This site makes extensive use of JavaScript.
Please enable JavaScript in your browser.
Live
PTR
10.2.7
PTR
10.2.6
Beta
Protribution DPS
Post Reply
Return to board index
Post by
Squishalot
(edited to make reading easier)
failblog - thanks, it's been fun =) Have you been playing with your protribution spec more recently? Any real world thoughts?
Hey svirve, questions on some of your assumptions/calcs...
Assumptions:
We're working of a 9-6 rotation, right?
I assume that you're substituting Exorcism/Avenger's Shield in lieu of Holy Shield?
Calculations:
Judgement DPS
is being calculated as (avg damage / 8) - this should be (avg damage / 9) to reflect the rotation.
Edit: Ditto
Consecrate
- avg damage / 9.
Edit 2: Ok, so for some adjustments to the formulae:
To make it more theoretical based (i.e. so you don't need to equip the items), I'm changing 'Value' to be the amount of stats as provided by the armour/equipment. Talents that increase stats (eg, Divine Strength) I'll apply by multiplying the (Base + Value) accordingly.
Eg: Strength
Base: 173 (human)
Value: 36 (from Titansteel Shield Wall)
Total: =ROUNDDOWN((E12+D12)*B19,0)
(where B19 = 115% being 5 points in Divine Strength)
This means that we can put the current equipment on one side, then simply add/subtract the stat point difference (eg, from Wowhead's comparison page) into the Value boxes on the other side, rather than having to equip to take talents into account.
Expertise
needs to be capped - adjusted the 'Total' column to be:
=MIN(ROUNDDOWN(E25/8.2,0)+D25+B24*100,26)
The B24*100 is because I've created a new talent item for Combat Expertise and wanted to add it into the total, rather than base. Base can be for racial bonuses.
Issue with the
SoB calculation
(I knew there was something that twigged my concern) - it doesn't work if you have hit% below 100%. It implicitly assumes that all melee attacks will connect, and then it'll calculate whether the SoB proc hits or not. I've now adjusted the DPS column formula to:
=((1/F28)+(1/6)+(1/6))*F38*(0.92+F23)
to take into account the hit rate of the melee attacks, to determine how often it procs.
Adjusting
crit chance
calculation:
=D22+E22/45.91/100+(B14)/5+<insert other talents here>
This converts E22 to be a rating instead of a 'value', again allowing easier comparison between gear.
Question on
Hit/Expertise
.
Current formula for DPS on white damage is
=(((D41+E41)/2)*(0.92+F23)*(0.935+(F25*0.0025)))+(((D41+E41)/2)*F22)
This assumes that the chance of hitting unbuffed and naked is (92%)*(93.5%) = 86.02%. Actual chance is (100% - 8% - 6.5%) = 85.5%. Should the formula be the following instead?
(1-(0.08-F23)-(0.065-(F25*0.0025)))
Values:
Edit 3: Ok, now for some random numbers.
On a naked human, properly spec'd, with nothing but a Titansteel Bonecrusher and a Titansteel Shield Wall (as every fresh 80 should easily be born with):
Stat Base Value Total
STR 173 36 240
STA 193 128 321
SPI 108 0 108
AGI 90 0 90
INT 98 0 98
SP 0 0 96
AP 220 0 700
Somehow, with a tight rotation, this should be able to put out:
Ability Min Max Avg DPS
AS 2695.14 3264.15 2766.90 92.23
HotR 1052.66 1052.66 977.50 162.92
ShoR 1258.11 1258.11 1168.28 194.71
Judge 242.84 319.19 273.88 30.43
SoB 210.44 331.33 251.54 184.47
Exo 1765.99 1947.60 1642.52 91.25
Cons 1351.07 1351.07 1322.36 146.93
White 438.42 690.27 524.05 209.62
Total: 1112 dps.
Puts a lot of pug dps to shame, eh?
Edit 4: Actually, that number is slightly on the low side - I haven't factored in the increased Stamina talents, and therefore the subsequent increase in SP. Still nice though!
Post by
149504
This post was from a user who has deleted their account.
Post by
svirve
Assumptions:
We're working of a 9-6 rotation, right?
I assume that you're substituting Exorcism/Avenger's Shield in lieu of Holy Shield?
