I don't know if this is the right place to ask but has anyone noticed with the latest live patch that death runes are reverting to normal runes prematurely.....ie reverting immediately when another rune is refreshed or on cessation of combat??

Notice the same and tried duplicate it on a target dummy, but with no succes to determine what it causing the problem.

Dangerously close wording to a handholding post, but I'll let it slide because the answer to your question does have merit.

I actually spoke to one of the item guys at this past Blizzcon and flat out posed the 'Why did you make items so brain dead now?' question. His response was "That instead of just looking at what stats you have on gear you have to think what you COULD have on gear." Of course higher ilvl items will be better, but we all hated having to do the odd 10-man or old dungeon for that one piece from the zone that eclipsed the similar item in the next tier.

Gear will probably now be more of a juggling act and more complex than it was before since you now have to starting looking at gear as malleable instead of static. Getting to that perfectly capped hit rating, as an example, is a lot more possible, but you have to huck some gold at it and move a bunch of other stats around.

Why would your personal dps increase when you spam deathcoils at the end of every transformation losing significatly in runic corruption uptime? Your method of testing seems pretty flawed.

One part of Consider's original post isn't making sense to me:

SS if both Unholy and/or all Death runes are up > FeS if both pairs of Blood and Frost runes are up

Doesn't it make more sense to FeS first then SS in this case? If you FeS, you do slightly less damage immediately, but you get two new death runes regenerating, which will more than make up for it. Using SS results in slightly more damage immediately, and a single unholy rune regenerating. If you FeS first, you will get more SSs out in the long run.

Although Kahorie's is still a bit buggy, it seems to agree with me that FeS makes more sense. The only case it doesn't make sense is if the boss is about to die.

Judging by current WoL dps rankings, there is still a noticeable difference between 2h frost and 2h unholy dps. Going by 25 man heroic saurfang --- highest unholy is 26k. 2nd place unholy is 25k. First and Second place 2h frost is only 21k.

In leading up to Cataclysm, many of us are wondering how to properly gear, gem, enchant, and reforge. The weights on the front page certainly lead us to pushing haste, but in actuality, I propose that if you are close to a certain haste value, it is better to gem for pure strength or crit/strength and to maximize expertise.

It is understood that haste has some cutoff point; basically as you approach and pass a point where you can fill every GCD, you begin to lose opportunities to use runes optimally -- something many have experience under Heroism with pre-pot'd speed potions and/or engineering gauntlets. It is not uncommon to be able to fill almost every GCD today, even in Unholy presence, barely ever having to touch Horn of Winter. The question is what is the cutoff for haste?

In working on and with the Unholy simcraft module, I sought to answer this question. I believe this module to be a fairly close representation of DPS at 80 under 4.0.1; in particular, I believe it properly handles rune regeneration and how crit and haste scale various sources of our damage. The result I found is that there is not a specific sudden point where haste becomes less valuable than crit, but instead, the curve gradually slopes off, starting significantly at as little as 700 haste:



To read the graph, think of the center as the current stats of a character (mine, in this case); the base values are on the right in the key. The X axis represents changes to a single stat, and the Y axis represents the DPS if that change were made. This is a very large graph, showing plus or minus 1,000 of each stat -- you can quickly see expertise cap at 0, and haste/crit cap just before the beginning of the graph (since they start at 1070 and 1180 respectively).

It is difficult to get the exact point where they meet up -- and indeed, every time you change one stat, the graph will change -- but at around -600 to -400 there is a clear intersection, after which crit becomes a more valuable stat. Even at just 1,000 haste, crit is performing roughly 25% better than haste. The real surprise is expertise -- once you get to the point where almost every GCD is used, a dodge or miss of a scourge strike is a significant DPS loss because you can't recover that attack. Expertise is, indeed, very valuable once you are at around 800-1,000 crit/haste because of the extreme cheap budget it has.

Zooming in on the same graph but plus/minus of only 150 points:



Here the graph is a bit less smooth, but the same details are present -- expertise is outperforming both crit and haste significantly, with these stat weights:

Once we hit 85 and haste is back down to sane levels (perhaps 20% or so), haste will once again dominate everything except hit and strength.

TL;DR version: Once you get to around 800 haste, maximize hit, expertise, and then dump into crit. Pushing more haste has less payoff, particularly compared to expertise (where the cap is 177 expertise rating, which is 23 expertise -- the in-game tooltip is incorrect, as has recently been confirmed by a blue post). Everyone should sim their own toon, be it via simcraft or kahorie's simulator, but haste definitely suffers from a gradual curve of diminishing value and at current insanely inflated gear levels, it is very possible to climb that curve to a point where haste is very suppressed in value.

Addendum: Why is it so easy to fill our GCDs? Obviously Unholy Presence is a big part of it. So is Unholy Frenzy. Runic Corruption is a huge, huge part of it as well. Add in Empowered Rune Weapon and Blood Tap for extra attacks, potentially some AMS absorption, heroism, speed potions, engineering gauntlet enchants and suddenly there are so many sources of extra runes (directly or indirectly via regen burst) that they can't all be spent optimially.

The results confuse me. Expertise is an absolutely terrible stat for us, having no impact on roughly 50% of our damage output (diseases, pets). Its value has always scaled with latency - more latency causes expertise to increase in value.

Crit doesn't affect gargoyle or SS shadow portion (directly), and frost fever is still bugged and only critting for 150% in current gameplay.

At over 1000 haste in a ~2 minute saurfang fight I had no issues with running out of GCDs even with haste pot/trinket proc/bloodlust/unholy frenzy uptime for over 50% of the fight. And on the much longer fights that matter such as LK, that won't be much of an issue since having UF/lust/DBW haste proc/pot up will be only 15% of the fight or so. On those fights, having faster rune regen is absolutely necessary above all else; more strikes and more runes available always trumps strikes that crit more often for us. The extra haste ensures I always have a rune to pestilence valks / adds when they land or spawn instead of having to wait, blood tap or hold off casting. In addition, DBW's proc can proc crit or str thus not always being an issue.

What sims won't show is that haste brings us more utility because we will have runes available more often to make decisions with such as pestilencing or DnDing adds, casting DT, extending diseases with FeS on a fight slated to end soon vs re-outbreaking, etc.

There would be a point where crit becomes more valuable than haste but I still don't believe it attainable in current gear. On very short sandbox fights where you have high bloodlust/UF/proc/pot uptime, perhaps, but in real world play, I don't believe so.

Much of your post here is pointing out is the general fallacy of sims or spreadsheets -- they never model real-world fights. This certainly is true, and yet, we still use stat weightings (which are the product of those flawed approaches). Also worth noting is that sims and models tend to amortize things like random procs from DBW and good/bad luck with avoidance and crits. It certainly is possible to, say, have an encounter where you are at 0 expertise and yet never miss. What a sim will tell you is that if you did the same fight thousands of times, on average, what difference a stat change may make.

The problem with dismissing expertise though is that it is all relative to where you are. 1% crit will make a larger relative portion of your dps when you have low crit than when you have high crit; the absolute difference is the same, but relative to your total DPS output, the difference is a smaller fraction of the whole of your damage (example: a hundred attacks doing 10,000 damage with a 2% crit rate will average to 10,000 * 0.98 + 20,000 * 0.02 = 10,200 damage; going from 0% crit to 2% crit is a 2% damage increase. Going from 50% crit to 52% crit, though, is still a 200 dps increase, but 200 is only 1.3% dps increase).

That really is the point; there are tradeoffs between stats and how they interact; more directly, ratings people use need to really come from a simulation of their own gear, not static values in the OP. I definitely agree with your point that such values need tempered with making decisions about what matters in a given fight trumping target dummy sim results, of course. I was just surprised at expertise's value and haste's curve shape.

When FF is fixed to properly crit 200%, and when ghouls gain our crit rate, crit will be even better; still, below 20% haste or so I imagine haste will always be superior, but the curve is there and it does become flatter the more you get. It isn't a hard cutoff like I had thought (and I believe many thought); instead it seems more gradual, probably due to interactions between high and low haste periods (heroism, DBW, potions, engineering gloves, etc).

In terms of fight specific mechanics like HLK, I certainly agree -- for all progression content, spec and gear should be tuned for what makes the most sense for that fight. But when we're talking about even 5% haste, noticing that difference in rune regen is very difficult when you're talking about a case of holding a rune for a key moment in a fight (much more important is knowing when the moment will come that you need the rune rather than that incremental 5% making the entire difference). I think the point stands, though, that the weights are what they are and while sims/models may not accurately represent real world fights, they still provide a guideline to understand what gear changes might change about our dps, and can change quite a bit based on each DK's own gear.

One thing I find a real issue in my rotation is the way death runes are currently used. I find that if I just hit double FeS and am due 4 new death runes coming up. Now my issue happens when I get some sudden doom procs just as my first set of runes come off CD. I now have 2D and 1U with the other 3 regenerating and around about enough RP for a DC. That sudden doom uses no RP but I have to use that proc. Then I would hit SS which will use an unholy rune first. Now if that SD DC had procced runic corruption, I will have my other runes racing towards being full. If the unholy rune comes up first, which often it does due to the SS in the previous rotation. I get that unholy rune regenerating before the other 2 death runes. Now I am sitting at 1U 2D and about 1 second to go before I have 4 death runes.
This in itself is not an issue. What is an issue however, is that the game will use both blood death runes before it starts to use any frost death runes. In this situation. I have some left over RP for a DC, I have just hit SS using an Unholy rune. I might get an SD proc and following all of this, I now spam some SS to get the blood out the way so that I can dip into my frost runes. What inevitably happens is that I am left with using 2 GCD's on SS that comes out of blood runes while my RP is getting close to cap and my frost runes are sitting capped.

This death rune priority system seems at odds with the new rune regen mechanic. It would seem far more natural to me that the death runes are prioritized as:
if (frost death runes number) > (blood death runes number) then use frost death rune first.

This should cause the blood and frost runes to stay closer in sync.

The other problem with this is that as your RP gets closer and closer to the cap due to some lucky SD procs. And you get into this rune situation. You are torn between 3 problems, all of which are not optimal in the unholy rotation, and all of which feed off each other, primarily due to this unintuitive rune usage.

1) Keep your frost death runes capped for as little time as possible by ignoring SD procs and RP cap and spam SS.
2) Allow your Frost death runes to remain capped for slightly longer because you need to either use the SD proc or not allow your RP to cap.
3) Spam DC's to dump your SD procs and RP, so that the following spam of SS does not cap your RP.

We are told to not let our runes stay capped if we can help it as it is a dps loss. So it makes sense to try to do option 1.
But then option 1 leads to the other no-no which is allowing your RP to cap.
Option 2 leads our runes to stay capped too long.
Option 3 overlaps possible RC procs, which is not optimal.

All primarily because we have to worry about getting the blood runes out of the way before we can dip into the frost runes.

If it ate the first blood rune, then went to the first frost rune. We would be able to naturally fit a DC in here as everything is regening. Currently after 2 death rune usages, we are torn between those 3 scenarios.

Am I the only one finding this a problem. With the new rune regen mechanic?

To make the problem really seem more obvious I use DDR. With the standard 6 rune ui it is not as apparent, but with DDR, it is very apparent when your frost line is full and you are still 1 or 2 gcd's away from even touching them.

I personally can't say I've noticed this happen. My guess is that you're having trouble following the priority perfectly, and as such, you're running into situations where your RP is in danger of capping as you have 3 or 4 death runes up. Even if this is the case, stick to the priority on the OP - where SS with multiple DRunes is prioritized over SD/RP capping. It's better to waste a bit of RP than let runes sit idle; unless you have math to prove otherwise.