It would be good if someone would /combatlog the Berserker testing, then post to WWS.


It is good that you confirmed that the enchant refreshes and has no cooldown.

This is certainly not a controlled test on practice dummies, but I did have the Berserk enchant in use during our Sarth+3 kill. Feel free to browse the log at your leisure -> Wow Web Stats.

16 procs in 525 seconds on your kill. However, it was only up for about 166 seconds because of it refreshing itself before the duration expired. So, it had an uptime of 166/525 = 31.6% which yields .316*400 = 126.5 static AP, still better than Massacre.

What I found more impressive was your Try 11, where it proc'ed 7 times in 148 seconds. It refreshed itself 4 times which led it to be up for 68 seconds and was up for nearly 46% of the fight. This was probably on the luckier side of the tracks, yet still evidence of it being possible.

Great to see the thread reborn with a great front page for new players. I was halfway through creating one myself but got sidetracked so it's probably a good thing someone else picked it up =]

I would assume the priority would mainly be based on which ability would be more beneficial to be off cooldown sooner, ie which ability would give more dps if maintained to its normal cooldown time as much as possible.

While the numbers supplied are probably quite far from the reality of things (shouldn't crusader strike be only 10% better than DS?), it should be easy to see that one ability out shines the other simply because you'd want it to be up off cooldown sooner.

CS (on cooldown)
1650 / 6 = 275 DPS

CS (after GCD delay)
1650 / 7.5 = 220 DPS
Loss of 55 DPS (20%)

DS (on cooldown)
1500 / 10 = 150 DPS

DS (after GCD delay)
1500 / 11.5 = 130.43 DPS
Loss of 19.57 DPS (13%)

Bit harder to get a comparative number for Judgments, but if we want to assume 5k average per judgment then:

Judgment (on cooldown)
5000 / 8 = 625 DPS
5000 / 7 = 714.29 DPS (4pc bonus)

Judgment (after GCD delay)
5000 / 9.5 = 526.32 DPS
5000 / 8.5 = 588.24 DPS (4pc bonus)
Loss of 98.68 DPS (15.79%)
or 126.05 DPS (17.65%) (4pc bonus)

For a real analysation, throw in some real numbers from your own WWS report.

Moving along, about the SoC change...

Plugging my woeful numbers in as a test for weapon ranges.
AP - 2578 / SP - 773
Colassal Skull-Clad Cleaver - 1237 - 1515

Old Proc -
0.45 * 1237 = 556.65
0.45 * 1515 = 681.75

Average proc = 619.2

New Proc
0.45 * 1237 + 0.45*0.23*773 = 636.66 (14.37% increase)
0.45 * 1515 + 0.45*0.23*773 = 761.76 (11.34% increase)

Average proc = 699.21 (average 12.92% increase)

Surely that gap between SoC and SoB just closed in... altho SoB is getting 5% increase (22% Weapon dmg -> 27%) as well, so maybe not so much. Even if SoC was slightly better, I guess we'd still use SoB for the recoil dmg + heal for mana trick. I guess the next question would be if the JoW change makes up for the SA mana loss if we switched to SoC.

FYI, SoB on live seems to have 27% weapon damage, whereas SoC seems to have a 12~% SP coefficient. So the SoB change is just a tooltip update, whereas the SoC change is a buff.

For reference, here are the values I observed. I went unarmed and whacked a few mobs.

Weapon damage: 360~361
SP: 703
SoB proc: 97 damage
SoC proc: 247 damage

Effective SoB weapon coefficient (Tooltip indicates 22% weapon damage: tooltip is wrong)
97 / 360 = 26.9%

SoC SP coefficient:
WD contribution: 360 * 45% = 162
SP contribution: 247 - 162 = 85
SP coefficient: 85 / 703 = 12.0%

Sorry, wrong thread.

This is pretty significant. If it's only a tooltip correction, they are giving us false "buffs". To buff SoB the same amount 22% -> 27% would've been (~12%), the new SoB should be around 33% weapon damage. This gives us roughly 2% overall dps boost.

If developers saw that we could use 12% boost to our SoB damage and all they do is update the tooltip, that goal obviously isn't met. Anyone with US account who could ask official forums what's up with this?

If I remember correctly they quickly reverted the last change as it was promised by Ghostcrawler that SoB wouldn't get nerfed but it was nerfed in the ptr/beta build anyway. Live never saw the 22% version it seems.
MMO-Champion BlueTracker - Retribution Changes - Please Read

@Duopoly and HamSlammer - excellent - exactly the sort of data we wanted to see. And Ham's already broken down the math.

Looks indeed to be the clear enchant of choice. Multi-proc is confirmed, no internal cooldown. As we wind up with much larger data in the future we can decide if it's PPM or just % chance - I leave that up to those actually able to run rigorous statsistics calculations. For now I'd just model it as a rough 1/3 (33%) uptime or about 132 AP on average.

I still say use Greater Savagery on "cheap" weapons and then when you're satisfied you have something worth the mats, hop straight to Berserking.

Note: Those of us who prefer plate to mail or leather - this actually raises the value of the lesser armours (if their stats are worthwhile). If Berserking is proccing then our armor is going down by 25% a good portion of the time anyhow (~5% mitigation loss, from 45ish to 40ish percent reduced). Until we get more comparisons I still expect a good portion of plate to be leaders due to large quantities of Str.

I can confirm what Fiola said, it is actually just a tooltip update. However, it's not the first time something that was "hotfixed" is later included in some patch notes as a new change, keep in mind not everyone follows blue posts religiously.

I guess adding a "previously hotfixed" label to their patch notes could be a better way to do things.

I'm assuming this is quoted from here.

Remember, these are the undocumented changes - not patch notes. Same goes for SoB at 27%. SoB and SoC are not mentioned in patch notes at all - these are likely to be tooltip updates to match hotfixes which previously occurred.

If it's not just a tooltip update (and tests were somehow skewed) and if we're currently getting 14% spellpower on SoC then this is a nerf, not a buff.

.45 * .23 = .104

SoC would go from 14% spellpower to 10.4% spellpower. That looks like a nerf to me. Not that SoC is our preferred seal.

This is given, of course, that the tooltip isn't wrong (.45 * .23? Why write it in that fashion?). If it's increased to 23% it's a buff.

Does the wording of this mean it might affect ALL seals and judgments now? I would like to believe yes but it seems like too big of a buff.

"Judgements of the Pure: This Holy talent now increases the damage done by Seals and Judgements."

edit: Confused it with the T1 talent of a similar name. Thank you for helping me see that.

On that front: a 25% increase to seals and judgements will equate to a roughly 5% DPS increase? Since HS remains the single largest DPS ability (and using it comes at a cost of healing mobility - unless it crits, and resilience should soon make sure it won't) we have, I don't see this number improving things to even where they were before the 3.0 reworking.

As it is, my SoR hits for 177 and Judges for 700 (Judging with SoW up actually isn't that far off and gives me the added benefit of infinite mana while soloing - though that will change too with the changes to SoW), a 25% increase would bring it to 220 and 875... still not the numbers we were getting in BT gear at 70. And, in PvP at least, these hits are going up against healthpools that have easily doubled.

The problem, of course, is that they had to gut whatever SP scaling we had on any ability because of Retribution's Sheath of Light talent. Reconverting an existing seal back to exclusively SP scaling and adjusting to make Holy DPS on par with other healers would be a nice, simple fix... though maybe too involved at this point in the game. SoR is an unlikely candidate as it would result in atrocious scaling while levelling. Late game Seals seem more appropriate for such treatment, but flipping SoB for SoR and scaling SoB with SP would probably be rocking the boat a little too much. I could see rolling Aura Mastery baseline and replacing it with a holy/SP scaling-only seal however.

Additionally, since they are changing JoW to scale of base mana rather than total mana, does that mean we should expect SoW to be similarly treated? Using SoW while soloing wasn't a noticeable DPS drop and allowed to spam every DPS ability we had on cooldown, which is pretty much the only way to make soloing tolerable as holy currently.

EDIT: I forgot which thread I was writing in... could this be moved to the Holy thread as it rather lacks Retribution specific content. Please and thank you.

Yes, the SotP buff for Holy is about a 5% damage increase.

Note Holy Pallies can use Rank 1 and 2 of Sot0R (I don't know if that is removed on PTR), and with JoW + Plea + SoR as Holy I could spam every attack without mana issues (this is before JoW is nerfed for Holy on PTR).

Another "attack" is being added (single target taunt that does a little holy damage), that should help a bit as well, especially for pulling mobs when Judgement is on cooldown, or working as a ranged pull for Ret.

SoW is likely staying the same, since Holy is encouraged to use it in raids.

Because the developers have to balance somewhat for leveling and three specs, the base damage of SoR isn't going to be messed with.

I was under the impression the additional hand spell was going to have a trivial amount of damage associated with it simply for tagging purposes... something akin to a warrior's throw ability, along with a taunt mechanic. I doubt it will be noticeable DPS. Additionally, if using both ranks of ShoR, I tend to run out of GCDs to use in some scenarios (undead mobs).

EDIT: just found it buried in the patch notes, 2 rank ShoR is in fact being fixed next patch.

First, thanks for this great post. It's nice to have all info placed within 1 thread for easy access.

However I have a problem with the spreadsheet from Bellator that is linked in the first post and was wondering if anyone else have this too. Whenever I press "Auto-Gem", "Show Enchants/Sockets" on the Character tab or "Analyze items" on the Item Calc tab I get a "Run-time error '1004': Method 'Range' of object '_Global' failed".

Anyone else seen this or know what is causing this and how to fix it? Sorry if this isn't the proper place to post this but I searched for a thread dedicated to the spreadsheet but couldn't find any.

Thanks in advance.

Edit: I am using MS Office 2007 btw.

Thanks for presenting this concept in this way. Was much easier for me to wrap my head around like this…

One point I’d like to throw out there is that you didn’t include Blood/Martyr damage in the CS/DS dps calc. Plugging in my own numbers (including martyr for cs/ds) from recent Naxx run, I come up with:

CS (on cooldown)
3129 / 6 = 521 DPS

CS (after GCD delay)
3129 / 7.5 = 417 DPS
Loss of 104 DPS

DS (on cooldown)
2967 / 10 = 296 DPS

DS (after GCD delay)
2967/ 11.5 = 258 DPS
Loss of 38 dps

Judgment (on cooldown)
6990 / 8 = 873
6990 / 7 = 998 (4pc bonus)

Judgment (after GCD delay)
4568 / 9.5 = 735 DPS
4568 / 8.5 = 822 DPS (4pc bonus)
Loss of 138 DPS
or 151DPS (4pc bonus)

Cons (on cd)
477x10/10 = 477 dps

Cons (after GCD)
4770/11.5 = 414
Loss of 63 DPS


This would lead me to prioritize judgment, except that the margin is so small that if prioritizing CS reduces the number of conflicts by even one, then CS is the way to go. So it seems pretty obvious that the 4pc priorities should be: CS=>Judge=>Cons=>DS. (which maybe everyone already knew, but I wasn’t sure. And if I screwed up the numbers somewhere, please let me know.)

Now to train my fingers to stop prioritizing judgments…

Edit: My judgment math was off. I forgot to account for Righteous Vengeance. Wws also had an abnormally low crit rate, so I increased that a bit as well. I think the conclusion remains the same though. As long as prioritizing CS illiminates two or more conflicts (which it does with 4pc bonus), then it's the way to go.

For those interested in the new Kirin Tor ring (http://static.mmo-champion.com/mmoc/...55kirintor.jpg) it's at a first glance an obvious sibling of Ruthlessness and Circle of Death, with slightly worse stats. Using Redcape's values puts it at a score of ~13500 (lootrank for quick reference), just above the normal Grobbulus ring (and the other Kirin Tor pdps ring) but with most of the higher end rings above it.

In short, if you'd be willing to spend the gold, haven't already done so on the Band and is doing normal mode Naxxramas only rarely or not at all it'd be a decent enough buy, but if you are doing heroic Naxxramas it'd only hold off upgrades in favour of other slots for a while.

Glad that it helped, the numbers weren't exactly real but just examples on how ppl could see it plainly whats the better options if they just plug in some of their respective numbers and also how the community decided that a J -> CS -> DS priority during clash was the way to go.

With your own numbers, looks like you've got a little error there that confused me at first.. but the math worked out to be correctly based on the first number for judgment.

I'd still choose Judgment myself, its crit rating is far higher, it gives mana back and with 4pc, it tightens up the rotation so there's less conflict (as long as you actually keep it on cooldown).

I guess the main point that needs to be made is that this priority is considering one GCD only and due to the staggering cooldowns, your abilities are -always- going to conflict, if not now then 30-40secs from now. The most amount of DPS in FCFS rotation generally follows that you keep all of your main abilities on cooldown and use secondary abilities (consecrate, exorcism, divine plea) when they're not going to come up for another GCD.

How pedantic you want to be is up to the user. The next step would be how much you should stall so another ability comes off cooldown that would potentially give better DPS.

This is, of course, in a glorious perfect environment where you can stand still and DPS consistently. Not the smelly realistic world we live in where latency, crit rating, raid makeup, RNGs, moving boss fights, computer lag, human brain patterns and spaghetti monsters affect your perfect rotations (which are 2-3mins long given the cooldown structure).

So i just had a question about geming. Would it be ideal to get hit/expertise cap first and then start to gem for str/crit? I've seen some paladins getting to the hit/expertise cap first. Seems like its more of a preference or would it actually be an increase in dps to go for the caps first?

I don't believe that you can gem expertise.
Hit rating, point for point is better than any other stat, it will increase your dps more than any other stat until cap.

For pvp, it's preference.

You can gem expertise but I personally wouldn't recommend it. I also wouldn't gem for hit because it is so easy to achieve without and strength gems are superior to use.

Anecdotal evidence: I replaced AP/Str/crit gems with Expertise and hit gems to the cap. I lost about 200AP and 1% crit, but went from 2000 DPS to 2400 DPS. It would appear that, if it is not ideal, ensuring 100% attack connection is certainly a very significant boost.