"Like I said, I believe it's a proc designed to see if healers can/will utilize procs that they can't control vs cooldowns that they can"

I don't even get that theory. Asiding the already mentioned fact that healers can't decide to need some kind of healing to be spread around just because the weapon proced, Val'anyr's proc entirely relies on some amount of damage incoming shortly. It doesn't change a thing for tank healing in most cases: spam healing is spam healing. It can't be used "reactively" for raid healing just because it proc-ed. If Frozen Blows is happening, yes, it'd be great had it proc-ed at the very strat. No, you can't control this in any meaningful way without weapon swapping.

While I'm sure there are a lot of interesting conspiracy theories running around that are based on merit, I suggest this one is much simpler: Blizzard didn't want the weapon to be good so that you needed one. They "owed" healers a legendary. They came up with something that was better than what's out there, but not gamebreaking. If this is part of a secret plan to see how healers in the best guilds react to procs rather than using cooldowns -- healers don't really use many cooldowns as a rule and most hate clickable healing output trinkets -- then it's a bad secret plan. They should've made an encounter with a healing proc that occurred as a buff and seen how all healers make use of it.

Of course, Ghostcrawler seems to think that the way to make healing more interesting involves constraining mana and -- perhaps -- making spells take longer to cast (per his latest missives on the topic). So I'm not sure I want to speculate too heavily on where there going -- it bears no relation to the parts of healing I find challenging or fun.

An example of what Tinwhisker might be referring to is the proc on [Crystal Spire of Karabor] from TBC and introduction of the resto Shaman talent in Wrath of the Lich King, Blessing of the Eternals, to see how they design a proc based weapon and implement it. Although the two difer in specifics they end up being very similar in their design.

This is the exact theory I subscribe to. They had a lot of trouble in the past with legendaries being too powerful, so I think they tried to play it safe and under shot on this one (they probably think they didn't undershoot because of their blind love for RNG and obvious lack of understanding of high level healing). They also failed to make it particularly cool/fun/original by borrowing a mechanic from two already existing sources used regularly (divine aegis and scarab brooch), and implementing it in an inferior way to even those existing sources.

Do I still want one? Sure, it's an upgrade and is a useful tool for helping me do my job better. Do I think it's poorly designed, should be changed, and is annoying to use effectively? Absolutely.

For disc, absolutely yes, I would rather have the furious weapon. For holy, valanyr(minus proc) edges out the furious mace, but not by a ton. If I had to choose one or the other to use all the time and couldn't have both, I would choose the furious weapon; on the whole, it is a better weapon.

In practice, you still need both and you still end up using the furious/hard mode ulduar weapon more than valanyr to control the proc.

The "more healing at low hp" effect was present even in nax40 on the warrior set. Same as a similar effect to the val'anyr was present in aq40.
There's also nothing to watch about how healers use the proc. You can't design a strategy around the proc because it's random (I assume they didn't design a legendary thinking it will be hot swapped) and you can't react when it procs because most of the time the spells you use are dictated by the encounter.

Strategic reaction to random events is part of what makes a good player good. During the proc, depending on your class/role, it can change which spells and which abilities are optimal to use, even in the same situation.

That's my main problem with healing in the game right now: a player who was able to take advantage of the proc effectively would likely not be noticeable at all over one who just ignored the proc. A DPS player in a similar situation could at least prove they did more total damage to the boss, but with a healer, even using a parser that tracks absorbs, you can only heal for as much damage as a player takes, and if they would have survived anyway it makes no difference if it was done "skillfully" or "sloppily".

And to make it worse, ildon, this proc has a real potential to come at a useless time -- when the extra healing/absorption is not needed. While that's true of dps procs, many fights do allow for fairly sustained dps on the main boss target and therefore using a dps proc is something that's more consistently available (yes, exceptions apply). "Using" the proc of Val'anyr when you're on raid healing but there isn't a raid damage spike or you're on tank healing during a dodge/parry string is pretty much impossible. And those are often real cases.

I provided a spreadsheet to simulate the usage (Val'anyr - Hammer of the Ancient Kings). It is a pretty uninteresting weapon, truth be told. Is it a good weapon? Sure. Is it worthy of orange text? That is debatable. It also takes a substantial time to build based on your guild's luck. We've had Ulduar on farm for a significant period of time, and are sitting at 21 fragments.

I just don't understand why they felt the need to throw an RNG factor on an item that will be used by players who, by nature of the class spec they've selected, must play reactively.

Well, here's a report for you guys.

World of Logs - Real Time Raid Analysis

'Divine aegis' is, I assume protection of ancient kings, the val'anyr proc, as I was holy this raid. As you can see, the absorb is approximately 1.1 % of my total healing in this raid, which is kind of not what I expected for a legendary.

Perhaps you number crunchers can find fault with this; or maybe world of logs is reporting the proc incorrectly. But from this information the proc looks indeed underwhelming. Thoughts?

Note: My priest is James. Filter 'healing by spell' to analyse the data.

Eh.... Divine Aegis is a priest talent.

Protection of Ancient Kings looks to be the correct buff. Look here for the procs.

Considering that you had 130 procs of "Val'anyr: Hammer of ancient kings" which resulted in roughly 12.6% uptime, you cast over 3000 (!) Protection of Ancient Kings shields (including 90 on yourself). I have no idea what the average absorb amount of these shields was, nor do I know if they were effective or not... it appears the parser doesn't (yet) recognize this as a shield.

Edit: Congrats on getting your hammer.

Ye of course I'm aware it's a talent, not too aware of how world of logs works though so i assumed aegis was how it's taking the proc into account.

But ye - if the report can tell you all anything then by all means analyse it, can upload a few more if you like!

If you look at your other holy priest he actually has 1.7% healing done by divine aegis. It's a known problem with parsing sites for holy priests when there's disc priests in the raid.

Its a shame it isn't being recorded on worldoflogs in terms of amount absorbed, I guess WWS is the same? I'll try it tonights raid and see. Failing that, perhaps the recount absorbs plugin will give us some input - though I'm sceptical about that.

On another point, having the hammer makes healing alot more fun when the proc is up, it was a blast playing with it last night. Although a few people with bad computers complained about the shiny bubbles all over the raid reducing their FPS!

So i finally got around to doing some testing, Val'anyr, Hammer of the Ancient Kings (the hammer proc) will proc off complete overheals, but on direct heals only including the resto 4 piece. Otherwise it procs as usual off hot ticks that actually do healing.

Protection of Ancient Kings (the absorption shield) procs in the same way, off of direct heals, regardless of overheal or not and only off hots that do actual healing. However, the shield amount is based off the full hot tick, not off the amount healed.

Well World of Logs is now adding the proc to the healing done. Here is a report from last night.

World of Logs - Real Time Raid Analysis

I was incredibly hungover this raid so not really playing my best, but it seems to account for quite a large amount of total healing done. Looks good.

It may be adding it to the parse but it is currently not calculating the overheal (wasted shields) percentage. Going from my experience of divine aegis which works in a similar manner I would expect the overheal to be in the 50-60% range which basically halves the contribution of the proc. Also given that the shield is 15% of the amount healed that requires that your average heal to proc it is 7760 which conflicts notably with the average hit size recorded for your other heals. given that CoH and PoM which are your top spells only get close to that size on a crit I think they need to continue to work on how they measure the shield effect. Finally if you add up all the healing done by you main spells and add them together and multiply by 15% you get a value smaller than the total stated as being absorbed and that is assuming you have 100% uptime on the proc, not the ~20% recorded by the log.

So in conclusion I would take those numbers with a large bucket of salt and would expect the amount healed to be far less than reported in actuality.

*Edit*

If we assume your buff uptime is correct we can estimate approximately how much the mace proc actually did. First we find out how much healing total was done. (total healing done - proc healing x 1/overheal%) which is 32,854,000 for the entire run. Now we have an uptime for the buff of 18.4% so this gives 6,045,000 worth of healing that could proc a shield. Given that the proc provides a 15% return this gives us a total of 906,000 healing done by proc max over the course of the run.

Now we pick favourable overheal (wasted sheild) percentage of 30% which gives us a total absorbed by the proc of 634,000 absorbed. If we then recalculate this as a % of the total healing done (corrected proc total/old effective healing total - incorrect proc total + correct proc total) we get almost exactly 4% increase in healing done by the proc. This is well in line with the theorycrafted predictions.

This is basically only 1/3 of the total reported by the log. This means that either the proc does not work as advertised by Blizzard OR the logging tool is not calculating things correctly. I leave it to you to draw your own conclusion as to which of these you consider the most likely.

Now that we know the stats/ilvl for season7 arena/tier9 weapons and more people are starting to pick up Val'anyr, I would be interested to see a comparison of the healing increase granted by Val'anyr's proc over Constellus, versus the healing increase granted by Relentless Gladiator's Mageblade (or a tier9 PvE equivalent) over Constellus.

If Ellyh's assessment in the post above is correct, and the proc provides about a 4% healing increase, I could see weapons as early as tier9 being unquestionable upgrades to the legendary. The 137 spellpower upgrade from Val'anyr to tier9 likely provides a similar percentage healing increase (likely smaller, but close), and the bonus is applied to 100% of your heals, instead of determined by RNG.

I am getting our guild's 2nd Val'anyr (5 fragments so far), and I expect to get a Relentless i258 weapon around the same time I finish my legendary (arena rating requirements are a non-issue), depending on our luck with fragments and how fast the patch gets pushed live. I will not be amused if I already have a better weapon before I even finish collecting fragments. Here's to hoping Val'anyr receives a buff when tier9 is released, assuming the numbers work out as I expect them to.

Minor correction: for the i258 weapon, it should actually be 701 weapon + 23 gem - 587 valanyr = 137 spellpower, because epic gems are coming out by the time it's available. I assume they'll have the same spellpower as [Runed Stormjewel].

The Relentless Gladiator's Mageblade is probably easier for comparison as well.

Using the mageblade, the net change versus [Val'anyr, Hammer of Ancient Kings] would be:
+137 sp
+2 int
+1 crit
-46 haste
-proc
+31 stam

You also obviously have the option of going for mp5 if you prefer it with the Gladiator's Salvation, giving the pvp weapon yet another advantage. "Legendary" indeed.

What's with this "they owned healers a legendary" thing anyway? Even before this mace, the % on healers to have ever held a legendary is much higher than the % of the warlocks (don't forget that atiesh was usable by 2 of the healing classes, and that warlocks back in the day were only brought to raids to cast curse of the elements for the mages).

Just some numbers I got as a Paladin on a Mimiron 25 hardmode (wasn't a kill, though we reached enrage at about 5% phase 4 so it's fairly accurate of a real kill)

Using recount guessed absorbs:
total effective healing (regular recount): 910,000
total Val'anyr absorbs (gussed absorbs): 222,000

So roughly 20% of my healing was Val'anyr. I'm at work right now so I'll grab the exact overhealing I did when I get home, but my HPS was typical of top 10 Mimiron hardmodes for paladins on WMO.

It's interesting to note that around 50% of the time my sacred shield will do the same or more healing than Val'anyr depending on the fight. Mimiron displayed some nice numbers though due to plasma blast guaranteed absorbs on nuke heals and the phase 2 AOE raid damage (with HL glyph).

Please post a log as most of the tracking tools are consistently recording higher heals for this effect than is mathematically possible given the explanation provided by Blizzard. Also given the widespread nature of this discrepancy it would be great if someone could run someone could run some controlled tests to see if the 15% of heals - bubbles is in fact correct.

example: for those numbers to be accurate with the 20% uptime which seems to be standard for this mace to get that much shield healing you would need 222,000/0.15 to generate the total healing that procced the effect = 1,480,000 but uptime is only 20% so we need to multiply this number by 5 to get your total healing throughput for the fight = 7,400,000 which translates into an overheal % of 88%!

I know pallies tend to overheal a lot but an 88% overheal rate for a hard mode strikes me as a little but suspect.

I think the most likely culprit is the parsers having no way to distinguish between Sacred Shield, Power Word: Shield, Divine Sacrifice, Ice Barrier and the Protection of Ancient Kings.

Previously, you could assume the contribution of Sacred Shield separate from Power Word: Shield because both had relatively fixed values, so if you saw an abnormally large absorption you could deduce which was which.

But with Protection of Ancient Kings being so random, it's hard to tell if the absorb was purely from the Protection of Ancient Kings or some combination of absorption effects.

We host the logs on wowmeteronline which says nothing but the number of procs and uptime of the Val'anyr proc.
80% overheal on Mim hard mode is very attainable by holy paladins. Check the top HPS: WoW Meter Online - Combatlog Replay
(especially true after the massive nerf to the fight)

This is the parse from last night I was looking at in my recount where I got those numbers from GuessedAbsorbs (it's since been overwritten :\): WoW Meter Online - Combatlog Replay
24% uptime of the actual shield on people, 29% uptime of the buff on myself.

As far as the accuracy of GuessedAbsorbs, I've compared the sacred shield number with another holy paladin's at the same time and they seem to be similar, and the Val'anyr absorb number is very distinct from the SS absorb so I tend to believe it's somewhat accurate. I know it's impossible to tell exactly "what" is absorbing but it's a pretty good ballpark.

Also, the uptime of Val'anyr can vary quite a bit it seems to me. The proc percentage is only 10% once off ICD... you can go a pretty decent string without seeing it at all after the 45 seconds.