Are there any specific EP values for items here or is it more of a personal per-item thing?
Your question probably belongs in the simple question/simple answer thread. You didn't specify your spec (or even your tree), and your profile links to a warlock, so nobody can help you anyhow. Oh, and if you'd searched before asking, you'd have found a fair chunk of discussion about the topic already. (Try the search terms: stat weights or pawn.)
Are there any specific EP values for items here or is it more of a personal per-item thing?
An EP system would imply a single output metric (e.g. DPS). Healers have at least two (throughput and longevity), and weighing the two is not something that can be done with a great deal of accuracy (you end up making a whole pile of assumptions about what fight you're talking about, so while you could say that, for example, 1 mp5 is equivalent to N spell power, that N will vary depending on the length of the fight, the exact damage pattern, your exact spell rotation, some assumptions about how adding 1 of each stat will affect certain talents, and so forth).
And it's my personal opinion that this question does not belong in the simple questions thread because it doesn't have a simple answer.
Promethia: thanks again. What I'm hoping to get is a number on overheal for PW:Shield and DA. Seems hard to get
Tattersail
Here is some info from my guilds first 25man kill of sapphiron. It wasnt my best fight by far numbers wise, but here is the info from it. Overall I was at 37% overheal (our OT respec'd holy for the fight).
13min40sec fight:
Power Word Shield Cast on MT 25 times (5569 Absorb per shield)
Damage to MT
1,106,026 Physical
302,148 Frost
80,818 Shadow
1,488,992 Total Damage
200,744 Absorbed or 13.48% Overall
5,569 x 25 = 139,225 Absorb from PW:S 69.35% of damage Aborbed
61,519 Absorbed thru Devine Aegis 30.65% of damage Absorbed
Looks like I was criting around 28.5% of the time.
The disc recount mod is interesting to be sure, I'll probably use it to assess my alt disc priests performance. One needs to be careful however in taking the results as gospel, since the potential for inflating the numbers artificially are there. From the wowhead post by the mod author:
Assumes all Shields are fully consumed.
Because of how the combat log reports damage absorption, there is no good way of isolating a specific effect's absorbed damage. Therefore all Shields and DAs are considered fully consumed; implying no "overhealing".
I am trying to assess usefullness of discipline build as well. There is way to distinguish between shields fully absorbed and shields that are used only partially, but as to amount of absorbed damage if it is not full it is mostly very fuzzy.
Fully absorbed shields are easily calculated by looking at start of power word:shield buff and its end. If it is less then 30 seconds, it is always fully absorbed shield. To accomplish this, just import text data of the log as obtained by wow into any sort of relational database (I am using MS SQL server) and filter it for Power Word: Shield. Then sort it according time and there is clear sequence of SPELL_AURA_APPLIED and SPELL_AURA_REMOVED records.
I am missing a lot of 'absorbed' records in the log though and that is true even for the shields, that have been removed 5 seconds after appearing. I suspect that logs actually miss the data. Question is though, whether they are missing from the log only, or whether shield mitigation is bugged. (I would say that former is valid)
What is quite clear is the amount absorbed by full shield. Results are showing that having approx 2250 SP raid buffed shielded value is approximately 7000.
With heal of approx 1300 gained from the glyph, effect of the shield is comparable to Greater Heal.
The problem is to do it right. A Disc's priest spellpower can vary considerable temporarily due to trinkets, coming in and out of totems, etc. Even when glyphed and the healing value can always be obtained, healing change effects on the priest and the target, ranging from Druid's Tree of Life to debuffs like Mortal Strike influence the healing but not the shield absorption.
So without considering _all_ possible effects, buffs and debuffs currently on all involved people, we can never be quite sure that what we try to calculate or count is 100% correct.
Careful observation of the events between AURA_APPLIED and AURA_FADE is of course what one needs to do, however there starts the problem that the log events are not always 100% correctly ordered. Specifically everything with the exact same timecode is almost guaranteed not in the correct order. There's been evidence of events with a clear sequence according to time being in reverse occurence in the game as well. Though that's things one just has to take and live with.
And all bets are off with partial absorbs where more than one effect caused absorption. Those are probably rather rare though.
In theory, it would be easy to just count absorbs between PW:S applying and fading. There's full effect absorbs that cause lines (SPELL|MELEE)_(EFFECT|SWING|..)_ABSORBED when the full amount gets eaten by absorption effects and the normal hits where only part of the effect gets absorbed which then is normally detailed as parameter.
When the second one happens, the shield/absorption effect should be exhausted.
The cursory glances I did showed me that often this is true, but sometimes it just isn't and I saw multiple partial absorptions before PW:S fades which is kind of hard to handle right, because it is (to me) not clear what handling right means in that case.
It would be great if the PW:S AURA_FADE would give as parameter the amount absorbed or something like that in order to be able to really assess the amount of damage mitigated.
Given Promethia's numbers, does anyone dare to make a guestimate of how close disc and holy will become in 3.08? Right now I'm lucky to squeeze out half the healing I can do in holy spec. (make that 65-ish %, given the shield numbers from Promethia).
You mean in effective healing done? Effective healing done should pretty much always be higher for multitarget raid healing, so it's not really a valid comparison.
Post 3.08, the question is more how holy priests compare to shamans and druids at raid healing. I'm not sure anyone knows for sure, but it certainly won't be so easy for priests to always top the meters.
Originally Posted by metapseudo
I am missing a lot of 'absorbed' records in the log though and that is true even for the shields, that have been removed 5 seconds after appearing. I suspect that logs actually miss the data. Question is though, whether they are missing from the log only, or whether shield mitigation is bugged. (I would say that former is valid)
Originally Posted by Cadfael
Careful observation of the events between AURA_APPLIED and AURA_FADE is of course what one needs to do, however there starts the problem that the log events are not always 100% correctly ordered. Specifically everything with the exact same timecode is almost guaranteed not in the correct order. There's been evidence of events with a clear sequence according to time being in reverse occurence in the game as well. Though that's things one just has to take and live with.
And all bets are off with partial absorbs where more than one effect caused absorption. Those are probably rather rare though.
...
The cursory glances I did showed me that often this is true, but sometimes it just isn't and I saw multiple partial absorptions before PW:S fades which is kind of hard to handle right, because it is (to me) not clear what handling right means in that case.
It would be great if the PW:S AURA_FADE would give as parameter the amount absorbed or something like that in order to be able to really assess the amount of damage mitigated.
Yes, there are several hitches that make it fairly painful to figure out absorb damage. Here's a basic synopsis of how I have done it:
1. Modify the log so it's easier to import into a database. Mostly the date header is different from the rest of an event line by being delimited by something other than a comma, so I delimit all fields with commas. Simple text seaches with replace can do this. Some database import tools might be able to do this as well.
2. Import the log into a database.
3. Select all SPELL_AURA_* events where the spell is "Power Word: Shield". Dump the query into a spreadsheet.
4. Select all SPELL_AURA_* events where the spell is "Divine Aegis". Dump the query into a spreadsheet.
5. Select all *_DAMAGE events where the absorb parameter is >0. Dump the query into a spreadsheet.
6. Select all *_MISS events where the absorb parameter is >0. Dump the query into a spreadsheet. [Some people fail to realize fully absorbed hits come out as MISS events, not DAMAGE events].
7. Add some columns to your SWING_ and _MISS spreadsheets so that the absorbed columns can line up with spell and ranged damage spreadsheets.
8. Paste all the source spreadsheets into a bigger spreadsheet
9. Create a new time stamp column which modifies the times for SPELL_AURA_ events. Because events can arrive out of order, you want to make it look like buffs are applied earlier and expire later. I modify them by 1.0 seconds.
10. Sort by the target first (!), then the modified timestamp.
11. With a little luck, the vast majority of absorb events will now be bracketed by buffs applied before and after the absorb events... but unfortunately, that might not always be true.
12. There are a lot of potential ways to proceed from here that might save you time. What I did was add columns for DA and PWS that tracked the state of that buff as well as who applied the buff. When no buff was up, the column is blank. I also added two columns to hold the amount absorbed by DA and PWS, respectively.
13. Now the fun begins since I know of no reliable way to go from here besides inspecting the spreadsheet and crediting the absorbed damage appropriately. Fortunately, most the time it's very obvious, but you'll see a variety of weird cases that require a little thinking.
14. I found that at least most of the time, general DA shields are absorbed before PW:Shield, even when the DA buff occurs after the PW:S buff. So when both buffs are up, the absorbed damage is usually assigned to the DA shield. But you can tell by seeing that the DA buff is consumed before the shield.
15. Sometimes both buffs may be up, and a hit will break through one shield and then be absorbed by the next. This will be apparent because the hit is fully absorbed (it's a miss event), but one buff fades (usually DA) while the other one does not fade. In these cases, I split credit in a 2:1 ratio with more credit going to the PWS. Don't stress over this since it's a rare event, but if you look over the data, that's a plausible ratio since DA shields are usually much smaller.
16. Once you get done going over every line in the spreadsheet, swear at Blizzard a while for failing to make this easier. It's a cathartic experience.
It's painful, but this produces pretty accurate results. Although there's some guesswork involved whenever two shields are up and a hit breaks through one into the other, there's enough context in most cases to get a reasonable estimate on those. For instance, you'll see that most DA shields absorb a similar amount. So when a hit breaks through it and into a PW:Shield, you know the DA shield was fully absorbed and it absorbed about what other DA shields from that priest would. If you apply a little reason, you won't be off by much on those.
The bigger problem IMO is that the combat logs are not timestamped with the time that the event happened on the server. The timestamp is when the event arrived on the client. That's messed up because network packets arrive out of order and give you "impossible" sequences of events. For instance, here's a real sequence from one of my combat logs:
12/29/08 12:19:24.299 AM SPELL_AURA_APPLIED,Promethiaep,Makmooka,Power Word: Shield
12/29/08 12:19:24.707 AM SPELL_PERIODIC_DAMAGE,Anub'Rekhan,Makmooka,Locust Swarm,6863,0,997,0,280
12/29/08 12:19:27.285 AM SWING_MISSED,Anub'Rekhan,Makmooka,Melee,5362
12/29/08 12:19:27.486 AM SPELL_AURA_APPLIED,Promethiaep,Makmooka,Divine Aegis
12/29/08 12:19:27.486 AM SPELL_AURA_REMOVED,Promethiaep,Makmooka,Divine Aegis
12/29/08 12:19:29.709 AM SWING_DAMAGE,Anub'Rekhan,Makmooka,Melee,4865,0,0,0,1567
12/29/08 12:19:29.896 AM SPELL_AURA_APPLIED,Promethiaep,Makmooka,Divine Aegis
12/29/08 12:19:29.896 AM SPELL_AURA_REMOVED,Promethiaep,Makmooka,Divine Aegis
12/29/08 12:19:29.896 AM SPELL_AURA_REMOVED,Promethiaep,Makmooka,Power Word: Shield
So what that basically says is that:
1. Promethiaep put a PWS on Makmooka.
2. Makmooka aborbed 280 (??) of the next hit and the rest was damage.
3. Makmooka absorbs 5362 and takes no damage from the next hit.
4. Promethiaep procs a DA shield on Makmooka which simultaneously fades.
5. Makmooka aborbed 1567 of the next hit and the rest was damage.
6. Promethiaep procs a DA shield on Makmooka which simultaneously fadesandPW: Shield simultaneously fades.
Sort that one out. The only lucky part is that if all I care about is who to credit the absorb damage to, then I'm probably ok, but imagine if those buffs came from different priests!
Fully absorbed shields are easily calculated by looking at start of power word:shield buff and its end. If it is less then 30 seconds, it is always fully absorbed shield.
I hate to be a stickler but this is not true for one case. If you overwrite your pws with another pws before the end of the duration of the buff.
just a tiny question, what do u guys think is the best(better) for general all around usage of the food buffs currently available for a 14/57 holy priest
just a tiny question, what do u guys think is the best(better) for general all around usage of the food buffs currently available for a 14/57 holy priest
I use Golden Fishsticks or Cuttlesteak, depending if I think stam is worth much on the fight or if I feel I need more regen. Hearty Rhino Dogs are also good until you get up into high int range (for the record I run around 1300+ int raid buffed and I usually sit at 1600/600 regen). The less int you have the more competitive the Hearty Rhino Dogs are. I guess you could go for Firecracker Salmon or the meat equivalent (spellpower/stam) if you feel like you're happy with your regen.
Originally Posted by Shaker
Hey guys, without dodge and evasion rogues are really weak to melee too.
I may just wade through a few 10-man WWS with me being the only Disc Priest and tally up all the absorbs. Now I was wondering if there any absorbs from other classes that could interfere. Do any tanks have any absorb effects? Does WWS count black as absorb?
I'm only aware of Mages and Warlocke being able to absorb some damage, so I'll check for their buff gains.
Any other input into what I may have overlooked could be good.
just a tiny question, what do u guys think is the best(better) for general all around usage of the food buffs currently available for a 14/57 holy priest
Cuttlesteak for regen, imperial manta steak for throughput. But that's more a question for the "simple" thread.
Anyone who cannot cope with mathematics is not fully human. At best he is a tolerable subhuman who has learned to wear shoes, bathe, and not make messes in the house. - R.A. Heinlein
The only other absorption effects I can think of are a Paladin's Sacred Shield, Anti Magic Zone, The Shadow, Frost, and Fire Wards, [Greater Ward of Shielding], Voidwalker Sacrifice. But I'm sure there are plenty of others.
I hate to be a stickler but this is not true for one case. If you overwrite your pws with another pws before the end of the duration of the buff.
That will actually general a renew event on the shield rather than removing the shield and then reapplying it.
Originally Posted by Tainter
I may just wade through a few 10-man WWS with me being the only Disc Priest and tally up all the absorbs. Now I was wondering if there any absorbs from other classes that could interfere. Do any tanks have any absorb effects? Does WWS count black as absorb?
I'm only aware of Mages and Warlocke being able to absorb some damage, so I'll check for their buff gains.
Any other input into what I may have overlooked could be good.
There are several other sources of absorbed damage, which is another reason I decided I had to inspect every absorb event before crediting it. In general other sources of absorbed damage are readily recognized because most are smaller in size and many of them absorb a very constant amount.
Another problem is that other holy priests (due to a bug) may proc DA shields if there's also a discipline priest in group, and of course all priests can cast PW:Shield. So credit for shields is only easy when there is one priest total.
One thing I'd like to know is the order of precedence for absorbing damage when there are multiple shields on a target. For instance, I saw multiple cases where a PW:S was applied, then a DA shield, but then the DA shield absorbed damage before the PW:S absorbed damage again. I suspect that DA shields generally absorb damage before PW:Shield, but I did see one case (and only one) where this didn't seem to be true.
There are alternative possibilities as well: it could be that the most recently applied shield gets used up first (a last in, first out stack), or it could be applied to the shield that is set to expire next. I just don't have enough cases to go on to really say definitely, but it seems to me DA shields generally get used before PW:Shields.
Despite seeing other absorbed damage, I didn't see clear cases where that was stacked with another shield, so I cannot comment on how non-priest shielding abilities stack with other shields. In the midst of a fight, shields just don't last very long, so it is rare that two shielding effects are active at the same time.
A couple things I think are pretty important with the 3.08 patch coming out.
If you are spec'd into holy enough to have 5/5 Providence you will be able to almost hash out a rotation of PoM, CoH, Flash Healx 3, PoM, CoH, Flash Heal x3 etc.
As a strong advocate for Healing Prayers over Surge of Light I won't push too much for the mana you save, however those who really like Surge of Light please take a look at the formula's because when you hit that CoH button I think you are all wanting a SoL to becoming from it.
Here are some quick numbers using the model I posted almost 2 months ago (I don't think above 40% crit is too realistic so I'm stopping there). Starting with 20% crit as it is easily attained; 5% from Holy Spec is a given, +5% from raid buffs is also a given, leaving 10% to be attained via intellect (~150 int per 1% crit), base crit chance and gear you get the following:
I guess I'm thinking out loud and didn't overly like the idea of gemming and/or gearing for crit. However, with the CoH cooldown era (completely imo) haste is a very sub-par stat now for holy priests. Perhaps the new wave is for *a lot* of flash healing and constant SoL procs via a bouncing PoM and a 6-target CoH every 6 seconds.
Anyone else feel that haste is no longer the throughput thoroughbred, both pre-WotLK and pre-3.08, it once was?
I never went to any level of haste to push CoH throughput (although it helped there). I primarily wanted it as a tool to speed up GHeal and FH casts. Given that we'll be doing this even more now, I don't see that some haste is necessarily a bad thing. Note that I can attain 28% crit, 21% haste raid-buffed at the moment ... it seems like a nice balance. I don't need the extra crit that badly.
Anyone who cannot cope with mathematics is not fully human. At best he is a tolerable subhuman who has learned to wear shoes, bathe, and not make messes in the house. - R.A. Heinlein
A couple things I think are pretty important with the 3.08 patch coming out.
If you are spec'd into holy enough to have 5/5 Providence you will be able to almost hash out a rotation of PoM, CoH, Flash Healx 3, PoM, CoH, Flash Heal x3 etc.
As a strong advocate for Healing Prayers over Surge of Light I won't push too much for the mana you save, however those who really like Surge of Light please take a look at the formula's because when you hit that CoH button I think you are all wanting a SoL to becoming from it.
Here are some quick numbers using the model I posted almost 2 months ago (I don't think above 40% crit is too realistic so I'm stopping there). Starting with 20% crit as it is easily attained; 5% from Holy Spec is a given, +5% from raid buffs is also a given, leaving 10% to be attained via intellect (~150 int per 1% crit), base crit chance and gear you get the following:
I guess I'm thinking out loud and didn't overly like the idea of gemming and/or gearing for crit. However, with the CoH cooldown era (completely imo) haste is a very sub-par stat now for holy priests. Perhaps the new wave is for *a lot* of flash healing and constant SoL procs via a bouncing PoM and a 6-target CoH every 6 seconds.
Anyone else feel that haste is no longer the throughput thoroughbred, both pre-WotLK and pre-3.08, it once was?
I had been thinking haste would count for more now, regen will be less of an issue with no CoH spam, then the next issue is how quickly can you react/cast - haste is good for this (and means your not being snipes every cast), IHC/SoL are proc based and SoL doesnt haste the GCD in any way.
For now I am trying to balance them more then any thing. I guess PoM favours SP and Crit.
the numbers around crit and how much haste on average does it provide via IHC is some thing worth looking at.
As a strong advocate for Healing Prayers over Surge of Light I won't push too much for the mana you save, however those who really like Surge of Light please take a look at the formula's because when you hit that CoH button I think you are all wanting a SoL to becoming from it.
Here are some quick numbers using the model I posted almost 2 months ago (I don't think above 40% crit is too realistic so I'm stopping there). Starting with 20% crit as it is easily attained; 5% from Holy Spec is a given, +5% from raid buffs is also a given, leaving 10% to be attained via intellect (~150 int per 1% crit), base crit chance and gear you get the following:
I guess I'm thinking out loud and didn't overly like the idea of gemming and/or gearing for crit. However, with the CoH cooldown era (completely imo) haste is a very sub-par stat now for holy priests. Perhaps the new wave is for *a lot* of flash healing and constant SoL procs via a bouncing PoM and a 6-target CoH every 6 seconds.
Anyone else feel that haste is no longer the throughput thoroughbred, both pre-WotLK and pre-3.08, it once was?
However, if you are raid healing, you appreciate the SoL proc. You won't save (a reasonable amount of) time, but the proc will save you mana. And if you are raid healing, you will want that flash heal anyway. While you are casting that CoH, you are already looking for your flash heal target. And thats the whole point anyway. The SoL flash saves time, ok. But the mana saved outweighs Healing Prayers on most fights (i'm just saying that here, proof me wrong). But even if it does'nt, the PoM may not even jump to the person who really needs the flash heal, so you have to cast it anyway.
In addition to that, going from say 25% to 30% raid buffed is huge (for me it is anyway). You may be able to do it, but you might gimp yourself on all other stats. The gain, in terms of SoL procs, may not even be noticeable.
So, i certainly do not feel like haste is no longer the throughput stat. For tank healing it most certainly will not be.
-snip-
The SoL flash saves time, ok. But the mana saved outweighs Healing Prayers on most fights (i'm just saying that here, proof me wrong). But even if it does'nt, the PoM may not even jump to the person who really needs the flash heal, so you have to cast it anyway.
-snip-
Surge of Light really doesn't save time. Over X consecutive casts, Surge of Light (instant) or not, you will still spend X*GCD(1.5sec or less depending on haste). The Surge of Light allows your heal to be less likely overheal (snipe-free!) but still you spend the same amount of time getting to your next target.
Is that correct? Or is my brain just fried from too much time at university today...
It doesn't save Execution or Cast time; but it still fires off 1.5s faster than normal and can be done while running.
It has the same time cost; but it does fire off faster. So in that regards it saved 1.5s in healing a target; but still costs you the 1.5s.
I think it's all terminology/semantics at this point. We should all know how it works and realize it's always a mana saver.
Originally Posted by arison
Everyone should start from the same place and rise based on their abilities, desires, and schedule. No one plays MMOs to *be* powerful, they play MMOs to *become* powerful. It's the journey, stupid. The rarer loot is, the more cherished it is when you get it, but only so long as there is a reasonable expectation to get it. The rarer loot is, the better it feels when you kill a boss or when $AWESOME_TRINKET drops.
Surge of Light really doesn't save time. Over X consecutive casts, Surge of Light (instant) or not, you will still spend X*GCD(1.5sec or less depending on haste). The Surge of Light allows your heal to be less likely overheal (snipe-free!) but still you spend the same amount of time getting to your next target.
Is that correct? Or is my brain just fried from too much time at university today...
Yes, that is correct, i think, it will just front load the healing done (ok, it can't crit). But the point of saved mana still stands. And while you are on global cooldown from the initial CoH which may, or may not, provide SoL, you are looking for the flash heal target (at least i do, when i am raid healing). If you do not get a SoL proc your target may still need the flash heal.
Edit: plus you still can't rely on PoM to jump to you prefered flash heal target.
It doesn't save Execution or Cast time; but it still fires off 1.5s faster than normal and can be done while running.
It has the same time cost; but it does fire off faster. So in that regards it saved 1.5s in healing a target; but still costs you the 1.5s.
I think it's all terminology/semantics at this point. We should all know how it works and realize it's always a mana saver.
Yeah definitely, I think possibly the best characteristic of SoL (completely blew by me hah!) is that it can be cast on the run, thanks for reiterating that. I think I'm definitely going to run my next couple weeks with SoL again (have not used it in over a month).
Also, anyone who uses Rawr on a regular basis not getting the theoretical stats to equal your actual raid stats? I seem to have been having that problem tonight when playing around with gear sets at work.
Undocumented patch change: [Majestic Dragon Figurine] now only stacks spirit on direct heals or direct damage spells. Direct heals can stack it multiple times however.
Will not proc from:
Prayer of Mending
Power Word: Shield
Renew
Devouring Plague
Shadow Word: Pain
Levitate
etc.
Will "bonus" proc on:
Binding Heal (2)
Circle of Healing (5 or 6 glyphed)
Haven't tested Mind Sear or Holy Nova
I'm finding it easier for 10 stacks to fall off, and I might have to reconsider the trinket as a result. I used to cast Levitate mid fight to keep the spell up (low mana, instant cast) but now the only instant options I've found are Shadow Word: Death or Desperate Prayer.
01/19/09, 2:46 AM
Similar Threads
