MoP Blood DK Discussion

85 Orc Death Knight
8105
My rotation with an RC build is something like DS>RS>HS>RS.
Ive tried DS>HS>RS but RC procs were not consistant enough.

I understand that mathematically RC is better, in the vacuum of a sim. I will probably be using RC for most tanking situations. This is beta though so im testing other stuff, not saying BT is better than RC.

I personally have been liking the implementation of blood tap because it allows me to use blood runes for threat or survivability. With RC I have no choice but to use my blood runes for threat, minus Rune Tap every 30 seconds.
Reply Quote
90 Draenei Death Knight
11090
It's not that RC is mathematically better. That's open to interpretation based on your definition of "better."

If I were to extend out the time period of the sim to infinity, RE would lead to the most DS's, and be very close to BT (or BT might even overtake it due to not sitting on blood runes). RC would be behind both. RC is about 4% less DS's.

The difference, though, is that 4% less DS's in a 10m fight doesn't lead to a significant decrease/increase in overall post-absorb damage taken. Your healers won't notice, and the logs won't definitively tell you, either.

So, no, RC is not the best for absolute survivability (in the theoretical vacuum).

However, these are the selling points for RC:
-Most damage dealt
-Negligible realistic damage taken difference
-No need to pay attention to rune states / BT stacks
-Most consistent DS interval
-Least penalized by player error
-Very little downtime

That's the reason why I say RC is better. You can fully disagree with me and that's just fine. I'm stating my opinion from the data that was generated.
Edited by Euliat on 5/2/2012 11:23 AM PDT
Reply Quote
90 Human Paladin
8390
http://us.battle.net/wow/en/forum/topic/4661688210?page=4#69

Hope you don't mind Euliat, figured it couldn't hurt to have your info on the beta forums too.
Reply Quote
100 Troll Death Knight
14920
Hey Euliat, is your TL;DR on Exp and Hit factoring in the possible change of rune behavior on missed attacks thats still in the beta?
Reply Quote
90 Draenei Death Knight
11090
05/02/2012 11:27 AMPosted by Deathtime
Hey Euliat, is your TL;DR on Exp and Hit factoring in the possible change of rune behavior on missed attacks thats still in the beta?


No, because I refuse to believe that Blizzard would do something so incredibly stupid.

Plus, the only person I've heard talking about it is you.

05/02/2012 11:26 AMPosted by Judgesyou
Hope you don't mind Euliat, figured it couldn't hurt to have your info on the beta forums too.


Thanks.
Edited by Euliat on 5/2/2012 4:03 PM PDT
Reply Quote
100 Troll Death Knight
14920
I have noticed I am the only one noticing this. Ill record it later and stick it on youtube.

Edit: assuming I can make it look semi decent my computer is kind of old.

It also still behaves a bit weirdly where if say you connect with one heart strike and then miss with a second then roughly half way through the rune cd it will reset. At times it seems like its just a display error but there are times were the GCD is clearly up and I still cant use those runes.

Edit2:Ok i recorded it and parts are smooth enough for it to be very clear what is happening.
Edited by Deathtime on 5/2/2012 4:36 PM PDT
Reply Quote
90 Draenei Death Knight
11090
05/02/2012 11:26 AMPosted by Judgesyou
Hope you don't mind Euliat, figured it couldn't hurt to have your info on the beta forums too.


Since the specifics of my sim are now in question:

The sim is right. I have several ways to demonstrate that.

Theoretically, with RC it should be expected that you can generate
(((60 / (10/1.2)) * 10 *3 + 30 + 60/(3.6/1.1) * 10)) / (1 - ((.45 * (3 / (10/1.2)) * 3 * 10) / 30)) = 512.330947 RP / min

(RP_From_Runes + RP_From_Horn + RP_From_Autos)*RC_Proc_Return_RP_Multiplier = Total_RP


My sim generates 495.2. Why is that not 512? Because my sim uses RT twice a minute (roughly).

A similar analysis for BT gives about the same result. The only difference is with RE, and that's because the equation being used makes a fundamental assumption: that all 3 rune types are always recharging. This is invalid for RE, and, thus, the result from the equation doesn't match up with the reality. If you are gaming RE properly, you will have two of the unwanted type of rune recharging for a duration of time.

Is the damage taken right? The results are within ~3-5% of my other sim under the same conditions, which is known to be right for Cataclysm. Given that runes are treated differently in both sims, 5% is an acceptable difference for me to reasonably infer that my sim is correct, as one uses a real-time analysis (this one), and the other uses an average analysis.

Thirdly, I have traced through by hand short-scale tests and found no inconsistencies.

Fourthly, as a note, I would not spend all of that time posting that information up if I were not confident in the accuracy of my sim. That would be a waste of my time.

