Neeb

Ahhh that was what I was worried about. I had gone back and looked at the % dmg of Necrosis (via WWS), and saw that it was much higher then what I was seeing just tooling around on the dummies. i.e. showing around 0.6% on dummy and WWS was showing around 7% actual raid damage so I knew there was something askew right there.


There is a pretty typical 51/0/20 build for blood, that I had played with in Beta that seems to be still around, need to look at the frost spec as well. I still want to try to think a little out of the box as to how I spec'd trying to look at where I could improve my DPS a little. I went a little too far outside it seems lol. I still want to tweek the spec some and see where to eek out a little more DPS. My gear is about as good as it is going to get for the level of content there is to play in but I feel like I am slacking some place, hence all the playing around.

I'll go back and try to catch up some more in this post since 3.0.8 see what else there is.

Edit: Bane of the test dummy. This should prove why sometimes dummies are just BAD to test with, Necrosis was down because my dummy was at near zero health, so I was getting 1 and 70 hits with necrosis because there was simply no health for the ability to take.

 

clairecakes

Thanks for clarifying. After you posted this I got to testing the ol' Haunted Dream. It does seem to have an awful proc rate. I'll take to testing every Tuesday in case a quiet buff comes down ( i can hope right).

And to TRIPLE check.... haste>ArmPen yeah?

 

Bloody_sorcerer

By my (see below) (and rawr's (see source)) math, ArP > haste, and an increasingly large amount better as you gain ArP.

 

Illundai

The maths on the previous page don't agree.

 

Bloody_sorcerer

I'm going to try this again.

haste-modified damage is ~35%, DC is ~13%. Armor Penetration-modified damage is ~81.5% (everything but DC and ghoul). 32.79 haste rating is 1% haste, 15.39 armor penetration rating is 1% armor penetration. Combat ratings are all 1 budget point per point of rating, so 1% haste is equivalent to 2.13% armor penetration rating.

Fully debuffed boss mobs have their armor determined as follows:
mitigation = 1f - (targetArmor / ((467.5f * (float)calcOpts.TargetLevel) + targetArmor - 22167.5f));
where targetArmor = (13083 - sunders and FF) * (ArPrating / 1539)
at zero armor penetration rating, this is 32.19% mitigation. 32.79 Armor penetration rating will reduce this mitigation to 31.73%, a change of .46%, a gain of .3749% DPS, while 1% haste will be .35% DPS.

Also, keep in mind that armor penetration rating improves as you gain it; the first 32.79 rating is only marginally better than an equal amount of haste, but the next 32.79 rating will be some amount better than the first 32.79 rating, and so on up to the cap.

 

Damarek

I tried the diseaseless blood yesterday (was standard blood 51/0/20 before) and I am really content with the results:
Wow Web Stats

Naxx 25 Construct Quarter + Saph/Kel averaging 4805 DPS with 8378 Spike at Thaddius.
My Equip is not that good (just got Bandit's Insignia and the epic bracers yesterday) and still I am seeing much better results than with diseases - and I always had managed to pull a good rotation without being RP capped etc. Diseaseless is ridiculously easier to perform and showing better results. Looking forward to getting the Betrayer

If they don't modify something with this, I am really going to love the changes that were posted recently :>

 

Archetype

I hope they fix the timer with DRW as well. Currently with 100rp, my DRW starts at 18 seconds... shouldn't it be 22?

 

Warrentt

im confused..is the disealess rotation OB OB HS HS rune dump OB OB HS HS..or OB OB HS HS rune dump HS HS HS HS HS HS rune dump

 

chiahatesyou

I have read this whole thread... jeez it's a lot to take in too, I have tried the Diseaseless Blood spec and I understand how it is supposed to work cept for one thing. On aoe packs what is the rotation? I don't think I have enough atk power to sustain the dps that most have shown on this thread but I should be at least somewhat competitive. I am trying to start distancing myself from 31/40 knowing that it will be nerfed soon enough.

So what should minimum Atk Power be for this spec to be top 5 effective?

And what do you do on AOE packs?

thanks

 

Zadkiel

I did a search on this thread on whether or not BCB was worth it but came up with nothing. I'm currently using the diseaseless 51/0/20 build and have seen my DPS increase quite a bit from 51/13/7 but looking at recount I'm not sure if I should have 3 points BCB. According to recount, BCB is only doing about 1.4% of my DPS. Should I stay with the build I have right now or should I switch out the 3 points in BCB for something like Outbreak since blood boil is suppose to be better in 3.1?

 

kurokaze

Looks like I had boss armor value incorrectly pegged at 10900 rather than 13083. Bloody_sorcerer's numbers appear to be correct. I can't for the life of me find where I got that number.

By the way, I'd like to mention that as far as I can tell, diseased 51/0/20 is the superior blood spec/rotation in theory. There's lots of discussion about the diseaseless rotation and how ridiculous it is but as far as I can tell, it's simply surprisingly competitive (and easier to pull off) rather than optimal. Granted, the fact that it is easier to pull off is probably worth the ~200 DPS difference between them on the spreadsheet, but there seems to be a pretty even split between diseased and not among the few blood DKs represented here.

 

Archetype

Depends if you take rune mastery or not. I go without, and simply do OB OB HS HS (rp dump) repeat.

 

◊ Abynthe

I tend to go [ob-hs-dc] plus an optional deathcoil when I have the additional runic. Decreases the chance of wasting free deathcoil procs. I found that 2xob, 2xhs was much easier to maintain than hs spam with death runes

 

pindle

It needs to be said tho that if you have the flexibility of changing your rotation for trash, 6x HS would do better damage versus 2x OB 2x HS, only using OB to keep the buff up. This is of course assuming you're in fact hitting 2 targets with the HSs. For that reason alone I would take Death Rune Mastery.

 

clairecakes

Stay with BCB. Bloodboil uses Blood Runes which you should be using on Heart Strikes.

 

Novaus

Several year lurker, first time poster here. I've been keeping an eye on this thread since it started and I wanted to share my two cents. I spent the first two days of 3.0.8 reading every piece of information that I could and when game time was available, respec'd several times and would also try and run several heroics to do more than simply "test dummy" run these builds. I've spent more money on doing this than I'm willing to admit. I evaluated well known theories and tested personal theories as well. I've tried every spec and every various rotation on every spec. I raid 4 days a week (2 days in 10's, 2 days in 25's) to try and find "the" spec that everyone was hunting for. By the end of the second day, I had determined a rotation and spec that was superior in raw application than any other build/spec I tested. This build now, largely due to Jmack's publicity, has been picked up and used by many DK's with great result.

According to WMO, my DPS on Patchwerk25 places me above the average in the World Top 50 DK's on Patch25.
5665 DPS, 3:31 parse, WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!

But Patchwerk parses are deceiving. In my 3/5 Heroes, 1/5 Valorous, I was sitting at 9600 DPS after 45 seconds, and at 90 seconds, I was at 7200 DPS. Would I have been in a group that killed him sooner, my numbers would have been much higher, easily putting me beyond the 6K DPS mark, enough to put me in the top ten, worldwide. As you can see, my parse was at 3:31, which means that the setup/rotation is very sustainable. Players are also stacking multilpe Hysterias, Tricks of the Trade, etc., to misguide people into a spec that doesn't give them anywhere near the same results.

What about AoE? How can Blood keep up with AoE? It can, and not through the normal conventional wisdom on AoE.

For Reference:
WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!
#1 on Noth as Blood.

Wow Web Stats
#1 on Gothik (Another important note is who is at #2, a person that I have tought to mimic what I do, but most importantly, he does it without a Sigil, for the readers who wonder about scalability. He's also lacking a few pieces that I have, but as you can see, the results are there. This scales well, regardless of gear level, it just so happens that when you're sitting at top notch gear levels (which I am not), this is borderline overpowered.)


So, I've demonstrated now that Blood can be superior on single target as well as AoE.

My Armory for reference: The World of Warcraft Armory

My spec is built around a diseaseless Blood Rotation. Emphasis is made on Obliterates, NOT Heart Strikes as commonly stated. My Obliterates crit for 10-13k. A double-critting, cleaving Heart Strike, if I'm that lucky *might* do equivelant damage to a lower end Obliterate crit. Because of this, even in AoE fights, you are still largely desiring those large chunks of constant damage through Obliterates, having a Cleaving Heart Strike simply become gravy beyond that.

The rotation is OBx2, HSx2, RP Dump, as is commonly pushed around in the last week or so. I've been using it for several weeks now and I feel I have it down pretty well. The largest reason why this wasn't on the radar pre 3.0.8 is because of Obliterate scaling. Glyph of Obliterate used to increased weapon damage, but now increases total Obliterate damage by a percentage amount.

