3.2 PTR - Retribution Discussion - Page 12

• Zurm · 07/08/09, 10:20 AM

Originally Posted by wighti

Think it's safe to assume it will, but if not - getting the Chaotic meta activated with those instead of the the soverign dreadstones should provide a boost of couple of DPS.

I actually don't think its safe to assume either of those points. Especially the second... crit and agi are VERY close in value, a 10 all stat gem might not be enough to make up the gap. We'll leave it up to rawr/spreadsheets to decide.

wighti · 07/08/09, 10:41 AM

Umm, 2x10STR and 10AGI surely come out on top against 2x10STR + 15STA?

That's what I was driving at. Or you mean deciding between the Chaotic and Relentless?

• Zurm · 07/08/09, 10:44 AM

Originally Posted by wighti

That's what I was driving at. Or you mean deciding between the Chaotic and Relentless?

Yes, the two assumptions I was talking about was only one prismatic being allowed, and the dps between chaotic and relentless. There is no doubt in my mind that the prismatic is better than a purple or green gem.

wighti · 07/08/09, 10:47 AM

Aaah, makes sense yeah.

One of those new +10's and then a +10STR +15 Crit might actually come on top with Relentless. Will need to hit the sheets for that.

Giulls · 07/08/09, 11:16 AM

You still won't need to use more than just one +10 to all Nightmare tear since it's prismatic, so it alone will count as your blue, yellow and red slot. The extra +10 str +10 crit would be redundant for meta purposes.

I'm rather surprised that the new + all stats gem is so strong, now it's such an obvious winner over two +10str + 15sta since it's almost the equivalent of only having to use one orange gem for your meta purposes.

Selenia · 07/08/09, 11:23 AM

Originally Posted by wighti

Anyone noticed that the new +10 Stats gem doesnt seem to have a 'only one can be socketed' text assosicated to it?

Think it's safe to assume it will, but if not - getting the Chaotic meta activated with those instead of the the soverign dreadstones should provide a boost of couple of DPS.

They're unique-equipped. Blizzard just did away with the flavortext reminder on top of it.

Dorvan · 07/08/09, 1:07 PM

Originally Posted by Giulls

I'm rather surprised that the new + all stats gem is so strong, now it's such an obvious winner over two +10str + 15sta since it's almost the equivalent of only having to use one orange gem for your meta purposes.

Yeah, I'm a bit surprised myself. For comparison, here are the item points associated with various gems

Rare gem -- 16
Enchanted Tear -- 28
Tear/Rare ratio = 1.75

Epic gem -- 20
Nightmare Tear -- 46.667
Tear/Epic ratio = 2.333

So as you can see the Nightmare tear has a much larger budget relative to standard gems than was present at the rare level. +7/8 stats would be in the right ballpark to maintain the same relative item budget power.

greatrichie · 07/08/09, 1:17 PM

But the Nightmare tear is made with a Dragonseye, so maybe they base it off the stats of the Dragonseye gems, which have more stats than the epic gems.

Dorvan · 07/08/09, 1:25 PM

Originally Posted by greatrichie

But the Nightmare tear is made with a Dragonseye, so maybe they base it off the stats of the Dragonseye gems, which have more stats than the epic gems.

....the materials required to make it are irrelevant. There's currently no indication that it's JC-only gem, so functionally it's just the epic version of the +all stats gem.

• frmorrison · 07/08/09, 2:07 PM

Originally Posted by Dorvan

Yeah, I'm a bit surprised myself. For comparison, here are the item points associated with various gems

When you stack many stats at once, the item cost is lower. So your numbers are inaccurate.

Anyway, even if the gem is nerfed to 8 to all stats, it is still better than 10 strength, so the point is moot, there will be an upgrade +to all stats gem.

Dorvan · 07/08/09, 2:26 PM

Originally Posted by frmorrison

When you stack many stats at once, the item cost is lower. So your numbers are inaccurate.