RE is being gamed flawlessly. I have done an analysis when RE is not being gamed at all and it is the worst of the choices. It led to less runes overall than RC, which means less DS's than RC and less HS's than RC.

RC scales properly with all sources of haste. %Rune regeneration modifiers do not affect RC's duration because they don't affect it on Beta. Haste modifiers do affect the duration because they do affect it on Beta.

With the exception of UB, the sim is completely up-to-date with the most recent Beta build. The results being displayed do not have RS with a 1.0s CD, but I can show you that even if I move RS's CD to 0.75s, it makes no difference whatsoever. Blood DKs are not bottle-necked with GCDs so changing the GCD to something under 1.50s makes no difference.

@Slant, re:
He says you get 0.45 from RE only if you don't waste it on a blood rune, but that is wrong. You get 0.45 of one rune with RE, period, independent of whether that rune is U, F, or B. If you want F/U only, gaming can only improve that number for F/U. And when do you have zero fully depleted runes? The first 3s of each combat?


Due to not being specific enough, yes, it is wrong for me to say that the 0.45 from RS decreases when you proc a blood rune (but not when you don't have any fully depleted runes and use RS). However, the intent of that statement is 0.45 F/U runes, since the typical behavior is to apply the full gain toward DS uses. In that regard, the statement is completely accurate. If you proc a blood rune, your F/U return is 0 for that proc.
Edited by Euliat on 5/2/2012 8:40 PM PDT
Reply Quote
100 Troll Death Knight
14920
http://youtu.be/uxiZf-XgtWk

Sorry parts of this are choppy, my computer is a bit old. Here is some proof that runes are behaving differently on beta on missed/dodged/parried attacks. Pay particular attention to

:54 Missed Death Strike and all 4 death runes are regenerating at the same time
1:06 Missed Heart Strike and both Blood Runes are regenerating at the same time.

The one at :42 is one of the odd times that it actually resets like it would on live.

Like I said earlier this could possibly be just a display error.
Reply Quote
90 Draenei Death Knight
11090
05/02/2012 05:06 PMPosted by Deathtime
Like I said earlier this could possibly be just a display error.


No clue.

I would think with an issue like that that more people would have talked about it. I can do it, but it'd be a pain and I wouldn't have time to implement and test it until Saturday.
Reply Quote
100 Dwarf Death Knight
15510
Finally vomited out a real feedback post:

http://us.battle.net/wow/en/forum/topic/4837663734

TL:DR, things are improving, but some things still need some work. Tier 5 and 6 suck. And I want to turn my ghoul into an abomination dammit.
Edited by Arcdeek on 5/2/2012 6:51 PM PDT
Reply Quote
90 Draenei Death Knight
11090
Quoting myself from somewhere else regarding SoB and pooling stacks.

Implemented holding off using DS until 'x' stacks. I wasn't originally going to post the data up, but, well, if I'm going to make this claim I might as well.

Scent of Blood should not affect when you use DS. At best, you are no better off than just blindly using DS without paying attention to your stacks. At worse, if you wait until a certain amount of stacks, you take significantly more damage and do significantly less healing. Respectively, you could be taking up to ~7-10% more damage and healing for ~20% less.

Why is this the case? Because Scent of Blood stacks are not that powerful. Scent of Blood is good because of the RP return. The healing addition is a bonus--not the focus. You must remember that Scent of Blood is a % of a %. DS heals for 20% of the damage you take. 1 Scent stack increases that to 24% of the damage you take.

First objection: Holding off until a certain amount of stacks to use DS is completely unrealistic because you're not going to wait until 'x' stacks every time.
Response: Agreed.

So, to counter that, I took it to the next level. I coded in logic to hold off on using DS if an auto-attack was coming within 0.75s. Since auto-attacks are evaluated first in my sim, this would grant the sim another stack before using DS.

Still no difference.

You can look at the data yourself here:
https://docs.google.com/spreadsheet/ccc?key=0AgEwKT9377f3dHNPMFNNWEJtbzRKM2R5NUhWTm96THc

...
[clipped]
...

...a few quirks when the sim starts waiting for higher stacks (affects > 3 stacks most).

For RE:
The sim hardly ever uses RS because the DS interval is so long that there are almost never any F/U depleted runes. This is why the RP generated and overflow look weird.

For BT:
The sim hardly ever uses RT because BT turns that blood rune into a death rune. Since the DS interval is so long, this blood(death) rune is hardly ever used. The sim will not use that blood(death) rune on HS/RT, so it just sits there.