This setup is more important than just the rotation. It's also about the choices you make as well. How do you approach AoE? How do you handle fights like Gothik or Noth, with adds? How to manage and time your cooldowns on single target fights like Patchwerk? It is at this point that with this spec goes beyond napkin math, beyond the spreadsheets and takes off. Without drastic changes to the way Blizzard scales this setup, it is simply not possible for a disease build to compete on multi-add fights, unless the adds accumulate to 6-8+. There is no "ramp up" time. There is no screwup in the rotation if you need to swap targets, or if a target dies early before diseases fall off, and unlike previous lazy builds which put emphasis on using more Heart Strikes, cleaving damage is more like "gravy" than it is a necessity.

Simplicity is king. One of the best things to do for a build like this is to take your Obliterate and Heart Strike macros and include the following:

/startattack

It sounds childish, and especially below the kind of expertise that you would expect out of a place like this, but this allows you to immediately resume DPS on the next target as soon as your current target is dead. No selecting a target, and no waiting for runes to become available for you to use a strike in order to force an autoswing. This spec is a combo of Obliterate damage, white damage (amplified by Necrosis and BCB), Heart Strike and Death Coil. While Obliterate is the "money shot", the remaining are all equally important and should not be neglected.

On fights with adds (Noth/Gothik/etc), you largely want to be hitting the highest HP person. Your rotation should not differ. You should not put more emphasis on Heart Strikes, or try to convert Death Runes into more Heart Strikes. A critting Obliterate will still hit as hard, or harder than a double-critting Heart Strike. 12k Oblit versus 2 5K HS's. You view AoE fights differently than anyone else should. AoE hits all targets equally. Your job is to focus on the target with the largest health and let your cleaves hit whoever they want. The largest health mob is generally the one picked up by a tank, which means its usually the one with armor penetration effects on it (sunders/FF/etc). This is where I see my gains on trash and AoE boss fights.



On Patch, this is what I do to hit my numbers and peak out at 9600 DPS for ~45 seconds.

AOTD before/during pull.
Ghoul while running in. Horn to generate RP. Arcane torrent for additional RP(no GCD)
OBx2, HSx2
Bloodlust
(Should be at RP cap at this point)
DRW + Hysteria
OBx2, HSx2, EMPOWER RUNE WEAPON, OBx2, HSx2, DCx3
From this point on, for the rest of the fight, it is: OBx2, HSx2, DCx2/3 (Depending on Sudden Doom procs).

Using Empower Rune Weapon to skip the RP dump phase, let yourself cap out, and immediately begin Obliterates again, while having it do the same for DRW, while under the effects of Hysteria is where the Godlike numbers come from. From there, you're set for the rest of the fight.

I have kept up with and contributed to a few various WoW/DK communities. I have converted dozens of people to this setup now and 99% of them came back and reported excellent results and a significant improvement in their own raid DPS.

You can see these discussions here (Some, under the name of Archivis):
[ Classes] DK best DPS spec - dslreports.com
Archivis, am I ready for your no disease build? - dslreports.com
Diseaseless Blood - What Non Blood Talents?

I am a large fan of "real world" results rather than target dummy or spreadsheet results. Target dummies help nail down the feel of a rotation, but are not an accurate measure of damage. While I appreciate their theorycrafting value, spreadsheets require only one statistic to be off in order to throw off the entire sheet and because of that, they have been inaccurate and misleading at times. Real World results are much harder to generate and require time in order to find the best results, but in the end, those results are what it boils down to.

I wanted to contribute my two cents, and despite this site being called "Elitist Jerks", I tried to keep the arrogance to a relative minimum.

 

tetracycloide

A parse you posted not more than a few lines away from this estimate shows your average obliterate crit as under 10k ( 9932) much less ranging from 10k - 13k. Just some quick napkin math shows your average obliterate 3624 non-crit and 9932 crit for a total of 3624 * (1 - 47%) + 9932 * (47%) = 6588.76 and your average heart strikes as 1788 non-crit and 4675 crit for a totale of 3144.89. Multiply by two for a one to one rune to rune comparison and we have 6588.76 vs 6289.78, at ~300 damage it would appear that the 20% chance for heart strike to proc a crit death coil (~2500 damage at a minimum) is a better use of your runes.

It's possible, maybe even probable, that after you've upgraded your weapon and gear to best in slot for 25 man runs that obliterate will be a better use of the runes but at the moment that certainly doesn't appear to be the case.

My vanity is justified.

 

Novaus

I should have reworded that my Obliterates crit "up to" 10-13k. On that parse that we're talking about, max Obliterate was 13292.

Obliterate crit 56% of the time. Heart Strike crit 44% of the time. You've overcomplicated trying to determine average strike damage, including crits.

