3.3.5 – Unholy BiS Lists

Rather self-explanatory title, I would think! I’m still working on additional variations (non-heroic, 10m only, etc), but figured I would throw these ones up, as it’s getting late and the others will take a bit of time (the more limitations you place, the longer it takes to perform the calculations), and thus probably won’t be ready until tomorrow.

The EJ thread will take slightly longer to be updated, largely because I have to format all of this into tables, and since I’m not entirely sure I’ll include all of these variations there, I have to make up my mind and pick and choose among the lot.

When looking at these, ignore the total score number. For set bonuses and such I don’t enter the actual value, but just a placeholder since I can simply force it to take the bonuses anyways (and it’s guaranteed that you’ll want to, of course).

Without further ado:

[A] All Items

[H] All Items

[A] All Items + Expertise Capped

[H] All Items + Expertise Capped

[A] All Items sans Shadowmourne

[H] All Items sans Shadowmourne

[A] All Items sans Shadowmourne + Expertise Capped

[H] All Items sans Shadowmourne + Expertise Capped

[A] All non-25-heroic Items sans Shadowmourne

[H] All non-25-heroic Items sans Shadowmourne

[A] All non-25-heroic Items sans Shadowmourne + Expertise Capped

[H] All non-25-heroic Items sans Shadowmourne + Expertise Capped

More to come tomorrow.


30 Responses to 3.3.5 – Unholy BiS Lists

  1. Vargr says:

    Argh, the return of leather BiS :-/
    Thanks for your listing, it’s very usefull

  2. Angvard says:

    I don’t really know how these are calculated, but I am pretty sure that the Alliance – All Items version is not that good, 5 expertise will often leave you wasting RP. Dunno, from experience about 15 expertise works really well in ICC heroic, more doesn’t seem to be much of an improvement, so personally I’ll stick with something close to 15 😉 Am I missing something here or is this just the way these lists are supposed to be, “in theory, but…”?
    By the way, you forgot to mention that these lists are for unholy, some readers might not know that.

    • Consider says:

      These are calculated how all BiS lists are calculated – by generating stat weights (which are done via a variety of simulators and spreadsheets), and then applying those stat weights to figure out what item is best in each slot (while keeping in mind caps and the fact that some stats get stronger/weaker the more/less of them you have). The underlying math behind it all is a bit more complex than that, of course, but such goes without saying.

      Expertise just isn’t that valuable *on paper*. I understand some people dislike not having it capped for various reasons, or find themselves in special situation (high latency and/or multiple resto druids/disc priests) where its value gets inflated, hence why I included options which cap it.

      But aside from those special circumstances, expertise doesn’t have some mythic value. On average, even accounting for the additional dodges, the gain in other stats will outweigh that decrease of RNG that expertise. You can have a yellow ability be dodged each ~20 second rotation without your rotation getting messed up, due to the grace period and the like. There will be times where you get multiple dodges in a rotation, of course, but there could also be times where you get a DC resisted ten times in a row – that’s no argument to cap spell hit!

      Thus, mathematically speaking, the first set is most optimal (meaning it will produce the most dps on average), and the expertise value is neither intentional nor unintentional – it’s just what it is.

      Don’t get me wrong, I see nothing wrong with capping expertise (as long as you do so intelligently; expertise gems and the like are just silly and suboptimal). It’s just not necessary to do so, and you’ll actually produce less dps (not much, but slightly so) on average doing so.

      • Angvard says:

        Thanks for the explanation, I just assumed you tried every possible (or reasonable) gear-combination and compared the simulated dps. I guess that would be a bit time consuming.

        The part about restos/disc priests is pretty much why I was talking about ICC heroic, you usually can assume at least two of those, in my caste most of the time three. In addition you can gain massive amounts of RP in most of the fights, usually a full 130 about every minute. I realize this is all very, uhm, non-scientific, I just don’t like wasting RP after having several dodges within 20 seconds. I currently (have to, damn drops) run with 10 expertise including talents, and I often get two dodges within a 20 second rune-set. Maybe I’m just unlucky, random is random etc. 😉

        Anyway, could you please point me to the tool shown in the screenshots, I’d like to toy around with stat-weights with added RP-gain and see how it changes things.

