Page 2 of 4 FirstFirst
1
2
3
4
LastLast
  1. #21
    I hope you realize that on Tortos it's primarily meter padding

  2. #22
    Mechagnome Italiandk's Avatar
    Join Date
    Mar 2012
    Location
    Dô yöû even liƒt¿
    Posts
    671
    Quote Originally Posted by Skullbound View Post
    festerblight isn't on top for those fights. I hope you know it's just aoe with dots. You could fb them but no one does. It's not worth it. So your argument is invalid.
    This.

    Typical single target normal fights with no damage modifier or AOE (on average):
    - megaera UH > Fr for about 2-3%
    - durumu UH > Fr for about 10%
    - iron qon UH > Fr for about 5-6%

    Now, mabybe UH is higher than other melees but overall it's in the middle of the pack (a bit higher on 25man tho) so I don't see why it should be too high. Probably a buff to SS is not really needed and probably it won't never go live for that matter, a nerf to FB, instead, it's just right because, imho, FB is killing the fun to play UH.

    Quote Originally Posted by Skarssen View Post
    Festerblight damage is too high already.
    Quote Originally Posted by Skarssen View Post
    WoL doesn't differentiate between festerblight and "standard" unholy, so it doesn't matter which playstyle is highest on a particular encounter. I am referring to unholy as a whole, and it is currently an outlier.
    Try to be more clear then...
    Death Knight -> Italiandk
    Shaman -> Cashrain
    Paladin -> Healyoutru
    Priest -> Ängel
    Monk -> Skadoosh
    Mage -> Icydeath

  3. #23
    Quote Originally Posted by Italiandk View Post
    Typical single target normal fights with no damage modifier or AOE (on average):
    - megaera UH > Fr for about 2-3%
    - durumu UH > Fr for about 10%
    - iron qon UH > Fr for about 5-6%
    I'm not interesting in data from normal modes, or "all parses". Both are worthless. Top 100 isn't good either, but it is less worthless than all parses. Either way, the delta is greater than what you posted. No non-cloth specs (don't compare DKs to cloth, they play by different rules) excel at both single and multi-target damage the way unholy does.

    This is just a comparison of frost and unholy currently;

    Jin'Rohk = 17%
    unholy - 260k
    frost - 217k
    Horridon = 15%
    unholy - 311
    frost - 267
    Council = 23%
    unholy - 224
    frost - 174
    Tortos = 41%
    unholy - 570
    frost - 337
    Megeara = 16%
    unholy - 208k
    frost - 176
    Ji-Kun = 10%
    unholy -212
    frost - 192
    Primordius = 11%
    unholy - 177
    frost - 158
    Durumu = 26%
    unholy - 211
    frost - 158
    Animus
    unholy - 181
    frost - insufficient data
    Iron Qon = 16%
    unholy - 175
    frost - 147


    And that isn't even the point. Many classes have similar or greater deltas in damage done, so we aren't unique. However, I don't see many other classes with a spec that is top 5 on 11/12 encounters receiving buffs. That is the issue, not so much the fact that frost stinks in comparison.

    ---------- Post added 2013-05-03 at 12:34 PM ----------

    Quote Originally Posted by bals View Post
    I hope you realize that on Tortos it's primarily meter padding
    If you are kiting bats, sure. If you are killing them, no.

  4. #24
    Quote Originally Posted by Skarssen View Post
    If you are kiting bats, sure. If you are killing them, no.
    which logs don't differentiate.
    also you can PS the crystal to get a new shield as well which is more padding.

  5. #25
    Dreadlord vmagik's Avatar
    Join Date
    Dec 2012
    Location
    Acherus
    Posts
    858
    Quote Originally Posted by Skarssen View Post
    *snip*
    This happens every tier though, remember how low unholy was last tier? That's not because it was shite (it wasn't) but the best players went frost because it had an edge. This tier plays to unholy's strengths, diseases rolling on adds out of AOE/cleave range so a lot of DKs see unholy edging ahead and decide that's the spec for them.

