I sincerely hope this ends the debate. He's essentially replicated the same math that he used in during the MoP beta, which lead him to his initial assertion then. Nothing has really changed since, and I have yet to see anyone satisfactorily challenge the conclusions he has drawn.
For those with little patience, I'll simply quote the conclusion he draws after his testing:
I would recommend that you do not stall to allow for more KM-OB’s. You are better off ignoring it.
There are three ways to get rid of unholy runes, let's see which is the best.
Death and decay: 48.2k dmg
Plague strike: 23.2k dmg + 100k (full duration disease, 10k per 3 seconds)
Obliterate: 78.1k dmg +45% chance for free HB = 113.2k dmg
HB: 78.0k dmg
Obviously your best bet is double HBs whenever possible, but let's say you have 1 unholy and 1 frost/death, what should you do?
In order from best to worst:
Using plague strike and HB gives 201.2k dmg IF diseases are about to fall off
Using plague leech, plague strike and 2 howling blasts gives 179.2k dmg
Using DnD and HB gives 126.2k dmg
Using obliterate gives 113.2k dmg
Using plague strike and HB gives 101.2k dmg
Ignoring the unholy rune and just spending the frost/death on HB gives 78k dmg
While it's true you can use plague strike and howling blast at half disease duration and count the 50% refresh as 50k dmg, it prolongs the ability to fully refresh diseases via the top option for 15 seconds, and is therefore not a damage gain. If blood plague is not just about to fall off, count it as if it isn't refreshing at all.
So, if diseases are about to fall off, plauge leech is off cd, or dnd is off cd, use the above options to remove unholy runes, otherwise just use obliterate if you have to.
However, I've noticed that simcraft has DnD very low for DW, and as such, I played around with it a bit and tried moving DnD up, which led to a slight overcapping which removed all the benefit. GCDs strike again. As such, I've come to the following conclusion, and updated the OP.
Note: Due to DW's current state of being close to GCD locked most of the time, we need to reexamine the way to deal with unholy runes. While the above-mentioned list is accurate in terms of numbers, obliterate is more complicated than it seems. Obliterate has two possibilities, it either does 78.1k in one GCD, or 156.1k dmg in 2 GCDs. While we can average this out easily, as done above, the GCD cost has to be considered. Using DnD/HB does slightly more than obliterate on average, but costs 2 GCDs. Using obliterate over DnD is a gain if you are GCD capped. Depending on luck, you'll either gain free damage with the rime proc, or gain a GCD which will avoid overflowing resources which will be a damage gain. If you are not GCD capped, DnD/HB wins out.
Updated the OP with some DW weapons, if you're curious about a weapon combo not listed PM me and I'll add it to the list. Also relocating simcraft profiles used here as it takes AGES to scroll through all of it when I'm editing the OP
And pestilence still works on elegon!
Profile and details for trinkets:
← Click Here
The number following a name is the upgrade level, if any.
50k iterations for every permutation of the top 3 trinkets and all possible upgrade levels
Both raid trinkets are the heroic version.
Using Default BiS DW profile, with the offhand changed to a scimitar, and the trinkets obviously moved around.
No items other than trinkets are upgraded.
Simcraft version 510-8
Patchwerk style no downtime
By request, adding weapon ranking for 2hand as well. Profile is below, here are details:
Ran both viable raiding weapons, normal and heroic versions, with every possible upgrade level. All sims are 50k iterations, using the default gearset and talents, being PL and RE, and are orcs with ench and engineering. Left the Shin'ka profile mostly alone, added both the legendary gem and a 160 strength gem in the prismatic socket. For Starshatter, the expertise had the potential to mess with the numbers, so I hard set the expertise to 2540, 200 higher than it would of been without a weapon, and converted all but 200 of the expertise on the weapon into haste. This yields accurate results, assuming you can find room in your gearset to reforge away the excess expertise you would gain by using starshatter, and turn it onto haste.
So during my downtime, when I dont have runes for obliterate, but do have a single rune. I.e. One unholy, one frost, or one death rune, what do I use them on?