Calculations:
Judgement DPS
is being calculated as (avg damage / 8) - this should be (avg damage / 9) to reflect the rotation.
Edit: Ditto
Consecrate
- avg damage / 9.
Yea i started out by assuming that but i think i noted a couple of posts back that i skipped this since we didnt know if we were gonna go for a 969 or FCFS style rotation.
Thus i added all the spells to be used ON cooldown so we can determine what kind of rotation we are to use.
To make it more theoretical based (i.e. so you don't need to equip the items), I'm changing 'Value' to be the amount of stats as provided by the armour/equipment. Talents that increase stats (eg, Divine Strength) I'll apply by multiplying the (Base + Value) accordingly.
Eg: Strength
Base: 173 (human)
Value: 36 (from Titansteel Shield Wall)
Total: =ROUNDDOWN((E12+D12)*B19,0)
(where B19 = 115% being 5 points in Divine Strength)
This means that we can put the current equipment on one side, then simply add/subtract the stat point difference (eg, from Wowhead's comparison page) into the Value boxes on the other side, rather than having to equip to take talents into account.
Sounds good to me i was planning on fixing this but been lazy lately ;p
Expertise
needs to be capped - adjusted the 'Total' column to be:
=MIN(ROUNDDOWN(E25/8.2,0)+D25+B24*100,26)
The B24*100 is because I've created a new talent item for Combat Expertise and wanted to add it into the total, rather than base. Base can be for racial bonuses.
It is and has always been capped ;p Expertise was the first number to get capped on the spreadsheet because it was giving me the most trouble.
How did you add Combat expertise if you need to multiply it by 100? oO
Issue with the
SoB calculation
(I knew there was something that twigged my concern) - it doesn't work if you have hit% below 100%. It implicitly assumes that all melee attacks will connect, and then it'll calculate whether the SoB proc hits or not. I've now adjusted the DPS column formula to:
=((1/F28)+(1/6)+(1/6))*F38*(0.92+F23)
to take into account the hit rate of the melee attacks, to determine how often it procs.
Hmm.. Arn't you double dipping the hit now? It's already added in the AVG dmg which is used to determine the DPS. Unless i'm missing something.
Adjusting
crit chance
calculation:
=D22+E22/45.91/100+(B14)/5+<insert other talents here>
This converts E22 to be a rating instead of a 'value', again allowing easier comparison between gear.
Sounds good to me. Makes it easier with talents.
Question on
Hit/Expertise
.
Current formula for DPS on white damage is
=(((D41+E41)/2)*(0.92+F23)*(0.935+(F25*0.0025)))+(((D41+E41)/2)*F22)
This assumes that the chance of hitting unbuffed and naked is (92%)*(93.5%) = 86.02%. Actual chance is (100% - 8% - 6.5%) = 85.5%. Should the formula be the following instead?
(1-(0.08-F23)-(0.065-(F25*0.0025)))
Right you are.
Values:
Edit 3: Ok, now for some random numbers.
On a naked human, properly spec'd, with nothing but a Titansteel Bonecrusher and a Titansteel Shield Wall (as every fresh 80 should easily be born with):
Stat Base Value Total
STR 173 36 240
STA 193 128 321
SPI 108 0 108
AGI 90 0 90
INT 98 0 98
SP 0 0 96
AP 220 0 700
Somehow, with a tight rotation, this should be able to put out:
Ability Min Max Avg DPS
AS 2695.14 3264.15 2766.90 92.23
HotR 1052.66 1052.66 977.50 162.92
ShoR 1258.11 1258.11 1168.28 194.71
Judge 242.84 319.19 273.88 30.43
SoB 210.44 331.33 251.54 184.47
Exo 1765.99 1947.60 1642.52 91.25
Cons 1351.07 1351.07 1322.36 146.93
White 438.42 690.27 524.05 209.62
Total: 1112 dps.
Puts a lot of pug dps to shame, eh?
Edit 4: Actually, that number is slightly on the low side - I haven't factored in the increased Stamina talents, and therefore the subsequent increase in SP. Still nice though!
Looks nothing short of amazing :)
I'll get around to making the changes to my spreadsheet in a while as well but atm i've got things to do.
