good question, I as going to ask about crit/haste balance to.
I'm having real trouble working out, with the new 'stacking' DA what the new crit target is raid buffed
well, from my experiences on the PTR, you don't have the regen or the time to it there overhealing the tank until you have a 9k DA on the tank. you'll run yourself out of mana real fast while raid members are getting killed (tons of raid damage going out). the other thing is that my PoM's were really traveling, and critting a lot. it was nice that there were so many DA's being fully used on the raid (this is why i still like crit). i'm torn between whether to keep stacking crit or if i should get more haste. I don't think we should stack more crit solely because of the stacking DAs though.
the only thing i really see stacking DA being useful for is double or triple penance crits.
Since crit increases throughput for the same mana cost, where increased throughput from haste obviously costs more, I've been of the mind for a while now that I'll aim for even higher crit/haste ratio when 3.1 comes, compared to now.
One could argue, however, that the same increased throughput could be obtained from haste by using the (in 3.1, only slightly) more mana effective Gheal instead of Flash heal...
I think I'll stick with penance + fh, and thus crit > haste, though ;-)
As what I've seen from 3.1 PTR there is so much crit clothies in the loot, some few parts that actually have been haste.
Aswell another class needing our spirit cloth will be mages, the only healing that matters imo is the boss healing having haste instead of crit in trash just to top meters ain't the goal.
My goal in PTR will probably be to just stack crit and have really low haste and rely on haste totem + enlightenment, I think those 11% will be ok and just go for crit, probably can reach 40% would be awesome to make hugh nice crits on the tank instead of having to spam 2 non crit flash heals I can pop a big GH instead with stopcasting macro.
Probably the 4 set bonus + imp healing will be kinda nice to my spec I guess. Haven't tried so much yet.
Not sure I agree with any of the "crit stacking" mentality. For the point of discussion I would suggest that clarity to what stat amounts you will be shooting for, why, how, ..etc. Including assumptions and expectations.
I will continue to maintain a raid buffed haste of ~25% not including Borrowed Time. From that level I will improve my crit as possible, currently I run ~30% raid buffed. Going beyond 25% raid buffed haste is not beneficial unless you use a significant amount of PoH or Greater Heal. Though I see no reason that having less than 25% is better. In fact, having less haste (only talents + BT) reduces the value of Crit.
In scenarios where overhealing is significant crit can be exceptionally valuable. Though I don't foresee the levels of current overhealing persisting into Ulduar. Most recently on PTR Iron Council "Hard Mode" I was running <18% overhealing.
0% Overheal
DA value to healing = 10.5% (30% Crit)
DA value to healing = 13.0% (40% Crit)
DA value to healing = 23.1% (100% Crit)
20% Overheal
DA value to healing = 12.7% (30% Crit)
DA value to healing = 15.8% (40% Crit)
DA value to healing = 27.3% (100% Crit)
40% Overheal
DA value to healing = 16.4% (30% Crit)
DA value to healing = 20.0% (40% Crit)
DA value to healing = 33.3% (100% Crit)
60% Overheal
DA value to healing = 22.7% (30% Crit)
DA value to healing = 27.3% (40% Crit)
DA value to healing = 42.9% (100% Crit)
80% Overheal
DA value to healing = 37.0% (30% Crit)
DA value to healing = 42.9% (40% Crit)
DA value to healing = 60.0% (100% Crit)
So if you run a reasonable amount of overheal 40% or less. DA is still not a the major component of total healing. It is valuable but not significant enough to give cause for stacking crit. It takes a significant amount of Crit >40% AND an extreme amount of overheal to cause crit stacking for DA to be worthwhile.
Meanwhile if you are running low on haste.. You negatively impact the channel speed of Penance and the GCD following a PW:S, both of which play a big role in our toolbox. Which seems like a extremely high cost for improving the DA presence in effective healing by .21-.36% per 1% of Crit.
Balance is the only answer, with slight favoritism for whichever stat you are light on and preference for Crit when you reach the Haste soft-cap(based on spell composition). Do what you like with your gearing but it doesn't make a ton of sense.
