I do agree that our primary roles in raids haven't really changed though. I find it unfortunate that we've been reduced to haste stacking/ch spam again... it was nice to get away from that for awhile. However, with the changes it's very clear that it is the most effective way for us healing. I too am going to wait a month or so when all the data is out before I really start changing my gear setup.
I agree with the first part of this statement, but I disagree with the second. I like Haste stacking(within reason due to how crit acts as a double stat effecting both throughput and regen), but only because it keeps Chain Heal low,and makes LHW and HW feel far more responsive. I still think we might see about a 50/50 allocation between Chain Heal and LHW/HW/Riptide; which is absolutely acceptable.
My Ulduar runs have become almost 80% CH spam for keepers. Especially during Mimiron and Thorim (arena), and Freya sees heavy use as well. These aren't hard-modes, but besides Mimiron-hard, the fights still will favor strong CH spam.
I'm in the process of regemming haste to replace Int, since I'm hardly going oom due to recent changes. I still use the deadly glad totem for LHW, since I often help tank heal, but I am definitely replacing it with the new spellpower from chain heal totem once I stack 25 triumph badges.
My Ulduar runs have become almost 80% CH spam for keepers. Especially during Mimiron and Thorim (arena), and Freya sees heavy use as well. These aren't hard-modes, but besides Mimiron-hard, the fights still will favor strong CH spam.
For me its almost been the same thing spamming 80% of my cast being chain heal except I'm on the hard modes. I haven't re gemmed just yet, but haste spell power is seeming more and more popular / appealing for most of the hard mode encounters I have done this week because I haven't had a single mana problem at all (now its the paladins). My heals per second went up by almost 1500 from just this patch not changing gems or any gear out. One thing I have noticed though is that I've decided to flask for Spell power instead of Distilled wisdom or Flask of Mojo and it seems to be working fairly well.
So at what point would you assume to start swapping from int to haste? I mean I know how wonderful and dangerous haste can be to output and the mana pool as I try to get haste on every gear-piece, but around when should the change be made?
The point where you're not having mana regen issues. YMMV.
I replaced virtually all my Int gems with Sp variations as I already had a haste-heavy enough set to get CH down to 1.8 seconds. I kept some Int, probably out of paranoia but I intend on gemming those off as soon as our JCs have the badges for the recipes. I find I'm often in a position to decline offers of Innervates since 3.2, and there's no good excuse for that.
(I have also seen a ridiculous increase in the amount I use CH and, personally, I really enjoy getting back to it being the backbone of our healing - especially now that we have a couple of ribs to attach to it. Having lost .4-.5 seconds off LHW has really made me feel utterly useless on Mim hard though.)
Having lost .4-.5 seconds off LHW has really made me feel utterly useless on Mim hard though.
I'm not really sure what you mean by feeling useless on Firefighter. My experience with that fight has been overwhelmingly positive this past week without even gemming for haste. I found that outside of Phase 1, the new and improved Chain Heal is an incredibly useful spell for the entire fight. In Phase two you Chain Heal the melee, and if your assigned range group is getting nailed by Rapid Fire Burst, then you Chain Heal your group. In Phase three you can pretty much spam Chain Heal on whoever is tanking the Assault Bots and keep a lot of people up. In Phase four Chain Heal becomes a godsend as everyone is moving around and taking random damage. It may seem like people are spread out too much in that phase but that is not the case - usually when someone takes damage there is someone else pretty close to them who also took that damage.
Here's my parse from last week. Remember this is before I switched to haste gems: Firefighter Parse
I am growing more and more weary of pure haste stacking, especially for hard mode. I think it's been blown way out of proportion, especially for hard modes. That said, I think you can have some fights with 500 haste (assuming WoA and SR and 4pT8) to yield 1.806 sec CHs and still have enough regen with IWS to nearly spam. I can't see more than 500 haste being necessary since crit is just so good for regen, especially for CH since it can proc all four bounces.
What I'm especially interested in the newer tank healing role. Most fights, it's not an issue to continue using 500 haste and spam CH on the tank, hit melee, and win. But on fights like Algalon, Shaman should be doing very well with glyphed ES, AH, and Healing Way. Taking that much haste seems no where near as optimized as stacking crit with nearly a HW spam. 259 haste will lower HW to 1.500 seconds (assuming WoA, SR and TW proc) and leaves a substantial amount of headroom for crit to add procs of IWS. We did Algalon this week with three tank healers - two shaman and one paladin. The Paladin was feeling especially pained for mana post-3.2. The other shaman ran 350 haste with 45% crit raid buffed with 3100 spell and spec'd Healing Way. I ran 439 haste with 37% crit raid buffed with 3320 spell and spec'd Elemental Weapons. He used a HW-heavy spam and I used a balance between LHW and HW. The results were astonishing given our overhealing was about the same how much better than was for Algalon.
So I really thing *most* fights are fine with having more haste and spamming CH more like back in BT and SWP. But hard mode challenges with VERY spikey tank damage - which I expect to see again in ToC and ICC - requires a different school of thought.
I am growing more and more weary of pure haste stacking, especially for hard mode. I think it's been blown way out of proportion, especially for hard modes. That said, I think you can have some fights with 500 haste (assuming WoA and SR and 4pT8) to yield 1.806 sec CHs and still have enough regen with IWS to nearly spam. I can't see more than 500 haste being necessary since crit is just so good for regen, especially for CH since it can proc all four bounces.
What I'm especially interested in the newer tank healing role. Most fights, it's not an issue to continue using 500 haste and spam CH on the tank, hit melee, and win. But on fights like Algalon, Shaman should be doing very well with glyphed ES, AH, and Healing Way. Taking that much haste seems no where near as optimized as stacking crit with nearly a HW spam. 259 haste will lower HW to 1.500 seconds (assuming WoA, SR and TW proc) and leaves a substantial amount of headroom for crit to add procs of IWS. We did Algalon this week with three tank healers - two shaman and one paladin. The Paladin was feeling especially pained for mana post-3.2. The other shaman ran 350 haste with 45% crit raid buffed with 3100 spell and spec'd Healing Way. I ran 439 haste with 37% crit raid buffed with 3320 spell and spec'd Elemental Weapons. He used a HW-heavy spam and I used a balance between LHW and HW. The results were astonishing given our overhealing was about the same how much better than was for Algalon.
So I really thing *most* fights are fine with having more haste and spamming CH more like back in BT and SWP. But hard mode challenges with VERY spikey tank damage - which I expect to see again in ToC and ICC - requires a different school of thought.
Stacking haste doesn't mean you have to spam a spell until you OOM yourself like a moron, either. It just gives more throughput when necessary and ultimately allows you to cancel / recast with less penalty. Stacking haste also doesn't require you to drop crit by any means, either. I had no problem healing Alg this week with my current setup and had something like 10k mana at the end of the fight - 4k effective hps using a nice balance of RT / LHW / CH and a few HW's when I knew the bigger heals would be needed.
Stacking haste doesn't mean you have to spam a spell until you OOM yourself like a moron, either. It just gives more throughput when necessary and ultimately allows you to cancel / recast with less penalty. Stacking haste also doesn't require you to drop crit by any means, either. I had no problem healing Alg this week with my current setup and had something like 10k mana at the end of the fight - 4k effective hps using a nice balance of RT / LHW / CH and a few HW's when I knew the bigger heals would be needed.
Haste =/= being an idiot.
Can you share more on how you did Algalon in terms of healer setup for tanks, your crit/haste/spell/MP5 numbers, and how you "knew" bigger heals would be needed? The damage is so spikey, how can you do anything other than spam and do reactive healing? Do you have a WoL for that fight?
Can you share more on how you did Algalon in terms of healer setup for tanks, your crit/haste/spell/MP5 numbers, and how you "knew" bigger heals would be needed? The damage is so spikey, how can you do anything other than spam and do reactive healing? Do you have a WoL for that fight?
My armory is a click away.
Setup this week was double holy pally and myself on tanks. Normally we do holy pally / disc priest / myself but we had to change it up due to a few vacations. I realize the damage can spike but if I see a cosmic form on one of the other tank healers chances are there might be a lull in the healing in which case I toss out a quick HW to compensate for the lack of heals over the next second or two. No real mystery as to how I'd "know" bigger heals would be needed. And until the healing way change I never once used HW on Algalon - purely RT / glyphed LHW.
Lately for tank healing I'll get RT up and then decide if I want to toss out a couple LHW's or two quick HW's - if melee are in tank range I can generally just bounce CH off the tank then quickly revert to LHW / HW if I need a little more single target healing over the next few seconds.
It's not that difficult - all stacking haste does is give you the means to be even more reactionary with LHW / HW while speeding up CH. And like I said, it doesn't require you to drop massive amounts of crit at all. Just glancing over the 245 / 258 gear in Coliseum, there's a nice variety of haste/crit pieces and haste/mp5 and crit/mp5. I'll actually be using several of the crit/mp5 pieces with gemmed haste. The final stats are pretty well balanced and given the nature of itemization in Coliseum, it's not going to be some crazy mystery on which items to take.
I'm currently sitting at 21k mana, 32.23% chance to crit, about 408 Mp5, and 2596 spell power, all unbuffed. Due to the recent changes to Lesser Healing Wave, my cast time has gone from about .626, to 1.3. Would it be worth it for me to regem for haste to get my cast time alittle lower for LHW or am I just wasting my time? As it sits now I hardly ever use LHW due to the fact that LHW cast speed is 1.3 and Healing Wave is right about 1.5, and heals at least twice as much. Please let me know what you think.
Sixthy - curious about your melee positioning on Algalon. In my guild the melee stand far away from tank so they don't have to move from each others' Cosmic Smashes. Did you specifically ask them to stand in range for Chain Heal bounces or did it just end up working that way?
Setup this week was double holy pally and myself on tanks. Normally we do holy pally / disc priest / myself but we had to change it up due to a few vacations. I realize the damage can spike but if I see a cosmic form on one of the other tank healers chances are there might be a lull in the healing in which case I toss out a quick HW to compensate for the lack of heals over the next second or two. No real mystery as to how I'd "know" bigger heals would be needed. And until the healing way change I never once used HW on Algalon - purely RT / glyphed LHW.
Lately for tank healing I'll get RT up and then decide if I want to toss out a couple LHW's or two quick HW's - if melee are in tank range I can generally just bounce CH off the tank then quickly revert to LHW / HW if I need a little more single target healing over the next few seconds.
It's not that difficult - all stacking haste does is give you the means to be even more reactionary with LHW / HW while speeding up CH. And like I said, it doesn't require you to drop massive amounts of crit at all. Just glancing over the 245 / 258 gear in Coliseum, there's a nice variety of haste/crit pieces and haste/mp5 and crit/mp5. I'll actually be using several of the crit/mp5 pieces with gemmed haste. The final stats are pretty well balanced and given the nature of itemization in Coliseum, it's not going to be some crazy mystery on which items to take.
You use the same gear/glyphs/spec on every pull? That is why i asked.
Also, would love to see a WOL so I can see how your raid heals that.
I've re-gemmed almost full haste (~750 unbuffed) and have sacrificed very little crit. I'm still looking at 34% unbuffed.
The haste isn't just for CH spam. It keeps your LHW fast enough to reactively heal the spike damage on the hard modes. Its not going to be as fast as pre-3.2 but it is noticable to me.
I've re-gemmed almost full haste (~750 unbuffed) and have sacrificed very little crit. I'm still looking at 34% unbuffed.
The haste isn't just for CH spam. It keeps your LHW fast enough to reactively heal the spike damage on the hard modes. Its not going to be as fast as pre-3.2 but it is noticable to me.
Yes, but what is your regen at? What is your healing role (raid, I assume)?
We are very good tank healers now too, and it sounds like you are gearing for non-hard mode raid healing - which is fine. But what I'm driving at is that I don't think there is an "ideal" gear/glyph/totem/spec that will work for every encounter. In SWP, we could map CH to every key on the keyboard and rollface on it to keep people up. Certainly for some fights (many?) in Ulduar you can. But I do not think so on tank healing fights like Algalon. And with this mind, I'm curious if Blizzard is going to push us more and more into a tank healing roll where LHW spam doesn't work and where regen from IWS crits would be better.
Yes, but what is your regen at? What is your healing role (raid, I assume)?
We are very good tank healers now too, and it sounds like you are gearing for non-hard mode raid healing - which is fine. But what I'm driving at is that I don't think there is an "ideal" gear/glyph/totem/spec that will work for every encounter. In SWP, we could map CH to every key on the keyboard and rollface on it to keep people up. Certainly for some fights (many?) in Ulduar you can. But I do not think so on tank healing fights like Algalon. And with this mind, I'm curious if Blizzard is going to push us more and more into a tank healing roll where LHW spam doesn't work and where regen from IWS crits would be better.
Why do you feel that increasing haste somehow decreases tank healing ability? I'm primarily a tank healer for just about every encounter in the game at the moment. I'll spot the raid when necessary but druids and priests can handle pretty much all of it for the most part. Healing melee falls under the tank healing category via CH when it's applicable. I gained 45,283 mana from WS on our last Alg kill; 15k from replenishment and then it trickles off from other various mana returns. I have 720 haste - so how exactly did that hinder my ability to regen mana?
Like I said before: stacking haste doesn't mean you have to mash your keyboard with flipper hands until an encounter ends. It increases throughput on every heal we have outside of RT and ES. We don't scale *that* well with SP, so it's not entirely necessary to go out of your way to acquire more of it via gemming - but I'm not completely against it. My SP is a little lower than I'd like at the moment, but with 245 / 258 itemization + a legendary I'll be back at respectable levels, although I still hit 3k raid buffed.
So I sacrifice nothing as far as I'm concerned to increase haste. I can take haste/mp5 gear and just swap enchants on my helm / shoulders to make up for the lost crit. I can take crit/mp5 gear and gem it purely haste to make up for the haste loss. I can take crit/haste gear and cover both bases. I still maintain roughly 45% crit raid buffed - so 70% on LHW when I need to get one off quickly (which is 1.1 seconds due to haste gemming), and I can still toss out a 1.28 second HW if I need a larger heal.
I just don't understand why you're equating haste with mindless CH spam to drain yourself OOM (which with the IWS change is pretty hard regardless) and that it somehow lowers tank healing viability.
Grats on the Val mace, Sixthy. I'm currently at 16fragments.
I agree with your stance on haste+crit for tank healing.
However, spellpower scales incredibly well with the Chain Heal mechanics (not to mention ES and it's the best stat for RT too, but if you're getting more than 18% of your total healing from those two spells... you probably don't need to heal at all). SP is much better than haste according to the CH section of shaman_hep.
So for shaman that find themselves spamming chain heal, I'd say mix up a lot of spellpower and some haste. I personally can't run out of mana on fights where I'm just using CH.
On Algalon (i've only succeeded in 10man, not 25 yet). My gear uses a bit more crit and a lot more haste. I heal it about the same way as what Sixthy said.
Watch for red circles or listen on vent if your guild announces them. You need to know if the other tank healers are being forced to move. If they are... I tend to just overheal the tank until I see the movements of the other healers coming to an end.
We used a bear and a paladin tank. I found that I could CH the bear unless the other tank healer had to run, then HWs on the bear. Tank-healing shaman+paladin, raid healing holy priest.
I found that I could choose either LHW or HW on the paladin, but a CH was risky. (I didn't have LHW glyphed, otherwise for 10man, i could probably have skipped using HW).
This coming week; I plan on the following additional technique.
RT - always keep it rolling on the tank (this is a concern in 10man with fewer other healers)
If RT is already on the tank, cast it on someone else in the raid... anyone else will take damage sometime.
Why do you feel that increasing haste somehow decreases tank healing ability? I'm primarily a tank healer for just about every encounter in the game at the moment. I'll spot the raid when necessary but druids and priests can handle pretty much all of it for the most part. Healing melee falls under the tank healing category via CH when it's applicable. I gained 45,283 mana from WS on our last Alg kill; 15k from replenishment and then it trickles off from other various mana returns. I have 720 haste - so how exactly did that hinder my ability to regen mana?
Like I said before: stacking haste doesn't mean you have to mash your keyboard with flipper hands until an encounter ends. It increases throughput on every heal we have outside of RT and ES. We don't scale *that* well with SP, so it's not entirely necessary to go out of your way to acquire more of it via gemming - but I'm not completely against it. My SP is a little lower than I'd like at the moment, but with 245 / 258 itemization + a legendary I'll be back at respectable levels, although I still hit 3k raid buffed.
So I sacrifice nothing as far as I'm concerned to increase haste. I can take haste/mp5 gear and just swap enchants on my helm / shoulders to make up for the lost crit. I can take crit/mp5 gear and gem it purely haste to make up for the haste loss. I can take crit/haste gear and cover both bases. I still maintain roughly 45% crit raid buffed - so 70% on LHW when I need to get one off quickly (which is 1.1 seconds due to haste gemming), and I can still toss out a 1.28 second HW if I need a larger heal.
I just don't understand why you're equating haste with mindless CH spam to drain yourself OOM (which with the IWS change is pretty hard regardless) and that it somehow lowers tank healing viability.
I think my point about haste is the diminishing return quality of it. So for example, running 750 haste - assuming raid buffs and 4pT8 - you're chain heal will take 1.681 seconds to cast. Running 650 haste under the same conditions will result in a CH in 1.729 seconds - a difference of 0.048 seconds or rough 1/20th of a second.
Let's say that gear for 750 haste was done with 5 epic haste gems. Should you replace those with 5 epic crit gems, you'd gain 2.178% crit.
Now I'm willing to accept that I'm looking at this the wrong way, but if I had a piece of gear that said "reduces your CH cast time by 1/20 of a second" and one that said "increases your crit by 2.178%", all other things being equal I know which one I'd choose.
My point is that if you look across these forums, Plus heal, WoW forums, and other places, there is no consensus at all on gearing right now. I choose to start looking at haste first (and perhaps this this a mistake) because it is the most easily calculable of the variable stats.
Edit: Also, Show of Faith and Sif's Remembrance do not share a CD?! I know I've tested Sif's and Spark of Life and they don't share a CD, so that's interesting.
I think my point about haste is the diminishing return quality of it.
Haste does not have diminishing returns, at least as a throughput stat. (I consider its value as a gib-prevention stat unquantifiable.) 1% haste means 1% more of a given spell cast in a given time period.
To put this in real terms, Healing Wave is a 2.5s cast. In 1000s of cast time, a shaman with zero haste can cast 400 Healing Waves. 1% of haste will increase that by 1%, to 404. In 1000s of cast time, a shaman with 50% haste can cast 600 Healing Waves, 50% more than the shaman with zero haste. 1% of haste will increase that to 51%, allowing 604 Healing Waves to be cast in that time period. Whether you have 0% or 50% haste (or any other amount), increasing your haste by 1% will give you the ability to cast 4 more Healing Waves in a 1000s time period.
1% of haste gain at 0% haste reduces Healing Wave cast time by 0.025s. At 50% haste, 1% haste reduces its cast time by 0.011s. This does not mean it is giving less value, it means cast time is the wrong way to measure throughput.
Adding haste does have reduced value once the 1.0s gcd is reached, but this isn't a real concern with reasonable gearing.
Haste does not have diminishing returns, at least as a throughput stat. (I consider its value as a gib-prevention stat unquantifiable.) 1% haste means 1% more of a given spell cast in a given time period.
To put this in real terms, Healing Wave is a 2.5s cast. In 1000s of cast time, a shaman with zero haste can cast 400 Healing Waves. 1% of haste will increase that by 1%, to 404. In 1000s of cast time, a shaman with 50% haste can cast 600 Healing Waves, 50% more than the shaman with zero haste. 1% of haste will increase that to 51%, allowing 604 Healing Waves to be cast in that time period. Whether you have 0% or 50% haste (or any other amount), increasing your haste by 1% will give you the ability to cast 4 more Healing Waves in a 1000s time period.
1% of haste gain at 0% haste reduces Healing Wave cast time by 0.025s. At 50% haste, 1% haste reduces its cast time by 0.011s. This does not mean it is giving less value, it means cast time is the wrong way to measure throughput.
Adding haste does have reduced value once the 1.0s gcd is reached, but this isn't a real concern with reasonable gearing.
To settle this, consider gear measuring haste versus crit (you can do whatever here, I'm using crit as my example.) There has to be a point at which adding more haste has less benefit than adding crit instead. That's what I'm trying to discover here. Now, with "rotation" variations, this can be a challenge. Assume for tank healing you want to do RT, HW, HW. Do I want 1% more haste to cast that rotation 1% more times or do I want 1% more crit to have all of those spells to have more chance to crit? I think it's safe to assume that starting from lower haste values, haste will have more benefit, but where is that point - which I believe exists - that more crit would have been beneficial?
Crit is a poor choice. First off, it is more expensive than Haste in terms of ilvl -> % calculations. Secondly, 1% crit is less than 1% extra throughput. If you grab a spreadsheet and crank some numbers around, you will always see Haste giving you more throughput value because it continues to scale all the way until you hit the GCD, which is something like 2000 Haste.
I'd expect that the setup of your tanks (whether they are geared for stamina or avoidance) affects whether quicker spells or more powerful spells are more efficient. An avoidance tank would be more suited to quick hasted LHW, whereas a stamina tank would be better served by the slower HW spells. Is there enough difference between the ways tanks arrange their gear to make altering your healing style useful?
Estimates need to be made in order for calculations to be done. Estimates are needed in terms of how many spells you are casting per minute, how much haste & crit you already have, how long a fight lasts, etc. Some aspects of the calculations are set in stone, but others have a range of possible values, requiring estimates to be made.
A good estimate for the mp5 from Crit is 10 crit = 3.57 mp5. The calculations are too lengthy to post here.
A good estimate for 10 intellect is 3.1 mp5, and we can calculate it at 2.76 crit rating, and 1.5 spellpower.
Spellpower only gives spellpower - no bonuses towards mana regen, crit, haste, or intellect.
Mp5 only gives mana regen - no bonuses towards spellpower, crit, haste, or intellect.
A good estimate for spellpower is that 72 spellpower = 2% increase in healing done.
A good estimate for haste is that 60 haste = 2% increased cast speed = 2% increased healing from channeled spells.
A good estimate on the mp5 cost of haste is that 30 haste = 15 mp5 more used.
A good estimate on crit's healing value is that 60 crit does about half the extra healing of 60 haste or 72 spellpower.
To sum it up, here are some options that could be attained through gemming, divided into just spellpower and mp5 components. 6 split gems worth of bonuses are assumed (12 spellpower, 10 crit, 10 haste, 10 int, or 5 mp5). I then divide the stats into their spellpower & mp5 equivalents since we can reduce Intellect, Crit, and Haste into what they contribute in terms of extra healing, and extra mana use & extra mana regen.
If mana regeneration is not an issue, haste is excellent. If mana regen is an issue, haste will screw you over majorly. It's that simple. All the math shows what seemed to be logical all along.
EDIT: removed all the math to make the post more readable.
Last edited by gabrielrockman : 08/12/09 at 3:19 PM.
If you are casting chain heal, it takes 66 spells per minute (not possible) to get that 8.2 mp5.
That's only proper if you consider your CH only heals one target. WS is able to proc multiple times per CH. So multiply the numbers for manareg caused by crit of CH by the average number of targets you're healing... Less than 17 spells per minute for 8.2mp5 per % of crit is the right value with CH-glyph.
gabrielrockman - Just to reiterate what Sixty said, Haste =/= OOM. Haste is a burst stat, it allows you to have a greater troughput when it's needed, SP will only do that to a certain extent and certainly not as good as Haste, Crit is far too random to rely on. Just because you can cast 10% more spells, you don't need to - and that is why Haste will not screw you over majorly. If Haste have a negative Mp5 then you're undergeared SP-wise for that particular encounter.
In short, if you're going OOM, and assuming your OH is not stupid, then it's time to get some more Int, Crit, Mp5 - if you're not going OOM, then you want SP or Haste, and Haste just happen to favor some people that like a more reactive healing style, and as long as we have to care about mana then that's what I prefer.
08/08/09, 10:32 PM
Similar Threads