Oh, I realize that for itemization budget you get more points if you spread them out among more stats, that's not what I said was surprising and not what my numbers where getting at at all. The spreading out of stats *doesn't* explain why a Nightmare Tear has more stat points than an epic gem by a much greater ratio than an Enchanted Tear has more than a rare gem. +7/8 to all stats isn't the point at which the stat points between Nightmare Tears and epic gems are equal, but rather the point at which the ratio is the same as that of the Enchanted Tear to a rare gem.

I agree that it wouldn't affect what we gemmed either way, I just find the budgeting of that gem to be interesting because it's not in fact in line with what you'd expect, even taking into account the fact that budget is spent on five stats.

Scorpiodrush · 07/09/09, 3:12 AM

Originally Posted by Dorvan

Oh, I realize that for itemization budget you get more points if you spread them out among more stats, that's not what I said was surprising and not what my numbers where getting at at all. The spreading out of stats *doesn't* explain why a Nightmare Tear has more stat points than an epic gem by a much greater ratio than an Enchanted Tear has more than a rare gem. +7/8 to all stats isn't the point at which the stat points between Nightmare Tears and epic gems are equal, but rather the point at which the ratio is the same as that of the Enchanted Tear to a rare gem.

I agree that it wouldn't affect what we gemmed either way, I just find the budgeting of that gem to be interesting because it's not in fact in line with what you'd expect, even taking into account the fact that budget is spent on five stats.

I believe he's saying you can't budget the Ratio of stats in that manner because ofr instance (purely theoretical)

16 Strength might actually be 20 Itemization points
while a 8Strength, 8 Agi would actually cost like 18.
1 strength=1 item point 10 strength might = 11 item points. 20 Str might cost 25 item points. 50 might cost closer to 70 points. etc.

He's saying that you can't gauge the "ratio of stats" without first knowing how they scale item level value when stacking a single stat.

About the only thing your math can ASSUME is that since Your math points to it being higher Ratio, so to should the Actual Stat value between the Gems, which Man not be true.

nethweniel · 07/09/09, 10:12 AM

Hej guys and gals. This is my first post here on these forums- so please be gentle with me.

I tried to come up with some sort of spreadsheet in order to get a clue of what kind of Seal to use with 3.2.

Well, this is with what I came up with (warning: LONG post!).

================
Executive summary

================

I have to make one assumption here: Addition of one DoT-Effect of the new SoV does not overwrite the ticks of previous stacks. I can't test it myself, Blizzard doesn't seem to want my character on the PTR *grr*

For Raids: Use the new Vengeance. The 5 stacks should be applied after 9 seconds and you will have additional 33% weapondamage with each attack. SoV and SoR are very close together- up till the point you have your 5 stacks. After that SoV is just better.

If you're out there farming/questing, I'd prefer SoR over SoV- very few mobs survive longer than 9 secs. If you're engaging elite monsters, use SoV. And definitely forget about command. Really. Do not take it. Even with 6000 AP and the Eartshaper-2h-Mace from Ulduar (670-1005damage, 3.6speed) SoR is far better than SoC thanks to sanctified wrath.

Spreadsheet:

I just give you the code, b/c I don't really know if it's ok (or possible) to add an excel-file to my post. If you want to have it added feel free to tell me. Oh, another thing: It was written with a german OpenOffice, so maybe you want to change "Summe" and "Quotient" for the appropriate english commands.