Post by
Squishalot
Yea i started out by assuming that but i think i noted a couple of posts back that i skipped this since we didnt know if we were gonna go for a 969 or FCFS style rotation.
Thus i added all the spells to be used ON cooldown so we can determine what kind of rotation we are to use.
Fair enough. We might need to just pick a rotation and stick with it, for the DPS number to be accurate.
It is and has always been capped ;p Expertise was the first number to get capped on the spreadsheet because it was giving me the most trouble.
How did you add Combat expertise if you need to multiply it by 100? oO
For some reason when I downloaded it, it wasn't capped. Regarding multiplying by 100, I had it as 6%, because it was being used for something else as a %.
Hmm.. Arn't you double dipping the hit now? It's already added in the AVG dmg which is used to determine the DPS. Unless i'm missing something.
Not really - that's to determine if the SoB proc hits. But SoB won't proc at all if the original melee attack hits - previously, we would be assuming that even if the melee attack misses, SoB will have its hit rate and its damage.
I'll get around to making the changes to my spreadsheet in a while as well but atm i've got things to do.
Don't worry too much about it - I've made quite a few cosmetic changes, and it's a pain to try to replicate each formula exactly, so I'll upload a new copy later.
I showed it to my brother tonight, and his first words were "OMG, no wonder you're so undergeared, you're wasting your time on this stuff, i mean, look at that formula! *points at avg dmg*..."
"... how many other people have this spreadsheet...? ... gah, man, we could've sold this online, do you know how many wow nerds there are?"
Edit:
Calculations:
Exorcism - why is damage multiplied by 1.2, before adding talents and whatnot? Are you assuming Glyph of Exorcism?
Post by
svirve
Fair enough. We might need to just pick a rotation and stick with it, for the DPS number to be accurate.
Indeed but we have to make sure it's the optimal rotation as well.
For some reason when I downloaded it, it wasn't capped. Regarding multiplying by 100, I had it as 6%, because it was being used for something else as a %.
Cant you just change it to an integer value? oO Not that it changes much but it's going to be wierd seeing it be multiplied by 100 ^^
Not really - that's to determine if the SoB proc hits. But SoB won't proc at all if the original melee attack hits - previously, we would be assuming that even if the melee attack misses, SoB will have its hit rate and its damage.
Aah right you are!
Don't worry too much about it - I've made quite a few cosmetic changes, and it's a pain to try to replicate each formula exactly, so I'll upload a new copy later.
Hehe sounds good =)
I showed it to my brother tonight, and his first words were "OMG, no wonder you're so undergeared, you're wasting your time on this stuff, i mean, look at that formula! *points at avg dmg*..."
"... how many other people have this spreadsheet...? ... gah, man, we could've sold this online, do you know how many wow nerds there are?"
Lol maybe we should sell it online ;p I'll take it off the filehost site and limit the damage to who's gotten a hold of it already ;p
Edit:
Calculations:
Exorcism - why is damage multiplied by 1.2, before adding talents and whatnot? Are you assuming Glyph of Exorcism?
Yea assuming glyph of exo.
Post by
Squishalot
Cant you just change it to an integer value? oO Not that it changes much but it's going to be wierd seeing it be multiplied by 100 ^^
I actually have, in my more final copy. It uses a more roundabout way to do it though - I've got a list of talents and associated stat increases. So Strength bonus is based on Divine Strength, Stamina bonus is based on Sacred Duty and Combat Expertise, Expertise is based on Combat Expertise.
The stamina bonus from Combat Expertise is now (talent points * 0.02), the Expertise bonus is now (talent points * 2), so there's no more "*100" in my formula.
Lol maybe we should sell it online ;p I'll take it off the filehost site and limit the damage to who's gotten a hold of it already ;p
I don't think many people can do much 'damage' from what's on the website, so you can leave it up for people to play with, as a template.
To make it a sellable product, we'd need to make all the advancements we've discussed to it - talents and all abilities, the ability to define a rotation, include a HPS measure, perhaps?
My brother asked me if it could be applied to a Druid or Lock - his two characters. The thought crosses my mind - it probably could, but the DPS would be more difficult to model, because they rely more on talent procs and the like - sorta like guestimating when we'd use Exo in 3.2 based on AoW procs.
