Theck's MATLAB thread - Cataclysm/4.x

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Kihra » Tue Dec 06, 2011 11:00 am

Nioknight wrote:Any help/advice would be really appreciated. It goes against my intuition to not use Resolve because it's higher, but I think it might be worse for me.


This is the wrong forum for gear questions. You should take this to the Gear Discussions and Advice forum.
Kihra
 
Posts: 340
Joined: Mon Dec 31, 2007 12:01 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Nioknight » Tue Dec 06, 2011 5:28 pm

Kihra wrote:
Nioknight wrote:Any help/advice would be really appreciated. It goes against my intuition to not use Resolve because it's higher, but I think it might be worse for me.


This is the wrong forum for gear questions. You should take this to the Gear Discussions and Advice forum.


Thanks, fixed it.
Nioknight
 
Posts: 6
Joined: Mon Dec 05, 2011 10:40 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Wed Dec 07, 2011 6:14 am

Nioknight wrote:Alright, this is my first time posting but I love the calculations and everything
First point:
I got Souldrinker on my Protection Paladin last night, it's amazing. (I could do logs for you guys but I'm not sure how)

EDIT:
Also the heals seem to definitely be more than 3% healing done from the proc off Souldrinker (1.5% doubled). I had 201,000 hp with flask and my own buffs, and it healed me for about 6700 hp a proc (instead of 6k hp if 3%). Divinity might be effecting it?
And overheals from Souldrinker do NOT count towards Scales of Life effect


If you have time, there's a list of tests that need to be performed in the Call to Arms thread. That post also has instructions for how to record and upload a combat log to WoL.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Sat Dec 10, 2011 7:57 am

Ret 2-piece sims are posted, see table of contents.

<edit> I also stumbled across a bug regarding the prot 2-piece. Judgement's absorb bubble isn't being modified by Inquisition properly. It's a simple fix, but I won't be able to get around to it until later today or tonight; I'll update the code and results at that point. This should only affect the rotation sims, as all of the others use a standard 969 rather than the Inq rotation.

<edit 2> Fixed, posts updated.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Jeremoot » Sat Dec 10, 2011 1:47 pm

Thanks for all the hard work Theck (as well as tlitp0, iminmmnni and anybody else involved with the matlabadin project), you've really been doing the community a great favor for years.
User avatar
Jeremoot
 
Posts: 423
Joined: Tue Feb 08, 2011 5:36 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby madmessias » Sat Dec 24, 2011 4:41 pm

If you have less then 1.5s left on Inquisition, and you have time to do crus strike+AS(grand crusader) (but the AS won't benefit from the inq this way) is it worth doing AS first? What about without GS procc? I assume it's CS this way, but not really sure what you should do in the situation of AS(GS+INQ) or just AS and go without inq (but still keep the GS procc)
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Sat Dec 24, 2011 6:27 pm

I think that entirely depends on what you're trying to optimize. If you want the most damage within those two GCDs, you'd use AS first. If you want the highest stochastic DPS, you'd use CS first, because otherwise you might be reducing your holy power generation rate. The answer will change slightly whether you start that scenario at 0, 1 or 2 holy power and how much time is left on Grand Crusader, and what time scale you're looking to optimize over.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby madmessias » Sat Dec 24, 2011 7:03 pm

theckhd wrote:I think that entirely depends on what you're trying to optimize. If you want the most damage within those two GCDs, you'd use AS first. If you want the highest stochastic DPS, you'd use CS first, because otherwise you might be reducing your holy power generation rate. The answer will change slightly whether you start that scenario at 0, 1 or 2 holy power and how much time is left on Grand Crusader, and what time scale you're looking to optimize over.

Ah forgot to say, i want optimized dmg overall on a fight. This is after a SHOTR btw incase it matters. So just do crus strike i suppose then, or did i get it wrong?

Merry christmas btw :)
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Sat Dec 24, 2011 7:22 pm

If it's immediately following a SotR, then yes, I suspect that CS first would be better simply because of Holy Power Generation. If you lead with AS, it takes you at least 4 GCDs to get to the next finisher (SotR-AS+-CS-X-CS-F). Leading with CS shortens that to a potential 3-GCD span (SotR-CS-AS+-CS-F). Increasing HPG almost always leads to a DPS increase.

I think the queue that would best test for this would be something like SDSotR>ISotR>Inq>AS+[buffInq<1.5]>CS>AS+>J>AS>Cons>HW. I can certainly throw that in the sims, but in all honesty it'll be such a rare event that I doubt it will make any difference in overall DPS.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby madmessias » Sat Dec 24, 2011 7:28 pm

theckhd wrote:If it's immediately following a SotR, then yes, I suspect that CS first would be better simply because of Holy Power Generation. If you lead with AS, it takes you at least 4 GCDs to get to the next finisher (SotR-AS+-CS-X-CS-F). Leading with CS shortens that to a potential 3-GCD span (SotR-CS-AS+-CS-F). Increasing HPG almost always leads to a DPS increase.

I think the queue that would best test for this would be something like SDSotR>ISotR>Inq>AS+[buffInq<1.5]>CS>AS+>J>AS>Cons>HW. I can certainly throw that in the sims, but in all honesty it'll be such a rare event that I doubt it will make any difference in overall DPS.

If you dont mind putting it in the sim i'd be very greatful, i just hate making choices that decreases your dps, or getting to "waypoints" in your rotation where you just randomize what you do ^^
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby madmessias » Sat Dec 31, 2011 9:21 am

Would it be worth tracking landslide and maybe not using Inquisition if it's a short duration left on Landslide (if it's up) and you will maybe only get a shotr with LS up, but if you use inquisition you will only get a crus strike.. Would be nice (if you think it even matters) if you do a list of at what seconds left it would be worth using Inquisition.
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Mon Jan 02, 2012 7:31 pm

It's not possible to track Landslide with the current version of the FSM code, and it's not worth trying to re-vamp it for what's probably a minor effect. Remember that 1000 AP only adds about 600 raw damage to SotR, which isn't going to be enough to make up for 30% of either AS/J, let alone seal/censure damage.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby econ21 » Tue Jan 03, 2012 9:48 am

On the weapon comparisons, is it worthwhile trying to model Soul drinker's proc? Since it seems to be one of only two viable end-game weapons for us (Hand of Morchok being the other), it would be interesting to know which gave higher dps. If it can't be modelled, do we have any data (e.g. from Logs) that could address the issue empirically or at least with a back of the envelope calculation?

I've read the thread comparing the two weapons in the gear section and the consensus seems to be that Souldrinker will give more dps, perhaps even if from a lower tier. From an armchair point of view, I'd be interested in comparing the LFR Souldrinker with the normal Hand of Morchok, but heroic raiders may want to compare normal Souldrinker with heroic Hand. I'd make the former comparison myself if I were lucky enough to have both weapons.
econ21
 
Posts: 1217
Joined: Tue Apr 01, 2008 8:53 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Tue Jan 03, 2012 12:33 pm

The proc is modeled in an updated version of the code, I just haven't had time to update the sims on the front page. With any luck, I'll get to that some time this week.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby madmessias » Tue Jan 03, 2012 12:54 pm

Sorry if im asking for to much, but a trinket comparison would be nice :) Using license to slay+essence of the eternal flame atm, duno if any DS trinkets are better. (dumb chimaeron one won't drop, heard it's alot better tho)
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Jeremoot » Tue Jan 03, 2012 1:30 pm

madmessias wrote:Sorry if im asking for to much, but a trinket comparison would be nice :) Using license to slay+essence of the eternal flame atm, duno if any DS trinkets are better. (dumb chimaeron one won't drop, heard it's alot better tho)


The VP Trinket is really nice, probably the best for us single target. Next would be the Creche.

I have no idea how well the whirlwindy trinket scales with vengeance, but if it gets the full effect it may be pretty awesome.
User avatar
Jeremoot
 
Posts: 423
Joined: Tue Feb 08, 2011 5:36 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Tue Jan 03, 2012 1:46 pm

madmessias wrote:Sorry if im asking for to much, but a trinket comparison would be nice :) Using license to slay+essence of the eternal flame atm, duno if any DS trinkets are better. (dumb chimaeron one won't drop, heard it's alot better tho)

Again, I can't easily sim out DPS values for trinkets that provide buffs because the FSM code doesn't track them. For things like License to Slay, which provide more-or-less constant stat bonuses, we can calculate estimated DPS values based on their average stat contributions using the coefficients in the Stat Comparison. For trinkets with temporary effects, it should be close enough to make time-averaged estimates of their total stat contribution.

I'm incredibly busy right now, so it's unlikely I'll get to any of that. If you take the time to post the details on the trinkets you're interested (static bonuses, buffs, buff durations+internal cooldowns or effective uptimes) then I can do the quick-and-dirty version based on the stat weights.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Awyndel » Tue Jan 03, 2012 3:46 pm

Slay and Chimaeron prolly still rate amongst the highest ones, specially if you consider you will have to swap less items for hit/exp, and thus lose less str swapping to lower items. T12 prot+t13 ret would be nice set bonuses here.
User avatar
Awyndel
 
Posts: 595
Joined: Sat Feb 14, 2009 8:49 am
Location: The Netherlands

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Jeremoot » Tue Jan 03, 2012 5:57 pm

Let me rephrase my last post. The VP trinket / Creche are best if you already have expertise soft cap. Otherwise the Heart of Rage is the way to go.
User avatar
Jeremoot
 
Posts: 423
Joined: Tue Feb 08, 2011 5:36 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Fri Jan 06, 2012 6:26 am

I've updated a bunch of the sims to include the Souldrinker proc (post-hotfix). There are still a few I need to run (Rotations and Enchants primarily), but those should be updated within a day or so.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby theckhd » Fri Jan 06, 2012 8:17 pm

All of the sims should be updated now. It looks like the model for Landslide wasn't taking into account Censure stacking anyhow, so no change was necessary.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 6.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 6211
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Glossary

Postby madmessias » Sat Jan 07, 2012 1:44 am

does Avenging Wrath change the way you should use Inquisition on singletarget? Since it's not much of a big increase, atm I am not using it at all during AW but got a feel that might be wrong, maybe only use it at 10 or more seconds (or perhaps 12, since that's how long Inq lasts)?
My reasoning to using it was you want to get as many finishers as possible off with the 20% buff.
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby Awyndel » Sat Jan 07, 2012 10:43 am

Thnx for the quick updates after the hotfix Theck, you're the best.
User avatar
Awyndel
 
Posts: 595
Joined: Sat Feb 14, 2009 8:49 am
Location: The Netherlands

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby tlitp » Sat Jan 07, 2012 1:13 pm

theckhd wrote:It looks like the model for Landslide wasn't taking into account Censure stacking anyhow, so no change was necessary.

We've tested the triggers of LS/WW one year ago, and then I've periodically glanced at random logs to check any anomaly. To the best of my knowledge, the application of Censure never triggered LS/WW during Cataclysm. The model was and still is "working as intended". :P

madmessiah wrote:does Avenging Wrath change the way you should use Inquisition on singletarget? (I) Since it's not much of a big increase (II), atm I am not using it at all during AW but got a feel that might be wrong, maybe only use it at 10 or more seconds (or perhaps 12, since that's how long Inq lasts)?
My reasoning to using it was you want to get as many finishers as possible off with the 20% buff. (III)

I. No. Using Inq in single-target scenarios is good practice, with or without AW.
II. You're probably underestimating the impact of weaving Inq between SotR casts. The ~400 DPS difference you're seeing between queues 19 and 38 (link) scales with hit/exp, and doesn't take into account AW/BL/HoW "execute" phase.
III. Do keep in mind that in the 12s timespan you're getting at least 1 J, 4 Cens ticks, 13 SoT procs. All of them are affected by Inq. That's not even counting the occasional AS/HoW, I've only listed stuff that will certainly be there. Yeah, sure, SotR is supposed to be Prot's big hitter. That doesn't mean that Inq isn't trailing it, output-wise.
User avatar
tlitp
 
Posts: 487
Joined: Mon Jul 27, 2009 3:25 pm

Re: Theck's MATLAB thread - Cataclysm/4.x

Postby madmessias » Sat Jan 07, 2012 6:33 pm

tlitp wrote:
theckhd wrote:It looks like the model for Landslide wasn't taking into account Censure stacking anyhow, so no change was necessary.

We've tested the triggers of LS/WW one year ago, and then I've periodically glanced at random logs to check any anomaly. To the best of my knowledge, the application of Censure never triggered LS/WW during Cataclysm. The model was and still is "working as intended". :P

madmessiah wrote:does Avenging Wrath change the way you should use Inquisition on singletarget? (I) Since it's not much of a big increase (II), atm I am not using it at all during AW but got a feel that might be wrong, maybe only use it at 10 or more seconds (or perhaps 12, since that's how long Inq lasts)?
My reasoning to using it was you want to get as many finishers as possible off with the 20% buff. (III)

I. No. Using Inq in single-target scenarios is good practice, with or without AW.
II. You're probably underestimating the impact of weaving Inq between SotR casts. The ~400 DPS difference you're seeing between queues 19 and 38 (link) scales with hit/exp, and doesn't take into account AW/BL/HoW "execute" phase.
III. Do keep in mind that in the 12s timespan you're getting at least 1 J, 4 Cens ticks, 13 SoT procs. All of them are affected by Inq. That's not even counting the occasional AS/HoW, I've only listed stuff that will certainly be there. Yeah, sure, SotR is supposed to be Prot's big hitter. That doesn't mean that Inq isn't trailing it, output-wise.

kk, but what if it's like 1s left on AW, still shouldn't do a finisher (even without SD)?
madmessias
 
Posts: 17
Joined: Sun Jun 12, 2011 6:45 am

PreviousNext

Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: No registered users and 1 guest

cron

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