I don't think that was the best of examples since some rogues would prefer one ability over the other instead of "switching it up" on the fly. I'd find myself agreeing more with the shaman weapon enchant that people will simply stick with which ever seal does the most damage regardless if its unreliable in pve (the statement of SoV on things that don't die very fast vs SoR on things that do for tanks makes sense to switch it up). I honestly do not feel its blizzard's intention to create seal of righteousness the optimal seal for everyone including retribution. That being said, we'll just have to wait and see what direction they plan on going with seals. There has been very little response on the paladin forum in beta, we can only hope that will increase in time.

After playing a rogue for two years, blizzard could nix every single instant save SS and I don't think I would have noticed or cared, especially if all the other instants were replaced with something that would get used instead of replacing an existing ability on my bar. Trust me when I say that having 4 different but mutually exclusive options for instant damaging attacks doesn't add very much to enjoyment (or maybe I'm just crazy, who knows). And seals aren't even on the level of rogue instants. They don't really open up new combat options like having a lower energy attack (hemo) or super fast CP generation (mutilate), they just make your autoattack behave slightly differently. Hell, there is only one talent that will encourage those who take it to favor one damage seal over the other, at least the rogue instants synergize with their respective trees.

And after playing ret on beta for a few weeks now, I could probably swap all my damage seal bindings around at random and not really notice or care. Our main damage delivery system has moved towards active abilities: judgment, CS, DS, HotR, HS, consecration, and SotR. All of those abilities are way more fun than looking at 4 seals and deciding which slight variation of autoattack damage buff I want to use. If blizzard wanted to drop some of those seals in favor of abilities that would get used, then I would welcome it.

Righteousness: Old Faithful; short fight tanking
Vengeance (corruption): Long fights or AoE tanking - certain pvp uses?
Command: PVP Ret
Blood (Martyr): PVE Ret

I believe thats the way it is supposed to be.

You were all complaining before that the seals were totally imbalanced: Righteousness was way better than vengeance. Blood was way better than command. Crusader was useless. Wisdom/light lowered damage prohibitively.

It also has to be balanced so that Tanks can do more damage whilst soloing.

I'm not totally sure, but i think that Vengeance may be a little overpowered in its present state: rough maths shows it could be over 1500 TPS including the DOT. On the other hand, DPS will probably be doing huge amounts of damage come lv80. we can't really tell until people begin lv80 instances/raids what is an 'acceptable' level of DPS.

In all honesty, I think the coefficients will be changed significantly before live. I'd expect some slightly lower AP amd higher SP coefficients on SoR and SoV, lower %AP->SP on Sheath, and higher %Stam->SP on TbltL,

Notice how you mention it yourself: "Daggers, or big swords?". That exact specialization/"being stuck in a nich" is what's lacking nowadays with Seals.

Instead of making one Seal for tanking with a 1h, one Seal that's good with 2h DPS and maybe one Seal for bursty PvP, they have all of them scale with both 2h as well as sword and board, for all roles (DPS and tanking) and all gear sets (Ret/Prot with lots of AP and some spelldamage, holy with lots of Spelldamage and some AP).

The problem is probably since they insist on keeping SoC a talent and at the same time are only giving SoB as late as level 66. SoR is "forced" to be great for everything.

A general change of approach is what's needed. They can't forcefully try to make every seal please everyone without ending up with a mess as they are now.


I have to say I really like the idea of making SoC baseline (as someone mentioned here). Make that the dedicated 2h leveling Seal available at level 1, SoR the dedicated tanking Seal with a 1h available sometime later (level 10 or 20). Restrict SoR for 1h.

Then offer SoV as a high level boss tanking Seal later on (also restricted for 1h), SoB as the endgame 2h DPS raiding Seal (possibly still talented in place of where SoC now is in the ret tree).

There, everything makes sense again. You have two seals for 2h and two for 1h, each with a distinct niche, without messing up people while leveling, or Holy/Prot who might need something good to use with a 2h. And no need to change any coefficients or how they scale.

I hadn't thought of it that way, but it makes perfect sense when you put it like that. It's pretty illogical that you get one seal at level 1 and if you don't go Ret, you don't get another damage seal until 64.

(If I were designing things, I'd probably still start off with SoR as the default, since it's the simplest seal to understand. SoB/SotM around level 10 would synergize nicely with Judgement of Light. But that's just details.)