      • Consider says:

        Zerack’s. One has to do some editting of it, as it doesn’t currently have the latest drops, the stat weights are slightly dated, and a couple other minor factors, but those (easily modified things) aside, it’s an excellent tool.

  3. Baadshah of Dawnbringer says:

    In my case, when you take into account my horrid latency and awful RNG luck (seriously, not kidding!), I just like staying on the safe side and using at least 15 Expertise. Plus we run with 3 (yes, 3) resto druids and 1-2 Disc priests on average. Hence Expertise goes way up in value for me.

    Two fights where I WILL equip the Tier 10 legs, as opposed to the Scourge Reaver’s Legplates, are Putricide and Sindragosa. Sindragosa in particular is one where I’ve noticed a LOT of dodges (and obviously I’m not attacking from the back or front).
    Unfortunately, if I end up getting Shadowmourne, I may end up having to use the Tier 10 legs a lot more often since currently I’m DW Frost for fights like Saurfang and the Aboms.

    On a completely unrelated and very annoyed note: I was hacked! The hacker deleted everything except for my sigil, Shadow’s Edge and Tier 10 armor and deposited my main toon on Kel’Thuzad server. He couldn’t have picked a worse time, too, seeing as our first Shadowmourne is about to be completed (and hence the second one is up for grabs).

    • Consider says:

      That sucks. Like really, really sucks. In general, but especially considering the timing.

      Blizzard is usually pretty good these days about restoring stuff, but it’s still a huge inconvenience ><. My sympathy.

      • Baadshah of Dawnbringer says:

        I’m currently unable to access my DK on Kel’Thuzad; apparently it’s locked until they can transfer it back. Of course they are taking their sweet time doing so, and the timeframe they gave me over the phone is approximately 2 weeks.

        Unfortunately, it looks like my being there or not wouldn’t have mattered… The next Shadowmourne in our guild is going to an mainspec Prot Warrior.

        I won’t even pretend to be happy about the decision, but it’s been announced so I guess not much can be done.
        Oh well…

        (Sorry if this isn’t the right place for this, just felt like venting)

  4. Dante says:

    (Sorry for my bad english) I’m really not that sure as regards the Horde BiS:

    _ First, you chose BS (which is not the best Profession any more according to your EJ post) instead of Engineering, so there is no Icewalker (12 hit) and no Precision on gloves ( 20 hit)

    _ The Meta isn’t the best one (if we take in account your ” Frost SM ” statsweights), and a Etched ametrine on a blue socket (chest) instead of a Sovereign Dreadstone doesn’t seems that good.

    _ With The Polar Bear Claw Bracers (heroic) and all yellow sockets filled by a Fierce Ametrine, it leads us with 282 Hit Rating (According Rawr) so I don’t think those Umbrage Armbands (heroic) are better.

    • Consider says:

      1) You have two professions, you know! I chose BS simply because it wouldn’t affect the results any which way. Nor would JC. Nor would alchemy. Nor would any profession save Engineerneering (which, yes, is best). If you use engineering, all you have to do is switch three Fierce (10str/haste) gems for three Etched (10str/hit) gems to cover for the lost hit. Simple as that. No need for another variation of each list!

      2) Yeah, that’s simply a screw up on my part. I forgot to check off the 10str/stam gem from that menu, and at the point I realized it, I didn’t want to go back through and start all over for that. Considering the difference is the equivalent of about 2 AP, I didn’t see it as huge deal. It will be fine when I make the tables on EJ, but for these screenshots, it just wasn’t worth the hassle of doing it from scratch.

