LM debuff bug, based on mob classification?

[vitamin-f]

I would like to share a thread from the LM forum regarding issues with debuffs based on mob classification (Normal, Elite, etc.), specifically looking at Fire-lore's -30% melee damage debuff.

The thread contains a lot of test data:

http://forums.lotro.com/showthread.php?t=325620

This sounds like a potential bug. If it's not, then.. well, yeouch. And it makes me wonder if it affects all similar type debuffs across classes.

[Raskolnikov]

Hrm... I thought I had a solid answer for you, but I don't! I double checked the debuff though and it hasn't change significantly since November of 2008 (when it was given a Resist Category).

[blueMorning]

Hrm... I thought I had a solid answer for you, but I don't! I double checked the debuff though and it hasn't change significantly since November of 2008 (when it was given a Resist Category).

Dear Raskolnikov. Because of your answer, I assume you didn't read the entire thread. Please, I beg you to do so. We belive that the skills aren't bugged themselves, but they are in the way they apply to the damage the mobs deal to us.

This is the way we think it's working:

Base damage + [(-FireLore)+(Mob buff)] =

When as the tooltip of the skill says, it should be something like this:

(Base damage + Mob buff) - FireLore/WindLore =

I hope I expressed this formula right, it's just too late, but anyway, please have a read to our thread...

http://forums.lotro.com/showthread.php?t=325620

[moebius92]

Dear Raskolnikov. Because of your answer, I assume you didn't read the entire thread. Please, I beg you to do so. We belive that the skills aren't bugged themselves, but they are in the way they apply to the damage the mobs deal to us.

This is the way we think it's working:

Base damage + [(-FireLore)+(Mob buff)] =

When as the tooltip of the skill says, it should be something like this:

(Base damage + Mob buff) - FireLore/WindLore =

I hope I expressed this formula right, it's just too late, but anyway, please have a read to our thread...

http://forums.lotro.com/showthread.php?t=325620

Not exactly. The best way to put it is - all mobs seem to have an inherent damage buff, based on the class - normal mobs run at +0%, signatures (at least the marauders) run at about +30%, elites, EMs, nemesis, etc. all go higher as you go up in class.

Damage = <base damage> * ((1 + <? of damage buff>) * <product of damage debuffs> + <inherent damage buff>)

(I'm not sure how damage buffs stack - haven't managed to get a FoS troll to stand still long enough to check it out - probably should've used a root to let him build up a momentum buff and then tried not getting killed while I checked his buffed damage. When I say product of damage debuffs, I mean the product of (1 - debuff magnitude) - i.e., if you could stack say fire lore on a mob three times, the product would be 0.7 * 0.7 * 0.7)

Anyways, both damage buffs and damage debuffs ignore the inherent damage buff - which makes the fact that marauders, guards, and brutes can all use the same +50% damage buff a bit weird, because while it's fairly effective in marauders, it's a much smaller relative increase in brutes (Marauders go from 130% damage to 180% damage (38% increase). Guards would go from 180ish% damage to 230ish% damage (28% increase), and brutes would have an even small increase (don't remember the typical inherent damage buff for an EM. Actually, I'm not even positive +80% is typical for elites, I'd've sworn FoS trolls were closer to +140%)). Upshot of that is that damage buffs and debuffs are less effective when applied to higher class enemies. Although oddly enough, debuffs become very effective at countering buffs.

(As a side note, as far as I can tell, the two equations you've written in your post say the exact same thing. I may be misinterpreting your notation however.)

(And yes, the behavior of the burglar's untraited disable and a tier six disabling gamble are consistent with the results found using fire lore. It probably works the same way with all damage debuffs. Unless they're applied to PCs in which case everything additively stacks.)

[blueMorning]

(As a side note, as far as I can tell, the two equations you've written in your post say the exact same thing. I may be misinterpreting your notation however.)

Thank you for doing it better than me, I'm still very sleepy (3 am).

About what I wrote, I'm considering the FireLore as a percentage, that's how it becomes different one from another.

[brasswire12]

Hrm... I thought I had a solid answer for you, but I don't! I double checked the debuff though and it hasn't change significantly since November of 2008 (when it was given a Resist Category).

Thank you for checking, but I don't think anyone ever thought it was actually the debuff itself that got changed. Rather, it seems that a rank-based buff for mobs has been added to the equation that is severely mitigating the contribution of the debuff to the final damage tally.

[rawlingsst7]

Just throwing this out there:

Could the Sig/Elite/EM/Nemesis/Arch-Nemesis damage buff be determined by parsing the Encounter mobs in skirmishes of increasing size?

ie. pick a particular skirmish Encounter mob, parse it on solo (Elite); Small Fellow (EM); Fellow (Nemesis?); etc etc.

Probably easiest to pick one that shows up early on, like the goblin in Thievery & Mischief, so the skirm can be reset until he spawns.

[aleczander]

closer to finding an answer yet, Rask?

[Stever1388]

I posted in the other thread, but I'm going to post this here as well because I think I can at least explain what might be going on, at least in terms of PVMP.

Players have ratings for nearly everything now; the only thing that doesn't have one that I know (or can't see) about would be miss chance and resist ratings (not Poison/Disease/etc, like Tactical/Phyisical/etc, Creeps have it on their stats, but Freeps don't). When you hover over any stat that uses a rating to determine the final percentage, it's always listed as such: "#.#% STAT (#.#% + #.#%)" though sometimes the first number and the STAT are reversed. Basically, the first percent is your total percent for that stat (be it Block Chance, Melee Crit Chance, Melee Offence, etc), the second percent is the percent that is calculated using your rating, and if there is a cap limit to the stat then this number is subject to the cap, and the third percent is calculated from any stats and skills that add direct percentages to that stat, and these are added to the second number to get the first number. This is also how you by-pass stat caps. for example, this is how guardians can have 20% Block Chance, even though the BPE cap is 15%; they can have 15% from rating (which would be the second number), and then 5% from the Block Stance (which would go in as the third number).

Now Monster Players have mostly the same stats as we do, and use ratings just like we do. Actual mobs must have ratings for at least certain stats, namely BPE and Critical chance because LMs and burgs have skills that directly lower the rating of mobs. I think it's safe to assume that mobs have ratings in other categories as well, such as Offence and Defence. This is where the problem occurs.

When you apply Fire-lore to a mob, it lowers the mobs Melee Offence by 30%, which is what it is suppose to do. However, it does this by taking the 3rd number from above, and reducing it by 30%. I tested this by having 2 players Spar and seeing the effects of Fire-lore and other skills. The problem with this is that a mobs Melee Offence is also calculated using rating, and the -30% doesn't get applied to it, just to the stats/buffs that add direct percents to Melee Offence. To clarify, let me show the results of using Fire-lore on my Captain.

First, his normal, fully buffed Melee Offence: "28.5% Melee Offence (18.5% + 10%)" The 10% comes from having a Hope Banner equipped. When Fire-lore is applied, the 10% goes down to -23%, which IS a 30% decrease in melee offence. But, you still have to add 18.5% back to it, so I end up at -4.5%. So after Fire-lore debuff, my Melee Offence looks like this: -4.5% Melee Offence (18.5% + -23%). Once you calculate what the total change in percent was, it comes out that my Melee Offence was only reduced by 25%! That equation is z = (x-y)/y, where x = debuffed total, y = non-debuff total, and z = total percent change. Plug in the numbers: (95.5 - 128.5)/128.5 = -0.2568. Times that by 100 to get -25.68%, which is the actual reduction. Note: When it says -4.5%, it really means your Melee Offence is at 95.5% of your base, which is where the 95.5 comes from. Basically, if a creep were to apply a Fire-lore-like debuff to my captain, the actual debuff would be less potent than intended.

Now, 5%, you may say, isn't that bad... and it's really not. The problem occurs when/if the Melee Offence Rating increases substantially above what any normal player character could get. My champ has 7.5k Melee Offence rating, which is 22.5%; I don't really think it's possible to push it much higher right now, and Fire-lore would still reduce his damage by 25% while in Fervour and 24.5% while not in Fervour, so that's not too bad.

But we can't see what mobs Melee (or Ranged/Tactical) Offence Ratings are, however. If they have enough Melee Offence Rating to make their rating percentage go to 100% (for example), their numbers would look like this: 200% Melee Offence (100% + 0%). Fire-lore is applied: 170% Melee Offence (100% + -30%). Anyone with basic understanding of fractions/percents can see that 30/200 is not 30% reduction, it's actually only a 15% reduction, half of what it should be.

This may be a bit extreme, but the other post was showing that Fire-lore was only reducing 12% of the damage, so it could be even worse than that on boss mobs like the Watcher, and in all honesty, though I didn't do any number checking other than what the equation shows, the 25% reduction to melee damage for normal (in my case, playable characters) mobs is exactly what they got.

The problem, however, may spread to other skills, anything that changes a stat based on a percentage. Oathbreakers, Reveal Weakness, Disable, and Telling Mark, just to name a few. If a mobs Melee/Ranged/Tactical Defence Rating is really high, Oathbreakers would not increase its incoming damage by very much. Just on a playable character with 15% Melee Defence from Rating (the highest you can go with the cap), Oathbreakers would only increase incoming damage to that person by 30%.

There's some other skills, which though I think could be considered "WAI", are, what I would say, worded poorly. Don't hate me champs (I play one), but the Fervour -Incoming Healing Penalty doesn't actually penalize you enough. On my champ, who has +15% incoming healing from relics and my class LI and +5% for the man passive bonus, fervour stance actually only reduces my incoming healing by 25%. However, I think the idea is that Fervour reduces your base healing (i.e.: before other buffs/stats are applied, or 100%) by 30%, in which case it does do that. Other skills that are like this are: To Arms, Ballad of War, Adamant, and LM Warding Circles. I can understand why a few of these skills would only apply the bonus based on the base (100%), otherwise some of them could become a bit OP (To Arms and Ballad of War especially).

I can see how this could have been missed, and I think it'll probably be a bit difficult to fix, because you have to make sure everything stacks right. It'll be interested to see how badly this actually affects the skills. I hope the dev posts again soon with more information...

[auximenes]

I posted in the other thread, but I'm going to post this here as well because I think I can at least explain what might be going on, at least in terms of PVMP.

Good post.

[aleczander]

I posted in the other thread, but I'm going to post this here as well because I think I can at least explain what might be going on, at least in terms of PVMP.

...

When you apply Fire-lore to a mob, it lowers the mobs Melee Offence by 30%, which is what it is suppose to do. However, it does this by taking the 3rd number from above, and reducing it by 30%. I tested this by having 2 players Spar and seeing the effects of Fire-lore and other skills. The problem with this is that a mobs Melee Offence is also calculated using rating, and the -30% doesn't get applied to it, just to the stats/buffs that add direct percents to Melee Offence. To clarify, let me show the results of using Fire-lore on my Captain.

First, his normal, fully buffed Melee Offence: "28.5% Melee Offence (18.5% + 10%)" The 10% comes from having a Hope Banner equipped. When Fire-lore is applied, the 10% goes down to -23%, which IS a 30% decrease in melee offence. But, you still have to add 18.5% back to it, so I end up at -4.5%. So after Fire-lore debuff, my Melee Offence looks like this: -4.5% Melee Offence (18.5% + -23%). Once you calculate what the total change in percent was, it comes out that my Melee Offence was only reduced by 25%! That equation is z = (x-y)/y, where x = debuffed total, y = non-debuff total, and z = total percent change. Plug in the numbers: (95.5 - 128.5)/128.5 = -0.2568. Times that by 100 to get -25.68%, which is the actual reduction. Note: When it says -4.5%, it really means your Melee Offence is at 95.5% of your base, which is where the 95.5 comes from. Basically, if a creep were to apply a Fire-lore-like debuff to my captain, the actual debuff would be less potent than intended.

Now, 5%, you may say, isn't that bad... and it's really not. The problem occurs when/if the Melee Offence Rating increases substantially above what any normal player character could get. My champ has 7.5k Melee Offence rating, which is 22.5%; I don't really think it's possible to push it much higher right now, and Fire-lore would still reduce his damage by 25% while in Fervour and 24.5% while not in Fervour, so that's not too bad.

But we can't see what mobs Melee (or Ranged/Tactical) Offence Ratings are, however. If they have enough Melee Offence Rating to make their rating percentage go to 100% (for example), their numbers would look like this: 200% Melee Offence (100% + 0%). Fire-lore is applied: 170% Melee Offence (100% + -30%). Anyone with basic understanding of fractions/percents can see that 30/200 is not 30% reduction, it's actually only a 15% reduction, half of what it should be.

...
The problem, however, may spread to other skills, anything that changes a stat based on a percentage. Oathbreakers, Reveal Weakness, Disable, and Telling Mark, just to name a few. If a mobs Melee/Ranged/Tactical Defence Rating is really high, Oathbreakers would not increase its incoming damage by very much. Just on a playable character with 15% Melee Defence from Rating (the highest you can go with the cap), Oathbreakers would only increase incoming damage to that person by 30%.

...