As a note, I caught an error with Heroism. I had it giving 45% haste for 45s, when it should be giving 30% haste for 40s. This has been fixed, so you may see the DS/min drop some. However, I double checked it with 5 of the 10 or so setups I posted previously, and the increase in damage taken and decrease in healing done is completely negligible. Heroism is only active about ~7% of the time. As such, I did not bother to update that data.


Regarding the Heroism part, if you want to claim that that makes the rest of my sim invalid, fine. Go write your own. If it even changes the damage taken data, it changes it by less than 2%, which isn't significant.

Regarding the actual Scent of Blood data, if you don't have the facts to back your claims up, then stop making them.

The extreme bluntness is because I've now seen 6 individuals claim that Scent of Blood should impact your DS use heavily, or that you shouldn't DS until you have a certain amount of stacks, under the guise that that's "optimal."
Edited by Euliat on 5/4/2012 4:12 PM PDT
Reply Quote
90 Night Elf Death Knight
12055
I can confirm that the rune bug (setting your remaining runes to the same CD as the other runes on a missed attack) isn't a display error.

Beta is ____...
Reply Quote
90 Draenei Death Knight
11090
Yay.

/pout
Reply Quote
90 Draenei Death Knight
11090
Just to clarify so I can code it properly,

If the pair of the rune is cooling down, and your ability misses (what about dodge/parries?), the rune that was just used inherits the CD of its mate. If the pair of the rune is not cooling down, and your ability misses, the rune that was just used has the normal CD. If the ability lands like normal, regardless of what the mate is doing, the rune goes on its normal cooldown (like Live).

I take it that RP is not generated on a miss/dodge/parry as usual.

Is this the only behavior, or are there other conditions or instances under which the "bug" happens?
Edited by Euliat on 5/4/2012 5:02 PM PDT
Reply Quote
100 Troll Death Knight
14920
If the pair of the rune is cooling down, and your ability misses, the rune that was just used inherits the CD of its mate.


Yes

05/04/2012 05:01 PMPosted by Euliat
what about dodge/parries?


It happens on dodges misses and parries

05/04/2012 05:01 PMPosted by Euliat
If the pair of the rune is not cooling down, and your ability misses, the rune that was just used has the normal CD.


It always inherits the CD of its pair and if the other of the pair is not cooling down then it will be available immediately. For Example both my blood runes are fully charged and I my heart strike is dodged/parried/missed, then the rune I attempted to use will be available again immediately. AKA the other rune has 0s on its cooldown so the one you attempted to use will inherit its 0s cooldown.

05/04/2012 05:01 PMPosted by Euliat
I take it that RP is not generated on a miss/dodge/parry as usual.


I think so but I would have to check to make sure.

05/04/2012 05:01 PMPosted by Euliat
Is this the only behavior, or are there other conditions or instances under which the "bug" happens?


Occasionally the runes that were on CD from missed/dodged/parried attacks will just reset and be available for use. This happens quite frequently but much less than the way I have been describing it. I would say this happens ~30% of the time while the other happens 70% of the time. I haven't figured out why this happens and its odd behavior makes me think that all of this is a bug.
Reply Quote
85 Blood Elf Death Knight
5940
05/02/2012 10:49 AMPosted by Euliat
BT lets you get a DS when you want it (provided you have stacks), but oddly enough, that doesn't let you have the most consistent DS interval.


Is having the most consistent DS interval actually ideal though? That's an honest question, I don't know the answer. We sort of assume a priori that regularity is positive on Live, but that's because there's no tradeoff. RE isn't very forgiving so you have at most maybe a second or two to use a Death Strike before you're wasting some variety of resource.

Once the tradeoff exists though, I don't think it's nearly as clear; my intuition would favor control at that point but I don't really know. I think that going an entire rune cycle without using a DS will feel a lot less punishing if you have the option of choosing that to happen during a cycle where you dodge and parry a lot.

The idea behind what I'm asking, just so it's clear, is that even if control doesn't net you any extra Death Strikes, it still might result in more effective Death Strikes. I think to truly answer that question would require a sim with a moderate amount of AI though.
Reply Quote
90 Human Paladin
8390
http://us.battle.net/wow/en/forum/topic/2140511400#20

Never forget.
Reply Quote
90 Draenei Death Knight
11090
05/04/2012 07:56 PMPosted by Anothriel
Is having the most consistent DS interval actually ideal though? That's an honest question, I don't know the answer.


And I think it's a perfectly valid question. We'll see if we end up in the land of two-shotting again after 1 raid tier.

I think there's a fair deal of value (to your healers) in having a very consistent DS interval in terms of the reliability of damage intake, and there's also a fair deal of value in being able to control specifically when that DS comes depending on fight mechanics. For Hagara's Focused Assault, you don't care about a consistent DS interval. You care about whether you have a DS or not.

