Sacred Shield vs Eternal Flame 5.4 [PROT]

[Wrathblood]

Wait a minute. Riemu, how on earth are you averaging getting 5 BoG stacks with still 10-15 seconds left on EF in 496 gear?

Lets say you're running with DP and you're Haste capped. Your rotation looks like CS - J - X - CS - X - J - CS - X - X, a cycle which would take 9 seconds and generate 5 guaranteed HoPo. Against a boss with a 1.5 second swing timer, you'll face 6 incoming swings in that time. Assuming 33% avoidance (high, but makes the math easy), you'll have 2 "avoids", and thus an average 0.6 GC procs. So that gives you 1.8-1.9 ShoRs per 9 seconds. Thus you'll get, on average, another 0.45 ShoRs from DP procs.

So, Haste capped, running a perfect rotation with 100% time on boss, you're looking at ~2.2 ShoRs per 9 seconds, meaning you can get 5 BoG stacks in 21-22 seconds on average. But that's on average. You might get a couple DP procs in a row or GC procs, letting you get there much faster. But you also might get zero GC or DP procs. You can still reach 5 BoG in 30 seconds, but only if you're Haste capped and playing perfectly. Plus, you lose all your timing flexibility. You need to fire off that EF as soon as you can, regardless of what damage you're taking.

I'm not exactly disagreeing with your approach, just noting that consistently waiting for 5 BoG stacks before firing off the EF might not be nearly as easy as you're suggesting.

Edit - Oh, forgot the most interesting part. When T16 4P comes into play, BoG stack management is a very interesting question. You get the free EF at 3 BoG which doesn't go away when you get BoG stacks 4 and 5. And the extra BoG stacks certainly make *that* EF (and ticks) more powerful.

However, BoG stacks 4 and 5 are taking away "potential" EFs you could be casting for free. If you had cast that EF as soon as you got your free one at 3 BoG stacks, the EF (and its ticks) would be less powerful, but you end up casting 66% more EFs (5 for every 3 you'd cast while waiting for 5 BoG stacks).

Lets think about the math. At 3 BoGs, each EF would directly heal for -40% less than a 5 BoG EF, both on the initial cast as well as each tick. Assuming you keep EF continually running, this means you'll get 40% less HoT healing over the fight which is a lot. On the other hand, instead of casting EF every ~23 seconds, you cast it every 15 seconds. This both likely reduces the overheal from those EFs as well as reduces the likeliness of you getting burst down. If a tank-kill spike damage period averages ~7 seconds, you'll get a big heal every second period instead of every third.

[Riemu2k3]

@ Wrathblood

I probably didn't put it right when I said that I can recast EF at 5 BoG stacks with 10-15 seconds left on EF. What I should have said was, the few times I looked at my EF timer when I had 5 BoG Stacks, it was hat 10-15 seconds left. I honestly can't tell you how often I looked, but that's what I remembered. You're math is probably spot on but then again, I'm not the mathy type. I'm more like the "feeling" type :-)

We tested Galakras (Multi Adds), Juggernaut (tank switching with 5 HP stacked after EF recast/switch), Dark Shamans (Multi Add fight, we tanked the blobs on PTR - But it was too hectic here to watch anything). Most of my observation was from Galakras and Juggernaut. And I'm quite sure (again, not the math type) when you tank switch and you have 5 hp pooled, it's totally possible to get to 5 BoG Stacks at average 10 seconds left, cause thats what it felt like. And when I'm not currently tanking, I don't really care about my EF tick as my SoI usually is enough to keep me topped (or else the AoE damage happening is so high, that there is plenty of raidhealing going on anyway). EF is just a bonus then.

HOWEVER - we went back yesterday and did a more controlled Dark Shaman / General Nazgrim. Here I actually noticed it quite often, that I had to recast EF with 4 BoG stacks, as it fell off / was about to fall off. Obviously if that happens, I don't wait till I get 5 Stacks.

But then again, compared to LIVE where I have 547 Itemlevel and 49% Haste in Offset Items (without 4pt15), that should hardly ever happen. I think it should be alright to reach 5 BoG stacks before EF falls off. And if it somehow still falls off, recasting it with 4 instead of 5 BoG stacks probably won't kill you. But that percentage should be quite low in 550+ Gear I think. Btw. Did you factor in the extra DP procc chance from casting EF in that window with 4pt16?

For clarification sake: I don't plan on using EF without 4pt16 as it just feels wrong to me rotation wise.