Also, I cant find a good rune addon that seems to work with 5.1, anyone have an addon they would recommend.
Use Acherus, it's by far the best one out there. It's not confusing to set up for someone who isn't UI Savvy, and it does what you want it to do, shows Runes, shows how many seconds they have left, spin when they are back up, and tracks Runic Power.
Set bonus evaluation: This numbers apply to DW frost only, and only single target.
2 set: Your Obliterate, Frost Strike, and Scourge Strike deal 10% increased damage.
Value: 4.77% dps increase
Note: Being nerfed to 4% in 5.2.
New Value: 1.91% dps increase
More notes: Single target only. Value calculated by taking the percent of DW's single target damage (with no set bonus) that is FS and oblit, and multiplying by the increase.
4 set: Your Pillar of Frost ability grants 5% additional Strength, and your Unholy Frenzy ability grants 10% additional haste.
Value: 1.17% dps increase
Notes: 3.51% increase when pillar (20 sec duration 60 sec cd) is up. Value calculated by Comparing a 50k single target sim with and without 4set.
2 set: Your attacks have a chance to raise the spirit of a fallen Zandalari as your Death Knight minion for 15 sec.
Value: 3.3% dps increase
Notes: Single target only, but apeares to proc more often during aoe despite using RPPM system. More testing needed. Value calculated by Hitting a training dummy for 20 minutes on the PTR, and comparing spirits damage to the damage that would of been done with no set bonus.
4 set: Your Soul Reaper ability now deals additional Shadow Damage to targets below 45% instead of below 35%. Additionally, Killing Machine now also increases the critical strike chance of Soul Reaper.
Value: 2% dps increase
Notes: Single target only. Value given by Euliat, from his own sim.
Going from 4 set t14 to double 2sets, after 5.2 hits, is a net 2.1% gain, ignoring the extra stats.
Going from double 2sets to 4 set t15 is a basically net gain, and a net loss, of 0%, ignoring the extra stats.
Once again, numbers apply to DW frost for my calculations.
RPPM system details for reference:
← Click Here
This new system, nicknamed Real PPM, aims to give the random nature of procs, the scaling with haste, and the ability for us to balance them assuming a standard proc frequency. Here’s the short version of how you can expect them to function:
Regardless of how you’re attacking or healing, slow or fast, with DoTs or direct heals, whatever, you can expect to get the same proc frequency, on average.
Dual wielding and having both weapons enchanted with the same enchant will double the frequency of procs that you get.
This Real PPM is increased by your haste %. (The highest of your melee, ranged, or spell haste is chosen).
Simple as that. Whether you’re an Affliction Warlock dealing very frequent DoT ticks and Malefic Grasp ticks, or a Holy Paladin casting purely Holy Lights, or a Combat Rogue quickly attacking and using specials, or an Enhancement Shaman attacking with slow melee attacks and spells, or a Shadow Priest channeling Mind Sear on fifty Onyxian Whelps, you’ll get 2*Haste Windsong procs or 10*Haste Elemental Force procs per enchant per minute.
We’re excited to see how this proc system works out. If it works well, we may start using it for more types of procs. Feedback about how it feels is most welcome.
Here are even more nitty gritty details, if you’re interested:
It can proc from any damage/healing event. It keeps track of the last time it had a chance to proc for that enchant.
It calculates the difference in time since the last chance to proc. It uses that time to determine the chance for that event to trigger a proc.
For example, if you have 22% Haste, it was 1.4sec since the last chance to proc, and you’ve got Windsong, then the chance to proc is 2(ppm) * 1.22(haste) * 1.4(time since last chance) / 60 (sec per min) = 5.693%.
The ‘time since the last chance to proc’ is capped at 10sec, so that your first attack of a fight isn’t a guaranteed proc.
Originally Posted by Ghostcrawler
Below are the scary guts of the new trinkets and metagems. This information is for theorycrafters. You don't need to read this to understand how the mechanics work or enjoy their benefits.