    Seriously, meter padding means bugger all... Couple that with how strong feather is for buffing diseases (whether you play true festerblight or not) and you're not going to see a lot of good frost DK parses, even if DW is still competitive.

    Also, you're all forgetting ghostcrawler has an unholy dk!
    Amazing signature courtesy of Shyama

  6. #26
    Field Marshal 1Snail's Avatar
    Join Date
    Sep 2012
    Location
    United States
    Posts
    61
    Quote Originally Posted by broodax View Post
    All of those things applied to the old Scourge Strike too. Any percentage buff from another source will mean exactly a 22% increase in SS damage from what it was before. If anything, the final difference should be less than that if there are any non-percentage modifiers to SS (I don't know if there are).

    x * 1.22 = 22% increase over x
    (x * a bunch of modifiers) * 1.22 = 22% increase over (x * a bunch of modifiers)

    Maybe if you'd adjusted the action list based on the change to somehow favor SS more, or use it with procs better, you could see a bigger change, but we're talking about the exact same action list played out, and the only difference is SS does 22% more damage.
    That just isn't how it works at all. A lot of things scale non-linearly or multiplicative. Things are rarely as static as you claim.

  7. #27
    Quote Originally Posted by Skarssen View Post
    I'm not interesting in data from normal modes, or "all parses". Both are worthless. Top 100 isn't good either, but it is less worthless than all parses.
    You're basing your argument on a dataset comprised of only 100 parses per spec. Please think about that for a moment before making your response.

  8. #28
    Problem with using rankings to judge a classes dps is that they are incredibly skewed by both RNG and cheese, especially with the RPPM trinkets, if you get a lucky streak of procs and/or have multiple rogues feeding you tricks you are gunna rank damn high regardless of the power of the spec.

    As for Tortos ranks, the top 200 parses for unholy (that's all I count none of this premium ranking stuff) pretty much all have the bats kited into range of Tortos near then end (when there are 60+) of them for a pestilence, basically "padding" the meters solely to get ranked, you can tell by the dps distribution and chart on the logs (there's always a steep spike near the end) so those logs are hardly indicative of the specs overall power when 60 - 80% of the damage is meter padding JUST to get ranked.

    As for Festerblight not being fun, that's subjective, I personally find classic Uh rather boring, where as for festerblight I actually have to pay more attention to what's going on to both me, the boss and my surroundings, I find it far more fun to play and will be sad if they nerf it to a point where its no longer worth playing.

  9. #29
    Quote Originally Posted by Zeeket View Post
    Problem with using rankings to judge a classes dps is that they are incredibly skewed by both RNG and cheese, especially with the RPPM trinkets, if you get a lucky streak of procs and/or have multiple rogues feeding you tricks you are gunna rank damn high regardless of the power of the spec.
    That's the main reason why you use the "all parses" dataset, to minimize the impact of RNG. If you're in a serious guild and only want to look at elite players, choose the 90th percentile rather than the median.

  10. #30
    Quote Originally Posted by Schizoide View Post
    That's the main reason why you use the "all parses" dataset, to minimize the impact of RNG. If you're in a serious guild and only want to look at elite players, choose the 90th percentile rather than the median.
    90th percentile changes little. Unholy drops to 6-7 on Primordius, and Animus, but is still top 5 on 5 others, still top damage overall (even without tortos) and still crushes frost. Any lower than 90th and you have players who are undergeared, underskilled, in guilds with inferior overall DPS, or some combination of the above. I don't know what Blizzards intent was with the latest round of changes, but seems strange that they would buff a plate DPS that is already so high.
    Last edited by Skarssen; 2013-05-03 at 06:36 PM.

  11. #31
    Yes, in other words, the vast majority of players in this game.

    Anyway, even though you came to the same conclusions using all parses at 90th percentile this time, that won't always be the case. Using the right dataset is important.

  12. #32
    Quote Originally Posted by 1Snail View Post
    That just isn't how it works at all. A lot of things scale non-linearly or multiplicative. Things are rarely as static as you claim.
    You are aware that this is the literal definition of a proportion, are you not?

    If the priority does not change, and the only change that is made is the damage of SS, then the sim will use SS the same number of times. If the sim uses SS the same number of times, but the post-change SS does (arbitrary) 39% more damage, then the relative difference in damage dealt by SS is going to be +39%.

    Let n = 500 (doesn't matter), d_a = 250, and d_b = d_a * 1.39.

    n * d_a = 125000
    n * d_b = n * d_a * 1.39 = 173750
    (n * d_b) / (n * d_a) = d_b / d_a = 1.39

    The DPET should follow a similar suit, as the execution time is fixed, but the damage changes (proportionally, might I add for emphasis). I don't even need to pull my sim out to verify this.
    "I have it all simmed."
    Euliat

  13. #33
    Quote Originally Posted by SSHA778 View Post
    You are aware that this is the literal definition of a proportion, are you not?

    If the priority does not change, and the only change that is made is the damage of SS, then the sim will use SS the same number of times. If the sim uses SS the same number of times, but the post-change SS does (arbitrary) 39% more damage, then the relative difference in damage dealt by SS is going to be +39%.

    Let n = 500 (doesn't matter), d_a = 250, and d_b = d_a * 1.39.

    n * d_a = 125000
    n * d_b = n * d_a * 1.39 = 173750
    (n * d_b) / (n * d_a) = d_b / d_a = 1.39

    The DPET should follow a similar suit, as the execution time is fixed, but the damage changes (proportionally, might I add for emphasis). I don't even need to pull my sim out to verify this.
    Thanks, I had given up. Scourge Strike has no hidden interactions with anything; it doesn't matter if things are rarely static, this is. If its damage is increased by 22%, then the damage it contributes will increase 22%. The non-crits will increase 22%, the crits will increase 22%, the shadow portion will increase 22%, all of the damage it does will increase exactly 22%, and the damage of every other ability will remain entirely unchanged.

    The new sim is wrong.

    If I had to guess I'd say that the damage in the sim was increased by 30% (which is an incorrect way of seeing what actually did happen. it went from 135% to 165% weapon damage).

    Quote Originally Posted by Skarssen View Post
    90th percentile changes little. Unholy drops to 6-7 on Primordius, and Animus, but is still top 5 on 5 others, still top damage overall (even without tortos) and still crushes frost. Any lower than 90th and you have players who are undergeared, underskilled, in guilds with inferior overall DPS, or some combination of the above. I don't know what Blizzards intent was with the latest round of changes, but seems strange that they would buff a plate DPS that is already so high.
    You apparently have some deeply held belief that plate dps should do less damage than other classes, and you believe that blizzard shares that belief. I'd contend that's entirely false, and a class that is often at the top of meters, but often overtaken by other classes, is in very good shape and might in fact see some adjustments, but much more would be required for it to be an outlier or to be broken or anything of that sort.
    Last edited by broodax; 2013-05-03 at 08:22 PM.

  14. #34
    Fury in particular historically always topped meters near the end of each expansion due to superior gear scaling. There is no anti-plate bias.

    That said, festerblight unholy is very strong right now. Not strong for a plate spec, or high for a melee, it is top-tier DPS overall. Any additional buffs will be short-lived, and looking back at history, likely hotfix overnerfed to compensate.

  15. #35
    Quote Originally Posted by Skarssen View Post
    90th percentile changes little. Unholy drops to 6-7 on Primordius, and Animus, but is still top 5 on 5 others, still top damage overall (even without tortos) and still crushes frost. Any lower than 90th and you have players who are undergeared, underskilled, in guilds with inferior overall DPS, or some combination of the above. I don't know what Blizzards intent was with the latest round of changes, but seems strange that they would buff a plate DPS that is already so high.
    Players are currently reporting D&D to be stronger single target than SS. The buff is to fix this, that is the intent. Keep the big picture in mind as well. Just because UH interacts so amazingly with the current tier of trinkets, doesn't mean that the next tier will be as so easy to roll super diseases.