Seal of Command
=============

A5: minimum weapon damage => Value in B5
A6: maximum weapon damage => Value in B6
A7: spell power holy => Value in B7
A8: attack power => Value in B8

A11: Damage melee (median) =QUOTIENT((B13+B14);2)

A13: Minimum additional damage => B12 =SUMME(0,36*B5)
A14: Maximum additional damage => B13 =SUMME(0,36*B6)

A16: Damage unleashing (median) =QUOTIENT((B18+B19);2)

A18: Minimum additional damage => B17 =SUMME(0,19*B5+0,08*B8+0,13*B7)
A19: Maximum additional damage => B18 =SUMME(0,19*B6+0,08*B8+0,13*B7)

Seal of Righteousness
================

D5: base weapon speed => Value in E5
D6: spell power holy => Value in E6
D7: attack power => Value in E7

D9: Points in „Seals of the pure“ => Anzahl in E9

D11: Damage melee:

D13: Additional damage =SUMME((1+E9*0,03)*(E5*(0,022*E7+0,044*E6)))

D16: Damage unleashing:

D18: Additional damage =SUMME(1+0,2*E7+0,32*E6)*(1+E9*0,03)

Seal of Vengeance
================

Ok, this is much more complex as with SoR and SoC. Therefore I tried to figure out when the stacks of SoV are applied and when the dot actually ticks

sec in fight Stacks
CS 0,00 Stacks:1
DS 1,50 Stacks:2
Judgement 3,00 Stacks:2
First DoT- tick 3,00 Stacks:2
Autoattack 3,x Stacks:3
CS 4,50 Stacks:4
Second DoT- tick 6,00 Stacks:4
Autoattack 6,x Stacks:5

If you have a weapon which is much faster then 3.0 secs, you will change the whole timetable a bit and have more damage for your first Damage off the first DoT- tick. But other than that this timetable should work for the theory- hopefully.

Calculation

G5: base weapon speed => Value in H5
G6: spell power holy => Value in H6
G7: attack power => Value in H7

G9: Points in „Seals of the pure“ => Value in H9

G11: Damage melee:

G13: Damage dot before 5 stacks, median =QUOTIENT((H15+H17);2)

G15: Damage 0-3sec =SUMME((0,013*H6+0,025*H7)*2*(1+H9*0,03))

G17: Damage 3-6sec =SUMME((0,013*H6+0,025*H7)*4*(1+H9*0,03))

G19: Damage 6-Xsec =SUMME((0,013*H6+0,025*H7)*5*(1+H9*0,03))

G21: Damage unleashing

G23: Additional damage, 2 stacks =SUMME(1+0,22*H6+0,14*H7)*(1,2+H9*0,03)

G25: Additional damage, 5 stacks ==SUMME(1+0,22*H6+0,14*H7)*(1,5+H9*0,03)

========================================
My values with my crappy, fresh 80, gear.
========================================

[top]
SoC:

Damage melee (median): 167

Minimum additional damage = 133,92
Maximum additional damage = 200,88

Damage unleashing (median): 289

Minimum additional damage = 289,8
Maximum additional damage = 325,14

[top]
SoR:

Damage melee:

Additional damage = 238,51

Damage unleashing:

Additional damage = 627,95

[top]
SoV:

Damage melee:

Damage dot before 5 stacks, median 183

Damage 0-3sec = 122,42

Damage 3-6sec = 244,84

Damage 6-Xsec = 306,05

Damage unleashing

Additional damage, 2 stacks = 513,43

Additional damage, 5 stacks = 627,53

All calculations based on the formulas given by wowhead.com

SoC: Seal of Command - Spell - World of Warcraft
SoR: Seal of Righteousness - Spell - World of Warcraft
SoV: Seal of Vengeance - Spell - World of Warcraft

Lesrek · 07/09/09, 11:51 AM

Originally Posted by nethweniel

For Raids: Use the new Vengeance. The 5 stacks should be applied after 9 seconds and you will have additional 33% weapondamage with each attack. SoV and SoR are very close together- up till the point you have your 5 stacks. After that SoV is just better.