---

That other thing you mentioned however depends on 10/25 man and the amount of damage intake I guess. Of course, if you are about to die you need to heal yourself back up with a 3 BoG EF. But, most of the paladins here (I'm abit too fast for my own good in that department I guess) hardly use WoG at all. Why should EF be any different? Especially if you know, that in 0.5 seconds comes the next 200k EF tick (assuming high vengeance fight when there is high damage intake).

My playstyle is more instinct based than anything else. If I feel I need the heal, I heal myself - I don't really care about how many stacks I have. I have to trust that if I make that decision, I made it for a reason. Playing with EF instead of SS just means, you have to be even more sure you needed that heal. And I don't think casting it every 3 stacks instead of every 5 stacks will change that.

---

But, what about the increased DP proc chance? I mean, if you cast it at 3 stacks, you will get more EF off and therefore get more DP procs. So I guess, in a low damage fight or a situation where you don't tank anything, you should be casting EF at 3 BoG stacks?

So I guess that adds quite a dynamic into our usual rotation...

What do you think?

[Nairobi]

In my testing, which was largely relegated to dummy tests, I spent more time casting EF @ 3xBOG instead of 5. Granted, due to DP procs, there were (quite a few) times that I was firing off 4-5xBOG EFs, but by and large, I focused on casting at 3x, to maximize DP procs and ShotR uptime (as well as Glyph of WOG uptime). Standstill rotation (IE no GC procs) netted me on average 75-80% ShotR uptime and 35% uptime on Glyph of WOG, with 95%+ uptime on EF. There were times that, thanks to DP procs, I'd be back at 3x BOG stacks before the Glyph effect even fell off!

Now, clearly, this is going to be lower on healing (via HOT effects), but it DOES have more frequent direct healing (via initial casts) which is an interesting position to be in. Additionally, it does have a higher uptime in both ShotR AND in GoWOG, which may offset the HPS decrease from holding out to 5x BOG.

I'm not advocating that one method is 100% right or wrong, but merely pointing out that we have a sliding scale to use:

Casting early and often will give more uptime on ShotR and GoWOG, at the cost of lower EF ticks (but more frequent EF initial casts).
Casting late will give more powerful HoTs, but fewer DP procs (ergo lower ShotR/GoWOG uptime).

Tailor to the situation, as always.

[Wrathblood]

As an aside, I spent a few hours last night running Simcraft across a couple dozen sets of conditions trying to map our healing generated, TMI, and the like. Didn't include T16 4P or haste capping (obviously the gear-set I was looking at had substantial haste, but it was a fair ways from capping) because I wanted to focus on where most raiders would be starting SoO (and I had enough variables to work with already, throwing in a couple more would have added a lot more situations). I used the "Light Movement" fight condition.

I do intend to get to Haste-capping and t16 4p, I just could only stand so many hours of Pet Battles (while running the sims in the background). Consider these numbers some thoughts on what to expect during the first few weeks while people are gearing up.

I'm not quite done running everything so I can't say this is 100% certain (and obviously it'll vary a bit with different specific gear sets, raid buffs, etc), but first off, EF > SS even without 4p or Haste capping. I haven't scaled it all the way up to T16 25H bosses yet, but at lower damage levels EF gets *stronger* vs SS as incoming damage goes up so I suspect EF will also win vs T16 25H.

Looking at EF, generally speaking your best results come from waiting to refresh your EF until you've got 5 or less seconds on the HoT remaining, at least 3 stacks of BoG and never refresh it with less than 3 HoPo. Holding out for 5 or even 4 stacks of BoG ends up being a loss of coverage time which is terrible for spike prevention.

Essentially, EF HoT up-time ends up trumping almost every other consideration. With conditions like "let the EF HoT wear off before refreshing and always wait for 5 stacks of BoG" you end up with HoT uptimes in the low 80s% which destroys you. The individual EF casts and the ticks heal for a TON under those circumstances, but your TMI goes through the roof.

The highest uptime came from "waiting for 3 stacks of BoG and refreshing with 5 or less seconds left (increasing the time from 5 seconds didn't affect things much, so I guess 5 seconds is typically enough time)", while using HA (more on this below). It cost a little bit of HPS (though less than I expected) but was the strongest in terms of TMI. Giving up ShoRs for EFs, as long as it increased your HoT up-time, appeared to be a win. With higher ilevel gear or 4p, we'll be able to ask the question of how giving up ShoRs for more EFs *without affecting HoT up-time* plays out. I assume you'll want the ShoRs but I don't know, and it may vary based on incoming damage.

I don't know how to incorporate tank-swaps into Simcraft so everything was a solo-tanked fight, making HA very attractive. With DP, it would often take 15+ seconds to get up to 3 BoG stacks and then another 3 HoPo to use that first EF, and occasionally that first 15 seconds would be pretty painful. If you have someone else do the pull (and there are 0-1 1 tank fights in SoO so hopefully its not a problem) then I think DP and HA will end up in the same ballpark (though HA is likely a better bet to take advantage of fight mechanics).

However, I suspect that once we get Haste capped and, more importantly, get T16 4P (though only our gloves are BiS and actually getting 4p requires clear itemization sacrifices. (The sacrifices aren't as painful as they've been in previous tiers, but there's still a cost) I'm personally leaning towards passing on tier in the early going. Once we start talking about heroic warforged pieces, I believe going for our 4p will still be worth it but I'm not *sure* that will be), then DP will pull clearly ahead and not look back.

[Deleted]

You should really not experience loss in uptime even if you wait for 4 or 5 stacks. You should be able to get far more than 3 stacks of BoG between each EF recast.


EF vs SS depends on the nature of the incoming damage, it is also a fairly more interesting comparison before you get the tier bonus. If you get the tier bonus than EF is probably gonna win out on almost every fight if not every fight, but before you get it the comparison is quite interesting.

[Deleted]

I don't know how to incorporate tank-swaps into Simcraft so everything was a solo-tanked fight, making HA very attractive. With DP, it would often take 15+ seconds to get up to 3 BoG stacks and then another 3 HoPo to use that first EF, and occasionally that first 15 seconds would be pretty painful.

That's a fascinating post, thanks. One question: how do you think SW would compare to HA and DP? Could it help cover the first 15 seconds instead of HA? The shorter CD of HA makes it more attractive for tank switches and trash pulls, I guess.

[Wrathblood]

You should really not experience loss in uptime even if you wait for 4 or 5 stacks. You should be able to get far more than 3 stacks of BoG between each EF recast.


EF vs SS depends on the nature of the incoming damage, it is also a fairly more interesting comparison before you get the tier bonus. If you get the tier bonus than EF is probably gonna win out on almost every fight if not every fight, but before you get it the comparison is quite interesting.

Depends on your gear and the fight mechanics. Haste capped against Patchwerk, certainly. With a light movement fight and ~14k haste on elite, reaching 5 BoG stacks between refreshes was a stretch and frequently resulted in the buff falling off briefly. 4 stacks was almost always reachable and impacted HoT uptime by 1% or so. My guess is setting the minimum to 4 stacks results in the HoT falling off perhaps once/fight.

Bear in mind that since I was also setting a maximum HoT duration for freshing (which I varied from 0 to 5 seconds), even with, say, a minimum of 3 BoG stacks, it would frequently refresh with more. So if I set the maximum refresh duration to 2 seconds, even if it reached 3 BoG stacks it would not recast EF until the HoT had 2 or fewer seconds remaining, potentially ending up casting with 4 or even 5 stacks. As a result, the shorter I made the maximum refresh duration, the more time the sim would have to potentially build BoG stacks before refreshing. As an example of the average BoG stacks consumed on each EF cast, you can look at how much healing the average EF cast did.

With a 0 second refresh max and 3 BoG stack minimum (using T15 25H damage intake), the average EF healed for ~330k so clearly it was frequently being cast with 4 or even 5 stacks of BoG. Raising the max refresh to 5 seconds (keeping the 3 BoG stack minimum) reduced the average EF heal to ~300k. Setting it to a 5 BoG minimum and 5 second max refresh raises the average EF heal to 378K.

[Nairobi]

So, just to make sure I'm taking away the right information here, Wrath you are only advocating HA in terms of using EF WITHOUT 4t16, right?

I just can't fathom how HA (or SW) would hold a candle to the ShotR uptime given by DP, once you factor in 4t16 and gaming stacks with re-casts at 3BOG. Every 3 ShotR's, you get 2 ADDITIONAL proc chances, often putting you back at 3 stacks in ~4 GCDs.

I, personally, won't be using EF until I have 4t16, so I guess that I just assume that anyone who discusses the talent will be doing so under those suppositions, but I still wanted to make sure that I'm not missing something egregious. As I posted earlier, (under 4t16) you're taking a slight hit in EF HOT strength to increase your uptime of ShotR, GoWOG, and additional "frontload" casts of EF. By almost all measures, that seems to be a win, in my book.

[Nobleshield]

How bad IS the SS nerf? I'm trying to debate between the two talents without factoring in 4pc T16 which obviously pushes EF higher. I kind of like the idea of having a HoT that I can keep rolling on me.

[Deleted]

I dont really think you can make a straight comparison between an absorb an a heal and say one is better than the other in all circumstances. It is apples and oranges. It is not a matter of which has the highest throughput, it is a matter of which works the best for a specific boss fight. They also got different tick speeds making it even weirder. The matter becomes even more complicated when you are not using the T16 4p, which means that you are comparing a heal vs an absorb + higher SotR uptime.

There is no easy answer to this question, you have to check your own damage in-take pattern on a boss to boss basis.

[Jackielope]

I dont really think you can make a straight comparison between an absorb an a heal and say one is better than the other in all circumstances. It is apples and oranges. It is not a matter of which has the highest throughput, it is a matter of which works the best for a specific boss fight. They also got different tick speeds making it even weirder. The matter becomes even more complicated when you are not using the T16 4p, which means that you are comparing a heal vs an absorb + higher SotR uptime.

There is no easy answer to this question, you have to check your own damage in-take pattern on a boss to boss basis.

There might not be an easy way to test one's damage patterns per boss, but there is certainly an easy way to get generalized statistics. Use SimC like Theck did at http://maintankadin.failsafedesign.c...art=45#p772465.

[Deleted]

There might not be an easy way to test one's damage patterns per boss, but there is certainly an easy way to get generalized statistics. Use SimC like Theck did at http://maintankadin.failsafedesign.c...art=45#p772465.

Did not need a SimC to predict that result. Still, SimC is comparing apples and oranges when you are making apple juice. The problem in raiding is that some bosses like orange juice. Some bosses can like both depending on yourself and your raid group. I have no doubt that EF is superior on a majority of the boss fights, especially with the 4p. Without it, I am not so sure before I do the fights in live. That it will be superior on every fight is quite an overstatement though.

Pretty much exactly what I was saying.

you cant make a straight comparison between an absorb an a heal and say one is better than the other in all circumstances.

[rhandric]

The difference between the two, prior to T16 4p, depends heavily on the fight and your healer team (and on you, as the tank). Once you get T16 4p, those will still have an impact, but the benefits of the 4p will reduce the impact of those factors, but aspects of the fight will still have a say into which is better.

[Deleted]

The difference between the two, prior to T16 4p, depends heavily on the fight and your healer team (and on you, as the tank). Once you get T16 4p, those will still have an impact, but the benefits of the 4p will reduce the impact of those factors, but aspects of the fight will still have a say into which is better.

Yeah, basically, the 4p changes the scales a lot. It goes from...

Shorter tick speed + healing vs absorbs + higher sotr uptime (and higher damage)
to
Shorter tick speed + healing vs absorbs

which makes it favor EF on some fights that it probably did not before. Another drawback of EF is that it makes emergancy WoGs harder and if you make a "small" WoG for whatever reason it will override your superior HoT.

[Nairobi]

Yeah, basically, the 4p changes the scales a lot. It goes from...

Shorter tick speed + healing vs absorbs + higher sotr uptime (and higher damage)
to
Shorter tick speed + healing vs absorbs

which makes it favor EF on some fights that it probably did not before. Another drawback of EF is that it makes emergancy WoGs harder and if you make a "small" WoG for whatever reason it will override your superior HoT.

Just nitpicking here, but that SHOULD read "Shorter tick speed + healing + higher ShotR uptime vs absorbs", since you're not wasting/using a GCD every 25-30sec to refresh SS, AND you can fish for DP procs. Also means more GoWOG uptime, obv.

[Deleted]

Just nitpicking here, but that SHOULD read "Shorter tick speed + healing + higher ShotR uptime vs absorbs", since you're not wasting/using a GCD every 25-30sec to refresh SS, AND you can fish for DP procs. Also means more GoWOG uptime, obv.

Yes, completely true. Higher SotR uptime with DP however. GoWoG as you mention too ofc.

[Wrathblood]

So, just to make sure I'm taking away the right information here, Wrath you are only advocating HA in terms of using EF WITHOUT 4t16, right?

I just can't fathom how HA (or SW) would hold a candle to the ShotR uptime given by DP, once you factor in 4t16 and gaming stacks with re-casts at 3BOG. Every 3 ShotR's, you get 2 ADDITIONAL proc chances, often putting you back at 3 stacks in ~4 GCDs.

I, personally, won't be using EF until I have 4t16, so I guess that I just assume that anyone who discusses the talent will be doing so under those suppositions, but I still wanted to make sure that I'm not missing something egregious. As I posted earlier, (under 4t16) you're taking a slight hit in EF HOT strength to increase your uptime of ShotR, GoWOG, and additional "frontload" casts of EF. By almost all measures, that seems to be a win, in my book.

First, in answer to your question, I'd say "Yes, assuming you can get someone else to pull. And if you can't, then probably still 'Yes' but a little dicier." In terms of through-put, once you get 4p and are Haste-capped, DP is going to pound HA and SW into the ground. There may well be strategies for particular fights where HA or SW line up so well with what you need to do that they're situationally superior, but overall DP will be your go-to choice.

HOWEVER, pulls aren't going to be pretty with DP. You can't pre-cast EF on the pull like you can SS, and while you *could* get EF up and running 3-4 seconds after the pull, its going to be a pretty feeble EF, it'll cost you an early SHoR, AND it'll push back your first free EF to boot. We'll probably have a free glyph spot when 5.4 hits, so I'm sort of resigning myself to glyphing DP (I mean, the cooldown DP, not the talent DP) and just assume that I'll be using it on the pull whenever I pull.

HA/SW don't have this "weak on the pull" issue, even Haste capped and with 4p there's still going to be a ~12 second period after the pull where HA/SW will be considerably superior to DP. If for some reason the pull is super-important and you *have* to do the pull, then HA/SW might be superior to DP.

- - - Updated - - -

How bad IS the SS nerf? I'm trying to debate between the two talents without factoring in 4pc T16 which obviously pushes EF higher. I kind of like the idea of having a HoT that I can keep rolling on me.

-30% to SS

Edit -

I dont really think you can make a straight comparison between an absorb an a heal and say one is better than the other in all circumstances. It is apples and oranges. It is not a matter of which has the highest throughput, it is a matter of which works the best for a specific boss fight. They also got different tick speeds making it even weirder. The matter becomes even more complicated when you are not using the T16 4p, which means that you are comparing a heal vs an absorb + higher SotR uptime.

There is no easy answer to this question, you have to check your own damage in-take pattern on a boss to boss basis.

I actually completely agree with Firefly here. There is no "one size fits all perfectly" answer, both because everyone has different raid compositions putting different requirements on the tank and because fight mechanics vary significantly. On Malkorok, when the Miasma is up, all heals are turned into absorbs up to a max of each raider's health. On the one hand, that seems pretty fantastic for EF and I'm sure it is for most folks.

However, for 25H the damage output on the tank is probably going to be pretty bursty and the tank is going to be getting massively bombed with heals. In a situation where the healers routinely cap the Tank's absorb, EF won't be bad since it'll be frequently adding to the absorb as well to replace damage taken. But if the Tank is getting hard enough to get their absorb repeatedly blasted through, SS might actually be even better since it can stack on TOP of the absorb the healers are already giving you. I don't know.

In general, I believe EF > SS even without 4p. But that's in a vacuum. Everyone should consider their raid composition and the mechanics of each fight before deciding for themselves what to do.

Edit #2 - Celyndrashad had a great quote, which I enjoy paraphrasing (and lead off the EJ post with) and I think is apt here as well: "I'm not here to tell you what's best for you. Only you can do that, because only you know exactly what your situation looks like. But to do that, you need to know how things work. And that's why I'm here: to help figure out how everything works."

[Nobleshield]

I'll have to experiment and see then, as a pretty much 100% 10-man raider. What still happens to scare me the most when tanking is a drop in health, so I think part of my thing with SS is that I don't really notice it there, since if I take a big hit my health still drops, while having a nice fat HoT ticking away is kind of reassurance.

[Jackielope]

Did not need a SimC to predict that result. Still, SimC is comparing apples and oranges when you are making apple juice. The problem in raiding is that some bosses like orange juice. Some bosses can like both depending on yourself and your raid group. I have no doubt that EF is superior on a majority of the boss fights, especially with the 4p. Without it, I am not so sure before I do the fights in live. That it will be superior on every fight is quite an overstatement though.

Pretty much exactly what I was saying.

you cant make a straight comparison between an absorb an a heal and say one is better than the other in all circumstances.

I said generalized statistics, which pretty clearly does not mean specific statistics for specific situations, though in some cases SimC can be set up for that. There is value in knowing what a baseline will typically be. Knowing when that baseline changes is likewise valuable, but tends to be more in depth work than simply figuring out what the baseline is.

[Deleted]

I said generalized statistics, which pretty clearly does not mean specific statistics for specific situations, though in some cases SimC can be set up for that. There is value in knowing what a baseline will typically be. Knowing when that baseline changes is likewise valuable, but tends to be more in depth work than simply figuring out what the baseline is.

And I said specific boss fights. Seems pretty silly to quote me with generalised stats.when I very specifically talked about specific boss fights