At the least, I suppose, it could tell you how much extra damage your Wrath spam would do?
Anyway, maybe it's time to stop spamming these forums and start taking it 'offline' to finish off this project, then come back when we've done the stat weightings and post some results.
Post by
svirve
I actually have, in my more final copy. It uses a more roundabout way to do it though - I've got a list of talents and associated stat increases. So Strength bonus is based on Divine Strength, Stamina bonus is based on Sacred Duty and Combat Expertise, Expertise is based on Combat Expertise.
The stamina bonus from Combat Expertise is now (talent points * 0.02), the Expertise bonus is now (talent points * 2), so there's no more "*100" in my formula.
Sounds great :) cant wait to see what changes you've made so far ;p
I don't think many people can do much 'damage' from what's on the website, so you can leave it up for people to play with, as a template.
To make it a sellable product, we'd need to make all the advancements we've discussed to it - talents and all abilities, the ability to define a rotation, include a HPS measure, perhaps?
My brother asked me if it could be applied to a Druid or Lock - his two characters. The thought crosses my mind - it probably could, but the DPS would be more difficult to model, because they rely more on talent procs and the like - sorta like guestimating when we'd use Exo in 3.2 based on AoW procs.
At the least, I suppose, it could tell you how much extra damage your Wrath spam would do?
Anyway, maybe it's time to stop spamming these forums and start taking it 'offline' to finish off this project, then come back when we've done the stat weightings and post some results.
I'm not really looking to sell it actually, it's a spreadsheet just like many others. It's not going to be hard for anyone with a high-school degree in math to do a new one.
Though fixing everything will be done. Right now i just feel like waiting for 3.2 before i put more time than i already have into this.
What spec is your brother?
Boomkin spread
,
Resto
,
Feral
,
Warlock
.
There are spreadsheets for almost all classes and specs. The reason to doing one for protribution is because it's (to my best knowledge) uncharted territory.
If we were to take it "offline" for now i say we should wait until 3.2 is released and hot fixed.
We could finish the cosmetic changes before (adding proper talents and such like you've been doing lately) 3.2 hits but wait with anything major until afterwards.
P.S. I apologize if i sound rough atm but it's 6:30 AM and i havn't slept yet >.< off to bed now though!
Post by
Squishalot
Haha, sleep is a wonderful thing, go get some!
I was thinking something more flexible than the spreadsheets that are out there. You're right, there isn't a protribution sheet, but that's because all the other sheets are basically designed for cookie cutter specs, not paper dolls that you can dress up and individualise.
Agreed that a spreadsheet isn't a sellable product, I just thought I'd float it if you were interested in taking the effort to turn it from being 'just another spreadsheet' (which is where it is now) into something that would be much more marketable. I didn't start work on this either with the intent of selling it (hell, I didn't even think of it until my brother pointed it out), but it just crossed my mind based on the amount of customisation it'd need to meet his requirements, that it's not worth it as an 'on the side' thing.
But yep, waiting for 3.2 will be a good idea. The Ret comparison shouldn't be too difficult to make though, so I'll plough on with that in the meantime.
Post by
TheJohan
Dear Sirs.
Last night i ran VH normal with my paladin. Standard "Consecrating your way to ding"-protbuild. As dps.
Reason was my friend wanted to tank with his shiny new DK.
I must say i was pretty amazed with the dps you can do, just changing from SoL to SoB, using a non-tank oriented weapon and using BoM instead of BoSanc. :)
As a lvl 77 paladin (build and gear in my armorylink, see below) i did about 1.1k dps through the zone.
Not that it really matters but i out-dpsed a lvl 79 dk (not the tank) and a lvl 76 hunter. (I also did more total damage)
Very amusing when we where done, and i mentioned that i had done more dps than the others as Prot. *muahahaha*
Post by
svirve
I was thinking something more flexible than the spreadsheets that are out there. You're right, there isn't a protribution sheet, but that's because all the other sheets are basically designed for cookie cutter specs, not paper dolls that you can dress up and individualise.
I'm not sure i am following you here. How should we make it more flexible other than (practically) embedding a ret spreadsheet for comparison?
Most spreadsheets i've used actually go so far as to add every single piece of worthwile gear. The ones that do also include enchants and go to the lengths of modelling trinkets, enchant procs etc..