- Brutal Talisman of the Shado-Pan Assault - 15% chance on landing harmful abilities and spells, and melee/ranged abilities and swings. 75 sec ICD.
- Fabled Feather of Ji-Kun - 0.56 RealPPM on landing harmful abilities and spells, and melee/ranged abilities and swings. 22 sec ICD.
- Primordius’ Talisman of Rage – 3.00 RealPPM on landing harmful abilities and spells, and melee/ranged abilities and swings. No ICD.
- Spark of Zandalar - 5.00 RealPPM on landing harmful abilities and spells, and melee/ranged abilities and swings. No ICD.
- Gaze of the Twins – 1.00 RealPPM on critical harmful abilities and spells, periodic spell, and melee/ranged abilities and swings. No ICD.
Update:- Gaze of the Twins – [0.8*(1+MeleeCritChance)] RealPPM on critical harmful abilities and spells, periodic spell, and melee/ranged abilities and swings. No ICD.
Trinket rankings (all done with 522 ilevel version):
Brutal Talisman of the Shado-Pan Assault: 1467 expertise * 1.8 = 2640 (you're going to be expertise capped with or without the trinket, and this value is the best case scenario, assuming you can reforge it all into mastery (expertise elsewhere). In all likeyhood, it will be closer to 1500) 8800 str for 15 seconds, with ~ 82 second intervals, 1610 static strength * 3.67 = 5907 + 2640 = 8547 dps
Fabled Feather of Ji-Kun: 1467 hit * 1.8 = 2640 (you're going to be hit capped with or without the trinket, and this value is the best case scenario, assuming you can reforge it all into mastery (hit elsewhere). In all likeyhood, it will be closer to 1500, or even 1k considering how much hit we have) "Your attacks have a chance to grant Feathers of Fury, granting 1333 Strength every 2 sec for 20 sec." Assuming the first proc starts at t=0, by t=18 the proc will be at 13330 strength, for an average of about 7331.5 strength * 3.67 = 26906. 0.56 rppm * 1.5 from melee haste = 0.84 ppm, 22601 with 1 ppm/4 = 5650 (with 1 ppm and 20 second uptime, and a 22 second ICD, it will proc once every 80 seconds on average, giving it a 25% uptime) _ 2640 = 8290 dps
Primordius’ Talisman of Rage: 1467 crit (587 mastery* 1.8=1056.6/880 crit*1.39=1223.2, reforged to mastery) = 2279 dps. Your attacks have a chance to grant you 1538 Strength for 10 sec. This effect can stack up to 5 times. 3 ppm * 1.5 meleehaste = 4.5 ppm, 75% uptime baseline * 1538 =1153.5 * 3.67 = 4233 dps, however each time a proc is overridden, the duration of the last proc is effectively extended to back to 10 seconds. There is a 75% chance of this happening each proc, with an average extension of 5 seconds, or 3.375 ppm with an uptime of 5 seconds, 16.875 extra seconds a minute, 28% uptime * 1538 * 3.67 = 1587 dps. 8099.5 dps
Spark of Zandalar: 1467 haste (587 mastery* 1.8=1056.6/880 haste * 1.65 = 1452, reforged to mastery) = 2508 dps. Equip: Your attacks have a chance to grant you a Spark of Zandalar. Once you have accumulated 10 Sparks, you will transform into a Zandalari Warrior and gain 7333 Strength for 20 sec. 5 rppm * 1.5 meleehaste = 7.5 ppm, 20 sec duration 0.75 ppm of actual buff, = 25% uptime, *7333 strength *3.67 = 6728 dps + 2508 dps = 9236 dps.
Gaze of the Twins: 1467 str * 3.67 = 5383 dps, Your critical attacks have a chance to grant you 2868 Critical Strike for 20 sec. This effect can stack up to 3 times. Following calculations work if you get 1 crit every 10 seconds, or more often, 1 ppm * 1.5 meleehaste = 1.5 ppm, 50% uptime baseline * 2868 crit * 1.39 = 1993 dps, however each time a proc is overridden, the duration of the last proc is effectively extended back to 20 seconds. There is a 50% chance of this happening per proc, or .75 times per minute, and the average extension will be by 10 seconds, or a 12.5% uptime, worth 500 dps. 7874 dps total.
These calculations were done before the change to the way the proc worked. It will now be slightly better with about 25% melee crit, slightly worse with below 25% melee crit, and the same with 25% melee crit. First look suggests this won't change ranking order.
I would personally rank them in this order:
Spark of Zandalar
Primordius’ Talisman of Rage
Gaze of the Twins
Brutal Talisman of the Shado-Pan Assault
Fabled Feather of Ji-Kun
Capacitive Primal Diamond – 15.00 base RealPPM on landing melee/ranged abilities and swings. No ICD. At 5 stacks, fires Lightning Strike, which deals [280 + 75% AP] Nature damage. That base proc rate is multiplied by an additional coefficient by spec:
Note: For DW frost, the legendary meta is worth about 5k dps in current gear. I'd imagine it's somewhat close for all specs. However, the cost of losing the normal meta gem is 2k, and I can't find the post I thought I read saying you get to keep both, so it's possible I'm crazy and that the net gain is 3k.
Meta gem value calculated by simming a full BiS DW setup with and without the meta gem, 2159 dps loss.
Legendary meta calculated for DW by using base AP of 48k, assuming procs will bring average ap to around 60k, 60k*.75+280=45280 dmg per proc.
15 base rppm*1.572 DW modifier * 1.5 melee haste modifier = 35.37 ppm/60 = 0.5895 pps/5 procs per hit = 0.1179 hits per second = 5338.512 dps * 15% meleel crit chance = 6139.29
All info for other classes and specs has been removed to clean up the post, and can be found at the link below.
Feng the Accursed
Gara'jal the Spiritbinder
Imperial Vizier Zor'lok
Blade Lord Ta'yak
Protectors of the Endless
It should be noted that just because you cannot extend from 100% to dead does not mean it isn't worth it. Tsulong is a good example of this as well as protectors and Elegon. AoE fights are not worth this and you should continue to blood boil as usual. Fights with constant target changes such as Lei Shi this is not worth using as well.
All changes are subject to more changes, and may not be final, and the results below are math from one person, and are unlikely perfect. Please help contribute to the findings by doing your own, and sharing your results.
Originally Posted by GC
- Might of the Frozen Wastes to 20% melee damage, up from 10%.
- Icy Talons increases attack speed by 30%, up from 20%.
- Blood Plague and Frost Fever damage +30%.
- Howling Blast damage to primary target +23%. Damage to secondary targets remains unchanged. Note: the tooltip for secondary damage says 0.5 but it has been 0.8 for some time.
- Scourge Strike damage +7.7%.
The only other dps changes for dks so far are gargoyle being free and benefitting from mastery.
And the 2-set being nerfed to 4%.
In 509 gear, simcraft currently has DKs ranked in the following order, single target.
← Click Here
Notes: Changes value calculated with 509 BiS sim, checking what % of the damage was made by up by the ability being buffed, and then buffing that particular amount of damage by whatever amount the ability was buffed, and then adding that on to the original total. I was unable to do this for the gargoyle being free change, and the extra km proc change, so I gave them what I judged to be an expected value. FB= festerblight. I used 509 gear for all sets, with the 5.2 version of the 2-set and 4-set t14, not 15. In addition, they are taken using 5.0 trinkets, and 5.2 trinkets will also change the numbers.
2-hand math not corrected, mistake I thought I made was a mistake, changed final number back, but don't have time to swap math back now.
One change at a time, set bonus nerfed:
Gargoyle changes added(gargoyle being free valued at 1600):
SS damage boosted:
Might of the frozen waste boosted(assuming it only effects white hits):
In the sims I've run, the stat values of everything remained about the same, and certainly within the same order. However, I've only run sims in the 509 ilevel range, and if anyone else has competing data, I'd love to see and compare it.
Short answer, no they don't seem to.