Added the 100% crit case to show that even at the maximum case for Crit, DA isn't the majority of healing until you cross 80% overhealing.
Last edited by TheDoctor : 03/23/09 at 1:25 PM.
Reason: Added 100% crit case...
Comparing Overheal to DA value doesn't show anything useful for gear selection, other than that you cannot heal using solely 3 points in DA -- something nobody was doing. You should be looking at Crit Rating versus Haste rating.
In a simplified math,
1 Haste rating provides 0.0305% of output (1/32.79)
1 Crit rating provides 0.0206% of output ((1.5*1.3-1)/45.91)
You have to factor in things like: Haste does not affect spell cooldowns. Crit doesn't work on every spell. Haste indirectly increases mana cost. Crit indirectly increases overheal.
Ultimately, haste is the output winner, and can make heals between powerful hits that you couldn't otherwise make.
However, critical builds can offer better protection against tanks getting 1 shotted, and has better longevity.
Comparing Overheal to DA value doesn't show anything useful for gear selection, other than that you cannot heal using solely 3 points in DA -- something nobody was doing. You should be looking at Crit Rating versus Haste rating.
In a simplified math,
1 Haste rating provides 0.0305% of output (1/32.79)
1 Crit rating provides 0.0206% of output ((1.5*1.3-1)/45.91)
You have to factor in things like: Haste does not affect spell cooldowns. Crit doesn't work on every spell. Haste indirectly increases mana cost. Crit indirectly increases overheal.
Ultimately, haste is the output winner, and can make heals between powerful hits that you couldn't otherwise make.
However, critical builds can offer better protection against tanks getting 1 shotted, and has better longevity.
My evaluation has value when Crit is being overvalued because DA stacks... DA isn't a significant part of your healing and hence stacking Crit isn't overly valuable. The outline was simply to show the error in looking at Crit because of DA's value. DA and hence Crit do become "more valuable" at higher levels of overheal because they are not part of the "overhealing equation". Arguing on indirect benefits/detriments isn't part of what I was doing either.
Some corrections on your "factors":
- Haste increases the rate at which you are capable of spending mana, not the mana cost.
- Crit increases the probability that partial overhealing will occur.
There are a ton more that can be manufactured as reasons for individual playstyle for either haste or crit.
I can't say that I agree with your simplified math for output. Don't have my values handy but I run a full breakdown of throughput impact per point per spell on both HPS and MPS. Your math assumes that both have an equal impact on throughput per point and only goes to show the stat weight variability between the two stats.
do any of you have any good methods for analyzing WWS to try to see how much mitigation is playing a role? the first thing i do is look at the healing our disc priest gets from Glyph of Power Word: Shield and divide it by 0.20. That should give the total "effective" mitigation that PW:S is providing. (I actually adjust the amount for crit so it's more accurate).
Any way to figure out Divine Aegis? I can look at the total healing from crits, and figure out the total shield that was applied. No real way to figure out how much was actually used, though, right? Anyone have a decent number that you multiply that by, say 30%? Or is it higher/lower?
So far, we've been using an undergeared disc priest and I've liked what I see. I still want to look into it further, though.
Roknroll; no, it's not currently possible to see the amount of absorption a disc priest adds to a raid, simply because the combatlog does not keep track of why X damage is absorbed. Hopefully it'll be changed, but till it is, there's no way you can perform accurate evaluation on disc priests performance from recount, wws, wmo etc. alone.
There's been made a modified version of Recount, but it only works for the disc priest, I think. If you want, you can ask him to install that and send his heal reports to you. But it's very flawed as well, as it makes it's effective healing/absorption calculations on assumptions, much like you've been trying - i.e., it won't *know* if a 3k DA or a PW:Shield has been absorbed by damage or not - it will just assume that it has, meaning you'll see somewhat exaggerated numbers from it. If he's (almost) purely maintank healing though, this error margin is lowered, as tanks tend to make good use of our shields
When I compare my modified recount to wws or a recount report from another player, my healing output is roughly 30-40% higher on my recount, with pw:shield being 70-80% of the difference, and DA the rest. Depending on how much shield-tossing and raid crosshealing I've been doing, I'll estimate the real effective heal output difference to be closer to +20%, though. But again, there's no way of knowing for sure.
Oh, and I think you misunderstood [Glyph of Power Word: Shield]. The heal from the glyph, when we cast a PW:S, lands immediately - not after the shield has been absorbed. This means that on pre-emptive shielding, such as a mage grabbing aggro for example, the heal portion might be 100% overheal, and the shield is likely to not be absorbed at all. The amount healed is thus calculated when we apply the shield, based on our current spellpower, and doesn't give any indication of the effective absorption of the shield.
Best way to evaluate a disc priest would probably be to stick him with some healing assignments where other healers wont be able to overlap. Sarth3.10, solo-healing Patchwerk 10, or something similar for example. It's very hard to get a feel of his performance by numbers alone. But I guess that goes for all healers, right?
If you "feel" he's a good player though, and he seems to keep the weakened soul debuff up on tanks, uses pom, penance and flash or greater heal, you can probably add 20-30% to his effective healing from wws, and that should at least give you a value that's closer to his real healing contribution. But it's still not accurate, sadly.
Last edited by Alv!ra : 03/24/09 at 9:15 PM.
Reason: item linking is hard
Roknroll; no, it's not currently possible to see the amount of absorption a disc priest adds to a raid, simply because the combatlog does not keep track of why X damage is absorbed.
This seems to have recently stopped being true. Im how it is trackign it as I thought it was also not possible, but I looked at our WMO log from the other night after the raid, and as if by magic, the apsorption numbers were counted towards my healing! Here's a link:
This seems to have recently stopped being true. Im how it is trackign it as I thought it was also not possible, but I looked at our WMO log from the other night after the raid, and as if by magic, the apsorption numbers were counted towards my healing! Here's a link:
It might just be counting all my shields as 100% since the overhealing for them is 0.... hard to say without a test though.
Ok well I definitely question this reporting. It appears fairly similar to what the pseudo-fix in recount does...
Some definite issues is that it shows 100% effective use of shields. It also shows that that for PW:S you had Max=8744, Min=42, and Avg=2092. It is possible that it is doing slightly better tracking via extrapolating the buff applied and absorbed amount following and the buff removed. Though this can be inaccurate due to combat log timing issues. Since no Disc shield would ever be an average of 2092, something is definitely up here.
For DA is is equally screwed up with showing Max=15447, Min=48, and Avg=2300.
Actually, I think it may be doing a slightly better job than Recount as I look at it. It is probably trying to base the min/max/avg based on the amount of absorbed damage between the buff applied and removed. Which means it probably shows the effective absorbed amount and just can't figure out how much excess you actually applied over the course of the fight. The issue would be how does it factor out amounts that show up as absorbed that don't source to a disc priest. It also won't be accurate when tracking during both a DA & PW:S being active.
Further how does it go about know that a given priest has the DA talent? If I remember correctly it shows that X gains Divine Aegis but doesn't reflect who the buff came from. Because apparently it doesn't do a good enough job, since it shows your Holy Priest as having DA healing as well.
Ok well I definitely question this reporting. It appears fairly similar to what the pseudo-fix in recount does...
Some definite issues is that it shows 100% effective use of shields. It also shows that that for PW:S you had Max=8744, Min=42, and Avg=2092. It is possible that it is doing slightly better tracking via extrapolating the buff applied and absorbed amount following and the buff removed. Though this can be inaccurate due to combat log timing issues. Since no Disc shield would ever be an average of 2092, something is definitely up here.
For DA is is equally screwed up with showing Max=15447, Min=48, and Avg=2300.
Actually, I think it may be doing a slightly better job than Recount as I look at it. It is probably trying to base the min/max/avg based on the amount of absorbed damage between the buff applied and removed. Which means it probably shows the effective absorbed amount and just can't figure out how much excess you actually applied over the course of the fight. The issue would be how does it factor out amounts that show up as absorbed that don't source to a disc priest. It also won't be accurate when tracking during both a DA & PW:S being active.
Further how does it go about know that a given priest has the DA talent? If I remember correctly it shows that X gains Divine Aegis but doesn't reflect who the buff came from. Because apparently it doesn't do a good enough job, since it shows your Holy Priest as having DA healing as well.
I was noticing all of these things too. What I suspect is happening, is that it is somehow counting the absorbs on a hit to hit basis. This would explain the 100% efficiency, as well as the large variation in absorb amounts.
Also, it says you procced DA 33 times, but your flash heal crit 45 times, and penance 15, so the DA procs should be at least 60, and not just 33. (your total amount of crits was 80)
The PW:S doesn't seem to be calculated from the glyph heals at least.
I found this topic on WMO's forum, where a disc priest asks for the crew to make wmo include DA absorbs too, and later reports the error with high DA shields. It seems like it's something they're currently working on developing. Would be nice if they can come up with a decent solution, perhaps we can help 'em :-)
Until 3.1 hits you can get a pretty good idea of effective healing + absorption from the Rapture returns. Take this WWS from my guild's recent naxx/sarth clear for example
It shows me as having done about 4.36 M effective healing over all the boss fights. My Rapture returns over those fights was 364446. Using the equation from the first post of this thread,
That means I did (364446*11460)/(0.025*25000) = 6.68 million effective healing+absoption from the Rapture spells. This is not exact however, because any portion of a heal that was effective past 11460 healing is not counted (so a Greater Heal crit that was mostly effective), so 6.68 million is an underestimation.
That doesn't include heals other than Penance, Gheal and Flash Heal, so those have to be added manually.
PoM did 1,009,974 healing and was 76% effective, so it did 0.75*1009974 = 757481
PoH did 603558 at 50% effectiveness, so 0.5*603558 = 301779
Renew did 284253 at 91% effectiveness, so 0.91*284253 = 258670
Glyph of PWS did 190170 at 38% effectiveness, so 0.38*190170 = 72265
Binding Heal did 51087 at 43% effectiveness, so 0.43*51087 = 21967
Glyph PoH did 49156 at 88% effectiveness, so 0.88*49156 = 43257
Divine Hymn did 28452 at 95% effectiveness, so 0.95*28452 = 27029
Add up those numbers and the number derived from the Rapture returns and you get 8.16 million effective healing+absorption, rather than the 4.36 million healing WWS shows, moving me from #5 to #1 by far. That means 46.6% of my healing was done through shields, which sounds a little high but I don't believe I've made any math errors. The 8.16 million is actually a low-end estimate too, because any Gheal crit that has more than 11460 effective healing isn't being counted fully.
Also, it says you procced DA 33 times, but your flash heal crit 45 times, and penance 15, so the DA procs should be at least 60, and not just 33. (your total amount of crits was 80)
If you look at the buffs though, there are 41 DA procs on Alocer alone. I had never noticed absorbs showing up until this weeks parse, so I guess there are still some bugs being worked out ><
The situation is probably thus for the tracking...
Similar to how a buff on WWS doesn't show as being applied again when it is refreshed. Their implementation is probably similar for shields. Meaning that lets say you proc DA on the target and before it fades you proc again... It probably continues to count all the "absorbs" and applies them to one DA occurrence. This creates inflated DA maximums. Same situation for PW:S and DA interactions. Because PW:S doesn't show over a ~8k value it would seem that it may have a coded limit and everything else goes to the DA line.
1.if Glyph of Power Word: Shield heal someone for 1000,there must be an shield could absorb 5000 damage on this guy.right?
2.for DA, on wowhead: Critical heals have a chance to create a protective shield on the target.only have a chance.why you guys say the crits always cause a DA?
and the number of DA is parsed from combatlog and I think it could not be wrong. The fact is that not all crits could proc a DA right now.anything i missed?
and the number of DA or PWS display on WMO is not the proc times of it but the times it works to absorb damage.
3.the shield heal amount over too much(8k?15k?) is a bug and has been fixed.
4.Shield would be always effheal and no overheal at all and equals the number it absorbed.It's the limitation of combatlog.sorry about that!
any more question please let me know!sorry about my engRish!
Last edited by wowmeteronline : 03/25/09 at 1:25 AM.
1) Yes, that's supposedly how [Glyph of Power Word: Shield] works. Remember the heal from the glyph can crit though, but the shield cannot. So if you see a crit glyph heal, the shield is not x5 that amount. Perhaps someone wiser on our crit mechanics can work out how much it'll be instead?
Also, there may be some priests who don't use the glyph, of course, but I'd think most do.
2) I'm afraid you've been looking at an outdated spell on wowhead. This seems to be the updated one, where you can see it is not refering to a proc-chance, but rather just says "critical heals creates a DA..."
Hm, it gets a bit tricky here. Apparently, if the heal effect from PW:Shield glyph crits, it will not proc a DA. But as far as I know, all our other healing spells do.
What it seems (and sounds like, from your description), what WMO is doing is actually track the absorptions made by DA and PW:Shield. That's, in my opinion, a lot better than showing us how many shields we put up. The interesting number is how much actual damage absorption we bring to the raid - not how much we can toss around unused. So please don't remove that. It's really a lot more valuable to know the effective absorption taken, than the total shielding done.
PS: I hope our kind moderators will bear with our asian friend here... It would really be nice if WMO can display our absorption more accurately.
Long shot here, but has anyone seen or know of a mod that can track approximate DA shield values? Who the shield is sitting on, and for what approximate value it would absorb for.
Probably not _necessary_ to have, but a nice to know.
Last edited by neld : 03/25/09 at 2:08 AM.
Reason: clarity
Long shot here, but has anyone seen or know of a mod that can track approximate DA shield values? Who the shield is sitting on, and for what approximate value it would absorb for.
Probably not _necessary_ to have, but a nice to know.
Grid can quite easily track who has a divine Aegis on them. I suggest using that to track who its on.
For approximate value, I can't think of any use for such a thing, as 3.1 will totally remove any purpose of it. I suppose you could program one to assume that all hits remove DA totally, and just count DA amounts added until the person next gets hit, but such an addon would be doomed to high levels of inaccuracy.
Long shot here, but has anyone seen or know of a mod that can track approximate DA shield values? Who the shield is sitting on, and for what approximate value it would absorb for.
Probably not _necessary_ to have, but a nice to know.
GridStatusShield (WoWAce) - Does that as a grid addon (I'm happy for any suggestions, bug-reports
ShieldMonitor (WoWInterface) - Show shields only on yourself. Hadn't been updated since a while but works.
1.if Glyph of Power Word: Shield heal someone for 1000,there must be an shield could absorb 5000 damage on this guy.right?
Roughly, yes. The exact factor is not precisely one, due to several misleading talents and weird spell computing.
(I recall that Twin disciplines is applied twice (once for the shield, and a second time for the glyph heal. And they are several other strange points like this, you can find them in this thread).
And remember that the glyph can crit, but PW:S can't.
Note also that the glyph heal occurs when the spell is used, and will be overheal is the target is already full (preemptive shielding).
2.for DA, on wowhead: Critical heals have a chance to create a protective shield on the target.only have a chance.why you guys say the crits always cause a DA?
and the number of DA is parsed from combatlog and I think it could not be wrong. The fact is that not all crits could proc a DA right now.anything i missed?
On live, all crits create a protective shield.
4.Shield would be always effheal and no overheal at all and equals the number it absorbed.It's the limitation of combatlog.sorry about that!
I know the difficulty there is. But that's basically the difficult part of evaluating the mitigation. DA at least currently is partially wasted, due to self-overwritting
And a significant part of shields/DA are wasted when they fade out.
1.if Glyph of Power Word: Shield heal someone for 1000,there must be an shield could absorb 5000 damage on this guy.right?
Actually the glyph is less than 20% of the absorb amount, because the healing portion doesn't seem to benefit from all the talents the same as the shield does. Also, as already pointed out the shield can not crit. It should apply a fairly steady amount of shielding. There was some math done that showed that the shield is ~5.5x the heal amount (not counting criticals), and not reducing the value do to overhealing.
Originally Posted by wowmeteronline
2.for DA, on wowhead: Critical heals have a chance to create a protective shield on the target.only have a chance.why you guys say the crits always cause a DA?
and the number of DA is parsed from combatlog and I think it could not be wrong. The fact is that not all crits could proc a DA right now.anything i missed?
and the number of DA or PWS display on WMO is not the proc times of it but the times it works to absorb damage.
That is definitely an old version. DA is a 100% application of 10/20/30% of the total healing amount as a DA shield. This also is an amount based on the entire heal not just effective healing.
A DA applied when another DA is active may prompt something different in the combat log. It may say refreshed or something. This currently causes DA to overwrite but in 3.1 they will stack.
How are you differentiating DA/PWS absorbs? That I know of there is no difference on the damage in message that provides how much was absorbed from each. That means that during times where both are active there could be inaccuracies.
Originally Posted by wowmeteronline
3.the shield heal amount over too much(8k?15k?) is a bug and has been fixed.
4.Shield would be always effheal and no overheal at all and equals the number it absorbed.It's the limitation of combatlog.sorry about that!
any more question please let me know!sorry about my engRish!
For shielding you could calculate the applied amount relatively easily. You can get an approximate effective amount by individually parsing the absorbed amounts. I would not bother with any of the "rapture monitoring" techniques as these will be obsolete in 3.1 live.
Example:
Priest A Glyph of PW:S heals Mage B for 1100. // This is a ~6050 shield PW:S
Mage B gains PW:S
Mob X hits Mage B (Absorbed: 750) // 750 effective absorption
Rain of Fire hits Mage B (Absorbed: 2100) // 2850 effective absorption
Mob X hits Mage B (Absorbed: 750) // 3600 effective absorption
... no damage
Mage B's PW:S fades
At that point you can see that 3600/6050, shielding was effective ~59%.
You can keep a parsed absorption total for each player and a parsed approx absorption provided for each player. That gives a total effective, minus the fact that there are some effects that count as absorption that aren't shields.
Unless there is some modification to the combat log it may be valuable to track things differently....
- Track mitigation of PW:S/DA applied as an approximate value.
- Have an absorption line item for the effective amount of absorbed dmg, calculate its sum effectiveness based on absorbed/(DA+PW:S).
That would make for a relatively close estimation of the amount, considering that most absorbed amounts are rather small from other sources (except DK's?).
Originally Posted by Elimbras
And a significant part of shields/DA are wasted when they fade out.
This is only true on non-tanks. For a tank during the fight ~0% of PW:S is wasted (tanks don't avoid all dmg for 30seconds) and for DA it is <15% (DA's are relatively small by comparison to PW:S ~50% or less, 8 seconds is still a long time in the tank world of dmg income). The overwriting for DA is a definite issue on live but not on ptr, and for tanks it will be <15% waste on ptr because a new DA refreshes the 8sec timer. At the crit of Disc the chance is very high that you will get at least 1 crit in the 8seconds to refresh a DA that would fade, though whether it would last that long in the first place is a big question.
How are you differentiating DA/PWS absorbs? That I know of there is no difference on the damage in message that provides how much was absorbed from each. That means that during times where both are active there could be inaccuracies.
When 3.0 was coming I've seen a blue post saying that DA will absorb before PW:S...
When 3.0 was coming I've seen a blue post saying that DA will absorb before PW:S...
My testing does indicate that is the case. Though with the potential out-of-order updates in the combat log it makes for some difficult parsing scenarios. It could be done to some level of accuracy but without combat log changes it will be difficult.
03/21/09, 8:18 PM
Similar Threads
