of multiplicative to keep power creep in a better place to allow content to be designed better for both elite and standard groups.

Throughout the handover of classes as staff changes occurred, this adjustment was missed for the Ranger, Warlock, Wizard, and Rogue. They are actually the classes in error here and the ones that will need to have their buffs adjusted.

asterdahl said:

those who fear that the current top tier classes will be nerfed, we'd like to avoid that. We'd also like to leave those mechanics alone, except in cases where there may be a bug or exploit. We have adjusted our expected DPS values, but those are hypothetical and we want to ensure that we are taking into account real world DPS numbers.

While I understand that apparently Ranger, Warlock, Wizard and Rogues have an overlooked "bug" in their intended mechanics, "fixing" it would end up do precisely what everyone feared - nerfing everyone instead of bringing up under-performing classes. Are you planning on also tweaking the damage formulas of the effected classes? Adjusting base magnitude of encounters? Hearing that what appears to be functioning, reasonable damage is due to an oversite in coding is not only frustrating, it's downright disheartening.

At this stage, the trial has been adjusted using data that will be irrelevant once this oversite has been fixed. I am really not hopeful for the community at large once the damage formula is... fixed.

elpide31

Right when balance was starting to get better... "Fixing" the buffs calculations on these classes will bring them down instead of working off the balance that is almost there. That being said the feedback that we have been giving on the trial is pointless.

nitocris83

Hey folks, jumping in this thread to let folks know that internal discussions are happening regarding classes. I will be monitoring this topic, especially once the trial hits live, more folks attempt it, and we start getting internal data. Changes to classes aren't decisions made rashly or impulsively so we will setup Q&As and continue to check in on player discussions among multiple channels.

As always, please keep feedback constructive. Emotions can run high on these topics on all sides but it helps when we can all focus on facts and not inflammatory speculation or personal attacks.

rjc9000

noworries#8859 said:

While there is a bug here, no one on this thread is going to like that it is the reverse of what is being called out.

[...]

We had made the decision in M16 to switch the buffs over to additive instead of multiplicative to keep power creep in a better place to allow content to be designed better for both elite and standard groups.

Hold up.
Multiplicative buffs were mathematically correct for as far back as Mod 6 (possibly even Mod 4/5 if my recollections of kaelac's works are correct).

So, tell us, where on any of the Mod 16 Official Feedback Threads was it mentioned that:

-buffs are supposed to be additive?

And especially:

-buffs are supposed to be included in the same damage multiplier as Critical Severity and Combat Advantage?

leviatano

I hope that this will be fixed soon. This isn't fair for DC, GF and GWF classes. Bring alla at the same level, burst the classes that are underperforming or bring down the others

mentinmindmaker

rjc9000 said:

So, tell us, where on any of the Mod 16 Official Feedback Threads was it mentioned that:

-buffs are supposed to be additive?

And especially:

-buffs are supposed to be included in the same damage multiplier as Critical Severity and Combat Advantage?

These are decisions I would say the devs are free to make without seeking any form of approval from the playerbase It however would be better if they informed us of such major changes.

I assume this change will be valid for all +dps buffs(also from gear), not just the self buffs that OP posted about.

It is fairly obvious that multiplicative dps buffs are bad - they makes the value of such buffs increase exponentially the more buffs you have. Things easily get out of control that way. My personal opinion is this is a needed and overdue change.

There however could be a discussion about how additive dps buffs are implemented, or rather what base damage they refer to.

Cryptic has chosen to let additive dps buffs refer to base damage before CA and crit severity. They could just as well let additive dps buffs refer to base damage including CA and crit severity, although that would introduce a small exponential effect by itself. It however would make dps buffs more valuable.

The way it is done now reduces the value of +dps buffs dramatically.

But either way, when his is implemented properly, of course there need to be a balancing pass to the classes. They are currently tuned with the current situation in mind, so cannot just remove that dps and say sorry

silverwolf

Just so that I'm clear ....

Asterdahl (GF/GWF/DC) - Performing correctly and not impacted by the bug.
NoWorries (SW, CW, HR, TR) - Overperforming due to the bug.
Not quite sure who did the OP.

If the solution to fix the problem will reduce the outgoing DPS of the NoWorries classes this might help to bring the Wizard, Rogue and Ranger into line with the Asterdahl classes.
Unfortunately it reduces the DPS output for the Warlock too, which is already woefully underperforming on their DPS side.

Asterdahl would like us to continue to still play our classes so that they can gather data and make more accurate changes to the classes to help them out. There are very few DPS Warlock's playing, the simply don't get into the end-game content unless they get a pity/carry run from friends/guildies. With the additional DPS reduction coming their way from this fix, a DPS Warlock will be as rare as rocking-horse s**t.

I understand that you need a lot of data to make accurate modifications, I fear that you'll not be gaining enough data for the Warlock. I'd suggest giving the class a little bit of an increase on the magnitudes to help them out so that you can collect more data about the class itself. It's a temporary measure, but one I suspect needs to be done to simply allow you to gather the right volume of info.

ramesh84

micky1p00 said:

weapons * magnitude * (1 + power / 100000) * ( 1 + crit * severity ) * (1 + ca) * ( arp_resistance) * (1- targetdeflectChance * deflectSeverity) * ( 1 + SUM OF ALL DAMAGE BUFFS) * (1 + SUM OF ALL DEBUFFS)

This is how additive buffs should work.

If when things where they belong too over-perform, then the can be adjusted, if it's lower CA ratio, severity, or adjusting the buffs or magnitudes.
But the entire things should have at least some consistency and logic first.

This please: logical formulas = easier balance too. Just mind not to fix everything without coming with balancing compensation too.
Have a nice day.

adinosii

silverwolf#7884 said:

Just so that I'm clear ....

Asterdahl (GF/GWF/DC) - Performing correctly and not impacted by the bug.
NoWorries (SW, CW, HR, TR) - Overperforming due to the bug.
Not quite sure who did the OP.

The warlock really feels like the odd man out here.I mean, there is a general consensus that Wizard is the best DPS right now, followed by Rogues and Rangers - sure, there are individual exceptions, but I'm talking about the class as a whole. So, bringing those three classes down a bit would probably level the field a bit. But Warlocks? If anything, they need a bit of a DPS boost, not a nerf.

DPS players of other classes have been somewhat worried that the discrepancy will make them "not wanted" for ToMM. Now, if the 3 classes mentioned before are nerfed a bit, that would make good Cleric, Fighter and Barbarians just as sought after for Tomm as the other 3 classes, but at the same time ToMM probably needs to be toned down a bit as well if the DPS pf Wizards, Rogues and Rangers is brought down to match that of the other DPSers.

rapidstar

The only classes I play are GWF and paladin. How much would this nerf affect TR/HR/CW ?

leviatano

I hope that these changes will hit neverwinter before console release of mod 17. If not, no one would bring a DC or GF as dps in ToMM. People are investing on dps path, that is a whole different thing from the healer or tank one. So, DC's, GF's, GWF's' and SW's deserves the same love as damage dealers.

rev

...

Is this on live or preview? because saying warlock is overperforming on live is probably the biggest joke i've heard in some time

mcfob

I get it that mistakes happen, apparently some of the dev team are not on the same page regarding buff multipliers or additives. What I don't understand is how this can not be fixed by now. rjc9000 reported on this, specifically due to the dismal dps of the Fighter, quite some time ago (I saw his post WITH data to prove the issue on July 18th), and nothing has been done. In my opinion, this is a significant issue and rjc9000 bothered to take the considerable amount of time to test and compile data to illustrate this issue nearly a month ago, I feel that at the least the dev team or community liasons could have addressed this in a more timely fashion, even if the fix would take some time, the communication seems to be lacking. Just my opinion, for what it is worth.

bgl

Please give me a real world example, how this huge number of multiplicative damage amounts to an unfair advantage for the warlock. (I mean, honestly, 10x 20% ???)

If you argue with fantasy numbers for fixing a problem, I get the feeling the problem is fantasy, too.

mentinmindmaker

The way additive dps buffs is implemented on Live now reduces the value of +3% dps to less than 1% actual dps:
(base 1 + CA 1 + crit sev 1,5 + dpsbuff 0.03) = 3,53
0.03/3,53 = 0.0085

3% nominal dps buff corresponds to a 0.85% dps increase.

Assuming 200k power, that is equivalent to 3000 * 0.85 = 2550 power increase.

With the additive dps buffs, the new mod 16 gear will be at least close to being competitive with the old dps buff gear.
With multiplicative dps buffs the old dps buff gear is a lightyear ahead as it will be worth 3% or more depending on how many dps buffs you can stack.

So the multiplicative buff error also is at the bottom of Cryptic missing the target of phasing out the old dps gear.

thefiresidecat

revenantpt said:

...

Is this on live or preview? because saying warlock is overperforming on live is probably the biggest joke i've heard in some time

I don't think this is because its overperforming but because it's bugged. hopefully they will adjust damage to be at least what it is already.

cherryman1

I like the additive formulas which is what should be happening instead of the multiplicative formulas of the past. Having those classes have those results does make a lot of sense of why they are doing so well. If memory serves the changes were being done at the same time as the issues with scaling. I can easily see oversights happening at that time. I am patient for fixes but it looks like if they remove those buffs we will have to see where those classes fall at in terms of the dps pecking order. Balancing will probably be a never ending task.

theraxin

I see the theoretical advantage of making it additive for the long term, but making it fixed for certain classes and push fixing for others until more data is gathered will make things pretty miserable for the fixed ones until that later time.

And because there is a direct correlation between the underperforming classes and those classes that got fixed, how about reverting the fix until mod18, where more data will be gathered and the problem is more understood?

dlynn334

At the end of the day, the Barb, DC, and GF were not up to par with the CW, HR, and TR. So I understand player frustrations which is acceptable. The SW is stuck in the middle of this so I do have some concern what will happen in the future. Either way I have invested to much time and resources into this class so I will ride through both the good times and bad. Have fun with release!

thefabricant

@asterdahl and @noworries#8859 The issue that I think many of the people in this thread have with this change isn't that buffs are becoming additive, it is that they are becoming additive with critical severity and combat advantage. There are a number of reasons that this is an issue for us, the first of which is it makes gear reduce the effectiveness of your feats. These 2 systems should not be adversarial, they should have synergy. The second reason is it makes feats less interesting, "oh, this feat does exactly the same thing as a piece of gear," is not a great feeling, especially when you have such a limited array of feats to begin with.

My suggestion is, that it works as explained by @micky1p00 above, but with the addition that item buffs are their own category (so the buffs from shadowstalker do not add with your feats) and they add with each other within the item buff category.

thefiresidecat

@nitocris83 just as a heads up. the facebook and reddit forums are blowing up wiht this. various facebook groups have a ton of people trying to sell their toons over this. I would like to suggest damage control of some sort some sort of statement assuring people that you don't really intend on nerfing without rebalancing damages. honestly... the viral on this is worse than I remember the viral for mod 16. people aren't taking it well at all.

thefiresidecat

adinosii said:

silverwolf#7884 said:

Just so that I'm clear ....

Asterdahl (GF/GWF/DC) - Performing correctly and not impacted by the bug.
NoWorries (SW, CW, HR, TR) - Overperforming due to the bug.
Not quite sure who did the OP.

The warlock really feels like the odd man out here.I mean, there is a general consensus that Wizard is the best DPS right now, followed by Rogues and Rangers - sure, there are individual exceptions, but I'm talking about the class as a whole. So, bringing those three classes down a bit would probably level the field a bit. But Warlocks? If anything, they need a bit of a DPS boost, not a nerf.

DPS players of other classes have been somewhat worried that the discrepancy will make them "not wanted" for ToMM. Now, if the 3 classes mentioned before are nerfed a bit, that would make good Cleric, Fighter and Barbarians just as sought after for Tomm as the other 3 classes, but at the same time ToMM probably needs to be toned down a bit as well if the DPS pf Wizards, Rogues and Rangers is brought down to match that of the other DPSers.

no one needs to be brought down. especially with the new raid.