Why Mental Quickness functions for Shamans is because none of their abilities scale with both AP and SP. Not a single one.

Now, that was mostly true with Sheath for Paladins too, until the Seal/Judgement/other damage spells remake in beta, where they did make them scale with AP and SP.

That caused problems. Seal of Righteousness, for example, now gains 60% damage from that talent, and JoR/JoV gain 48%. Way too much for a single talent.
The question is to fix the scaling coefficients on abilities, or fix the root of it, Sheath? In my opinion it's Sheath which needs some kind of change.

What was really the main objective of Sheath? I think the primary objective was to scale caster abilities like Judgements, Consecration and Exorcism for example, just like the Shaman talent. The secondary effect was to improve the Ret Paladin's offhealing(and arguably giving Holy Paladins an alternative spec), which the HoT effect also shows. Now, when AP scaling has been given to all offensive Paladin spells, the primary objective is void.

What says healing spells can't also directly scale with attack power?

However, personally I agree with the person who suggested a 100% Str-SP conversion. However I think it should be baseline. One doesn't get monstrous amounts of Str from raid buffs, so that problem not exist. As an effect the dual scaling of offensive spells could simply be removed, and spells could scale with either AP or SP, not both.

As a cool secondary effect, with Divine Strength Str would have 98% of the effect of spell power, and with BoK it would surpass SP in item value.

Rather, isn't the HOT-on-a-crit effect a way to encourage Paladins to still use spell crit healing gear even when not specced into Illumination?

No, SoL (Ret) and Touched by the Light (Prot) are rather attempts to improve Rets/Prots healing in general.

Since in WotLK there is no more separate spell crit/melee crit and both talent trees got some spell power granted by talents, both ret and prot paladins are able to do some emergency healing. Those talents are encouraging players to do that.

Problem is, while no holy paladin will go 43 points deep into prot tree to make healing crits 20% stronger (and gain maybe 300-400 spell power), going 20 points into ret tree to get rolling HoT and about 500 spell power (and +5% crit to spells on the way) is too good to miss.

Linking HoT effect to the talent deeper in Ret tree (similarly to the protection) would probably do the trick, but it's hard to tell what is present devs vision of the holy paladin - are they considering deep holy talents essential, or just an option.

Comparing other classes it seems in most cases they are aiming at someone of "Spec XYZ" to actually have the 51pter in that tree, which in this case means Sheath of Light needs to be moved deeper, changed or otherwise made less useful for someone offspeccing into Retribution.

Another thing to tackle, IMO, would be the discrepancy between the Spellpower/Offhealing talent Protection gives versus the one in Retribution.

Retri offers the talent which is both easier to reach, and more powerful. It has a higher return on spellpower, plus it's "special effect" is superior, proccing HoT, more useful for emergency healing where you're unlikely to spam for long before you return to DPSing, compared to the 20% increased critsize for Protection where Prot is unlikely to crit in the first place.


Point being that the talents need to... well... switch. Kinda. Have the lower return (in spellpower) on the Retri talent, and give it larger healcrits, have the Prot talent have the higher return (the numbers you quoted would mean Retri gives ~70 spellpower per point instead of ~165, Prot instead gains ~250 per point instead of currently ~100), and Prot procs HoTs instead.

This'd still not be balanced, but it would be less spec-breaking, and more senseful to the placement in the tree. It's faster to stack Attackpower compared to Stamina, after all.

People you have to understand that you can't fix this with coefficients unless you make up some really stupid formula that scales the damage back as you have more of both SP and AP. Think about it,

a) You have a holy paladin with like 1500 SP and like 1500 AP,
b) Then you have prot, and prot numbers are going to be a bit different now for AP, but going off current block value and getting it off strength and endgame prot warriors having like 700 block value, you'll have what would be intended prot gear with something like 3500 AP maybe more and like 750 SP.
c) And then you have ret with AP going upwards to 4500 and 1500 SP from sheath.

If sheath is the 'suggested place to look at to fix this' the only thing they can do is to remove it entirely. SoR should be doing more damage for ret than it does for holy and prot because the numbers just work out that way. If you take seals like SoR and SoV and give it sensible values for its intended use, which are good numbers for holy and prot, you have the problem arising when you apply all the multiplicative talents in deep ret, you get all this crazy crap. Proposed solutions are:

a) Drastically nerf damage on SoR/SoV so that it does crap damage even after all the multipliers in deep ret. Works out great for rets.
b) Drastically nerf damage on SoR/SoV on 2h weapons. Probably easiest approach, but not pretty to be honest, it doesn't solve the issues. Bandaid solution.
c) 'Look into sheath' what that means other than removing sheath entirely I'm not sure. Suggestions are to move it down the tree which solves nothing, make it scale off str instead of AP which solves nothing again. I hope you guys can see that sheath has nothing to do with solutions here.
d) Have all of the talents from vengeance and above exclude SoR/SoV. Everything. So that's vengeance, sanctified seals, fanaticism, righteous vengeance. Have the instants never proc them. If it needs further damage cutbacks I'll concede to 2h nerf.

What if:
Increase spellpower coefficients on SoR/SoV, decrease AP coefficients.
Increase Stam->Sp conversion % for prot. (to 60/70+%)
Decrease AP->SP conversion % for ret (to 15%)

Result? if done correctly, SoV/SoR damage doesn't change much for Prot but goes down for Ret.


Also, I'm for TbtL's secondary effect getting changed. Ret are going to be decent healers: good amounts of Spellpower, Haste, Crit and Spellpower as well as fantastic mana return. Prot have about 1/2 the spellpower, no haste/crit to speak of, yet a top-tier talent increasing our critical heal effect? IMO would be much better suited to a mana cost reduction.

Also, does anyone else see melee plate on a sheathbot healer? Str->AP-Spellpower, Haste and Crit benefit both spells and melee....

And one last thing: Doesn't the Vengeance DOT mechanics look a little odd? most other abilities are either equal scaling of AP/SP, or SP has almost twice the coefficient (see SoR swing (10% SP 5% AP) or Judgement (58% SP + 36% AP). However SOV seems to be the other way around; 7% of AP and 3.5% SP. Error maybe?

O wait; wowhead is saying something different to start of this thread. Is it 5*(0.07*AP+0.035*SP) per tic of full stack, or 0.07*AP+5*0.035*SP ??

Only too good to miss at the moment. I don't get the complaints, shouldn't we be welcoming the variations instead of hating everything that's not the new 41/20/0? Speaking from a PVP point of view, sheath isn't really that gamebreaking seen with todays eyes. WoTLK arena might be unchartered territory, but right now i see sheath specc in 2v2 and 3v3.

If the holy tree's upper tiers were a "tad" bit better, then you'd probably get two different speccs for different uses. To get sheath you are passing up some really nifty protection talents: 30% dispell immunity, 30% shorter stuns, 30sec CD on HoJ, 30% of your party's damage redirected to you while in bubble and all the other usuals. All it takes is a better 51pt talent in holy or just the upper tiers, and we'll be left with two viable choices instead of just the one you'd have if you moved sheath further down the retri tree.

Wtb variation

As far as SoR goes wouldn't a simply diminishing returns on weapon speed be the best method of nerfing the damage a bit on two handers with out screwing over one-handers? Say the cut off is 2.0 attack speed. That may hit a tank a bit if they are using a 2.6 swing speed weapon, but not nearly as much as it would diminish the damage on a two-hander at 3.6 or 3.8 swing speed.

Some thing along the lines of Integral from 0 to WS of f(x) where f(x) = 1/(1+e^(2*x-6)) should get the diminishing returns I am thinking of.


Edit: Added Formula

How well these broken seals will scale with ap. Currently all melee classes gain about 0.3-0.4dps per attack power with raid buffs.
If retri paladin use this spec and Seal of Vengeance every ap contribute to SoV with sheath of light before multipliers:
[0.034 + (0.3 * 0.34)] * 5 = 0.0802 * 5 = 0.401 per tick or 0.13366 damage / s.
Every melee attack hit also 1.2% per spell power per second: 0.012 * 0.3 = 0.036 damage / s.
Judgement every 8s: 0.36 + (0.3* 0.58) = 0.534 or 0.06675 damage per / s * crit% * crit bonus damage

Total(ignoring judgements crit%): 0.204 * multpliers.
SoV and JoV get:
+15% seals of the pure
+15% vengence
+2% sanctifed retribution
JoV also get:
+36% crit from talents
+25% bigger crits

SoV + JoV gain 0.275dps per ap when counting only self buffs and ignoring crit.
With these coefficients retri paladin will scale almoust well as other melee classes now with only counting SoV/JoV.

The first version of Beta SoV was 7% AP + 5*3.4% SP.

Then in the patch which brought "smart" tooltips, SoV was changed to 5*(7% AP + 3.4% SP).

My feeling is that the second one is a mistake with the brackets, and the first equation is meant to be correct. Or possibly, it should be 5*(1.4% AP + 3.4% SP), if Blizzard wants the spell to fully scale with stacks. (The first version had the odd property that it would do the same damage for different numbers of stacks if you had no SP.)

In my mind, 33% does seem like a bit much for something you need to be in melee range to use, and can use every 8 seconds. Shadow Priests have been eating 100% recoil on SW: Death for ages now, but I guess the developers even thought that was too high and are reducing it via talents.

It also might be the way that they're balancing Judgements of the Wise. It might be that it isn't intended to be used every cooldown with SoB (SoM) without a significant drawback.

What level of gear are those numbers supposed to represent?


SoR only needs to be roughly balanced between all 3 specs in green/blue gear, self-buffed. If an epic'd out Ret paladin does more damage with a damage seal than an epic Holy paladin, that'd be working as intended. (ie: damage spec does more damage than a healing spec) If you're comparing raid-buffed numbers, then of course it's going to be hard to balance between the 3 specs. (that represents 1000+ AP and 300 SP for the Ret paladin right there)


The more important question is how SoR can be balanced against SoC/SoB/SoV, and for that purpose, we can tweak their gain from weapon damage. For an example of other equivalent abilities that work that way, Bloodthirst only scales with AP while Mortal Strike scales with both AP and weapon damage.

Maybe SoC could be removed and replaced with a PVP friendly ability instead.

One thought I had was to replace it with Improved SoV. To retain some of the flavor of SoC, maybe it would increase the baseline damage of SoV plus add in the increased judgment damage now associated with SoC.

SoR becomes the 1H Seal
SoV becomes the 2H Seal (given at a lower level than currently, like 20) and is improved through the tier 3 talent for 1 pt
SoB is the raiding seal


Just an idea.

Those are very rough sunwell gear raid buffed numbers. Prot is way off on AP of course, but I'm going by block value on prot warriors and 2 str = 1 block value. For my solution ret would still do more damage with SoR/SoV than holy or prot but only because he's stacking offensive stats, nothing much you can do about that, but SoR/SoV don't benefit from deep ret scaling.

It might be nice to include the result of the integral. If I have it right, it should be F(x) = x - 0.5*log(e^6 - e^2x).

I'll make a little graph of the function from 0-3.8 once my computer stops freezing and add it as an edit.

Edit: Here it is. It doesn't seem to make sense; I may have possibly made a mistake somewhere. It goes to infinity at 3.0 weapon speed, and after that becomes imaginary.

Graph the function and figure out the area, you did your integral incorrectly, but I am not sure where. The function basically stays at 1 until weapon speed hits 1.6 and goes to 1/2 at 2.5 seconds. Sorry I would put a graph of it out but I don't have a program that will do it.


WS WS coefficient
1.0 1.00
1.1 1.10
1.2 1.20
1.3 1.30
1.4 1.40
1.5 1.49
1.6 1.59
1.7 1.69
1.8 1.78
1.9 1.87
2 1.96
2.1 2.05
2.2 2.12
2.3 2.19
2.4 2.25
2.5 2.30
2.6 2.34
2.7 2.37
2.8 2.40
2.9 2.41
3 2.42
3.1 2.43
3.2 2.44
3.3 2.44
3.4 2.44
3.5 2.45
3.6 2.45
3.7 2.45
3.8 2.45
3.9 2.45


Ended up doing it mechanically in excel.

EDIT: Formating

I have to say that i don't see the point of suggesting such a diminishing return. There are simpler way to obtain the result you're aiming for, and you know they won't use such a function...

By the way dekkar, you're graph isn't correct, the proposed function is increasing.

my calculator give me F(x)=x +0.5* (ln(1+e^6)-ln(e^2x+e^6))

Any way, it's a hammer to slam a fly (or the equivalent expression in english...)