Total Obliterate Damage: 293,682
Total Obliterates: 41
Average Obliterate Damage(Damage Divided by Total): 7162
This result is much higher than your Napkin Math.

Total Heart Strike Damage: 125,287
Total Heart Strikes: 41
Average Heart Strike Damage: 3055

As you can see, single target, 1 Obliterate does more damage than 2 Heart Strikes. 1 obliterate, while requiring additional runes, uses one less Global Cooldown. Obliterate's RP generation through talents and the set bonus, allows for me to get 2-3 DC's per rotation. Using Heart Strikes instead of Obliterates means that runes will be active sooner than you can use them. Yes, that means more Sudden Doom procs, but you cannot manage a rotation that allows you to make full use of your RP and plug in additional 1-rune strikes without extending the rotation beyond 10 seconds, leaving runes unspent OR leaving RP unspent... Or you can go through the Unholy Presence, which is largely unfavorable from everything I've been able to determine.

Even in 3.1, when Sudden Doom gets changed to proc off of Heart Strikes, it's still a close call. We're not completely sure that these new free auto-proc'ing Death Coils are guaranteed to crit. But my average Death Coil crit'd on Patchwerk for 4829. Sudden Doom has a 20% chance to proc. With that, you could assume to add that damage to HeartStrike (4929 * 20% = 965). You can add that as a flat bonus to Heart Strike, which would make that a 4020 average Heart Strike (+SD proc) on that parse.

That puts two of those above Obliterate, but not in 3.0.8. Also, take into consideration that this is damage, not damage per second. It takes 3 seconds to accomplish an average of 4020 on a 3.1 HS/SD average proc, while it only takes 1.5 seconds to do a 7162 Obliterate.

Obliterate DPS? (7162 / 1.5) = 4774
Heart Strike DPS? (4020x2 / 3) = 2680

At 10 seconds, in Blood Presence, OBx2, HSx2, DCx2 provides me with a clean 9 second rotation. Add in latency and you're looking at just barely making that 10 seconds. Swapping an Obliterate for an additional 2 Heart Strikes provides unecessary rune-uptime.

If I am wrong, please correct me. Ultimately, I'm just trying to find the best setup.

 

Warrentt

even while running with an unholy dk and having ebonplauge up diseaseless rotations are better? i mean with ebon plauge up your diseases are ticking 30% more dmg.. and dpsing on patchwork doesnt really even count as a true parse imo.. i mean a blood dk can get ridiculous numbers in the first few moments of the fight because drw is crazy now..the dps will spike ridiculously high..

 

kurokaze

Claiming the observed crit rate of the two abilities over one shortish fight is rather fallacious. Only the theoretical critrate difference (or the difference over a long period of time) makes a difference, and that should be limited to the 5% bonus from 2PT7.

Your Heart Strike damage seems extremely low relative to your Obliterate damage. You did have an FFB mage in raid; at a guess, I would suppose that you may not be using Glyph of Blood Strike. If this is the case, it's a huge boost to HS (and to 8xHS 2xOB) that you're missing out on.

Personally, I haven't been able to produce a noticeable DPS difference between the 'faceroll' 4xOB 4xHS rotation you advocate, the slightly more complex 2xOB 8xHS, and the much more complex diseased rotation. Since I'm in the early stages of dabbling and often run with PvP or PvE 0/20/51 instead, I generally flub the diseased rotation, but there's not much opportunity for player error in yours and it doesn't produce better numbers. Why could that be?

 

Novaus

I'm not sure what other parses to use for a standstill even comparison. I know that in the short fights, quick burst means you get an edge, but my parse where I was at ~5600 DPS was at around 3:30 in length.

But here are some parses of other fights:
Noth #1: WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!

Heigan #1: WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!

Loatheb #1: WoW Meter Online - The Best World of Warcraft Combat Log Analyse System!Fully support Wrath of the Lich King!
(Please Keep in mind that actual DPS is very low for the raid because we attempted to go for the "Spore Loser" achievement. After having someone moonfire a spore and killing it, we resumed killing spores)

Thaddius #1: http://www.wowmeteronline.com/combat/1633933
(Due to Comcast connectivity issues as well as... uh... well let's just say the same reason why we'll never see Heroic Safet Dance... I was not able to maintain anywhere near max stacks. You can see my damage relative to others).

Gothik #1:Wow Web Stats

I tried to include a balance of parses that include AoE as well as single target damage. My rotation stayed the same on every fight.

 

tetracycloide

The 12% difference in crit rate is what's driving the higher average obliterate damage for that parse. There are other boss fights even in that same raid parse where the 12% swings in the other direction. Loatheb, for example, has 78% for obliterate 90% for HS. So I took your actual character page crit and added in typical raid buffs and arrived at an average crit rate that fell somewhere between the two (47%), which makes actual sense. The RNG is fickle and makes it impossible to determine differences of a few % total DPS from parses alone, you need spreadsheets, you need real averages.

The DPS argument is a straw man, in a full 10 second rotation the power of abilities that use runes is not damage per second but damage per rune. You have ~10 secs between rune cooldowns which is fixed so it doesn't matter if you have one strike that deals 600000 damage and uses all 6 runes in 1.5 secs (40,000 DPS by your estimation) or 6 strikes that deal 100000 damage and uses one rune in 1.5 secs (6,666 DPS) because your actual DPS is still 6 runes * damage per rune / 10 secs or 6,000 DPS either way. Now if it's an in and out fight and you can predict when you can jump in or jump out then yes, the single front loaded strike is obviously better, but on tank and spank it's the damage per rune that matters.

If the rotation is that much easier for you to pull of then it's probably still optimal for your but that's not the same thing as saying it's absolutely optimal and the two should not be confused.

edit - I didn't check for the 2pT7 bonus and that was my mistake. The point still stands that assuming a 12% crit descrepancy is not justifiable.

My vanity is justified.

 

Novaus

I do not disagree with you in regards to your first statement. In a 10 second window, you have to fit as many 1.5 second GCD's in as possible. Some of those are reduced slightly because they're spells (haste), and some of that is increased by latency. In my personal Experience, I can get 4 Strikes, and 2-3 DC's off before my first U/H runes activate. That's 6-7 GCD's. I'm not sitting there with "dead time"

What you're suggesting is to sub out an Obliterate for 2 HS's. Effectively increasing the amount of strikes to 5, also increasing the chance for Sudden Doom to proc, so safely say 7-8 GCD's being used for Strikes/Spells. How are you fitting that into a 10 second window? You can't. Runes are activating faster than you can spend them. Your "rotation" takes 11, 12, even 13 seconds depending on your RNG SD procs. Let's take SD out of the picture, because soon, it won't trigger a GCD. You're talking about doing OBx1, HSx4, RP Dump, right? Assuming 2 DC's in an RP dump, that's 7 * 1.5 = 10.5 seconds + latency. At 150ms latency, you're adding nearly an extra second on at 11.5 seconds. And that's just 3.1 we're talking about. 3.0.8 Sudden Doom procs means much more than that.

So, in today's setup, you're talking about edging out an extra 5-10% damage in your rotation, but taking 15-30% longer to do it.

 

tetracycloide

You don't have to fit it into a 10 second window. The game automatically shortens your rune cooldowns for up to 2 secs if a rune goes unused so every 6 runes + dumps need not fit in 10 seconds as long as the average duration is 10 seconds per 6 runes. With one rotation of OB OB HS HS and one of HSx6 including dumps and SD procs will still net an average of ~10 secs per 6 runes. Now sometimes you'll have more time than you need because no SD procs and sometimes you'll have less time than you need because it procs on every HS but the average is somewhere in the middle. So in reality never changing from OB OB HS HS is droping some percentage of DPS as insurance against the edge cases in a HSx6 rotation where there won't be enough time.

Come 3.1, assuming the auto SD goes live, there won't even be an issue of time since the maximum time of 11.5 quoted is already guranteeded to be less than 12 secs. Meanwhile the other half, the OBs to generate the death runes, is always going to be less than 8 secs. So the average between the two will always be less than 10 i.e. maximized rune usage over time.

My vanity is justified.

 

Leaflock

The problem I'm having with the HSx6 part of the diseaseless rotation is that the "edge cases" of not having enough time seem to happen more often than not. The issue is that the latency/error cushion on cooldowns doesn't help here-- when your blood runes come back, HS will use those before death runes, even if you haven't used up all your death runes yet.

If you open with HS HS OB OB, your second round will be HS 0, HS 1.5, HS 3.0, HS 4.5, HS 6.0, HS 7.5.

1 SD proc or runic power dump pushes the last HS back to 9.0 seconds, and a second dump makes it 10.5. It's within the buffer zone for rune cooldowns, but a HS at 10.5 will use a blood rune, not a death rune, so you actually start to lose runes per cycle.

Additionally, you get a delay in the next cycle because the UUFF runes come back in that order-- you have to wait an extra GCD for the first F rune to Obliterate again.

I think that the 3.1 change will make this a moot point, but for now, more Obliterates probably do make a better rotation.