Agreed that a spreadsheet isn't a sellable product, I just thought I'd float it if you were interested in taking the effort to turn it from being 'just another spreadsheet' (which is where it is now) into something that would be much more marketable. I didn't start work on this either with the intent of selling it (hell, I didn't even think of it until my brother pointed it out), but it just crossed my mind based on the amount of customisation it'd need to meet his requirements, that it's not worth it as an 'on the side' thing.
To be honest i feel this is far from 'just another spreadsheet' ^^ there's still alot to be done.
If we were to make it into a marketable product we should definitely step out of spreadsheets and make it into a real program like rawr, which would require
alot
of work.
Once again i'm staying up late (for some odd reason, perhaps cause my parents are out of town) so i might sound harsh but my smilies seem to go to bed before i do >.< (hmm i guess he's to grumpy to). Anyways it's all in my best intent.
Post by
Squishalot
I'm not sure i am following you here. How should we make it more flexible other than (practically) embedding a ret spreadsheet for comparison?
Most spreadsheets i've used actually go so far as to add every single piece of worthwile gear. The ones that do also include enchants and go to the lengths of modelling trinkets, enchant procs etc..
Talent specs, rotations, in addition? Modelling for random procs is something that could be simulated with more complexity. Think something more along the lines of my "27% chance of SoC being better than SoB" calculations, rather than a more basic "average uptime" calculation.
To be honest i feel this is far from 'just another spreadsheet' ^^ there's still alot to be done.
If we were to make it into a marketable product we should definitely step out of spreadsheets and make it into a real program like rawr, which would require alot of work.
I don't know that stepping out of spreadsheets is necessary as such, unless you've got a programming background. I do most of my RL work modelling in Excel and use VBA for running scripts and designing the GUI. Ugly, but incredibly functional. Even without marketing it, I'm still thinking of trying to create a GUI to make the presentation neater for my own purposes.
Post by
svirve
Talent specs, rotations, in addition? Modelling for random procs is something that could be simulated with more complexity. Think something more along the lines of my "27% chance of SoC being better than SoB" calculations, rather than a more basic "average uptime" calculation.
I was under the impression you had added most talents already (with an on/off function to a degree), and yea rotations is a good point. Though all of which i think should be put on hold until 3.2 hits live and is hotfixed a couple of times. We might as well lose all the viability of this spec when it hits. After that though and if it still proves to be viable, i'll be more than happy to model random events.
I don't know that stepping out of spreadsheets is necessary as such, unless you've got a programming background. I do most of my RL work modelling in Excel and use VBA for running scripts and designing the GUI. Ugly, but incredibly functional. Even without marketing it, I'm still thinking of trying to create a GUI to make the presentation neater for my own purposes.
Ofcourse it wont be necessary but it would open up for a more user friendly GUI and most of all improving the looks.
If you would've asked me about a programming background 3 years ago i would've said 'hell yea!' but that's behind me know and i barely remember anything.
I love excel but the problem is it's hard to make it very informative and pretty (i hate that word when it comes to math) but still being functional.
VBA i'm not familiar with so i'm not aware of what it's capable of.
The problem remains for it to be a marketable product it has to be pretty (ugh), functional, easy to use and informative. But nevermind it's not going to be a marketable product anyways so no need to nit pick over this =).
Post by
Squishalot
To make things marketable, you have to take the ugly out of it. In Excel, the formulas are ugly, so you basically do everything in a flashy front-end and don't let users even get to the back end calculations. From a model presentation perspective, users shouldn't need to open up an actual spreadsheet, they should be able to get everything within a few clicks in the GUI.
But nevermind it's not going to be a marketable product anyways so no need to nit pick over this =).
Too true. A mate of mind reminded me of the numerous skill calculator spreadsheets that were produced for Diablo II - game dynamics have come a long way since then...
Calculations...
I'm not sure if it's something that was there originally, or if it's just because of something I've modified...
WPDMG min/max formulas include the bonus from one-handed weapon spec and crusade.
WPDMG total uses the WPDMG min/max formulas.
WPDPS uses WPDMG total.
HotR uses WPDPS,
and multiplies by bonus damage
Are we double counting here?
Post by
svirve
I think we mentioned this a couple of posts back and didn't come to a real conclusion.
To me it feels like double counting the bonuses should be very correct in this situation.
Why:
The bonuses are applied on your paper doll to your damage it even sais "XXX - XXX x
103%
"
Crusade
- Increases
all damage caused
by 3%.
Though it's not unlikely that HotR doesn't use the paper doll DPS but instead uses the unmodified DPS.
I'm not sure what to do to be honest. I made it that way because it felt right to me.
Best way would be to test it. And see if you get a 6% increase or 3% when adding/removing Crusade.
Since HotR does a fixed damage it shouldn't be that hard to determine either.
Too tired to test myself to be honest would most probably just screw up ^^
Post by
299425
This post was from a user who has deleted their account.
Post by
Squishalot
I'm still cant believe that i can get better boss dps as prot spec LOL
Just a quick question tho, should i use a 3% crit damage meta or a 5% SBV meta, I have 28.26% crit chance in my protribution gear atm if that'll help at all
I would have thought generally that 3% crit would up your damage overall, whereas 5% SBV only helps ShoR. Remember to get a Relentless Earthsiege Diamond, so that you don't need to equip two blue gems, and can focus on hit/strength.
Post by
Squishalot
Since svirve is bugging me to post stuff, here goes....
... ok, need to redownload it from my office. So give me a little while, but in the meantime, just a tidbit that I found today:
Calculations...
Why haven't you added in (agility_total / 52.09/100) to crit%? Does this need to be added?
Just realised that it wasn't included when I was shoving in all of the Rogue stats this morning.
Spreadsheet edits
Ok, here goes...
Crusader Strike:
=((1.1*O27)*(1+'Talents After'!B81)*(1+'Talents After'!B85))*(1+'Talents After'!B73)
O27 = MinWPDMG
B81 = Art of War increased damage
B85 = Sanctity of Battle increased damage
B73 = Overall increased damage talents
Ok, so I've branched out across sheets now... 'Talents After' is the name of the sheet that compiles all of the talents that you would select for the 'After' comparison. Overall increased damage talents includes One-Handed Weapon Spec, Crusade, Sanctified Retribution and Vengeance. It doesn't include Two-handed weapon spec, since that only applies to melee weapon based attacks.
Post by
svirve
Since svirve is bugging me to post stuff, here goes....
I like bugging people ;p
Why haven't you added in (agility_total / 52.09/100) to crit%? Does this need to be added? When i was adding crit by a percentage it did not since you would just copy it from your paper doll which already included agility. But now that we've switched to adding crit by rating it should be added! We should also add SPCRIT for exo (if it uses spell crit, i think it does).
Ok, here goes...
Crusader Strike:
=((1.1*O27)*(1+'Talents After'!B81)*(1+'Talents After'!B85))*(1+'Talents After'!B73)
O27 = MinWPDMG
B81 = Art of War increased damage
B85 = Sanctity of Battle increased damage
B73 = Overall increased damage talents
Ok, so I've branched out across sheets now... 'Talents After' is the name of the sheet that compiles all of the talents that you would select for the 'After' comparison. Overall increased damage talents includes One-Handed Weapon Spec, Crusade, Sanctified Retribution and Vengeance. It doesn't include Two-handed weapon spec, since that only applies to melee weapon based attacks.
Ok i'm gonna need a copy of your sheet before i understand how that fits together but i get how you've done it.
Looking forward to the upload =)
Post by
Squishalot
Quick q.... You can open up Excel files in OOo, right? Haven't tried at home yet...
Not sure how the formulae will convert, so here's hoping that it all works out ok! I'll upload it later tonight once I get out of work and can get access to file sharing websites (damn internet filtering here).
Post by
svirve
Quick q.... You can open up Excel files in OOo, right? Haven't tried at home yet...
Not sure how the formulae will convert, so here's hoping that it all works out ok! I'll upload it later tonight once I get out of work and can get access to file sharing websites (damn internet filtering here).
I think i should be able to. Used a very good rogue spreadsheet that was made in excel before so unless you've got some wierd version of excel it should be fine :)
Post Reply
You are not logged in. Please
log in
to post a reply or
register
if you don't already have an account.