      3) Nah, it’s not better. You could go ahead and multiply it out by the stat weights and see. Umbradge win out, mainly by virtue of their higher ilvl.

  5. Dante says:

    Well, I gave it a try and this is what I got:

    Frost SM Statsweights

    _ Polar Bear Claw Bracers (Heroic)

    108 (Strength) x 3.11 = 335.88
    60 (Hit) x 3.70 = 222
    68 (Crit) x 2.11 = 143.48

    Fierce Ametrine => 14 (Str) x 3.11 = 43.54
    10 (Haste) x 2.97 = 29.7

    335.88 + 222+ 143.48 + 43.54 + 29.7 = 774.6

    _ Umbrage Armbands (Heroic)

    109 (Agility) x 1.41 = 153.69
    73 (Crit) x 2.11 = 154.03
    73 (Haste) x 2.97 = 216.81
    113 (PA) x 1 = 113

    Bold cardinal Ruby => 20 (Str) x 3.11 = 62.2
    (If we assume the Socket Bonus) 4 (Agility) x 1.41 = 5.64

    153.63 + 154.03 + 216.81 + 113 + 62.2 + 5.64 = 705.37

    Polar Bear Claw Bracers Heroic (774.6) > Umbrage Armbands Heroic (705.37)

    Now, with the Umbrage Armbands (Heroic) we are at 222 Hit rating so we need 4 Etched Ametrine (JC/Engineering as professions); a lost of 4 Fierce Ametrine , which mean 40 Haste.
    Finally, I might be wrong but I think you forgot the fact that we need 8% Hit rating (Horde) and not 7% (Alliance).

    • Zibidideath says:

      But 282 hit means that of that 60 hit on the Polar Bear Claw bracers is going to waste. 8% hit is still only 263 hit rating.

      19 (Hit overcap) x -3.7 = -70.3

      774.6 – 70.3 = 704.3

      Polar Bear Claw Bracers Heroic (704.3) < Umbrage Armbands Heroic (705.37)

  6. Dante says:

    19 Hit rating sure, but after 8% it isn’t worth 3.70 anymore if i’m not mistaken

    • Zibidideath says:

      No, but you counted it as worth a positive 3.7 for every one of those 19 points, I was merely removing that deficit.

    • Zibidideath says:

      Honestly though it isn’t about what’s better based on the “stat weights” because as any theorycrafter/min-maxer will know, as your gear changes so do stat values, including crit and haste. Just run the sims based on your gear and you’ll be fine.

  7. Branith says:

    any chance of a BiS list that doesnt include any heroic loot what so ever. My guild doesnt do 10 man nor do they care about progression in 25 man. Basically jsut want a BiS list with 25 man regular only.

  8. Dante says:

    Argh ! I forgot to mention the Spell hit calculation !!!

    774.6 – 70.3 = 704.3 But
    19 (Hit rating) x 0.63 (Spell hit) = 11.97
    704.3 + 11.97 = 716.27

    Polar Bear Claw Bracers Heroic (716.27) > Umbrage Armbands Heroic (705.37)

    (I know it depends on the gear but I really want to be sure about the True best in slot 🙂

  9. Wesser says:

    Noticed on the expertise capped set for alliance that you used http://www.wowhead.com/item=50728
    as opposed to http://db.mmo-champion.com/i/54581/penumbra-pendant_/. according to the sim that set sits at 31 expertise, wouldnt it be better to use the pendant since expertise basicly has no value for dps past 26 expertise?

  10. Jstar says:

    Isn’t this basically ‘True BiS’?

    If it’s not, what are these stat weights based on? Average of a few different sets or average of the ‘True BiS’ set?

    It’s confusing since to calculate stat weights you need gear, yet if your gear changes so do the stat weights. There’s never a constant, or am I missing something?

    • Zeran says:

      In my understanding, these are the BiS lists because (given their stat weights) there are no improvements left. That is to say that they’re the true BiS Lists.