  16. #36
    That's true, simcraft shows DnD hitting harder than ScS. It starts out hitting 9.8% harder in T14N, then 14% in T15N, and 16.7% in T15H. Not clear if that's a simcraft artifact or something actually in the game, though, and of course it only applies if the target stands in the DnD for its full duration. DnD definitely wasn't worth using on single targets in 5.0.

    That said, from the GC tweet I posted in the festerblight thread, it looks like the change is aimed at making festerblight less of an obvious choice, not DnD vs ScS. But that could be a target as well and GC just didn't mention it, he only has 140 chars after all.

  17. #37
    Quote Originally Posted by Djouga View Post
    You'd have a hard time convincing anyone festerblight is retarded, since you're probably stuck at 6/13 at best.
    yes i stuck at 6/13, cuz da hc ist simply too hard, sorry. festerblight is the worst spec ever, even obli unholy was better

  18. #38
    Field Marshal 1Snail's Avatar
    Join Date
    Sep 2012
    Location
    United States
    Posts
    61
    Quote Originally Posted by broodax View Post
    The new sim is wrong.
    Definitely not. The changes were implemented to the spells, nothing else changed. I'm starting to doubt whether I should even bother informing the community on new changes for classes. The blizzard forums appreciate it, people like you just argue for the sake of arguing and it's really annoying.

  19. #39
    I'm guessing you typo'd something or used a really low iteration count or whatever, because I just did the overrides to match SimC's scourge strike + festering strike with mmo-champion's/wowdb's datamining info, and I'm not seeing the same leap you are for scourge strike.

    override.spell_data=effect.48018.average=0.7903
    override.spell_data=effect.48019.base_value=165
    override.spell_data=effect.87374.average=0.7582
    override.spell_data=effect.87375.base_value=175

    Live
    http://dread-gaming.com/simc/unholy_mop_02.html

    PTR hypothetical
    http://dread-gaming.com/simc/unholy_mop_03.html

    +23.27% DPET for scourge strike (compared to your +29.97% and the theoretical +22.22%) and -11.76% DPET for festering strike (compared to your -12.38% and the theoretical -12.5%).

    As a side note, the reason why my scourge strike numbers have a slightly higher than theoretical gain and my festering strike numbers have a slightly lower than theoretical loss is because wowdb's +damage values for scourge strike and festering strike are different than wowhead's / simc's.

    wowdb:
    FS: 200% weapon damage + 1080 (live) / 175% weapon damage + 945 (PTR)
    SS: 135% weapon damage + 806 (live) / 165% weapon damage + 985 (PTR)

    wowhead:
    FS: 200% weapon damage + 540 (live) / 175% weapon damage + 540 (PTR)
    SS: 135% weapon damage + 597 (live) / 165% weapon damage + 597 (PTR)


    If I override the +damage values in SimC to be consistent with what mmo-champion/wowdb believe they are on live, then:

    http://dread-gaming.com/simc/unholy_mop_04.html

    Compared to the hypothetical PTR results from before, that's +22.7% DPET for scourge strike and -12.7% DPET for festering strike, which is roughly consistent with the theoretical gain/loss for each ability.

    I'm not sure whether wowhead+simc or wowdb is correct for the +damage values on live, but I'm *guessing* it's wowhead+simc (wowdb's value appears to be wowhead+simc's values multiplied by the weapon damage % inheritance for each ability).
    Last edited by Nitwit; 2013-05-04 at 02:58 AM.

  20. #40
    Quote Originally Posted by 1Snail View Post
    Definitely not. The changes were implemented to the spells, nothing else changed. I'm starting to doubt whether I should even bother informing the community on new changes for classes. The blizzard forums appreciate it, people like you just argue for the sake of arguing and it's really annoying.
    I'm not arguing for the sake of arguing, I'm telling you that it is impossible for a 22% straight damage buff to an ability that has no effect on any other abilities to have more or less than a 22% damage increase to that ability overall. Nitwit was kind enough to run it again and show that you must have done something wrong, but we didn't need to do that to know your sim was wrong because math isn't hard.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •