Casting PWS during spirit shell.

[Deleted]

According to my calculations and confirmed by game testing it is possible to increase the absorption stack you build during spirit shell by using PWS, while spirit shell is active. This is very dependent on your haste.

Definitions:

H = (1+(hasterating)/4.25)*PI, where PI = 1 if you have PI up or 0 otherwise
M = %mastery from tooltip expressed as a fraction of 1
Tn = Spirit shell duration time available for casting non borrowed time spirit shell PoH after casting n PWS
Xn = total number of spirit shell PoH casts when casting n PWS
An = total absorb value of spirit shell + n PWS
BT = 1 if you have borrowed time up before activating spirit shell or 0 otherwise
PoH spirit shell absorb*5 = k
PWS absorb = k*0.42*(1+M)/(1+0.5M)


The decrease in casting time for PoH due to borrowed time is 0.15*2.5/(1.05*H), thus we can assume all casts of PoH have the same cast time and just increase the total casting by by this factor for each borrowed time proc.

Xn = int[Tn/2.5*(1.05*H)], where int is a function that returns the integer part of a real number

An = k*[Xn + n*0.42*(1+M)/(1+0.5M)]

Thus maximising the absorption buffer you can create during spirit shell requires maximising [Xn + n*0.42*(1+M)/(1+0.5M)]

This assumes that you are pre-shielding i.e. the raid is not currently taking high levels of damage, so you are just interested in the amount of absorb you can put on top of ppl in anticipation to the damage spike. Under conditions of continuous damage PWS/PoH 1:1 is automatically better HPS than straight PoH spam anyway.


Tn = 15 +(0.15*2.5)/(1.05*H)*BT - n*(1.5-0.15*2.5)/(1.05*H) = 15-(n*(1.5-0.15*2.5) - 0.15*2.5*BT)/(1.05*H)

Xn = int(6*1.05*H +0.15*BT - 0.45*n) = int(6.3*H + 0.15*BT - 0.45*n)

Xn = 6 with n = 1, requires H = 1, but that is with 0 loss of time from latency. For me I need at least 3% haste in order to get 6 spirit shell casts with zero haste so all values need to have 3% added to them to account for this.

Xn = 6 with n = 2 requires H = 1.0715

That means 7.15% unbuffed haste allows you to cast 2 extra PWS during spirit shell, which means it raises the amount of absorb by 0.9 of a full spirit shell cast at 25% mastery

Xn = 7, with n = 0 requires H = 1.0874

Xn = 6 with n = 3 requires H = 1.143

Xn = 7 with n = 1 requires H = 1.159

Xn = 7 with n = 2 requires H = 1.231

Xn = 8 with n = 0 requires H = 1.247

Thus in general there is an intermediate breakpoint for the absorb with n = 1.

With PI on

H needs to be divdied by 1.2, which means anything with H below 1.2 is baseline.

Xn = 8, n = 0 requires H = 1.039

Xn = 8, n = 1 requires H = 1.098

Hence up to 10% unbuffed haste + a latency factor (this is ~3% for me) allows you to add 7 casts of spirit shell baseline and 8, n=1 casts during spirit shell.

[Deleted]

Seems quite obvious considering that there's very few haste amounts where you can fit in exactly X PoH's without any spare time, simply spending that spare time on shields (and said spare time will then be increased slightly by the borrowed time) is obviously more efficent than to start a PoH cast that'll be finished after SS has expired.

However unless you can fit in two PW:S the difference between doing this and just throwing a shield (or something else, a solace/penance might be more useful) at the end of your SS (and speeding up your next cast after) should be very slim, so I don't really see any reason to gear specifically for any haste amount based on this, but it might be useful to keep in mind if you end up with said haste anyways.

In regards to the speciifc breakpoint where you can fit in 2 shields you are already quite close to the next PoH during shell, so wouldn't it simply be more convenient/useful to gear for that breakpoint instead since that's a far more efficent way to gain roughly the same amount of absorbs. If you just care about HPS without any regard for mana you should just weave PW:S->PoH->PW:S during shell either way.

To me shadowfiend breakpoints are the only ones worth considering for 10 man, for 25 these breakpoints may have some use but I doubt you'd care more about these than a simple SS/PoH breakpoint. I usually think you make some great posts, but this just feels like you came up with an idea that you like way too much. To me the actual value of these breakpoints are incredibly slim and I'd never consider gearing for them.

[Deleted]

I think you missed the points. Stopping your spirit shell cast early and then going to PWS spam, is only slightly less effective that using the correct number of PWS between PoH casts, in terms of theoretical throughput, but it is completely different functionally:

Knowing that you can add 1 PWS in the middle of your PoH casts means that if you are interrupted for whatever reason and you have to move you can throw a PWS out and not lose a spirit shell cast. If you are below the breakpoint maybe you can spare 1s to move, but you might not have the time to cast a PWS. The result of NOT knowing where you are makes the difference between zero time lost or replacing one PoH cast with a PWS.

Adding PWS in the middle of your sequence, also means you have leeway to save someone in danger, who might not be grouped up for spirit shell.

It means more rapture procs

The +2 breakpoint might be close to the extra cast breakpoint, but it is better, because it is less subject to interruption, leads to less wasted time when you active PI and you can spend more of your budget on crit/mastery.

1:1 PWS-PoH during spirit shell is highly ineffective. You will most likely end up wasting spirit shell uptime.

Whether you choose to gear for these breakpoints or any other breakpoints, is not really the subject of my post. The subject is to tell you what you can cast given how much haste you have. Knowing that give your haste you can actually hit a PWS mid spirit shell without losing any spirit shell casts is invaluable for me personally, so I make sure to calculate it. Since it changed in 5.3 I recalculated and decided to get a proper formula for it. I posted the results so others can benefit.

[Deleted]

I think you missed the points. Stopping your spirit shell cast early and then going to PWS spam, is only slightly less effective that using the correct number of PWS between PoH casts, in terms of theoretical throughput, but it is completely different functionally:

Knowing that you can add 1 PWS in the middle of your PoH casts means that if you are interrupted for whatever reason and you have to move you can throw a PWS out and not lose a spirit shell cast. If you are below the breakpoint maybe you can spare 1s to move, but you might not have the time to cast a PWS. The result of NOT knowing where you are makes the difference between zero time lost or replacing one PoH cast with a PWS.

Normally you just cast an instant regardless if you have to move during shell.

The +2 breakpoint might be close to the extra cast breakpoint, but it is better, because it is less subject to interruption, leads to less wasted time when you active PI and you can spend more of your budget on crit/mastery.

Sometimes you need to move once, twice, thrice or not at all during shell. Gearing for one of these specific situations isn't something I think is worthwhile, but if so the one assuming zero movement is the most common occurance.

1:1 PWS-PoH during spirit shell is highly ineffective. You will most likely end up wasting spirit shell uptime.

Hpm wise yes, and so is 2 additional PW:S over an additional PoH.

Whether you choose to gear for these breakpoints or any other breakpoints, is not really the subject of my post. The subject is to tell you what you can cast given how much haste you have. Knowing that give your haste you can actually hit a PWS mid spirit shell without losing any spirit shell casts is invaluable for me personally, so I make sure to calculate it. Since it changed in 5.3 I recalculated and decided to get a proper formula for it. I posted the results so others can benefit.

Fair enough. I agree that it's useful to know the specific breakpoint and if that was your intention with the post I have nothing to disagree about. I simply got the impression that you advocated gearing for this specific breakpoint (+2 PW:S) which I really don't think is worthwhile unless you already are incredibly close (if even then).

[Deleted]

Normally you just cast an instant regardless if you have to move during shell.



Sometimes you need to move once, twice, thrice or not at all during shell. Gearing for one of these specific situations isn't something I think is worthwhile, but if so the one assuming zero movement is the most common occurance.



Hpm wise yes, and so is 2 additional PW:S over an additional PoH.



Fair enough. I agree that it's useful to know the specific breakpoint and if that was your intention with the post I have nothing to disagree about. I simply got the impression that you advocated gearing for this specific breakpoint (+2 PW:S) which I really don't think is worthwhile unless you already are incredibly close (if even then).

If you don't have the right amount of haste then you just shot yourself in the arse by casting an instant. You just wasted a spirit shell cast. Moving out of shit takes less time than a GCD and you might just make the extra cast. That is why instead of criticizing you should be paying close attention to my post, if you DONT have the haste then it might be better to not cast that instant as long as you only move a short distance. If you do have the haste, there is no need to bother thinking about that you just cast PWS automatically and you can be sure that you are losing no time.

Again the point of my post is that you can gear for moving twice and get THE SAME RETURN OR BETTER as gearing for zero PWS casts. Even being able to move once dramatically reduces the time you will lose from moving twice, since having to move once is far more frequent than having to move twice. Even if you have to move twice at least you lose only one cast not two, which again is much better. Gearing for these breakpoints either eliminates or reduces your losses from moving. Since it is either slightly lower or higher healing overall what possible reason is there it not gear for it.

On your last point you made a critical error: 1:1 PWS-PoH outside spirit shell is higher HPS than PoH alone, but during spirit shell it is not because it will lead to you not getting completely overhealing PoH instead of a spirit shell PoH at the last cast. Adding the 2 PWS during spirit shell leads to effectively zero loss in HPS and it can actually be better HPS depending on your mastery.

So basically you are getting almost zero benefit from the extra 800 haste rating you need to get to the 7 casts breakpoint. All you are doing is shooting yourself in the arse by making your biggest CD extremely inflexible.

Being able to cast a PWS during spirit shell AT ZERO LOSS of spirit shell casts adds an immense amount of flexibility. You can move during spirit shell, you can PWS someone about to die without losing spirit shell casts, you build an absorb buffer on a group that you can't hit 5 targets in.

Basically getting that extra haste gives you more HPS overall since haste is a great stat and it is much much much better for spirit shell due to the added flexibility. Basically the 1 PWS haste break point is best for a non-haste build and the 2 PWS haste break point is best for a haste build. The extra cast breakpoint is only best for a full crit build, but anyone going for a full crit build doesn't really care about optimising their healing.

[Deleted]

If you don't have the right amount of haste then you just shot yourself in the arse by casting an instant. You just wasted a spirit shell cast. Moving out of shit takes less time than a GCD and you might just make the extra cast. That is why instead of criticizing you should be paying close attention to my post

Wouldnt the added 15% reduced cast time for the next poh more and less make up for the penalty of casting a shield while having to move anyway, though? Starting to think the differences there are so minuscule time-wise that latency will bite you in the ass with these calculations, hard.

[Deleted]

I already covered a fair bit of the things you mention here in the disc reforging thread, so I'll just add a few things.

If you don't have the right amount of haste then you just shot yourself in the arse by casting an instant. You just wasted a spirit shell cast. Moving out of shit takes less time than a GCD and you might just make the extra cast. That is why instead of criticizing you should be paying close attention to my post, if you DONT have the haste then it might be better to not cast that instant as long as you only move a short distance. If you do have the haste, there is no need to bother thinking about that you just cast PWS automatically and you can be sure that you are losing no time.

Depends on the length of the movement, but if you want to maximize your absorbs like you keep talking about, throwing a PW:S over delaying a cast will be a hps gain regardless of the movement length (if you then miss a whole PoH for the SS, use more shields in this time). I never disputed that knowing if you can fit in an instant or not without losing a PoH during SS was helpful, I was saying that it isn't worth gearing for a specific cap since the amount of movement varies depending on the fight.

Again the point of my post is that you can gear for moving twice and get THE SAME RETURN OR BETTER as gearing for zero PWS casts. Even being able to move once dramatically reduces the time you will lose from moving twice, since having to move once is far more frequent than having to move twice. Even if you have to move twice at least you lose only one cast not two, which again is much better. Gearing for these breakpoints either eliminates or reduces your losses from moving. Since it is either slightly lower or higher healing overall what possible reason is there it not gear for it.

Already went over this in the other thread, but why not gear for 3 PW:S+PoHs? 4? Yes, because the increased mana cost. Your choice is arbitary and in most fights you can use SS without moving.

On your last point you made a critical error: 1:1 PWS-PoH outside spirit shell is higher HPS than PoH alone, but during spirit shell it is not because it will lead to you not getting completely overhealing PoH instead of a spirit shell PoH at the last cast.

Depends on your haste amount, or you could simply replace one PoH/PW:S for essentially the same effect as long as all the PoH's are fit into the SS window.

Adding the 2 PWS during spirit shell leads to effectively zero loss in HPS and it can actually be better HPS depending on your mastery.

Adding 3 PWS during spirit shell leads to effectively zero loss in HPS and it can actually be better HPS depending on your mastery.
Adding 4 PWS during spirit shell leads to effectively zero loss in HPS and it can actually be better HPS depending on your mastery.
etc. The reason why you don't is due to the large mana increase for a negligible hps increase.

So basically you are getting almost zero benefit from the extra 800 haste rating you need to get to the 7 casts breakpoint. All you are doing is shooting yourself in the arse by making your biggest CD extremely inflexible.

I'm personally not gearing for any SS caps but the gain is quite obvious, mana.

Being able to cast a PWS during spirit shell AT ZERO LOSS of spirit shell casts adds an immense amount of flexibility. You can move during spirit shell, you can PWS someone about to die without losing spirit shell casts, you build an absorb buffer on a group that you can't hit 5 targets in.

Yes, using other spells during SS than PoH adds flexibility but nullifies any particular haste caps, which is why I don't find them very valuable.

Basically getting that extra haste gives you more HPS overall since haste is a great stat and it is much much much better for spirit shell due to the added flexibility. Basically the 1 PWS haste break point is best for a non-haste build and the 2 PWS haste break point is best for a haste build. The extra cast breakpoint is only best for a full crit build, but anyone going for a full crit build doesn't really care about optimising their healing.

Haste is a shitty stat. Yes it's obviously better for SS/PoH than most of our other spells since PoH is a filler and hastes biggest weakness is the fact that it does jack/very little for our cds.

Regarding crit I suppose this should've been talked about in the other thread, but crits additional healing being only DA is definitely an argument for it being stronger than a formula balancing the two stats indicates (even excluding the dps gain). At some point mastery obviously becomes stronger if you keep stacking crit, but how you should balance them is hardly that obvious. Regarding overhealing a higher crit amount will generally let your DA shields be refreshed and last longer, it means that you dpsing in low damage situations has some additional benefit (more da shields up for higher damage periods) and in most (not all) scenarios where they drop without doing anything mastery wouldn't have been that helpful either.

[Deleted]

Depends on the length of the movement, but if you want to maximize your absorbs like you keep talking about, throwing a PW:S over delaying a cast will be a hps gain regardless of the movement length (if you then miss a whole PoH for the SS, use more shields in this time). I never disputed that knowing if you can fit in an instant or not without losing a PoH during SS was helpful, I was saying that it isn't worth gearing for a specific cap since the amount of movement varies depending on the fight.

Already went over this in the other thread, but why not gear for 3 PW:S+PoHs? 4? Yes, because the increased mana cost. Your choice is arbitary and in most fights you can use SS without moving.

Depends on your haste amount, or you could simply replace one PoH/PW:S for essentially the same effect as long as all the PoH's are fit into the SS window.

Adding 3 PWS during spirit shell leads to effectively zero loss in HPS and it can actually be better HPS depending on your mastery.
Adding 4 PWS during spirit shell leads to effectively zero loss in HPS and it can actually be better HPS depending on your mastery.
etc. The reason why you don't is due to the large mana increase for a negligible hps increase.

I'm personally not gearing for any SS caps but the gain is quite obvious, mana.

Yes, using other spells during SS than PoH adds flexibility but nullifies any particular haste caps, which is why I don't find them very valuable.

Haste is a shitty stat. Yes it's obviously better for SS/PoH than most of our other spells since PoH is a filler and hastes biggest weakness is the fact that it does jack/very little for our cds.

Regarding crit I suppose this should've been talked about in the other thread, but crits additional healing being only DA is definitely an argument for it being stronger than a formula balancing the two stats indicates (even excluding the dps gain). At some point mastery obviously becomes stronger if you keep stacking crit, but how you should balance them is hardly that obvious. Regarding overhealing a higher crit amount will generally let your DA shields be refreshed and last longer, it means that you dpsing in low damage situations has some additional benefit (more da shields up for higher damage periods) and in most (not all) scenarios where they drop without doing anything mastery wouldn't have been that helpful either.

Casting a PWS and losing a spirit shell cast instead of moving and getting an extra spirit shell cast is a loss of HPS, because when you are setting up spirit shell health bars are mostly full. Thus any lost spirit shell casts cannot be replaced by PWS casts or PoH casts AFTER spirit shell. Try and model it as much as you want you will see that just moving without casting leads to a gain in HPS over casting a PWS if you don't have haste above a certain level. Using any other instant is very frequently an HPS loss due to overheal.

There is no 3 and 4 PWS breakpoints. The only breakpoints are 1 PWS, 2 PWS, extra cast. You can potentially squeeze more PWS casts into the sequence, but that is counter productive. You aren't gaining anything in terms of flexibility and you are no longer able to maintain spirit shell on more than two groups. You really don't want to have less than 6 casts of spirit shell available.

Casting a PWS during spirit shell pays for itself due to increased rapture proc frequency. Even 2 PWS are a neglible mana loss considering you spirit shell at most once a minute.


Haste is a damn good stat. Just mana hungry.

Let us calculate the chance that aegis will be refreshed if someone is hit n times by a heal during the aegis duration

It is simple 1-(1-crit)^n

Lets add some values n = 2 crit = 25% --> 43.75% chance

n = 2 crit = 30% --> 51% chance

n = 2 crit = 35% --> 57.75% chance

n = 2 crit = 40% --> 64% chance

The higher the n value the faster crit loses value. A 60% increase in crit rating is a 46% increase in the refresh rate under conditions when you are PoH spamming 3 groups. Basically the return is negative, because you are shifting more healing to crits, which means more crits during those times when aegis can be wasted, but the refresh rate does not go up proportionally.

Now lets factor in the chance that someone will be hit by additional spells during spirit shell. We can getting an approximate value by assuming that getting hit a cast has a fixed probability P then the probability of getting hit by a crit is P*Crit. Thus the chance of refreshing aegis is

1-(1-P*crit)^n

for n = 2 crit = 25% P = 10% --> 4.94% chance
n = 2 crit = 30% P = 10% -->5.91% chance
n = 2 crit = 35% P = 10% --> 6.88% chance

I trust that you understand the problem. Basically the only thing that crit really does is reduce your overheal sensitivity, which is of course very good. Mastery increases healing and those two work best when combined together.

The burst nature of the encounters where most of the healing is concentrated in short bursts means that balanced mastery/crit is always better and that is pretty much inescapable. At the end of the day though because most of our crit comes from buffs and intellect, even if you gear for as much crit as you can you won't get far enough from the breakpoint to make the difference noticeable.

---------- Post added 2013-05-31 at 04:58 PM ----------

Wouldnt the added 15% reduced cast time for the next poh more and less make up for the penalty of casting a shield while having to move anyway, though? Starting to think the differences there are so minuscule time-wise that latency will bite you in the ass with these calculations, hard.

It doesn't. You need 5% haste on top of the latency threshold to do that and the latency threshold can be quite high. For me it is 3% haste. Basically if you have less than 1k haste rating at a latency of 40ms or so you will fail to get an extra cast if you cast a PWS when moving, but you actually can spare nearly 1s for moving and still get a cast off.

I tested this a lot and it works exactly once you add the latency threshold.

There is no question that having that small amount of extra haste is immensely beneficial and it is good to get enough haste to fit 1 PWS into spirit shell even if you don't really want haste (unless of course you are going for full crit). It costs neither mana, nor HPS but it is an immense gain in flexibility.

Even if one wants to go for extra haste there is no reason to hit the extra cast threshold, the two PWS threshold is what one needs to aim for.

[Deleted]

Casting a PWS and losing a spirit shell cast instead of moving and getting an extra spirit shell cast is a loss of HPS, because when you are setting up spirit shell health bars are mostly full. Thus any lost spirit shell casts cannot be replaced by PWS casts or PoH casts AFTER spirit shell. Try and model it as much as you want you will see that just moving without casting leads to a gain in HPS over casting a PWS if you don't have haste above a certain level. Using any other instant is very frequently an HPS loss due to overheal.

Obviously one PW:S won't replace one PoH SS cast, but two however pretty much will and takes about as much time, or even slightly less with BT used (if this was untrue your entire basis for 6 PoH+2 PW:S being as good as 7 PoH would be incorrect). One PoH SS cast can essentially be replaced by 2 PW:S casts if you need to move, at the cost of mana.

There is no 3 and 4 PWS breakpoints. The only breakpoints are 1 PWS, 2 PWS, extra cast. You can potentially squeeze more PWS casts into the sequence, but that is counter productive. You aren't gaining anything in terms of flexibility and you are no longer able to maintain spirit shell on more than two groups. You really don't want to have less than 6 casts of spirit shell available.

Yes you are gaining something, the ability to move even more during SS which is the entire premiss for your argument. Why is it worth so much to move one or two globals during SS but nothing to move three globals? If less than 6 SS is an issue depends on how far away the damage intake is, but the SS that falls off the earliest if you use 5 instead of 6 PoHs falls off about 3 seconds earlier, that'll rarely be a big deal unless you screwed up your SS timing entirely.

Casting a PWS during spirit shell pays for itself due to increased rapture proc frequency. Even 2 PWS are a neglible mana loss considering you spirit shell at most once a minute.

and for 3 instead of 2 PW:S you pay the additional cost of one PW:S once a minute, not a big deal either.

Haste is a damn good stat. Just mana hungry.

and since that forces you to spend extra stats into spirit to compensate it can hardly be considered such an amazing stat.

Let us calculate the chance that aegis will be refreshed if someone is hit n times by a heal during the aegis duration

It is simple 1-(1-crit)^n

Lets add some values n = 2 crit = 25% --> 43.75% chance

n = 2 crit = 30% --> 51% chance

n = 2 crit = 35% --> 57.75% chance

n = 2 crit = 40% --> 64% chance

The higher the n value the faster crit loses value. A 60% increase in crit rating is a 46% increase in the refresh rate under conditions when you are PoH spamming 3 groups. Basically the return is negative, because you are shifting more healing to crits, which means more crits during those times when aegis can be wasted, but the refresh rate does not go up proportionally.

The refresh rate goes up slightly slower but the shields being refreshed will also be larger with the high crit rate. Not really sure if this is significant enough to lower it's value that much.

Now lets factor in the chance that someone will be hit by additional spells during spirit shell. We can getting an approximate value by assuming that getting hit a cast has a fixed probability P then the probability of getting hit by a crit is P*Crit. Thus the chance of refreshing aegis is

1-(1-P*crit)^n

for n = 2 crit = 25% P = 10% --> 4.94% chance
n = 2 crit = 30% P = 10% -->5.91% chance
n = 2 crit = 35% P = 10% --> 6.88% chance

I trust that you understand the problem.

Not really following you here, I'm tired with a fever, feel free to clarify what you mean.

Basically the only thing that crit really does is reduce your overheal sensitivity, which is of course very good.Mastery increases healing and those two work best when combined together.

I'm well aware of both these things, that's kinda my point. I don't think that blindly stacking crit and ignoring mastery will yield the maximum hps, but in realistic situations I think more crit and less mastery than what a forumla showing the theoretical max indicates, exactly because crit has a lower chance of ending up with overhealing.

The burst nature of the encounters where most of the healing is concentrated in short bursts means that balanced mastery/crit is always better and that is pretty much inescapable.

That's kinda why I feel like crit has a slightly higher value, that means that a slightly bigger portion of the atonement spells I'm spamming when a very low amount of damage is going out will come in handy.

At the end of the day though because most of our crit comes from buffs and intellect, even if you gear for as much crit as you can you won't get far enough from the breakpoint to make the difference noticeable.

Considering that it's one of our premier output stats I fail to see how it isn't noticable.

[Deleted]

Obviously one PW:S won't replace one PoH SS cast, but two however pretty much will and takes about as much time (if this was untrue your entire basis for 6 PoH+2 PW:S being as good as 7 PoH would be incorrect). One PoH SS cast can essentially be replaced by 2 PW:S casts if you need to move, at the cost of mana.

Yes you are gaining something, the ability to move even more during SS which is the entire premiss for your argument....

and for 3 PW:S you pay the additional cost of one PW:S once a minute, not a big deal either.

and since that forces you to spend extra stats into spirit to compensate it can hardly be considered such an amazing stat.

The refresh rate goes up slightly slower but the shields being refreshed will also be significantly larger with the high crit rate.

I'm well aware of both these things, that's kinda my point. I don't think that blindly stacking crit and ignoring mastery will yield the maximum hps, but in realistic situations I think more crit and less mastery than what a forumla showing the theoretical max indicates, exactly because crit has a lower chance of ending up with overhealing.

Considering that it's one of our premier output stats I fail to see how it isn't noticable.

Nope, my premise is that 6 casts + 2 PWS is better than 6 casts and 1s of the spirit shell duration wasted. Not that you can replace 1 PoH with 2 PWS. That does not work. You need extra haste to do that.

You can add an infinite amount of PWSs and claim that it is just 1 more. 1 PWS is for partially offset by increased rapture. 2 PWS is one more, 3 PWS is 2 more. Every PWS you add makes the mana loss tougher and tougher to bear.

6 POH with 1% more healing from extra crit/mastery + 0.91 of a PoH from the PWS means that you have the equivalent of 6.97 PoH casts compared to 7 with the extra haste. That does not work out the same way when you use more than 2 PWS.

However 4 PWS and 4 PoHs with 1% more healing are 6.84 PoH casts compared with 7 casts.

Significantly larger shields also leads to more overcapping and wastage. Having more crit leads overall to bigger wasting of aegis. This is only partially offset by the higher refresh rate, in a senario where everyone gets hit by one of your heals at least once every 15s. In a situation where the chance of that happening is only a tiny proportion of your shields is refreshed, but the refresh is roughly proportional to crit.

Either the return from refreshing is negative or rather small. Basically crit hits the wall of negative returns pretty hard and the higher refresh rate does not really do anything. Most of the aegis you put during a low damage phase when atonement healing is on the tank.

Basically refresh offers no real advantage to crit.

If you have 10k crit rating that is 16% crit. You get [edit] 14-15% from int and the crit buff [/edit]. The benefit of your entire crit rating is 10-14% more HPS at most. The differences from pretty much any random combination of secondary stats is not that noticeable. It becomes more noticeable if a stat becomes ineffective though. The difference between one combination of mastery and crit and a different combination is even smaller. Actually the best point for me on 25man is between the high mastery low crit PWS breakpoint and the high crit low mastery balance point for heals. I still have more crit than mastery though.

[Deleted]

Nope, my premise is that 6 casts + 2 PWS is better than 6 casts and 1s of the spirit shell duration wasted. Not that you can replace 1 PoH with 2 PWS. That does not work. You need extra haste to do that.

Wasn't the haste required for 6 PoH+2 PW:S lower than 7 PoHs according to you? If so 2 PW:S with borrowed time used should essentially replace a PoH, assuming that they aren't more than the number of PoHs (borrowed time gets consumed).

You can add an infinite amount of PWSs and claim that it is just 1 more. 1 PWS is for partially offset by increased rapture. 2 PWS is one more, 3 PWS is 2 more. Every PWS you add makes the mana loss tougher and tougher to bear.

Exactly^^.

6 POH with 1% more healing from extra crit/mastery + 0.91 of a PoH from the PWS means that you have the equivalent of 6.97 PoH casts compared to 7 with the extra haste. That does not work out the same way when you use more than 2 PWS.

It does, it's just other haste points, just as arbitary but you didn't chose to look into them.

However 4 PWS and 4 PoHs with 1% more healing are 6.84 PoH casts compared with 7 casts.

I'll assume that you mean 4 PW:S and 5 PoH's? And it's more mobility, isn't that always better despite the mana cost?

Significantly larger shields also leads to more overcapping and wastage. Having more crit leads overall to bigger wasting of aegis. This is only partially offset by the higher refresh rate, in a senario where everyone gets hit by one of your heals at least once every 15s. In a situation where the chance of that happening is only a tiny proportion of your shields is refreshed, but the refresh is roughly proportional to crit.

If the shields are overcapping/expiring frequently there's so little damage going out that the mastery would be practically irrelevant as well.

Either the return from refreshing is negative or rather small. Basically crit hits the wall of negative returns pretty hard and the higher refresh rate does not really do anything. Most of the aegis you put during a low damage phase when atonement healing is on the tank.

Basically refresh offers no real advantage to crit.

I'm not saying that the gain is huge, I'm saying that crit offers more than mastery/haste in a situation with low damage, both due to the damage and the absorbs. Since mastery and crit are really close (unless your crit amount is way higher) that's worth taking into account.

If you have 10k crit rating that is 16% crit. You get 19-20% from int and the crit buff. The benefit of your entire crit rating is 10-13% at most. The differences from pretty much any random combination of secondary stats is not that noticeable. It becomes more noticeable if a stat becomes ineffective though. The difference between one combination of mastery and crit and a different combination is even smaller. Actually the best point for me on 25man is between the high mastery low crit PWS breakpoint and the high crit low mastery balance point for heals. I still have more crit than mastery though.

Yes the secondaries are relatively close, doesn't mean that the differences are irreleveant or there wouldn't be a point in discussing it.

[Nekked]

Nope, my premise is that 6 casts + 2 PWS is better than 6 casts and 1s of the spirit shell duration wasted. Not that you can replace 1 PoH with 2 PWS. That does not work. You need extra haste to do that.

You can add an infinite amount of PWSs and claim that it is just 1 more. 1 PWS is for partially offset by increased rapture. 2 PWS is one more, 3 PWS is 2 more. Every PWS you add makes the mana loss tougher and tougher to bear.

6 POH with 1% more healing from extra crit/mastery + 0.91 of a PoH from the PWS means that you have the equivalent of 6.97 PoH casts compared to 7 with the extra haste. That does not work out the same way when you use more than 2 PWS.

However 4 PWS and 4 PoHs with 1% more healing are 6.84 PoH casts compared with 7 casts.

Significantly larger shields also leads to more overcapping and wastage. Having more crit leads overall to bigger wasting of aegis. This is only partially offset by the higher refresh rate, in a senario where everyone gets hit by one of your heals at least once every 15s. In a situation where the chance of that happening is only a tiny proportion of your shields is refreshed, but the refresh is roughly proportional to crit.

Either the return from refreshing is negative or rather small. Basically crit hits the wall of negative returns pretty hard and the higher refresh rate does not really do anything. Most of the aegis you put during a low damage phase when atonement healing is on the tank.

Basically refresh offers no real advantage to crit.

If you have 10k crit rating that is 16% crit. You get [edit] 14-15% from int and the crit buff [/edit]. The benefit of your entire crit rating is 10-14% more HPS at most. The differences from pretty much any random combination of secondary stats is not that noticeable. It becomes more noticeable if a stat becomes ineffective though. The difference between one combination of mastery and crit and a different combination is even smaller. Actually the best point for me on 25man is between the high mastery low crit PWS breakpoint and the high crit low mastery balance point for heals. I still have more crit than mastery though.

That depends on how much raid damage is going out. We're healers not dps. Theories on max output is pointless.

[Deleted]

Wasn't the haste required for 6 PoH+2 PW:S lower than 7 PoHs according to you? If so 2 PW:S with borrowed time used should essentially replace a PoH, assuming that they aren't more than the number of PoHs (borrowed time gets consumed).

It does, it's just other haste points, just as arbitary but you didn't chose to look into them.

I'll assume that you mean 4 PW:S and 5 PoH's? And it's more mobility, isn't that always better despite the mana cost?

If the shields are overcapping/expiring frequently there's so little damage going out that the mastery would be practically irrelevant as well.


I'm not saying that the gain is huge, I'm saying that crit offers more than mastery/haste in a situation with low damage, both due to the damage and the absorbs. Since mastery and crit are really close (unless your crit amount is way higher) that's worth taking into account.

Yes the secondaries are relatively close, doesn't mean that the differences are irreleveant or there wouldn't be a point in discussing it.

5 PoHs with 4 PWS requires H = 1.0556. That is effectively 1:1 PWS:PoH, which means that you will have just 2s to absorb the first PWS cast. Now consider that PWS can crit and that you can also have spirit shell on the same group on 10man. It also increases the time between spirit shells on different groups to 3.8s and removes your ability to cast it on 3 groups. That sacrifices flexibility instead of gaining it, as makes timing spirit shell very awkward.

A small amount of mana is definitely worth a lot of flexibility, but a large amount of mana is not worth a small amount of flexibility and a lot more awkwardness

7 PoHs = 13k mana *7.

6 PoHs + 2 PWS = 13k mana * 7 + 13k mana - extra rapture

5 PoHs + 4 PWS = 13k mana *7 + 39k mana - extra rapture

8k mana a minute is worth the added flexibility of 2 PWS casts

34k mana a minute on the other hand is actually pretty tough to swallow, considering you get less flexibility.

Basically you quadruple the mana sacrificed to make your CD less flexible and harder to use.

As I said you never want to sacrifice the 6th spirit shell cast. It just messes up timing and wastes mana for no added benefit. If you need to move 3 and 4 times you have enough haste to do so and get PWS off even if you are at the 6+2 breakpoint.

There is nothing arbitrary at all about the 6+2 breakpoint and the +1 and +2 breakpoints are always the only ones worth considering. There only question is do I want enough haste to just neutralize latency on spirit shell or do I want enough haste to get to 6+2 breakpoint. It is not worth considering the 7 casts breakpoint in my view, especially on 10man.

===================================================================================

The low damage phases are a smaller part of your healing than the high damage phases and the added crit does not really help that much considering that the vast majority of your absorbed aegis during low damage phases is from the tanks.

Also consider that absorbs from heals are given by the formula crit*(1+mastery)*(1+0.5*mastery). So mastery also increases the amount of aegis you put out. Lets see how this works out

The rate at which crit rating increases absorbs is (1+mastery)*(1+0.5*mastery). The rate at which mastery increases absorbs is crit*(1.5+mastery)*1.6

So at 25% crit and 28% mastery the rate at which crit increases absorbs is 1.28*1.14 = 1.4592. The rate at which mastery increases absorbs is 0.25*(1.5+0.28)*1.6 =0.712

Thus mastery increases absorbs at roughly half the rate at which crit increases absorbs, however you also forget that PWS is a significant part of your absorbs during low damage phases. The rate at which mastery increases PWS is (1+crit)*1.6, while the rate at which crit increases PWS is (1+mastery). Using the same values as above you get 1.25*1.6 = 2 for mastery and 1.28 for crit.


So if 30% of your healing during a low damage phase is PWS, 50% is aegis and 3% is your non overhealing heals the rate at which mastery increases your healing would be 2*0.3+0.712*0.5+0.5*0.03 = 0.971, while the rate at which crit increases your healing is 1.28*+1.4592*0.5 = 0.933888

Surprised? You should be. Mastery is nowhere near as bad as you think during low damage phases.

Lets say you cast 1.5 penance, 5 smites and 1.5 holy fires per PWS. Then the amount of raw healing done by atonement is about 4 times that of PWS. With 25% crit rate and 28% mastery that means aegis will be equal to 1.28 of a PWS, so the values I picked for my example are pretty reasonable. I cast twice as much PWS during low damage phases though.

Basically crit has little benefit for low damage phases over mastery because PWS is an important source of heals and both crit and mastery increase both aegis and pws, but with opposite effectiveness.

Balanced mastery crit is always better than full crit and full mastery no matter what. The balance point is dependent on your casting pattern and healing style, but typically it lies somewhere between the balance point for heals and for PWS.

[Deleted]

5 PoHs with 4 PWS requires H = 1.0556. That is effectively 1:1 PWS:PoH, which means that you will have just 2s to absorb the first PWS cast. Now consider that PWS can crit and that you can also have spirit shell on the same group on 10man. It also increases the time between spirit shells on different groups to 3.8s and removes your ability to cast it on 3 groups. That sacrifices flexibility instead of gaining it, as makes timing spirit shell very awkward.

A small amount of mana is definitely worth a lot of flexibility, but a large amount of mana is not worth a small amount of flexibility and a lot more awkwardness

7 PoHs = 13k mana *7.

6 PoHs + 2 PWS = 13k mana * 7 + 13k mana - extra rapture

5 PoHs + 4 PWS = 13k mana *7 + 39k mana - extra rapture

8k mana a minute is worth the added flexibility of 2 PWS casts

34k mana a minute on the other hand is actually pretty tough to swallow, considering you get less flexibility.

Basically you quadruple the mana sacrificed to make your CD less flexible and harder to use.

As I said you never want to sacrifice the 6th spirit shell cast. It just messes up timing and wastes mana for no added benefit. If you need to move 3 and 4 times you have enough haste to do so and get PWS off even if you are at the 6+2 breakpoint.

It's not harder to use or less flexible. Every PW:S past the first (rapture) adds the same amount of movement and the same amount of mana cost during SS, if they cost you a PoH or not just depends on whatever haste you are going for. If you don't need to move for two globals during the 6+2 "breakpoint" that additional haste has no special value (there's no actual advantage besides the normal hps gain from haste to fit in 2 PW:S inside the SS window instead of dumping one after). If you need to move more than two globals that additional haste has no special value. The only time it's worth more than standard haste is if you have to move exactly two globals during SS, doesn't happen that often for me at least. Fitting in one PW:S during SS does have some merit since it's essentially free, due to rapture, any PW:S beyond that has no additional/special value compared to the next (regardless of it's the second, third or fourth), this is what I was trying to illustrate.

Regarding the sixth spirit shell cast you've yet to explain why it has some special value, it's an additional SS'd PoH and has no more value than the fifth/seventh. There's no realistic chance of a SS expiring due to you not refreshing it if you didn't use it way too early. There's no special need for every grp to have the same amount of absorbs (or hell, you could even arrange that by doing a IF PoH on the grp that just gets one shell). I fail to see why you are so obsessed about it.

===================================================================================

The low damage phases are a smaller part of your healing than the high damage phases and the added crit does not really help that much considering that the vast majority of your absorbed aegis during low damage phases is from the tanks.

Also consider that absorbs from heals are given by the formula crit*(1+mastery)*(1+0.5*mastery). So mastery also increases the amount of aegis you put out. Lets see how this works out

The rate at which crit rating increases absorbs is (1+mastery)*(1+0.5*mastery). The rate at which mastery increases absorbs is crit*(1.5+mastery)*1.6

So at 25% crit and 28% mastery the rate at which crit increases absorbs is 1.28*1.14 = 1.4592. The rate at which mastery increases absorbs is 0.25*(1.5+0.28)*1.6 =0.712

Thus mastery increases absorbs at roughly half the rate at which crit increases absorbs, however you also forget that PWS is a significant part of your absorbs during low damage phases. The rate at which mastery increases PWS is (1+crit)*1.6, while the rate at which crit increases PWS is (1+mastery). Using the same values as above you get 1.25*1.6 = 2 for mastery and 1.28 for crit.

So if 30% of your healing during a low damage phase is PWS, 50% is aegis and 3% is your non overhealing heals the rate at which mastery increases your healing would be 2*0.3+0.712*0.5+0.5*0.03 = 0.971, while the rate at which crit increases your healing is 1.28*+1.4592*0.5 = 0.933888

Surprised? You should be. Mastery is nowhere near as bad as you think during low damage phases.

Lets say you cast 1.5 penance, 5 smites and 1.5 holy fires per PWS. Then the amount of raw healing done by atonement is about 4 times that of PWS. With 25% crit rate and 28% mastery that means aegis will be equal to 1.28 of a PWS, so the values I picked for my example are pretty reasonable. I cast twice as much PWS during low damage phases though.

Basically crit has little benefit for low damage phases over mastery because PWS is an important source of heals and both crit and mastery increase both aegis and pws, but with opposite effectiveness.

Balanced mastery crit is always better than full crit and full mastery no matter what. The balance point is dependent on your casting pattern and healing style, but typically it lies somewhere between the balance point for heals and for PWS.

I fail to see why you'd ever cast PW:S outside of rapture in a low damage situation, even with a lmg proc, assuming that you aren't looking to do some silly padding. The additional damage from atonement combined with the spot healing vs a shield that (outside of the tank, the rapture'd shield) might not even get consumed (far more likely for a PW:S on a non-tank to not get used up in a low damage situation than small DA shields, mostly on the tank). If you for some reason have 30% of your healing from PW:S in a low damage situation, sure, crit isn't far ahead (fairly certain that you mean 20 and not 3% from your actual heals in the calc btw). For me this isn't even close to the case, in a low damage situation I'd never cast PW:S outside of rapture (assuming that I'm not loading up for a high damage period) and if my mana is at or close to full I wouldn't even use it when rapture is up.

I'm not disputing that a balance of mastery and crit will be superior to purely stacking crit or mastery when it comes to healing output (I've sad this several times). I'm saying that the there's more things to factor in when you chose to balance them than the balance giving the most raw hps based on your PW:S/other heals usage. The added dps and less overhealing which, assuming that all of the DA gets consumed, is the case regardless of how you swing it unless you have litterarly zero overhealing from your actual heals does make crit slightly (or if you value the dps highly significantly) stronger than the raw hps numbers indicate. Regarding the overhealing it's really quite easy to determine this, crit/mastery are relatively equal in hps output, one achives it's full hps in the situation I suggest (all the aegis gets consumed) one doesn't (not all the healing boosted by master is used). For mastery to still as much healing as crit it'd simply have to be ahead in raw hps, which isn't the case.

[Deleted]

1.5 penance, 5 smites, 1.5 solace 1 PWS , 1.5*3+6.5*1.5+1.5 = 15.75 seconds. This is a FULL atonement cycle with PWS every 15s mate. It will lead to the kind of distribution I just posted aegis to PWS will be just about 1:1.

I didn't just make that up I looked at various low damage phases from various fights.

You are just thinking intuitively, but your intuition is wrong. High crit is worse for healing but just adds a tiny about of DPS more.

[Deleted]

1.5 penance, 5 smites, 1.5 solace 1 PWS , 1.5*3+6.5*1.5+1.5 = 15.75 seconds. This is a FULL atonement cycle with PWS every 15s mate. It will lead to the kind of distribution I just posted aegis to PWS will be just about 1:1.

First of all, despite not being a fan of haste I'm stuck at over 4k (playing a fair bit of shadow as well so I'm stuck with a few of those pieces, I guess most discs are closer to ~3k), I also doubt that anyone who hasn't been farming 25 man hcs for weeks can avoid all the haste. You'll generally also get slightly more than 1,5 penance/PW:S due to ToT which considering how strong penance is skews the numbers a bit further. This is also not even including that you can sacrifice PW:S usage entirely if your mana is at a good level. Meaning that your FULL cycle is incorrect for most players.

Secondly. The raw healing from from the rotation you mentioned (1.5 penance, 5 smites, 1.5 solace) is more than four times the raw healing from PW:S, especially if you don't assume a higher mastery than crit % (which I don't have, but if your rotation has a significantly larger portion of PW:S I guess that makes sense for you).

I didn't just make that up I looked at various low damage phases from various fights.

If I look at random logs in WoL those aren't the ratio's/amounts I find in most low damage phases, nor are they correct for my logs.

You are just thinking intuitively, but your intuition is wrong. High crit is worse for healing but just adds a tiny about of DPS more.

No, I'm thinking logically, there's a difference. You are purposly using math/numbers that confirm your statements. If the dps increase for shifting some (again, I'm not talking about going full crit/ignoring mastery entirely) mastery into crit is tiny I don't really know what I should call the healing difference, abyssmal might be a good word:P. The additional healing from crit will essentially never result in overhealing (if it does the damage situation is so low that the additional healing from mastery would as well), while mastery is prone to at least some overhealing in every situation but the most extreme.

To clarify, they both add roughly the same amount of healing (obviously varies on your spell usage and stats), for one 100% of it is effective healing (generalising, but if da is dropping without absorbing there's so little damage that your healing is irrelevant, meaning that crit wins out from the added damage) for the other a portion (yes a big part of mastery is PW:S/da/SS, but not all of it) is going to end up overhealing to some degree in almost every situation. So assuming that your stats/spell usage indicate that the stats add the same amount of healing (which they do, if you balance the stats like you have/suggest) crit is simply a slightly superior stat to mastery (dps+less overhealing, what's masterys perk?). If we go pure crit and no mastery, mastery adds significantly more healing than crit, resulting in it being a stronger stat. The truth/correct way to gear is, to me, somewhere in between (and arguably closer to the crit/mastery "balance" that you suggest).

I'm not saying that I know exactly where/how crit/mastery should be balanced according to this (there's way too many factors to consider, not just spell usage and overhealing, but spell usage/overhealing for different portions of the fight where healing is more or less important, etc), I'm saying that the balance should be somewhere between those two points.

---

That being said, I'm honestly more interested in your reasoning why the sixth PoH/second PW:S is so important during SS. I can understand your reasoning regarding crit/mastery (even if I think your numbers are a bit skewed, at least for me personally) and either way the difference between the amounts is incredibly slim (at my current stats that's shifting an extra ~2k mastery to crit). However your reasoning regarding SS just makes no sense to me, at all.

[Basmothh]

High crit is worse for healing but just adds a tiny about of DPS more.

This isn't true. Regardless of which stat you stack, the difference in total average output is trivial. The only difference is the ratio of your raw healing to DA, with high Crit builds obviously having the it skewed towards DA. Still, high crit builds tend to do better in a practical raid setting due to DA having lower overhealing.

[Deleted]

First of all, despite not being a fan of haste I'm stuck at over 4k....

Secondly. The raw healing from from the rotation you mentioned (1.5 penance, 5 smites, 1.5 solace) is more than four times the raw healing from PW:S, especially if you don't assume a higher mastery than crit % (which I don't have, but if your rotation has a significantly larger portion of PW:S I guess that makes sense for you).

You want to add an extra smite in there and make it 1.7 penance? No problem. Make penance 50k per tick and everything else average 50k. so 1.7*150 = 255, 7.5*50k = 375k, for 630k total. 25% of that is 157.5. No crit PWS at 100k with 25% crit it becomes 125k. so aegis to PWS of 1.26:1. I am assuming 30% PWS and 50% aegis which is 1.66:1 aegis to PWS and that is pretty high especially for a low damage phase. I expected an argument like this so I took a pretty high aegis to PWS ratio. Look at some logs and you will see that I am right. PWS is reasonably close to aegis in most cases and during low damage phases PWS usally comes on top due to almost zero overheal on PWS, but significant overheal on aegis. Some examples:

Picking some 10H top logs at random (so I didn't scan a lot of logs and cherry picked) I am picking high parses and ones that look english. These are probably 5.2 values, so pre-nerf.

Maegera: http://www.worldoflogs.com/reports/r...8&e=2396#Steni
1:1 Aegis PWS. Look at the low damage phases to see that PWS is the top heal. 25% odd crit rate 20% overheal on aegis, 4.5% on PWS 17% on atonement healing.

Same person on council 1:1 aegis:PWS 15.8% overheal on aegis.

Council: http://www.worldoflogs.com/reports/2...e=304#Amabella
1.74:1 Aegis PWS ~40% crit rate 27.5% overheal on aegis, 16% on PWS, 22% on atonement healing.

I dont see this person ranked in the first page on maegera

Another top log on maegera

http://www.worldoflogs.com/reports/r...240#Ka%C3%ADri
again 25% crit rate approximately. 20% PWS, 15% aegis (more PWS than aegis) 30% overheal on aegis. 2.4% on PWS

Top disc log on ji'kun
http://www.worldoflogs.com/reports/r...?s=5114&e=5456
Again roughly 25% crit rate, 20% aegis, 16% PWS 25% overheal on aegis 26% overheal on PWS Atonement overheal 22% average or so

Every log I see I rarely see a anything higher than the 1.66:1 aegis/PWS and on average aegis has higher overheal the higher your crit rate just as I predicted mathematically. More importantly aegis does not have anywhere near the low overheal rate ppl expect. PWS does, but aegis no. Because aegis is random even with extremely high crit rates, refresh for long periods of time is impossible.

Many of the top logs don't use excessive crit rates, which means they are running pretty much balanced crit/mastery.

Only on places like horridon with a massive damage buff do you see high aegis/PWS ratios and even then not universally. People who say high crit rates work better than balanced mastery/crit are the same like those who were claiming PoH double dipped on crit in cataclysm. It never did, people just couldn't do the calculations correctly.

Basically here is the problem: crit scales nicely with aegis, and moderately well for PWS, while mastery scales extremely well with PWS and about 40-70% what crit does on aegis. On top of that it helps your heals. The higher your crit rate the better the scaling for mastery on aegis and on PWS.

The system just does not behave like you think. The optimal point is probably between the PWS optimal and the heal optimal. Going above 30% crit rate is not necessarily a great idea.

That being said, I'm honestly more interested in your reasoning why the sixth PoH/second PW:S is so important during SS. I can understand your reasoning regarding crit/mastery (even if I think your numbers are a bit skewed, at least for me personally) and either way the difference between the amounts is incredibly slim (at my current stats that's shifting an extra ~2k mastery to crit). However your reasoning regarding SS just makes no sense to me, at all.

It is very important to be able to prolong your spirit shell buffer for as long as possible and the 6th cast is essential for that, the 2 PWS break point is not important at all. I find it very useful, someone else might not. What is extremely useful is the 1 PWS haste breakpoint, which actually only requires taking enough haste to counteract latency, since the 5% haste buff is enough to reach that point.

---------- Post added 2013-06-02 at 01:45 AM ----------

This isn't true. Regardless of which stat you stack, the difference in total average output is trivial. The only difference is the ratio of your raw healing to DA, with high Crit builds obviously having the it skewed towards DA. Still, high crit builds tend to do better in a practical raid setting due to DA having lower overhealing.

Actually it is true. The difference is 3-5%, which is not trivial but it is small. Both mastery AND crit healing decrease the raw healing to DA ratio, because mastery incerases DA much more than it increases heals. but we don't really care much about that. What matters is your PWS to DA ratio. Mastery has exceptionally high scaling for PWS that just gets bigger and bigger with crit. So mastery increases your PWS absorbs massively and your DA absorbs about 60% as much for a high crit rate build, while crit increases both aegis and PWS absorbs moderately at low mastery. Because mastery has such a high scaling for PWS it sneaks quickly ahead of crit in the battle of absorbs. On top of that you are going to get the mastery benefit for many of your regular heals.

If it was just heals yes you would see a slightly better result from crit, but because you also have PWS and spirit shell in the equation, mastery gets ahead very quickly as soon as your crit rate goes above 25%.

[Deleted]

....Every log I see I rarely see a anything higher than the 1.66:1 aegis/PWS

I'm finding that roughly half of the logs I look into (just randomly clicked on some in top 20) have more than a 1.66 aegis/pws total ratio:
http://www.worldoflogs.com/reports/d...?s=4749&e=5009 (low damage period close to 57% DA, 7% PW:S)
http://www.worldoflogs.com/reports/w...?s=3026&e=3400 (close to 38% DA, 5% PW:S)
http://www.worldoflogs.com/reports/s...?s=5751&e=6138 (close to 49% DA, 17% PW:S)
http://www.worldoflogs.com/reports/r...?s=8410&e=8974 (close to 28% DA, 10% PW:S)
http://www.worldoflogs.com/reports/2...s/4/?s=0&e=304 (22% DA, 20% PW:S) high ratio of PW:S!
http://www.worldoflogs.com/reports/0...?s=6827&e=7179 (14% DA, 0% PW:S)

and on average aegis has higher overheal the higher your crit rate just as I predicted mathematically. More importantly aegis does not have anywhere near the low overheal rate ppl expect. PWS does, but aegis no. Because aegis is random even with extremely high crit rates, refresh for long periods of time is impossible.

If da ends up overhealing additional mastery would be essentially useless as well, and crit wins due to the added damage. PW:S does have a lower overheal ratio than da, but da is way ahead of our normal heals on most fights (especially non-smart/big ones). It's not about refreshing them infinitely, it's about any da applied having a chance to be useful for (at least) 15 seconds if damage is coming in (and if it doesn't, well no harm), while the non da/pw:s part of mastery will be wasted a lot of the time.

Many of the top logs don't use excessive crit rates, which means they are running pretty much balanced crit/mastery.

People who say high crit rates work better than balanced mastery/crit are the same like those who were claiming PoH double dipped on crit in cataclysm. It never did, people just couldn't do the calculations correctly.

No clue about cata, never played during that. There's a difference in a high/excessive crit ratio and balancing mastery/crit at a slightly different level. I've never advocated purely stacking crit and disregarding mastery entirely, just that if they are entirely "balanced" looking at the raw healing crit has a slightly higher value.

Only on places like horridon with a massive damage buff do you see high aegis/PWS ratios and even then not universally.

Worth noting that there are several fights with damage buffs in this tier (jin'rokh, horridon, tortos, which admittedly does favor mastery, and ji-kun), so it can't be disregarded entirely.

Basically here is the problem: crit scales nicely with aegis, and moderately well for PWS, while mastery scales extremely well with PWS and about 40-70% what crit does on aegis. On top of that it helps your heals. The higher your crit rate the better the scaling for mastery on aegis and on PWS.

and vice-versa. I'm well aware that a large (but far from all) of masterys benefit comes from da/pw:s, or I would actually be advocating that we stack full crit.

The system just does not behave like you think. The optimal point is probably between the PWS optimal and the heal optimal. Going above 30% crit rate is not necessarily a great idea.

I'm fairly certain that it behaves exactly like I think, I'm simply looking at different numbers/benefits than you. Yes, the optimal point is probably somewhere in between the pw:s optimal and the heal optimal (although closer to the heal optimal). Crit doesn't really lose a significant value going higher, it's just that mastery gains, which yes makes stacking crit blindly a bad idea.

It is very important to be able to prolong your spirit shell buffer for as long as possible and the 6th cast is essential for that

Are you refering to the actual length the shields applied by SS lasts? If so, the sixth SS adds less than 3 seconds until the SS applications starts dropping (admittedly not from the same grp as it would with 6). You essentially gain one PoH cast in time to wait for the damage to start going out before your shells starts to drop (and in 5's case the shell dropping at that point would be a small one, which gets consumed quicker), it's really not that significant.

the 2 PWS break point is not important at all.

Good, then I'm on board.

What is extremely useful is the 1 PWS haste breakpoint, which actually only requires taking enough haste to counteract latency, since the 5% haste buff is enough to reach that point.

Yep, I agree that having enough haste to fit in one PW:S during your SS applications is somewhat handy (your focus seemed to be on 2 PW:S being so important, which I found strange). Considering that you generally want to/can throw a PW:S right before applying the SS (meaning that rapture usually is on cd during most of it) I don't really think having enough haste to fit another one in is 'that' important, but yes it has some value (especially since it's well, hard to avoid).

[Basmothh]

Actually it is true. The difference is 3-5%, which is not trivial but it is small. Both mastery AND crit healing decrease the raw healing to DA ratio, because mastery incerases DA much more than it increases heals. but we don't really care much about that. What matters is your PWS to DA ratio. Mastery has exceptionally high scaling for PWS that just gets bigger and bigger with crit. So mastery increases your PWS absorbs massively and your DA absorbs about 60% as much for a high crit rate build, while crit increases both aegis and PWS absorbs moderately at low mastery. Because mastery has such a high scaling for PWS it sneaks quickly ahead of crit in the battle of absorbs. On top of that you are going to get the mastery benefit for many of your regular heals.

If it was just heals yes you would see a slightly better result from crit, but because you also have PWS and spirit shell in the equation, mastery gets ahead very quickly as soon as your crit rate goes above 25%.

False. Basing it on http://us.battle.net/wow/en/characte...Siory/advanced and adjusting the stats to account for heavy Crit (37.03% crit/13.68% heal mastery/27.35% shield mastery) and heavy Mastery (29.62% crit/19.6% heal mastery/39.2% shield mastery), you get these values for:

SS
Crit stacking - [1 * 1.1368] + [0.3703 * 1.1368 * 1.2735] = 1.67
Mastery stacking - [1 * 1.196] + [0.2962 * 1.196 * 1.3962] = 1.69

Mastery results in ~1.2% higher SS values on average.

PW:S
Crit stacking - [0.6297 * 1.2735] + [0.3703 * 2 * 1.2735] = 1.75
Mastery stacking - [0.7038 * 1.3962] + [0.2962 * 2 * 1.3962] = 1.81

Mastery results in ~3.5% higher PW:S values on average.

DA:
Crit stacking - [0.3703 * 1.1368 * 1.2735] = 0.54
Mastery stacking - [0.2962 * 1.196 * 1.3962] = 0.49

Crit results in ~10% higher overall DA values on average.

Heals:
Crit stacking - 1.1368
Mastery stacking - 1.196

Mastery results in ~5.2% more healing on average.

The difference between stacking both stats is rather negligible, especially when it comes to SS and PW:S. The main difference is whether you want higher raw healing (Mastery) or DAs (Crit). Ultimately, the overall difference in total output is pretty trivial, and Crit may actually come out ahead once overhealing is factored in. You could do this for various values of Crit/Mastery stacked builds, and you'd still reach the same conclusion.