I can see how this could have been missed, and I think it'll probably be a bit difficult to fix, because you have to make sure everything stacks right. It'll be interested to see how badly this actually affects the skills. I hope the dev posts again soon with more information...

the fact that you have to post all this math is a testament to the fact that something is broken. The -30% should come off after all other calculations are done, and taken from the actual total (if mob has +200%, the final result of Fire-lore should be -60% of 200, which would be the true -30%)

You could make a case about the tool-tip, but the fact of the matter is, 30% should always be 30%. Other parameters on a mob should not take precedence or otherwise chance that number, especially when the average player can do nothing but read the tooltip and assume its correct. Any other outcome is BS, and I'm pretty sure Turbine knows it. I'm also hoping that they do want to fix it, but it may take a long time to rework a new CU in order to do so. it is what it is

[Raskolnikov]

closer to finding an answer yet, Rask?

Still investigating/discussing, but I'm also trying to get my monster behavior stuff done (amongst other random tasks). Hopefully next week I'll be able to make some time.

[aleczander]

Still investigating/discussing, but I'm also trying to get my monster behavior stuff done (amongst other random tasks). Hopefully next week I'll be able to make some time.

you need a combat-monkey assistant! Want me to write your boss and ask for more funding?

jk

[john_anthony]

The -30% should come off after all other calculations are done

this is the issue, very well and succinctly stated.

the problem is that the debuff was not updated to take into account new additions to the damage formula that came with mirkwood.

what needs to happen is that the two affected debuffs (fire-lore, wind-lore) have to be put back "outside the parentheses," calculated last so that they actually do what they say they're supposed to do. (for fire-lore, that would mean taking 30% off melee damage, instead of the 5% it currently does against pve boss monsters as a result of this problem, for example. ditto for wind-lore on ranged damage.)

[DopplersRadar]

The burglar forum has a post that shows this affecting the burglar skill disable also. To my knowledge this does NOT affect reveal weakness or telling mark. So I am guessing it does not affect buffing skills, only de buffing skills

http://forums.lotro.com/showthread.php?t=327142

[scrubmonkey]

I tested an EM and came up with -6% melee damage from untraited disable for the US side as well. Posted here:

http://forums.lotro.com/showthread.p...62#post4569862

[aleczander]

Still investigating/discussing, but I'm also trying to get my monster behavior stuff done (amongst other random tasks). Hopefully next week I'll be able to make some time.

It's next week, so you officially need an assistant. Tell your boss we said so.

[Raskolnikov]

It's next week, so you officially need an assistant. Tell your boss we said so.

I won't argue with you.

But yeah, the rabbit hole is deep indeed (and I was out on Wednesday).

[aleczander]

Since this seems to be a bug that affects all mob debuffs, it made me think about an old thread: the one from late december that talked about the Watcher being "harder".

The figuring was, since the devs confirmed no changes were made to the boss, then the difference was a result of some mysterious buff (the devs won't give specific numbers, hence the mystery)that all sigs/elites + got from the SoM update. Since LM debuffs are a big part of Watcher damage, this could help explain the differences in parses pre and post Dec. 1
Or, is this new math to calculate the buffs differently the exact and single source for the "Bosses will be tougher" talk that was advertised with the SoM release?

/conspiracy rant off

[Galahadur]

I won't argue with you.

But yeah, the rabbit hole is deep indeed (and I was out on Wednesday).

Someone decipher this for me?

Does it mean that things are not WAI and that there is indeed a bug?

[aleczander]

Someone decipher this for me?

Does it mean that things are not WAI and that there is indeed a bug?

I think rask's quest to officially tag this as a bug is long and challenging, and hidden in wonderland

[BIZL]

I would say if they have to look into this at all then it's not WAI.

You don't intentionaly modify damage calculations and not know about it.

The question is, will they do anything about it.

You can bet your *** that if this was reversed and disable somehow made the mob do 15% damage instead of -15% damage there would be a pretty rapid hot fix.

[findorin-gilrain]

Someone decipher this for me?

Does it mean that things are not WAI and that there is indeed a bug?

I guess so. An adventure much longer and more curious than originally expected....

[JohnnyManeuvers]

There's also a chance he's only playing in his sandbox. So it might be WAI because another team made the change that he didn't know about.

For example, the team that designs raids or balances the game felt a full -50% melee is too effective on the boss or sig mobs. So they do a change to the skill that reduces it based on level. Nobody tells Ras who in charge of designing monsters not overall game balance or instance design.

So you get a situation where it's possible it's not a bug and Ras is trying to figure out why the change was made. Hence going down the "corporate" rabbit hole.

[Elfheart]

Well, if this is WAI then I fully expect a DPS buff to both LM and Burglars since their primary roles of debuffs and CC has been made obsolete and only DPS seems to matter anymore.