Although BT is going to get some penalty on the overall scale for timing it and also not being able to use it within 0.01s like my sim, I think in-the-moment the use will be worth it. The question is just how often is that use going to be truly worth it, though. If you're just using it to primarily combat auto-attacks, I don't think that's a really truly effective use for it and I think a more consistent DS interval would be better there.

However, if there's a boss that has a breath on like a 20s timer, I could see BT start to shine some. A PL-Glyphed Outbreak-BT (we should probably come up with acronyms or something) setup generates somewhere around 494 RP / min, but that's with hit/expertise. I think I said it was something like 10% less without hit/expertise, so ~450 / min. You need 5 stacks (2.5 RS) to get 1 rune, so on average you get a 2x BT-runes every 20s, which will line up pretty nicely.

But, I have to ask myself, if RC has a really consistent DS interval, shouldn't that also mean that it will more readily be able to sit on DS runes to time them? The sim is, after all, just blindly using DS right now (with the exception of the Scent stuff I posted).

edit: The good part, though, is that at least we can see that between RE/RC/BT, for the overall scale, it's not like 1 is a clear cut dramatic winner. They're all pretty close to even. So, that does give you the freedom to make that choice yourself. Although many people seem to consider this tier a failure, I don't think they really understand what Blizzard is trying to shoot for with their talent tiers. They want 3 choices that are unique but yet still equivalent in the end (aka. overall). That's exactly what this tier is.
Edited by Euliat on 5/4/2012 9:21 PM PDT
Reply Quote
85 Blood Elf Death Knight
5940
But, I have to ask myself, if RC has a really consistent DS interval, shouldn't that also mean that it will more readily be able to sit on DS runes to time them? The sim is, after all, just blindly using DS right now (with the exception of the Scent stuff I posted).


I would think that it would be hard to pool runes with RC because you'll end up unpredictably GCD capping yourself. I haven't really tried it though, to be honest.

If you're just using it to primarily combat auto-attacks, I don't think that's a really truly effective use for it and I think a more consistent DS interval would be better there.


I think this depends on how spiky things end up being at the end of the expansion. If we're talking full on ICC then that would probably heavily favor regularity (at the same time though, I think DKs would be incredibly weak regardless compared to rage tanks, since we necessarily have downtime on our mitigation). If things end up like the end of Cataclysm though, even though you spike down potentially fatally, it doesn't happen that often. The opportunity to immediately recover might outweigh the increased spike frequency in that scenario, even if you're just talking auto-attack damage.

edit:

edit: The good part, though, is that at least we can see that between RE/RC/BT, for the overall scale, it's not like 1 is a clear cut dramatic winner. They're all pretty close to even. So, that does give you the freedom to make that choice yourself. Although many people seem to consider this tier a failure, I don't think they really understand what Blizzard is trying to shoot for with their talent tiers. They want 3 choices that are unique but yet still equivalent in the end (aka. overall). That's exactly what this tier is.


Yeah, after looking at the numbers, I'm inclined to agree. I really doubt that I'll ever use anything but Blood Tap the way the numbers are right now, but I wouldn't assume that someone who uses RC doesn't know what they're doing. I think RE has a non-negligible (if not exactly measurable) weakness in real-world play for Blood, since it perversely incentivizes keeping FU runes on cooldown, but that doesn't really apply to DPS specs.
Edited by Anothriel on 5/4/2012 10:01 PM PDT
Reply Quote
90 Draenei Death Knight
11090
05/04/2012 09:58 PMPosted by Anothriel
I think RE has a non-negligible (if not exactly measurable) weakness in real-world play for Blood, since it perversely incentivizes keeping FU runes on cooldown, but that doesn't really apply to DPS specs.


I think RE is only "good" for Frost, and that's because all of their runes are equally useful. For Unholy, I don't think it's as powerful since they're primarily going to focus on getting U/D runes, which is a really odd combo to try and get. Of course, that's without any modeling at all, and I might take the structure from my sim now and see what it does with Unholy/Frost, but probably not.

Anyway, what I was trying to get at is, if anything, I think RE needs to be buffed. Except in PvP, I really don't see you ever taking RE as Blood. Maybe that's fine with the developers, but, to me, it just seems like a gimped version of BT. Same DS throughput using unintuitive mechanics and having to give up a pretty decent amount of pure blood runes. It's really just a lose/lose.

That being said, I don't think modifying RE itself is necessarily the way to go. I think the underlying problem is and always has been that not every rune is equally useful.
Reply Quote

Please report any Code of Conduct violations, including:

Threats of violence. We take these seriously and will alert the proper authorities.

Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.

Harassing or discriminatory language. This will not be tolerated.

Forums Code of Conduct

Report Post # written by

Reason
Explain (256 characters max)

Reported!

[Close]