  11. Volger says:

    Sort of off-topic, but can i have some clarification on how to gem an unholy DK? Looking at your BiS Lists, you list the Ashen Ring rep ring with a Fierce Ametrine gem, instead of ignoring the socket bonus and using a Bold Cardinal gem. I tried looking over your EJ post and saw the section about gemming strategies, but I can’t seem to find the theorycrafting behind doing so. Would someone be able to break it down for me?

    And also, regarding some of the above posts, what does having resto druids and/or disc priests have to do with stacking Expertise?

    Thanks for the help.

    • Baadshah of Dawnbringer says:

      I thought I’d respond to your second question regarding Disc Priests and Resto Druids. The answer to this is the talents these two classes have: Revitalize and Rapture. They interact amazingly well with DKs and provide lots of free Runic Power With multiple resto druids for instance, I have an almost guaranteed, steady stream of outside RP available to me.

      For Unholy DKs, this means lots more Death Coils, which in turn mean less empty GCDs. This tightens up our priorities quite a bit (assuming you don’t want all that RP to go to waste), so dodges start costing you more DPS when they happen.
      As such, Expertise’s value goes up.

      In my case, bad latency also makes it more valuable due to longer communication time with the server.

  12. Zeran says:

    To answer the first part, haste is that good for us if you can pick up a proper sized socket bonus. In short, if you multiply the stats out by their weights you get bigger numbers gemming the way Consider posted.

    And this is why I don’t post on Ej.

    • Volger says:

      Thanks for the responses. I’m wondering how much of a dps increase switching from a Bold Cardinal to a Fierce Ametrine in my Ashen Verdict rep ring will be.

      *glaces at Consider*

  13. Dante says:

    It depends of your Stats weights so it’s not very accurate but if we take the most common profil (Frost ilvl 264):

    _ Bold Cardinal Ruby: 20 (Strength) x 3.11 (Strength weight) = 62.2

    _ Fierce Ametrine: 14 (Strength + Socket Bonus) x 3.11 = 43.54 | 10 (Haste) x 2.75 (Haste weight) = 27.5

    43.54 + 27.5 = 71.04 | 71.04 – 62.2 = 8.84

    8.84 dps increase.

  14. tjn says:

    I understand the theory behind BiS lists and the like, but some of the gemming for these lists don’t pass the sanity test and that undermines confidence in the rest of the choices, specifically when dancing around the hit cap.

    For example in my own case I am unlikely to get Shadowmourne in the near future and while I am alliance, we have a shortage of draenais in raid, but my guild is 10/12 heroic in ICC, so I went to the All items Horde BiS sans Shadowmourne list.

    Now, I assume that Strength > Haste, point for point. But in the above list it has gemmed in the chest slot a Bold and an Etched (assuming in for hit cap reasons), ignoring the socket bonus. If we were to trade that Etched with a Fierce in a different slot to preserve our hit, the socket bonuses wouldn’t change at all.

    But now we’re left with a Str/Haste gem that could be replaced with a full Str gem without any other effect. Which leads to the conclusion that 10 Str and 10 Haste is somehow better than a full 20 Str and that isn’t consistent with the rest of the gemming.

    • Consider says:

      That’s an odd error, and I’m not really sure how it comes about. The items matter more than the gemming, of course, and some quick math shows that the item don’t change even if you swap that etched to a bold and one of the fierce to an etched.

      It’s a problem with the program, but not sure how it produced it.

      • tjn says:

        Similarly, the horde all items + expertise capped list uses a rigid in a colorless socket which could be traded out to full strength by ditching some fierce gems for etched.

        It’s minor nitpicks, but BiS lists should be BiS lists. =p

        The program looks like it prioritizes gemming in colorless or sockets that break the socket bonus first when it should prioritize the current most valuable stat for it’s respective gem color. The program would probably need to exclude the rigid cut from consideration, or add additional directions to the program in order to optimally deal with the hit cap.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: