Remove Advertisements

A new protection rotation

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Re: A new protection rotation

Postby Martie » Mon Dec 20, 2010 3:29 pm

theckhd wrote:This isn't correct. If you're ignoring Cons, it'll end up being closer to 66% AS, 33% HW, and will vary significantly with your hit and expertise. See the Analytical Model of 939 post in the MATLAB thread if you're curious about the details.

I know there are several things in my spreadsheet that has flaws in it due to assuming Sacred Duty and Grand Crusader procs regardless of the skill that triggers it hitting, but best I could figure, incorporation those would probably favor the HotR rotation. I guess I should just figure out a way to simulate combat.

The 33%-66% number doesn't seem right. Especially for the HotR rotation. I mean, one out of every two fillers will be a guaranteed Avenger's Shield. The other is a proc, which works out to hit chance * (1-(0.8^3).
I'm about to drop the hammer...
...and dispense some indiscriminate justice!

This is my armory profile
Martie
 
Posts: 246
Joined: Tue Jul 29, 2008 9:27 pm

Re: A new protection rotation

Postby theckhd » Mon Dec 20, 2010 4:27 pm

Martie wrote:The 33%-66% number doesn't seem right. Especially for the HotR rotation. I mean, one out of every two fillers will be a guaranteed Avenger's Shield. The other is a proc, which works out to hit chance * (1-(0.8^3).


Except that's not what the proc works out to be, for reasons that are explained in the post I linked you to:

Theck wrote:Avenger's Shield, Holy Wrath, and Consecration
Grand Crusader makes dealing with the "X" in the rotation a bit complicated. It's not as simple as just saying "we cast AS every other cycle, and Holy Wrath in the alternate ones if we don't get a GC proc." This is because if you do get a GC proc, you've put AS on cooldown and the next cycle becomes the "HW if no GC" portion.

Instead, you have to ask the question, "for any given cycle, what's the chance that AS will be off of cooldown when X occurs." It's fairly easy to figure this out, starting from an initial cycle where AS is not on cooldown.


Go read the post I linked to see what the actual solution is.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7710
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: A new protection rotation

Postby theckhd » Mon Dec 20, 2010 6:05 pm

Nikachelle wrote:I might be looking at my buffs wrong, but I was finding that if I use 3 HoPos to refresh Inquisition BEFORE Inquisition has actually fallen off, it doesn't actually refresh it but instead continues to countdown from the same seconds it was at before. I only noticed this a few times and stopped paying attention to this afterwards, but does anyone know if this is accurate? If it is, it doesn't make much sense to me.


I wasn't able to reproduce this. It seems to be reapplying fine for me.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7710
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: A new protection rotation

Postby balloonknot » Mon Dec 20, 2010 7:45 pm

martie - maybe i'm missing something here but theck's code is freely available and while i've not tested it on anything other than matlab, it likely runs in freely available software such as octave. If your intension is to locate a missing rotation or an error in the simulation code you should direct your efforts toward that endeavor rather than making random statements about what is best/better/whatever without identifying specific issues.
balloonknot
 
Posts: 13
Joined: Wed Jan 27, 2010 6:30 pm

Re: A new protection rotation

Postby tlitp » Mon Dec 20, 2010 10:31 pm

Martie wrote:numbers

When people ask for your work, you're expected to post code, not raw numbers.
Anyway, just by looking at what you've provided, some glaring issues :
a. HotR doesn't proc SoT
b. J (through JotJ's application of the aura) does proc SoT
c. handling of GrCr

I suspect there are many more glitches in your calculations (n.b. : I'm not trying to dwarf your efforts, I'm merely emphasizing that you've worked on them for a few days; matlabadin's models were refined during several months). Some people (*cough* Knaughty) think that it's unreasonable to ask a neophyte to RTFsource_code. Citing the good Francis Bacon : "Mahomet cald the Hill to come to him, againe, and againe; And when the Hill stood still, he was neuer a whit abashed, but said; If the Hill will not come to Mahomet, Mahomet wil go to the hil." :P

We really support the idea of people bringing in new stuff to "chew on". However, we can't create miracles out of anecdotes - that's the reason we ask for a methodical approach (testing/brainstorming/code).


The OP is mostly fine, it just needs to be updated with the correct model of GrCr/SD ([RM] 19-31).
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

Re: A new protection rotation

Postby theckhd » Mon Dec 20, 2010 11:50 pm

tlitp wrote:

The OP is mostly fine, it just needs to be updated with the correct model of GrCr/SD ([RM] 19-31).

I thought I already did update that. The binopdf treatment of CS is in there already. The only thing from those lines that's missing is line 22, which is trivial enough that it's probably not needed.

Am I missing something? (Possible, as it's 2am and I'm nodding off).
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7710
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: A new protection rotation

Postby Martie » Tue Dec 21, 2010 12:44 am

tlitp wrote:When people ask for your work, you're expected to post code, not raw numbers.
The numbers show the general idea I had and numbers I ran - I posted it here because I needed feedback. I'm a horribly arrogant ass, I know, but I don't really need feedback from someone who can't see what I did from those numbers.[/quote]

Anyway, just by looking at what you've provided, some glaring issues :
a. HotR doesn't proc SoT
b. J (through JotJ's application of the aura) does proc SoT
c. handling of GrCr

Hrm, that means SoT works differently than I expected. Still, you are right, and it's easily enough modified.

I suspect there are many more glitches in your calculations (n.b. : I'm not trying to dwarf your efforts, I'm merely emphasizing that you've worked on them for a few days; matlabadin's models were refined during several months).
If I was sure I had no glitches in there, I wouldn't have asked for help or opinions and I would have used some very definitive language in my original post.

Some people (*cough* Knaughty) think that it's unreasonable to ask a neophyte to RTFsource_code. Citing the good Francis Bacon : "Mahomet cald the Hill to come to him, againe, and againe; And when the Hill stood still, he was neuer a whit abashed, but said; If the Hill will not come to Mahomet, Mahomet wil go to the hil." :P
Some people object to being called a neophyte by people that know nothing about them. (Hint: My account here is older than yours.) Regardless, I came here looking for help, with a well thought out idea. I'm getting the help I needed.


We really support the idea of people bringing in new stuff to "chew on". However, we can't create miracles out of anecdotes - that's the reason we ask for a methodical approach (testing/brainstorming/code).

I'm not providing anectdotal evidence, and I'm almost offended at the suggestion that I'd believe that anecdotal evidence is anything more than anecdotal.

If you want to know my method? It's simple. I dug through logs checking my ability damage. I noticed that Shield of the Righteous is doing low enough damage that it's possible that SotR*0.3+HotRnova*4*0.3+Judgement*0.3+Filler*0.3+Censure damage over 12 seconds*0.3+Melee seals over 12 seconds*0.3 > SotR. I started running numbers to confirm or deny it, and after I thought I saw it confirmed, I decided to check it with the community.
I made some mistakes and had some wrong information (and worked off of unconfirmed ideas, silly me.) That doesn't mean there's no method, that doesn't mean I'm some misguided idiot who thinks he's reinvented the wheel.

Here's my spreadsheet. Yes, it's messy.
Martie
 
Posts: 246
Joined: Tue Jul 29, 2008 9:27 pm

Re: A new protection rotation

Postby Nikachelle » Tue Dec 21, 2010 7:33 am

theckhd wrote:
Nikachelle wrote:I might be looking at my buffs wrong, but I was finding that if I use 3 HoPos to refresh Inquisition BEFORE Inquisition has actually fallen off, it doesn't actually refresh it but instead continues to countdown from the same seconds it was at before. I only noticed this a few times and stopped paying attention to this afterwards, but does anyone know if this is accurate? If it is, it doesn't make much sense to me.


I wasn't able to reproduce this. It seems to be reapplying fine for me.

I double checked this last night and it seems I was wrong. My mistake!
User avatar
Nikachelle
Maintankadonor
 
Posts: 11000
Joined: Mon Mar 23, 2009 10:39 am
Location: Toronto, Canada

Re: A new protection rotation

Postby theckhd » Tue Dec 21, 2010 9:20 am

Martie wrote:I'm a horribly arrogant ass, I know, but I don't really need feedback from someone who can't see what I did from those numbers.


Interjecting as a moderator here, I think the issue is less "I'm a horribly arrogant ass," and more "I post unsubstantiated opinions without reading the relevant source material."

For example, you obviously at least skimmed the single-target rotation simulation post before starting your calculations, but you didn't read it thoroughly enough to realize that it already contained data for most of the situations you were considering. Even if there was some confusion about the Inquisition trigger conditions, a simple post asking about it ("Does having Inq at the top of the queue mean you recast it at 3 holy power even if it clips the duration?") could have cleared that up and saved you some work.

Similarly, a few posts ago I noted that your Grand Crusader model was incorrect, and linked you to the correct one. Instead of reading that, you just replied stating "that seems wrong," despite the fact that both the reasoning and the error in the simple model you presented are very clearly laid out in the post I linked.

I'm not suggesting you stop posting, just that you should slow down a bit and do a little searching and reading. I try and set the bar a little higher for this sub-forum, because I think that AT&C should be more like EJ (though less caustic) and less like Tankspot or the rest of Maintankadin. As such, you don't gain reputation with AT&C by posting hastily and inaccurately, you gain it by reliably posting solid results that show your work in as much detail as possible.

As an example: Your account may be older than tlitp's (and mine), but he knows more about paladin mechanics than anyone else on this forum, me included (and I consider myself pretty well-informed). The quantity of posts may be low, but the quality is always superb. That's why people listen to him - he's very rarely wrong

(Trust me on this one, he's depressingly accurate. My only gripe about working with him on matlabadin is that he's right all the damn time. Rarely do people make me feel stupid as often as he does.)
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7710
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: A new protection rotation

Postby johninnewport » Tue Dec 21, 2010 10:36 am

First off I want to thank you for all your work. Not only can I not do this math, I am far too lazy...

I have been applying the basic principles you outlined as well as the rotation. I have successfully cleared all the Heroics and have gear that is average item level 350. I am expertise 26 and hit around 8%.
I have no problem with threat.

I am however disappointed a little bit in the DPS I am putting out. The most I can get to is 7k assuming its a trash pull with multiple mobs and I dont have to use WOG ever.
But more realistically I am in the 5-6k range on bosses and trash. Whereas on those same fights I am seeing DPS from ranged and melee over 8k and as high as 10k. Are you guys experiencing the same thing?
johninnewport
 
Posts: 8
Joined: Tue Dec 14, 2010 9:21 am

Re: A new protection rotation

Postby Malthrax » Tue Dec 21, 2010 11:12 am

johninnewport wrote:First off I want to thank you for all your work. Not only can I not do this math, I am far too lazy...

I have been applying the basic principles you outlined as well as the rotation. I have successfully cleared all the Heroics and have gear that is average item level 350. I am expertise 26 and hit around 8%.
I have no problem with threat.

I am however disappointed a little bit in the DPS I am putting out. The most I can get to is 7k assuming its a trash pull with multiple mobs and I dont have to use WOG ever.
But more realistically I am in the 5-6k range on bosses and trash. Whereas on those same fights I am seeing DPS from ranged and melee over 8k and as high as 10k. Are you guys experiencing the same thing?


I don't think anything in normal 5's or heroic 5's is capable of doing enough damage to us ( w/o outright killing us ), and/or dies too fast, to allow Vengeance to stack to a point where we do "ridiculous amounts" of DPS.
User avatar
Malthrax
 
Posts: 726
Joined: Tue Feb 03, 2009 11:23 am

Re: A new protection rotation

Postby johninnewport » Tue Dec 21, 2010 11:24 am

Malthrax wrote:
johninnewport wrote:First off I want to thank you for all your work. Not only can I not do this math, I am far too lazy...

I have been applying the basic principles you outlined as well as the rotation. I have successfully cleared all the Heroics and have gear that is average item level 350. I am expertise 26 and hit around 8%.
I have no problem with threat.

I am however disappointed a little bit in the DPS I am putting out. The most I can get to is 7k assuming its a trash pull with multiple mobs and I dont have to use WOG ever.
But more realistically I am in the 5-6k range on bosses and trash. Whereas on those same fights I am seeing DPS from ranged and melee over 8k and as high as 10k. Are you guys experiencing the same thing?


I don't think anything in normal 5's or heroic 5's is capable of doing enough damage to us ( w/o outright killing us ), and/or dies too fast, to allow Vengeance to stack to a point where we do "ridiculous amounts" of DPS.


I guess my question is more specific - Are these numbers in line with what other protection paladin tanks are experiencing?
johninnewport
 
Posts: 8
Joined: Tue Dec 14, 2010 9:21 am

Re: A new protection rotation

Postby theckhd » Tue Dec 21, 2010 11:53 am

Those numbers seem reasonable based on my experience in heroics. On bigger trash pulls, where I can take full advantage of Inq, HotR, and Cons, I've managed to break 10k.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7710
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: A new protection rotation

Postby inthedrops » Tue Dec 21, 2010 11:55 am

I don't think there's any reason for you to be concerned.

I don't pay too much attention but I'm probably around 4k single target roughly. Depends on if I use wings. On AoE packs I might hit above 10k but that's only when CC broke and I have more targets on me than I actually want. I WoG a lot too and am not terribly concerned about tanking DPS these days.

I used to be a big DPS whore when tanking but Blizz broke that around the time they introduced Vengeance so now I'm just happily doing what I can to help keep things smooth instead of worrying about DPS.
inthedrops
Maintankadonor
 
Posts: 1281
Joined: Mon Oct 29, 2007 9:19 am

Re: A new protection rotation

Postby econ21 » Fri Dec 24, 2010 4:54 am

Nikachelle wrote:I might be looking at my buffs wrong, but I was finding that if I use 3 HoPos to refresh Inquisition BEFORE Inquisition has actually fallen off, it doesn't actually refresh it but instead continues to countdown from the same seconds it was at before. I only noticed this a few times and stopped paying attention to this afterwards, but does anyone know if this is accurate? If it is, it doesn't make much sense to me.


That happened to me last night while soloing as Ret. No idea why.
econ21
 
Posts: 1373
Joined: Tue Apr 01, 2008 8:53 am

PreviousNext

Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: No registered users and 1 guest


Remove Advertisements

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest