I don't know how you run 5-mans but I tend not to pull the whole instance without going out of combat for at least some of it - certainly true before you outgear the instance. Do you mean that the cooldown only ticks while you're OOC?

I think that the situation where you're unwilling to use pots on cooldown but you're also unable to use SoW to regen mana once in a while in a 5-man is a rare one, I definitely can't think of anywhere in TBC that's true - closest example I can think of is the Shattered Halls timer, but that's a one-off, so I'd be willing to pot on cooldown for it. (I'd also be willing to stack a group for regen, but I acknowledge that's ducking the question.)

If you want to rule out using SoW, you also have to put us in the same bucket as hunters, mages, and (to a much lesser extent) warlocks, all of whom have active mana regen that lowers their DPS.

Your model doesn't include OOC regen, by they way, which isn't a huge source of mana for us but in a 5-man isn't negligible either, and also doesn't include SA regen from S/JoB. I also don't think factoring in the cost of blessings is particularly reasonable, since in any situation that matters you'll greater buff (or use the minor glyph), drink, and then start the timer. Mind changing your model to reflect those and seeing what we look like?

Edit: actually it's not really fair of me to ask you to change your model to accomodate sliding values from SA regen and OOC. That said, without your tool taking them into account, your tool model the lower bounds: not time-to-OOM but "you won't possibly need to stop inside this time". Given that the worst-case rotation, in a worst-case group composition, with a worst-case mechanics is OOM in 3 minutes, and with level 80 water we refill mana in just over 10s...I don't think we're in a terrible state.

They're not melee DPS. Different mechanics.

I'm actually trying to put together reasonable level 80 numbers to calculate 5-second-rule regen and Spiritual Attunement regen from damage taken by Seal of Blood. I'll hopefully have this done over the weekend.

5SR regen? With our spirit being abismally low and having CS in a 6s cooldown, you won't be getting much out of it. At least that's what I think. Our rotations are so tight right now that I just can't see a benefit coming from it.

On the other hand, I don't see why it's so hard to calculate at least the minimum amount of SA regen that you're getting from an encounter. It's 30% of the damage you deal with Seal and Judgements, so you could at least get a ballpark figure of the minimum regen you should be getting. Is there something I'm missing?

I agree 5SR will be minimal if even noticeable. The reason why SA is difficult is because the model is built around nothing but mana - no damage. I'd like to model damage from SoB to be controllable, but I have to determine some base statistics and code damage calculation as well as health. =P It's just outside of the current scope of the tool until I get more time to spend on it (like when I'm not at work =X).

Admittedly my math skills are not so advanced, but my understanding of FCFS is that you move from ability CD-based actions to GCD-based actions. I don't understand how we "lose" casts of CS if every GCD is filled. I am not experiencing a lot of dead space waiting for a bunch of abilities to finish their CD but rather seem crowded by the GCD. I may be way off base, but if your GCDs are filled, you aren't losing DPS by chosing the higher damage skill over the lower damage skill.

It's all about clash resolution. When ability A and ability B come out of cooldown at the same time, which do you press? Assuming you press A everytime first, ability B would lose 1.5 sec DPS time (1 GCD) every clash resolution over the course of the fight.

At the end of the fight, you delayed B by X times 1.5 seconds, where X is the number of clash resolutions.

In our specific example, delaying CS 8 times means you would lose 8 x 1.5 = 12 seconds of CS time, or 2 casts of CS.

If you delay DS 8 times, you would lose 8 x 1.5 = 12 seconds of DS time, or 1 cast of DS. Going for the limit, it boils down to losing 5 : 3 casts of CS : DS.

Hence, it's more favorable to delay DS than CS, as you would only lose 3/5 as many casts.


Edit: Can 3 DS casts be worth more than 5 CS casts at a high enough crit rate?

It would look like this:

Meaning 3 DS casts will never outdo 5 CS casts no matter how much crit you have. Also note: Higher crit actually increases the gain of 5 CS casts over 3 DS casts even further.

Edit: Some BS corrected. Blame notepad.

I understand this. What I don't understand are the assumptions, the field of play for this scenario, as it were. If we have a wide open space with a 6 sec CD and a 10 sec CD where they clash every once in a while, your point makes sense. But if we have a field of GCDs and we are talking about keeping them filled, it would seem to work differently. We have 40 special attacks per minute. If we use every one of them (which I think is possible on an undead boss), they using the highest damage attack for each one should yield the highest damage because there are no "lost" attacks. It seems that to make sense of the "lost" attacks, you need to treat each attack as a dot when it is really an instant attack with a damage-dead CD.

On a non-undead (which is going to be the default assumption), you'll be doing 31 attacks per minute so this is not the case.
On an undead (+Exorcism) you'll be doing 35 attacks per minutes, still not the case.
If you want to use Holy Wrath (extreme example), you'll be doing 37 attacks per minute, still under 40, so again it's not the case.

In all those examples higher DPS first > higher Damage first.

The last 20% of a bossfight where you use HoW is a special case, but we should really be worrying about what we're doing 80% of the time, right?

I agree that we should be examining what is to be done the majority of the time. But I am still trying to figure out how much "free time" it takes to make treating instant attacks like DOT uptime a valid assumption (especially based on Dalm's post here: http://elitistjerks.com/f31/t17193-p...40/#post936133). That is the math I would like to see. I apologize for hanging out my own assumptions.

As explained previously there are 2 levels to break this down.

Logic and simulation: Logic we've explained above: FCFS DPS

Simulation (to figure out lucky "clash avoidance" patterns) is done in the post you linked. However, it's is unrealistic to map out a 2 min long chain of attacks and except 0 mess ups/latency at the hopes of gaining 1 less GCD clash here or there, it doesn't happen and you lose any perceived advantage.

I really don't know how much more we can talk in circles about this, it has been discussed over dozens of pages previously (the addition now being the new mechanic of RV and that has been resolved) and that conclusion was reached:

-Casting CS is in almost all cases more beneficial than casting DS first (unless multiple mobs obviously)
-FCFS DPS is the way to go

It occurs to me that if we are switching back to SotM/SoB from Casino that the Felspine would pull ahead of Torch once again, as Torch was only better due to its speed and SoComm correct? Interestingly, Maxdps is listing Cat's Edge as better than both with SotM, but I'm unsure if that is still valuing ArP too highly.


I've also been thinking about leveling, and I'm pretty sure it will be easiest to level in pvp gear because of the int. It will be less damage, but with the extra int should cause less down time due to greater returns from JoW. Am I off on this evaluation?

Don't trust Maxdps when things are close. With SoB, Felspine should be better.

Using PvP gear is good and resilience removing crits isn't a bad thing, so why not use some? Note special cases, like great mana return from 2 Tier 6 and the superior stats of Sunwell items, even though they do not have int on them.

It shouldn't change anything. SoB procs have the same overall mechanics of SoC, in that the proc damage (and thus weapon scaling) is proportional to your weapon speed. This matters because we get 16 bonus SoB procs per minute with CS & DS. A slower weapon speed increases the amount of bonus SoB proc damage from CS & DS. Since CS & DS bonus procs account for roughly half your swings and half your seal damage, lowering weapon speed by 10% results in a 1/2 * 10% = 5% reduction in seal damage.


edit: On SoC vs. SoB, wouldn't SoC's ability to proc off of JoC even things up on single targets? (Assuming that's going to stay in the game)

Do holy attacks work off of the same hit mechanics that physical attacks do?

Seal of Blood can proc off of Judgements (SoC does as well).


Things for Paladins that use spell hit (everything else is melee hit):
Consecration (only the first tic I believe)
Shield of Righteousness

That isn't the case however at the moment, only SoC is proccing off Judgements.

Has this changed again/been confirmed as an unintended bug or what are you basing this statement on?

It doesn't happen 100% of the time, but I cast a few Judgements with my back turned a got some Seal of Blood procs.

I didn't test it very much, just noted it.


I also added up Stasis of attacking a Boss that shows Seal of Blood and got around half my Judgements causing SoB procs.

I had 91 Seal of Blood procs, but I had 87 attacks that could proc it (melee, CS, DS), so Judgement must be the source.


More Ret nerfs incoming to be hotfixed:

We are looking at some way of making Avenging Wrath and Divine Shield mutually exclusive. This may be through Forbearance, if we can reduce the Forbearance duration through a hotfix. The bubble is intended as a defensive measure. Yet bubble+wings currently is used a lot in BGs and Arenas and helps contribute to the feeling of being destroyed by a Retribution paladin while you are unable to respond.

My apologies to the Ret paladins who already feel picked on. We are really trying hard to not overdo anything but even the last round of pretty severe nerfs didn't seem to do enough to Ret's dominance of PvP.

We are also sympathetic to the notion that if Ret can't do insane burst damage, that the spec might need other tools to be viable in PvP. We're exploring this now, possibly though the missing 11-point Prot talent, possibly through other avenues. We'll let you know when we've made a decision here.

On Live, All Seals proc from all Judgements(I'm pretty sure, I haven't tested them all).

In the current Beta build(9155), only Command will proc from any Judgement. No other seal will proc from any other Judgement.

Back pre-3.0 I might have wanted to kick a fuss about this, considering how our burst damage (especially on stunned mobs) were prone to the equivalent of insinuating negative things about the mobs' respective mothers.
But with the threat buff the tanks got in 3.0, I don't believe this to be anything but a situational issue in PvE (unless they for some reason design an encounter where the entire raid takes massive damage, and you can't possibly time AW's Forbearance to still allow Divine Shield).

Of course, I do hope they decide to go with the Cone of Cold vs. Dragon's Breath (one triggers a 20s cooldown on the other) approach rather than Forbearance.

One could argue, though, that Forbearance being back on a 1 minute duration would actually be a PvP buff (though against melee only), what with the ability to use/receive Hand of Protection more often.
I know it wasn't intended as a second self-bubble, but I play Alliance and I don't play organised PvP, so I always used Blessing of Protection as a self bubble.


Being that I was one of the whiners who flamed and cancelled account during the time of the major nerfbat clobber, thinking Blizzard had just yanked my chain yet again, I for one am glad that Blizzard is actually actively developing Retribution to ensure that we are far from useless in PvE (which is what matters when deciding on a raiding main, after all).

The hotfix was applied. Avenging Wrath causes Forbearance, Forbearance now lasts 2 minutes.

This effectively removes Divine Shield from our usable PvE abilities. AW cooldown is 2 minutes with talents, Forbearance lasts 2 minutes. We will be on permanent Forbearance unless we choose to not use AW. There are plenty of other ramifications for other specs or situations but this is the major consideration I came up with for Ret PvE. It seems odd that they don't simply prevent the usage of AW/DS with shared cooldowns as Nefarium mentioned: either one puts the other on a 12/20s CD respectively. Ah well.

One thought was to put it on a shared cooldown, similar to how trinkets work, so you can't use both at the same time, but you could use one right after the other.

It was really nice for PvE to be able to save bubble for bad things (like Bubble Taunt and nasty debuffs).

This also nerfs PvP Ret and Holy (since they need to save bubble most of the time) so it is hard when to know when to use AW when you have to pick between the two.


My recommendation:
Put 3 minute debuff back (instead of current 2 minutes)
Put AW/Bubble/50% Bubble on Shared Cooldowns

Change in this area was fairly expected. I've only played in a handful of battlegrounds since 3.0 but the shift in emphasis towards powerful active attacks and away from swing meant that the paladin was hardly experiencing any of the old drawbacks of the bubble and could practically continue doing the same damage to players. There's an argument for saying you can currrently do too much damage in divine shield even without avenging wrath. A more logical solution would be to forget forbearance and simply reduce damage by 50% during divine shield's duration, rather than halving attack speed. That way, it isn't locked out in pve and is still available for emergencies.