The only time I don't put a straight spell power gem in is if the socket bonus is also spell power. In those cases I am giving up a small amount of spell power for overall stat balance weighting. It's not that I NEVER choose regen, it's that I do my best to prioritize spell power. I also have Darkglow instead of Lightweave embrodiery. I have thought about switching that up for Lightweave and probably will once I get Solace as well. I am riding a line on some of the more aggressive hard modes right now and using Innervate at every 3 min cooldown on those so I have to be thoughtful about how far I push the throughput envelope.

Also, I'm slowly replacing whatever normal gems remain with epics -- alchemist so I am transmuting as I go. I'll probably just buy the rough gems though and then transmute to make the money back so I can finish these off before next raid night.

You're contradicting yourself. You agreed that Ember Skyflare should only be used when it is the FINAL sacrifice of regen for throughput.

Thank me for this math: after epic gems, but following your current gemming theory in [Conqueror's Nightsong Robe], you're choosing 19 spellpower + 10 int over 23 spellpower. That is exchanging 4 spellpower for ~5 mp5*. In [Conqueror's Nightsong Handguards] and [Boots of Hasty Revival], you're choosing 19 spellpower** + 10 spirit over 23 spellpower. That is exchanging 4 spellpower for ~4 mp5***. So, if those gem choices are:
Then certainly a higher ratio would be better? *1 Int is worth ~0.5 mp5
**1 spirit is 0.194 spellpower, I rounded up
***1 spirit is worth 0.374 mp5 (these #s are from the 1st post in this thread)

Doesn't it do the animation every time it procs? Seemed like it did the animation each time it restarts the 9 seconds. I'm at work so can't check.

I'm pretty sure I didn't say FINAL but we're splitting hairs. At any rate, I've made it a general methodology on item gemming to match gem sockets with spell power as a socket bonus and, according to current stat weights, there's nothing wrong with that -- it actually works out to be correct from a stat weighting perspective. Here's the inconsistency, currently a pure stat weight methodology would also rate Earthsiege above Skyflare as well but on the meta, I've ignored stat weights and gone pure throughput.

Could I reverse methodology and go with the stat weight choice on the meta and max throughput on ALL other gems. Well, no because of the meta requirements of Earthsiege I couldn't. I would need to keep at least 2 non-red gems. So lets say I went 12 sp 10 INT on chest (7 sp socket bonus) and gloves 12 sp 10 SPI (5 sp socket bonus) to meet those requirements. So, now I can swap a foot gem and a belt gem to the tune of 12 additional sp more than I have in my current gemming methodology. But I lost 25 sp on the meta by using Earthseige. So I'm down 13 sp from my current methodology and I have some extra wasted regen that I don't need.

So going Earthsiege doesn't really do anything for me but cost me some money in gems. Could I possibly go all red gems and Skyflare. Maybe, possibly I could. We're not talking about any significant regen here though I do like the mixed SP / INT gems for Vezax (again not that these are make or break). It is simply the methodology I've chosen. If I found I really needed some extra regen to make it through fights, Earthsiege is the first change I would make. Until then, I'm fairly comfortable with where I'm at whether you find it to be consistent or not.

Well the whole point is that IED vs. ESD gives you a gigantic amount of regen for how much throughput you lose, and it is better than almost every other possible tradeoff in the game. Which means that if you are using ESD and darkglow, you should ASAP switch to lightweave even if it means you have to switch metas to keep your mana up. This trade will gain you about 22 spellpower *and* 35-55mp5. It's a no brainer. IED gives you 2-3 times the regen of darkglow, and costs you just over 1/2 the spellpower vs. the alternative throughput option. If after getting rid of darkglow, getting rid of all your regen trinkets, equipping the new idol, and resocketing all your gems for pure sp except as needed to activate your meta or for extreme socket bonuses (like +8sp or more), you *still* have no mana issues ever, *then* you swap metas.

It's better in some cases to wear hitless *caster dps* gear in place of natural healing gear for more throughput than to swap out IED. You really need to be in a place where you have made very dramatic tradeoffs of regen for throughput elsewhere before swapping to ESD makes sense.

Thinking in terms of tradeoffs is critical when making these decisions. For every possible tradeoff, look at how much throughput you gain for how much regen you lose. What kind of mp5 : sp ratio is it. Make a list of all the places you can do this and make your changes in the appropriate order until you have just enough mana to feel comfortable.

The Lightweave / Darkglow piece is something I honestly didn't take enough time to look at previously and only recently had begin to think about swapping it out. Clearly I should have spent some more time on that because, prior to swapping my idol out yesterday, honestly I probably could've put Lightweave in place of Darkglow and probably even still kept ESD in place of IED. If it made people happy, I could've also made every gem an SP gem as well to be fully consistent throughput and still without regen issues.

On the flip side, if it made folks happy, I could also make a simple swap of IED for ESD and be full consistent stat weighting. But, I don't believe you have to be 100% consistent stat weighting or 100% consistent throughput but can run a hybrid of the two -- at some point, you are fine-tune tweaking your setup one direction or the other. The way I described a few posts above was how I was running that hybrid setup. The only thing I really regret about it is not looking at Lightweave sooner.

That said, with the recent addition of Flaring Growth idol to my setup, at least until I can get Solace, I have a sneaking hunch that I'm probably going to have to take a look at adding in some regen to my gemming if I want to keep both throughput trinkets in place of regen trinkets. I haven't run a raid since I got it, so I'm not sure how it's going to go. But based on my calculations, losing the Rubber Ducky is about a ~150 mp5 hit or so given a sustained cast rate of around 25 Rejuvs per minute (which is what I've seen average on the parses of the most RJ heavy fights).

Due to this hit, I am going to have an IED pre-cut in my bags for the next raid as IED replacing ESD the first change I will make if I find I do need a significant regen boost. But, once I get Solace and/or other T9.5 gear evens out any regen hiccup from using Flaring Growth, I can very much see putting Lightweave/IED combo as you've described it in. Then, perhaps with double Solace (25 normal and 25 heroic versions) making any regen issues a thing of the past, I can see dropping IED for ESD again as well.

In my opinion, it will be a minor buff to crit for trees but not much more (so maybe 0.2 instead of 0.15 in stat value). Unless you gear in a really unusual manner, you're still not gonna get a whole lot of listed crit. This set bonus really needs a talent or glyph to make it shine. As it stands, even with around 500 crit rating from gear (which would be very high for us), this set bonus is still FAR worse than that of the tier 8 set. Only the increase in other stats make tier 9 worth taking. Personally, I don't plan to transition to tier 9 until I have 4 pieces of iLevel 258 tier 9.

More to the point on T9 crit - I'm terrible at game maths, but it feels like breaking 4 t8 bonus is going to take a lot of itemisation/much higher ilvl gear to equal the gain of the set bonus. I've seen the proc accounting for 10% of my healing on random WWS logs.

Any thoughts on when, or more accurately which group of ilvl (232, 245, 258) t9 it'll take to drop 4/5 t8?

I doubt many of you haven't ended up using the 4T8 heal as a means to snipe heal where it isn't needed ie a raid member just took 4k random damage and you just RJ them instead of leaving it or using Nourish for example.

Just going by the meter results for the healing done from 4T8 is misleading due to the above and also the fact you will often have just cut off a RJ tick by having the initial heal early (so you heal at 0 and 3 instead of 3 and 6) and while sometimes that is useful many times it is not. We could even test this now that we get overheal results in the combat log.

i dunno i'm having mixed feelings about the t9 bonuses. to me crit was my least interested stat, due to my top heals being rejuve/wg and occasionally lifebloom if i end up tank healing which i dont let it drop. so for now i'm thinking stickin with 4pc t8 and waiting until my 4th piece of t9 to switch. maybe even waiting for the heroic 25 ilvl258 t9 even.

If a lot of your healing is sniping (defined as ''healing something which would have been healed anyways very quickly'') it means either your healers are uncoordinated, or you are bringing too many healers for the encounter, or both. When healers are pressed for HPS, the 4t8 fully counts for increasing this HPS (and is less likely to overheal than a hot crit if you are intelligent about where you place new Rejuvs). When healers are not pressed for HPS, it doesn't matter what you wear.

It is really annoying that we shouldn't break our 4pT8 until 4pT9. I got a trophy today and have 32 badges, but I guess I will buy one of the rings-they are big upgrade for me. We go from have 2 bonuses that are good to maybe a 4pT9 that might be decent. They should have at least given us a 2pT9 that was decent to help with the transition. I guess I could get the gloves, but the ring seems better.

Well at least they put up a good idol and some rings to give us some upgrades before the 4pT9.

I'm still using the idol to reduce RJ mana cost and it's cheaper to cast than a nourish, and can also provide a minor buff through revitalize. So even if most of the ticks are overheal, the GCD was best spent on RJ. If the group is bringing too many healers then yes another heal might land in that 3 second window, but normally a balanced group will have the healers on the same page there, but the target might take another bit of damage that the remaining ticks can heal as well.

No one idol is ideal for every fight. If you aren't switching them in and out based on your past experiences with the fight you probably aren't doing yourself any favors. The new idol is great for fights like mimiron or yogg and just a bad idea for fights like freya or xt. (all hardmodes of course.) Part of being a healer is to compensate for the unforeseeable. If you are going in with pure thoroughput you aren't going to be able to compensate for possible mistakes / bad rng you aren't any better than a meter padder.

Overall, I am happy to have 2 idols to choose between, and as for the tier bonus I am not as excited for critting hots as I thought I would be. I think they will be great on paper and not so great in play. I haven't decided when I will drop T8, but I don't think it will be for a good while.

Well, is 4tp really worth wearing? Sure, it will bring you some additional HPS, but with ~50% OH most of the additional crit is wasted anyway. Instead you could wear full Ulduar / Coliseum hardmode loot (nonset pieces) an get lots of SP (and better stats weighting since you can choose excatly what pieces of gear you want to wear).

Maybe I'm just bad, but I don't really see any massive benefit to 4t8 like everyone else, and I'm looking forward to 4t9..

Except for a select few fights: IC, Thorim... and probably mimi but we haven't done that hard mode yet..

I usually already have a rejuv ticking when someone takes damage, so with the exception of those fights where everyone is constantly taking massive damage, the initial tick is just wasted. I think the rejuv crits are going to be more throughput in more situations.

Personally, even if you aren't having any mana problems, I would still say take spirit+haste over haste+crit items. At least until you hit the 4-piece T9 bonus, crit doesn't mean too much to most resto druids. I don't know what kinda gear you have with your druid, but offhand, I would guess your trinkets and idol are efficiency models. With the new flaring growth idol and two throughput trinkets, it would be tough to make it through IC hard mode or heartbreaker without mana problems (especially since the replenishment nerf).

Of course this is just a guess without knowing anything about your druid. If I knew the character's name and server, I might be able to give more specific advice.

Wait...why exactly is the new idol a bad idea for freya/xt?

With 458 haste I am finding just how usefull haste is, but the diminishing returns of the haste is getting worse.

At how much Haste should I try and avoid haste and go for crit or regen? It would probably get to the point where the haste has almost no effect.

My nourishes are already a 1 second cast now with shaman nearby, have I hit the "hard cap" of haste?

There is no diminishing return on the absolute value of haste -- each point will reduce cast time by the same amount. This is really more about a diminishing return on the relative value of haste.

Relative value is subjective but here's what it boils down to for me. I just finished a night of ToC and Ulduar. Parsing the log, no less than 97% of my throughput came from RJ, WG, LB, and Swiftmend. Only 3% of my throughput came from Nourish or Regrowth. So up until 359, each point of haste is increasing my efficiency on all spells and has a positive effect on 100% of my throughput. Past 359 and a 1 second GCD, each point of haste is only increasing my efficiency on the spells that make up 3% of my throughput.

Based on the above, as a druid raid healer, assuming 5/5 GotEM and the buffs that are commonly available in any serious raiding guild, my opinion is that haste past 359 has far less value to me. I'm not saying it has no value or to avoid it (it does continue to show value on cast time spells still as well as reduce the GCD of decurse, abolish). But, as haste past 359 affects less than 5% of what I do, it absolutely becomes less of a concern for me to purposefully stack or seek out.


EDIT: Now, the above is all assuming I'm raid healing in 25s. If I am healing 10s (just Firefighter and 1 Light Yogg from 10-man meta achievement), haste does have more relative value to me. In 10s, Nourish/Regrowth peak up to over 10% of my throughput and I definitely get value from them channeling faster to land on a target in time. However, my primary focus in gear / spec decisions revolves around my role as a raid healer in 25s.

You can easily get oom on xt/freya even with old idol and spark, so new idol is bad idea on them.
But on jaraxus or something like it the new idon owns, so i thing ist "must have" anyway.

Every throughput boost will contribute to overhealing. A crit is more likely to do some than SP, yes, but since rejuv ticks aren't as big as other classes heals then it should be pretty smooth.
The heroic T9 gear is ilvl-258 so the hardmode non-tier loot isn't going to be a lot better. 4T9 will surely be BiS gear until IC.
And lastly and most importantly, our top healing spell scales with crit (assuming 4t9 is actually based on your real crit-%). This is a huge difference, and the right method will be to gear until you're haste-capped with 5/5 GotEM and then stack crit. Right now neither haste (past 359) nor crit offer any meaningful gains. Getting gear upgrades that actually mean something == good.