If you have a weapon which is much faster then 3.0 secs, you will change the whole timetable a bit and have more damage for your first Damage off the first DoT- tick. But other than that this timetable should work for the theory- hopefully.

sec in fight Stacks
CS 0,00 Stacks:1
DS 1,50 Stacks:2
Judgement 3,00 Stacks:2
First DoT- tick 3,00 Stacks:2
Autoattack 3,x Stacks:3
CS 4,50 Stacks:4
Second DoT- tick 6,00 Stacks:4
Autoattack 6,x Stacks:5

You should have 5 stacks up in 3 seconds, the time it would take to Judge, CS, DS and 2 Auto attacks.

When raid buffed, most weapons will drop significantly below the 3.0 mark. 20% haste from DK or Shammy alone will drop weapons underneath 3.0 and that doesn't count our own 3% haste or any haste on gear.

For the third part I quoted, you are using CS, DS, J. Even under the current priority system this is wrong. On the PTR, more and more evidence is beginning to show that J, CS, DS will be the way to go, but at the very least, CS, J, DS. Under very few circumstances will DS ever jump ahead of Judge.

EDIT: I thought about why you were using CS, DS, J as the beginning and I remember that would have been the rotation to get to 5 stacks quickest. Since you aren't on the PTR, you should note that Judgement is currently procing seals and therefore contributes to stack building. Also, it is hitting like a truck.

• Zurm · 07/09/09, 12:18 PM

Originally Posted by Lesrek

For the third part I quoted, you are using CS, DS, J. Even under the current priority system this is wrong. On the PTR, more and more evidence is beginning to show that J, CS, DS will be the way to go, but at the very least, CS, J, DS. Under very few circumstances will DS ever jump ahead of Judge.

Incorrect. Rawr (the dev version, not necessarily the most recent released one) is repeatedly showing HOW > CS > Judge > DS > Consec > Exor, and my own tests prove it.

Just because judgement hits hard doesn't mean its #1 (it's an easy assumption to make, I've done it). CS is still higher DPS, and pushing CS back one gcd for judge is a bigger dps hit than pushing judge back one GCD for cs. Note that CS procs 2x the number of seals over the same time frame.

Lesrek · 07/09/09, 1:17 PM

Originally Posted by Zurm

Incorrect. Rawr (the dev version, not necessarily the most recent released one) is repeatedly showing HOW > CS > Judge > DS > Consec > Exor, and my own tests prove it.

Just because judgement hits hard doesn't mean its #1 (it's an easy assumption to make, I've done it). CS is still higher DPS, and pushing CS back one gcd for judge is a bigger dps hit than pushing judge back one GCD for cs. Note that CS procs 2x the number of seals over the same time frame.

Granted, I haven't used or seen the dev version of Rawr, but last I checked, Rawr didn't have judgement applying seal damage. If it is, and it is still showing CS > J, then my mistake. Anyhow, what I said above still stands either way.

Aenae · 07/09/09, 1:18 PM

Originally Posted by Zurm

Incorrect. Rawr (the dev version, not necessarily the most recent released one) is repeatedly showing HOW > CS > Judge > DS > Consec > Exor, and my own tests prove it.

Just because judgement hits hard doesn't mean its #1 (it's an easy assumption to make, I've done it). CS is still higher DPS, and pushing CS back one gcd for judge is a bigger dps hit than pushing judge back one GCD for cs. Note that CS procs 2x the number of seals over the same time frame.

Well, if judgements hit harder than CS, and both proc the seal, why should i prioritize a ~3k crit over a 5k crit* + sealdamage? (*numbers done by testing for almost 10 seconds with selfbuffs!). Seals hit for around 1k-2k for me, and judgement has a lot higher critchance than CS, am i missing something?

Also, as start to get up the stack, i think (not based on any math) that it might be better to cs-ds-j, as your judgement will be quite a bit higher if you have 0-1 stacks or 4-5 stacks.

Nicki · 07/09/09, 2:01 PM

Originally Posted by Aenae

Well, if judgements hit harder than CS, and both proc the seal, why should i prioritize a ~3k crit over a 5k crit* + sealdamage? (*numbers done by testing for almost 10 seconds with selfbuffs!). Seals hit for around 1k-2k for me, and judgement has a lot higher critchance than CS, am i missing something?

Also, as start to get up the stack, i think (not based on any math) that it might be better to cs-ds-j, as your judgement will be quite a bit higher if you have 0-1 stacks or 4-5 stacks.

Hitting hard != dps

Your crits:

3k/4(Cooldown) = 750dps (CS i assume)
5k/8(Cooldown) = 625dps (Judgement)

Assuming judgement crits 15% more than CS;

(70*1500)+(30*3000) = 195k
(26*2500)+(24*5000) = 185k

And sorry but im inclined to believe rawr devs over 10 second numbers pulled from a hat in a non raid situation. Either way it seems your argument falls flat on its face when you do papernapkin math for dps...

Edit: just stating this maths is very likely wrong and doesn't include latency, clashes or combat roll tables like i said napkin maths...

flyingtoastr · 07/09/09, 2:19 PM

Originally Posted by Aenae

Well, if judgements hit harder than CS, and both proc the seal, why should i prioritize a ~3k crit over a 5k crit* + sealdamage? (*numbers done by testing for almost 10 seconds with selfbuffs!). Seals hit for around 1k-2k for me, and judgement has a lot higher critchance than CS, am i missing something?

3.2 PTR - Retribution Discussion

Again, it comes down to Damage per Second per Effective Cooldown, not raw damage.

Aenae · 07/09/09, 2:35 PM

Originally Posted by flyingtoastr

3.2 PTR - Retribution Discussion

Again, it comes down to Damage per Second per Effective Cooldown, not raw damage.

You are right, it just felt a bit counter-intuitive at first, but by reading a bit more, you (both) have a point - nice explanation in that post btw.

• Zurm · 07/09/09, 2:50 PM

Originally Posted by Lesrek

Granted, I haven't used or seen the dev version of Rawr, but last I checked, Rawr didn't have judgement applying seal damage. If it is, and it is still showing CS > J, then my mistake. Anyhow, what I said above still stands either way.

As a former developer of Rawr.ret and Rawr.DK, I have access to the code base, and Endoscient (the current Rawr.ret dev) is a guild mate and I work with him on formulas for it. I can assure you the build *I* used had the right changes, the version publically available may not.

In addition, there was no effective change to the priorities between the rawr you probably used and the dev one, with my gear. Even on live, CS > judge > DS.

flyingtoastr · 07/09/09, 2:54 PM

Paladin QA is up. Relevant PvE ret goodies:

Q: All classes vary in dps from one encounter to another; however, some paladins may feel their dps can be less competitive at times in comparison to other classes, more so in straight stationary single-target dps encounters. How do we feel paladins are doing in terms of dps across the board?

A: Retribution dps is too low in PvE in 3.1. We are buffing it in 3.2 through the new way Seal of Vengeance / Corruption will work. This Seal is designed to really deliver damage once the paladin gets five stacks up, which will make it the Seal of choice for boss fights. Seal of Command will be used in PvP or PvE for short fights. We expect overall for Corruption / Vengeance to be the “go-to” Seal much of the time, perhaps even in PvP, provided you can keep the buff up.

Seems like they're ok with the SoV + UA shenanigans in PvP, which is perfectly cool with me.

Otherwise nothing new here, our DPS is indeed looking fairly competitive with the SoV changes.

Q: Consecration seems to utilize a sizable portion of mana per application of this spell; do we have plans on making this a bit more mana-efficient?

A: We think the mana cost is appropriate. Retribution and Protection paladins have enough ways to earn mana back that it doesn’t seem to be slowing them down much.

I can't say I like this one. Aside from the obvious "bleh" factor of an AoE ability in a single target rotation, Consecration usage does cause fairly large mana problems when you aren't fully regen buffed (i.e. not in a 25-man raid).

Q: Exorcism will be usable in both PvE and PvP once more which is great; aside from allowing this ability to be used against other players, what were the reasons to go down this route in redesigning this attack?

A: Exorcism never did a ton of damage to players, but it was an instant attack which meant paladins could use it while closing with an enemy. It was essentially just free damage and never a decision of any kind. The new approach to the spell prevents it from being used while closing, and also makes Retribution paladins have to pay a little more attention to their combat rotation -- you want to use Exorcism when Art of War procs, and generally not at other times.

AoW + Exorcism seems to have been their plan for making us "think" in our rotation. Again, somewhat disappointed with the response here.

And that's really it for PvE (there's a slew of PvP stuff though, but it doesn't particularly belong in this thread). There were a couple nice general class points that were addressed (such as visible librams) as well.

Lesrek · 07/09/09, 2:56 PM

Originally Posted by Zurm

As a former developer of Rawr.ret and Rawr.DK, I have access to the code base, and Endoscient (the current Rawr.ret dev) is a guild mate and I work with him on formulas for it. I can assure you the build *I* used had the right changes, the version publically available may not.

In addition, there was no effective change to the priorities between the rawr you probably used and the dev one, with my gear. Even on live, CS > judge > DS.

I agree that on live it is CS > J > DS. I was only stating that the last version of rawr I used for the PTR testing didn't have judgements applying seal damage, so I didn't know if that would affect the Rawr rotation. I have no doubt you know exactly what is going on and it is clear I was basing information on outdated information.

EDIT: UPDATE, Paladin Q&A

So, the paladin Q&A is up and it was confirmed that our dps buffs on the PTR are intended.

"Retribution dps is too low in PvE in 3.1. We are buffing it in 3.2 through the new way Seal of Vengeance / Corruption will work. This Seal is designed to really deliver damage once the paladin gets five stacks up, which will make it the Seal of choice for boss fights. Seal of Command will be used in PvP or PvE for short fights. We expect overall for Corruption / Vengeance to be the “go-to” Seal much of the time, perhaps even in PvP, provided you can keep the buff up."

It also seems they are intending SoC to not be great, and only see situational use.

OnTheHissay · 07/09/09, 5:26 PM

Anyone seen them commenting on where they think ret aoe dps is in 3.1 and should be in 3.2? Because from what I've read about the SoV change aoe dps is gonna go down, when we're almost dead last in aoe dmg on multiple targets (2 and 3 targets we do fine atm), only worse aoer is probably a shaman.

One fix to this would be to make SoV and Divine Storm to work together in a way that the aoe dmg is determined on how many stacks of SoV you have on your main target. Still keeping burst down, but after you get your stacks up Divine Storm will do damage as if all aoe targets had the amount of stacks your main target has. That way when adds die and new spawn, you'd be ready with 5stacks on the boss e.g. and hopefully keep up competitive aoe dps as well.

Edit: Update: I've tested aoe dps on some random lowlevel (70+) instances on PTR and to me it shows as it's gone down substantially from 3.1. This might be an issue because they die to fast, but doing competitive dps on trash should be a factor in our dps scheme as well.

• Zurm · 07/09/09, 5:29 PM

Originally Posted by OnTheHissay

One fix to this would be to make SoV and Divine Storm to work together in a way that the aoe dmg is determined on how many stacks of SoV you have on your main target. Still keeping burst down, but after you get your stacks up Divine Storm will do damage as if all aoe targets had the amount of stacks your main target has. That way when adds die and new spawn, you'd be ready with 5stacks on the boss e.g. and hopefully keep up competitive aoe dps as well.

I think we have yet to see SoV work quite right. At the moment, if you were to DS 4 nearby targets with 5 stacks on them already, all 4 procs would hit your currently selected target... even if its not one of those 4 DS'd mobs and is 30 yds away.

[top] SoC:

[top] SoR:

[top] SoV:

[top]
SoC:

[top]
SoR:

[top]
SoV: