Theck's MATLAB thread - MoP/5.x

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Re: Theck's MATLAB thread - MoP/5.x

Postby Schroom » Tue Dec 10, 2013 5:32 am

Thanks a lot for the work, but I am also a bit confused about some stuff...

WoG highest DPS glyph?

doesn't the WOG glyph only procc when you cast wog/Ef on a friendly target and not on yourself?

I mean, we are wasting 3 Holy power, thereby using one less SOTR.

so the damagebuff from the glyph is more DPS than one SOTR? (also it should have implications on AS)


if so, how often do we have to cast Wog/EF on a friendly target for it to be on that spott?

similar question as Thels.

for TMI what defensive impact has AS and even more IT? we don't even use SoT.... at least not for TMI...
User avatar
Schroom
 
Posts: 618
Joined: Mon Aug 15, 2011 2:03 pm
Location: Luxembourg

Re: Theck's MATLAB thread - MoP/5.x

Postby Thels » Tue Dec 10, 2013 7:12 am

Schroom wrote:Thanks a lot for the work, but I am also a bit confused about some stuff...

WoG highest DPS glyph?

doesn't the WOG glyph only procc when you cast wog/Ef on a friendly target and not on yourself?

I mean, we are wasting 3 Holy power, thereby using one less SOTR.

so the damagebuff from the glyph is more DPS than one SOTR? (also it should have implications on AS)

if so, how often do we have to cast Wog/EF on a friendly target for it to be on that spott?


I naturally assumed the user would be specced for Eternal Flame, and wouldn't actively WoG outside of keeping EF up. This is of course debate-able, as people use different 'rules' for keeping EF active. If you don't use Eternal Flame, you probably do not want to use the WoG glyph.

I just checked (I'm EU, so 5.4.1), and with the Glyph, if I cast EF on myself, I do receive the following buff: Glyph of Word of Glory



Schroom wrote:similar question as Thels.

for TMI what defensive impact has AS and even more IT? we don't even use SoT.... at least not for TMI...


IT should have zero effect, since we're not using SoT. AS might have a minuscule effect, since it increases our damage output, and thus reduces the length of the fight, but I don't think that's how the sim works.

The sim does include RNG, so if 2 setups are supposed to give identical results, one might have gotten just a tad luckier than the other on the rare occasion. These minor differences don't matter if one setup is actually stronger than the other, since the difference in strength usually outweights the difference in RNG, but if two setups are identical, RNG will favor one over the other.

If Theck were to run things again, I'm pretty sure we'd get the results in a different order, with some results replaced by others. They'll all have DS, and none of them will have BH. Other than that, they'll be unpredictable.

Though it makes me wonder why AW's not listed in the top one. Is it's benefit so insignificant that RNG knocks it off the top spot, or does the rotation never pop AW?
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby Schroom » Tue Dec 10, 2013 9:52 am

Code: Select all
I just checked (I'm EU, so 5.4.1), and with the Glyph, if I cast EF on myself, I do receive the following buff: Glyph of Word of Glory


ok. thanks just came home and wanted to test it myself (and i will in a coupleof minutes)

allthough this is new then? or did I miss this completely and messed it up? :P
User avatar
Schroom
 
Posts: 618
Joined: Mon Aug 15, 2011 2:03 pm
Location: Luxembourg

Re: Theck's MATLAB thread - MoP/5.x

Postby Thels » Tue Dec 10, 2013 10:08 am

I don't think it has been changed, but I'm not sure. I never actually used the Glyph as pre-EF it wasn't worth it.
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Tue Dec 10, 2013 1:30 pm

Thels wrote:So for DPS, our best glyphs are, in order:

- Focused Shield (Of course, this would only apply against a single target.)
- Word of Glory
- Alabaster Shield
- Final Wrath

So take the top 1, 2 or 3 depending on how many utility glyphs you take, and for fights where you want to damage more than 1 target, replace Focused Shield by the next glyph on the list, or perhaps with Double Jeopardy?

Yep. I haven't simmed DJ yet because it's not properly implemented in SimC yet. On the to-do list, but low priority.

Thels wrote:Though I wonder why the Focused Shield-Holy Wrath-Word of Glory glyph combo has such a high TMI in comparison to the others.

Probably not true to within error margins. The sims are already suffering from the "you generate more HPS than you take in DTPS, on average" problem which exaggerates RNG a bit. I'd have to look back at the raw data, but I think the TMI uncertainty is in the hundreds for those runs.

Thels wrote:The glyphs for lowest TMI are a mess. All 10 entries list DP and 2 other entries at random. The probable cause is that DP is the only glyph that affects TMI.

Yeah, at this point DP is the only glyph that has any significant effect, so the "low TMI" sets are just all DP + random stuff. I haven't decided what (if anything) to do about that yet. In some sense, it's sort of irrelevant since we only have one glyph that matters for TMI, so we could really get by with two setups to see the effect: E/E/E and E/E/DP.



Thels wrote:Also, in either top 10 glyph lists, if the first slot is empty, it includes that line twice.

That's... odd, but I think I can guess why that's happening. Shouldn't be hard to fix.

Thels wrote:For DPS and TMI, does it assume you cast HP on yourself or your opponent? For ES I assume you cast it on your opponent. Perhaps it would be interesting to include two variants of HP and ES. One self-targeted and one boss-targeted, but I don't know how tricky it is to pull that off.

Cast on an enemy. At this point there's no way to do that in the sim itself without some restructuring, unless we define some conditional that can be implemented on the APL to do it for us (i.e. cast it on self if incoming_damage_5s>0.3*max_health or some such). If you want to compare "always cast on enemy" to "always cast on self," we'd need to change APLs while we're changing talents. Could do that by having 5 talent options that tier to include "ESS" and "HPS" variants of the talents (and automatically switch APL when those are chosen). Or just hard-code the one or two special cases we need (i.e. X-Y-Z-ESS, and then compare to the X-Y-Z-ES line to see what changes).
"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 - MoP/5.x

Postby theckhd » Tue Dec 10, 2013 1:31 pm

Thels wrote:I don't think it has been changed, but I'm not sure. I never actually used the Glyph as pre-EF it wasn't worth it.


It has always affected self-heals. Maybe Schroom is remembering when they hotfixed it to stop it from proccing off of offensively-cast WoGs via Harsh Words?
"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 - MoP/5.x

Postby Schroom » Wed Dec 11, 2013 1:25 am

yeah I guess I misinterpreted it back than, and read the hotfix wrong maybe, or something stuck, but wrong context. I tested it myself yesterday and it works great.
User avatar
Schroom
 
Posts: 618
Joined: Mon Aug 15, 2011 2:03 pm
Location: Luxembourg

Re: Theck's MATLAB thread - MoP/5.x

Postby Halide » Wed Dec 11, 2013 3:49 am

Unless i'm interpreting this wrong, wouldn't that mean there is basically no situation to use ES in, single target dps HP is better, suitability (TMI)+ AoE LH is better?
Code: Select all
| Talents | L45 | L60 | L75 | L90 |    DPS |    HPS |   DTPS |   TMI |    Err |   %Err |  SotR |
|  312231 |  EF |  US |  DP |  ES | 388627 | 149833 | 150132 | 272.4 |  54.20 | 19.90% | 73.0% |
|  312233 |  EF |  US |  DP |  HP | 399801 | 149415 | 149725 | 329.6 |  65.80 | 20.00% | 73.0% |

And just checking, ES has the lower TMI, that is just statistical error or whatever? I can't see why ES would be better than HP if both are cast on enemy?
Halide
 
Posts: 6
Joined: Wed Oct 09, 2013 3:15 am

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Wed Dec 11, 2013 7:03 am

Note that the "error" (which is really just a measure of the variance of the distribution in this case) on the TMI measurement is ~60. Thus, the two TMI values are within simulation error of one another, and are indistinguishable for statistical purposes. That's part of the "self-sufficiency" problem when simming against heavier-hitting bosses. I'll probably have to pull some tricks to artificially dampen Vengeance in order to curb that problem.

HP pulls ahead in DPS because of (a) better scaling, and (b) more free GCDs. This is more noticeable with EF than with SS.
"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 - MoP/5.x

Postby econ21 » Thu Dec 12, 2013 4:01 am

The glyph results are intuitive, to me at least. If I am reading the table right, FS is a about a 4.2% dps increase, WoG a 1.9% increase, AS a 1.4% increase and FW a 0.9% increase. The WoG result is close to the 1.8% I got from a back of the envelope calculation:

posting.php?mode=quote&f=4&p=773571

The AS and FW gains are somewhat lower than I inferred from earlier matlab results - maybe that's a gear scaling issue. I often favor FW over AS as it seems that there are quick a few fights where more dps at the end of a fight is more important than more dps at the start (often fights where you would not want to bloodlust on the pull). It might be good to link these results in your glyph and talents guide in due course.

The HP result is surprising - I used to think I was bad for not using ES on some fights for more dps, although I often justified this by my tendency to often forget to use ES enough with its awkward 1 min CD; I find surprisingly many openings to use HP in my rotation (often liking it for its heal, its AOE aggro or its range).
econ21
 
Posts: 1217
Joined: Tue Apr 01, 2008 8:53 am

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Thu Dec 12, 2013 8:36 am

Note that the FW glyph is averaged over the entire fight. So it's really 1/5 of its value during execute phase. Multiply by 5 to approximate how much DPS it grants sub-20% (i.e. around a 4.5% DPS increase if your 0.9% value is correct - I haven't checked your math).
"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 - MoP/5.x

Postby Thels » Thu Dec 12, 2013 8:51 am

That's not really an exact science, as a lot of classes deal more damage in the sub-20% area. Since the boss receives more damage during the last 20%, a Patchwerk style boss will be in this phase for a shorter period of time.

But yeah, for final burn phases like the Dark Shamans, Final Wrath seems excellent.

You might want to take Focused Wrath, so that in encounters with multiple opponents, all damage goes to the mob below 20%.
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Fri Dec 13, 2013 8:23 pm

Thels wrote:That's not really an exact science, as a lot of classes deal more damage in the sub-20% area.

Including us, hence "approximate." But note that what other classes do is irrelevant here - the sim is just a single paladin against a single boss, so the estimation of the DPS gain of Final Wrath is just (encounter-averaged value)*(fight length)/(time spent sub-20%).

edit: and that's roughly (encounter-averaged value)*5 here. If we account for the higher DPS, it's probably a little lower (say 4.5 as a rough guess?).
"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 - MoP/5.x

Postby Thels » Sun Dec 15, 2013 8:02 am

It does make it interesting, though. Right now, it's our 4th DPS Glyph, but on fights where the last 20% is more important than the rest of the fight (Korkron Shamans?), it might very well be worth to take over another DPS Glyph.

It seems unlikely at this time that we'll ever see a Vaelastrasz encounter again.
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby Darielle » Sun Dec 15, 2013 2:53 pm

Not without some massive revamping, no. Any encounter like that would devolve into "STack ALL the Warriors"
Darielle
 
Posts: 853
Joined: Mon Apr 19, 2010 2:41 am
Location: Auckland, New Zealand

Re: Theck's MATLAB thread - MoP/5.x

Postby Schroom » Mon Dec 16, 2013 7:46 am

I prefer final Wrath on fight where you have a lot of execute pahses. like spoils. lots of adds have lots of execute phases.

And I know those DPS. if an add is sub 5% they do NOTHING to it anymore. because "the DoTs will do the job"

"yeah... sure... why did we wipe again?"

so here I clearly prefer final wrath to make sure those things DIE.
User avatar
Schroom
 
Posts: 618
Joined: Mon Aug 15, 2011 2:03 pm
Location: Luxembourg

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Sun Jan 19, 2014 12:28 pm

Rotation sim is nearly complete. Posting results here first so that you can comment on them.

Especially interested in:
-Anything that looks fishy
-Any rotation that you think would be interesting to add / compare

Note that this is all single-target; for AoE I need to modify this file, which could be either very easy or very difficult depending on how specific I want to get.
(Ex: if I just change it to 5 targets, it's easy. If I want to do a table of DPS vs. # of Mobs like I have in the past (see first page of this thread), then it's significantly more involved. I haven't decided what to do yet.

Note also that, as I mentioned on my blog last week, there was a significant bug with ES for prot. This explains why ES was so low in the talent sims. For this, I'm using my own latest build of SimC from the trunk, which has this bug fixed. So you may not be able to completely replicate these results in 542-1.

Code: Select all
basic Rotations
Max DPS Error: 87
Talents: 312232
Glyphs: focused_shield/word_of_glory/final_wrath

|               Rotation |    DPS |    HPS |   DTPS |    TMI |    Var |  SotR |  Wait |
|        CS>J>AS>Cons>HW | 377966 | 160063 | 160402 |  17016 |  26130 | 71.0% | 14.5% |
|        CS>J>AS>HW>Cons | 384511 | 159552 | 159891 |   5434 |   3570 | 71.4% | 13.2% |
|   CS+W0.3>J>AS>HW>Cons | 378839 | 157713 | 158030 |    426 |     73 | 73.2% | 12.7% |
|       CSw>J>AS>HW>Cons | 378607 | 157655 | 157973 |    360 |     42 | 73.2% | 12.7% |
|       CSw>AS>J>HW>Cons | 379101 | 158535 | 158857 |    545 |    189 | 72.4% | 12.8% |
|       J>CSw>AS>HW>Cons | 372725 | 162592 | 162962 |  68314 |  71730 | 68.5% | 17.4% |
|       J>AS>CSw>HW>Cons | 377386 | 163147 | 163520 |  28202 |  15002 | 67.9% | 17.0% |
|       AS>J>CSw>HW>Cons | 376938 | 163959 | 164336 | 265226 | 175401 | 67.2% | 17.4% |
|       AS>CSw>J>HW>Cons | 383576 | 159833 | 160169 |   1847 |    594 | 71.1% | 13.0% |
|      HotR>J>AS>HW>Cons | 376862 | 159510 | 159844 |   2844 |    894 | 71.4% | 13.2% |
| AS+GC>CSw>J>AS>HW>Cons | 379705 | 158225 | 158545 |    499 |    160 | 72.8% | 12.9% |
| CSw>AS+GC>J>AS>HW>Cons | 378862 | 158048 | 158368 |    455 |    128 | 72.9% | 12.8% |
| CSw>AS+GC>J>HW>AS>Cons | 378733 | 157901 | 158219 |    369 |     57 | 73.0% | 12.7% |
| CSw>AS+GC>J>HW>Cons>AS | 378692 | 157852 | 158174 |    370 |     41 | 73.1% | 12.5% |


Code: Select all
execute Rotations
Max DPS Error: 92
Talents: 312232
Glyphs: focused_shield/word_of_glory/final_wrath

|                   Rotation |    DPS |    HPS |   DTPS |   TMI |   Var |  SotR |  Wait |
|       CSw>J>AS>HW>Cons>HoW | 383928 | 157717 | 158033 |   407 |    94 | 73.2% | 11.2% |
|       CSw>J>AS>HW>HoW>Cons | 384332 | 157714 | 158034 |   348 |    30 | 73.2% | 11.0% |
|       CSw>J>AS>HoW>HW>Cons | 383724 | 157579 | 157889 |   391 |    66 | 73.3% | 10.9% |
|       CSw>J>HoW>AS>HW>Cons | 383508 | 157907 | 158233 |   410 |    57 | 73.0% | 10.9% |
|       CSw>HoW>J>AS>HW>Cons | 383459 | 158009 | 158340 |   498 |   160 | 72.9% | 10.9% |
|       HoW>CSw>J>AS>HW>Cons | 383838 | 158363 | 158732 | 30787 | 42566 | 72.5% | 11.0% |
| CSw>J>HW+FW>AS>HW>HoW>Cons | 384853 | 157781 | 158103 |   373 |    61 | 73.1% | 11.0% |
| CSw>HW+FW>J>AS>HW>HoW>Cons | 385007 | 158118 | 158455 |   372 |    32 | 72.8% | 11.0% |
| HW+FW>CSw>J>AS>HW>HoW>Cons | 385100 | 158116 | 158452 |   495 |   148 | 72.8% | 11.0% |


Code: Select all
defensive Rotations
Max DPS Error: 88
Talents: 313232
Glyphs: focused_shield/word_of_glory/final_wrath

|                            Rotation |    DPS |    HPS |   DTPS |    TMI |   Var |  SotR | Wait |
|             CSw>J>AS>HW>HoW>Cons>SS | 368484 | 111889 | 121224 | 215875 | 16872 | 68.8% | 3.3% |
|          CSw>J>AS>HW>HoW>SS+R1>Cons | 370223 | 113782 | 122093 | 149299 | 13747 | 69.9% | 9.0% |
|       CSw>J>AS>HW>HoW>SS+R1>Cons>SS | 367149 | 111763 | 118761 | 110363 |  9521 | 68.8% | 3.3% |
|       CSw>J>AS>HW>SS+R1>HoW>Cons>SS | 366926 | 111697 | 118260 |  97275 |  6456 | 68.7% | 3.3% |
|          CSw>J>AS>SS+R1>HW>HoW>Cons | 368772 | 113144 | 119982 | 100653 |  8645 | 69.9% | 9.1% |
|       CSw>J>AS>SS+R1>HW>HoW>Cons>SS | 367003 | 111638 | 118020 |  95814 |  5862 | 68.8% | 3.3% |
| CSw>J>AS+GC>SS+R1>AS>HW>HoW>Cons>SS | 366846 | 111678 | 118024 |  94221 |  6260 | 68.8% | 3.3% |
|       CSw>J>AS>SS+R2>HW>HoW>Cons>SS | 366893 | 111655 | 118031 | 101362 |  7371 | 68.8% | 3.3% |
|       CSw>J>AS>SS+R3>HW>HoW>Cons>SS | 366877 | 111682 | 118088 | 101001 |  9483 | 68.8% | 3.3% |
|       CSw>J>AS>SS+R4>HW>HoW>Cons>SS | 366619 | 111671 | 118064 |  96593 |  5492 | 68.8% | 3.3% |
|       CSw>J>AS>SS+R5>HW>HoW>Cons>SS | 366806 | 111669 | 118041 | 100004 |  7406 | 68.8% | 3.3% |
|          CSw>J>SS+R1>AS>HW>HoW>Cons | 368637 | 112840 | 119195 |  81745 |  5018 | 70.0% | 9.1% |
|          CSw>SS+R1>J>AS>HW>HoW>Cons | 368086 | 112756 | 119022 | 100434 |  7898 | 69.7% | 9.2% |
|          SS+R1>CSw>J>AS>HW>HoW>Cons | 368628 | 112673 | 119008 | 102389 |  8215 | 69.7% | 9.1% |


Code: Select all
talents Rotations
Max DPS Error: 91
Talents: 312232+custom
Glyphs: focused_shield/word_of_glory/final_wrath

|                 Rotation |    DPS |    HPS |   DTPS | TMI | Var |  SotR |  Wait |
|  CSw>J>AS>HW>HoW>Cons>ES | 404722 | 157823 | 158143 | 358 |  34 | 73.1% |  9.6% |
|  CSw>J>AS>HW>HoW>ES>Cons | 406964 | 157829 | 158146 | 384 |  83 | 73.1% |  9.8% |
|  CSw>J>AS>HW>ES>HoW>Cons | 407021 | 157837 | 158159 | 416 |  67 | 73.1% |  9.8% |
|  CSw>J>AS>ES>HW>HoW>Cons | 406642 | 157771 | 158089 | 473 | 100 | 73.2% |  9.9% |
|  CSw>J>ES>AS>HW>HoW>Cons | 405797 | 157818 | 158138 | 373 |  48 | 73.1% | 10.0% |
|  CSw>ES>J>AS>HW>HoW>Cons | 405046 | 157868 | 158189 | 549 | 203 | 73.0% | 10.0% |
|  ES>CSw>J>AS>HW>HoW>Cons | 404993 | 158098 | 158418 | 514 | 137 | 72.8% | 10.1% |
|  CSw>J>AS>HW>HoW>Cons>LH | 393473 | 157951 | 158268 | 286 |  28 | 73.0% |  9.6% |
|  CSw>J>AS>HW>HoW>LH>Cons | 394203 | 157942 | 158250 | 273 |  22 | 73.1% |  9.8% |
|  CSw>J>AS>HW>LH>HoW>Cons | 394012 | 157996 | 158304 | 279 |  28 | 73.1% |  9.8% |
|  CSw>J>AS>LH>HW>HoW>Cons | 393837 | 157966 | 158269 | 302 |  42 | 73.2% | 10.0% |
|  CSw>J>LH>AS>HW>HoW>Cons | 394221 | 157945 | 158248 | 295 |  65 | 73.1% | 10.0% |
|  CSw>LH>J>AS>HW>HoW>Cons | 394428 | 158084 | 158388 | 401 | 189 | 73.0% | 10.0% |
|  LH>CSw>J>AS>HW>HoW>Cons | 394936 | 158307 | 158614 | 432 | 115 | 72.7% | 10.0% |
| CSw>J>AS>HW>HoW>Cons>HPr | 396435 | 158261 | 158576 | 356 | 111 | 72.8% |  7.7% |
| CSw>J>AS>HW>HoW>HPr>Cons | 396078 | 158371 | 158677 | 420 | 167 | 72.7% |  7.8% |
| CSw>J>AS>HW>HPr>HoW>Cons | 396037 | 158373 | 158684 | 310 |  55 | 72.7% |  7.9% |
| CSw>J>AS>HPr>HW>HoW>Cons | 394804 | 158164 | 158471 | 309 |  30 | 72.9% |  8.0% |
| CSw>J>HPr>AS>HW>HoW>Cons | 394359 | 158132 | 158440 | 313 |  37 | 72.9% |  8.0% |
| CSw>HPr>J>AS>HW>HoW>Cons | 393417 | 158399 | 158703 | 342 |  56 | 72.7% |  8.2% |
| HPr>CSw>J>AS>HW>HoW>Cons | 395484 | 159561 | 159883 | 481 |  77 | 71.6% |  8.3% |
"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 - MoP/5.x

Postby Thels » Tue Jan 21, 2014 3:59 am

What does CSw stand for?

I thought it used to be a slight DPS gain to place AS on the top of the priority list, but from the top table, this no longer seems to be the case?

For the Execute rotations, there's HW and HW+FW. Does that mean over the course of a fight or entirely during Execution phase?

Having a CS>J>AS+GC>HW+FW>AS>HW>HoW>Cons rotation might be interesting, because that might very well be the highest DPS rotation without sacrificing any HoPo, from the looks of it.

Also, can we have a T90 comparison outside of Execution phase, as the addition of HoW takes away our filler spots.
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Tue Jan 21, 2014 10:37 am

Oh, right, I guess I should explain the nomenclature.

The basic idea is (Ability)[+(Option)], where you can string options together. So for example, "W" is a "wait" option, such that "CS+W0.3" means "Use CS, or wait if CS's cooldown is less than 0.3 seconds." It translates into the following APL lines:
Code: Select all
actions+=/crusader_strike
actions+=/wait,sec=cooldown.crusader_strike.remains,if=cooldown.crusader_strike.remains>0&cooldown.crusader_strike.remains<=0.3


"CSw" is a shorthand for "CS+W0.35" because I wanted to condense things, and since we're almost always going to want to implement that wait.

Other options are things like:
"+FW" ("If Final Wrath is glyphed and target is in Execute Range")
"+GC" ("If the Grand Crusader buff is active")
"+R#" ("If the remaining duration of this buff is less than # seconds")

There are a bunch of others ("+ex" for execute range) but I think those three are the only ones I'm currently using here.

It looks like it's still a slight DPS gain to do AS>CSw>J>HW>Cons or AS+GC>CSw>J>AS>HW>Cons vs. CSw>J>AS>HW>Cons, but not very much. (Also note that the forum formatting here is somewhat limited; the blog post will let you sort each table by any column you want, which is very convenient.)

My earlier comment should clarify HW vs. HW+FW, but just in case, HW just checks if HW is available (i.e. /actions+=holy_wrath), HW+FW checks for execute range and glyph (i.e. actions+=/holy_wrath,if=glyph.final_wrath.enabled&target.health.pct<=20).

Should be simple to add that rotation; I thought it was in there before, but I may have removed it at some point. Easy enough to fix.

Note that all of these are results from regular sims, i.e. not artificially set to execute phase. So the T90 comparisons are for an entire fight, including execute phase. I can re-list them without HoW to see how they compare if you ignore execute range, but I'm not sure that's very useful. You have both HoW>L90 and L90>HoW versions, which should already give you all of the relevant information.
"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 - MoP/5.x

Postby Thels » Thu Jan 23, 2014 5:20 am

The +GC and +R marks were clear. The +W wasn't, but it makes sense.

The thing about T90 is that we earlier determined that outside of Execute range, it was best to save T90 for last, so it would fill an empty GCD. However, during Execute range, we have no empty GCDs, so that was a good reason to bump T90 up the priority list. Not sure if that's still the case.

It would therefor make sense to me to first establish the optimal rotation without using the Final Wrath glyph or Hammer of Wrath, much like you're already doing in the first post, but with the T90 abilities, and once we got the optimal rotation, use variants of that rotation with HW+FW, HoW and T90+EX placed in various locations.
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Thu Jan 23, 2014 10:08 am

That should be sufficiently accomplished by adding the following two rotations to the mix, I think?
Code: Select all
                    'CSw>J>AS>HW>ES+ex>HoW>ES>Cons';
                    'CSw>J>AS>ES+ex>HW>HoW>ES>Cons';
                    'CSw>J>AS>HW+FW>ES+ex>HW>HoW>ES>Cons';


(With associated versions for LH and HPr).
"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 - MoP/5.x

Postby theckhd » Thu Jan 23, 2014 11:26 am

Results; ignore the TMI column, it's using a new definition I'm testing, so it's not directly comparable to the values posted earlier in this thread:

Code: Select all
|                              Rotation |    DPS |    HPS |   DTPS | TMI | Var |  SotR |  Wait |
|               CSw>J>AS>HW>HoW>Cons>ES | 404778 | 157835 | 158154 | 732 |   3 | 73.0% |  9.6% |
|               CSw>J>AS>HW>HoW>ES>Cons | 406739 | 157784 | 158102 | 728 |   3 | 73.1% |  9.8% |
|               CSw>J>AS>HW>ES>HoW>Cons | 407066 | 157820 | 158142 | 728 |   3 | 73.1% |  9.8% |
|               CSw>J>AS>ES>HW>HoW>Cons | 406621 | 157717 | 158036 | 725 |   3 | 73.2% |  9.9% |
|               CSw>J>ES>AS>HW>HoW>Cons | 406142 | 157816 | 158137 | 728 |   3 | 73.1% | 10.0% |
|               CSw>ES>J>AS>HW>HoW>Cons | 405073 | 157864 | 158183 | 733 |   3 | 73.0% | 10.0% |
|               ES>CSw>J>AS>HW>HoW>Cons | 405626 | 158151 | 158474 | 759 |   3 | 72.7% | 10.0% |
|         CSw>J>AS>HW>ES+ex>HoW>ES>Cons | 407265 | 157770 | 158091 | 727 |   3 | 73.1% |  9.7% |
|         CSw>J>AS>ES+ex>HW>HoW>ES>Cons | 406901 | 157838 | 158156 | 729 |   3 | 73.0% |  9.8% |
|   CSw>J>AS>HW+FW>ES+ex>HW>HoW>ES>Cons | 406992 | 157816 | 158136 | 734 |   3 | 73.1% |  9.8% |
|               CSw>J>AS>HW>HoW>Cons>LH | 393515 | 157924 | 158240 | 686 |   3 | 73.1% |  9.6% |
|               CSw>J>AS>HW>HoW>LH>Cons | 394249 | 158001 | 158304 | 677 |   3 | 73.1% |  9.8% |
|               CSw>J>AS>HW>LH>HoW>Cons | 393747 | 158043 | 158346 | 676 |   2 | 73.0% |  9.8% |
|               CSw>J>AS>LH>HW>HoW>Cons | 394315 | 157977 | 158278 | 673 |   2 | 73.2% |  9.9% |
|               CSw>J>LH>AS>HW>HoW>Cons | 394188 | 157955 | 158259 | 668 |   3 | 73.1% | 10.0% |
|               CSw>LH>J>AS>HW>HoW>Cons | 394436 | 158061 | 158365 | 669 |   3 | 73.0% | 10.0% |
|               LH>CSw>J>AS>HW>HoW>Cons | 394946 | 158310 | 158617 | 688 |   3 | 72.7% | 10.0% |
|         CSw>J>AS>HW>LH+ex>HoW>LH>Cons | 394050 | 158036 | 158342 | 677 |   3 | 73.1% |  9.8% |
|         CSw>J>AS>LH+ex>HW>HoW>LH>Cons | 393880 | 158044 | 158349 | 676 |   2 | 73.0% |  9.8% |
|   CSw>J>AS>HW+FW>LH+ex>HW>HoW>LH>Cons | 394057 | 158011 | 158317 | 677 |   2 | 73.1% |  9.8% |
|              CSw>J>AS>HW>HoW>Cons>HPr | 396639 | 158192 | 158512 | 707 |   3 | 72.9% |  7.6% |
|              CSw>J>AS>HW>HoW>HPr>Cons | 395690 | 158304 | 158613 | 689 |   3 | 72.8% |  7.9% |
|              CSw>J>AS>HW>HPr>HoW>Cons | 395583 | 158351 | 158659 | 692 |   3 | 72.8% |  7.9% |
|              CSw>J>AS>HPr>HW>HoW>Cons | 394853 | 158238 | 158544 | 681 |   3 | 72.9% |  8.0% |
|              CSw>J>HPr>AS>HW>HoW>Cons | 394588 | 158159 | 158466 | 683 |   3 | 72.9% |  8.0% |
|              CSw>HPr>J>AS>HW>HoW>Cons | 393779 | 158397 | 158702 | 701 |   3 | 72.7% |  8.1% |
|              HPr>CSw>J>AS>HW>HoW>Cons | 395843 | 159532 | 159858 | 759 |   3 | 71.6% |  8.2% |
|       CSw>J>AS>HW>HPr+ex>HoW>HPr>Cons | 395850 | 158402 | 158711 | 692 |   3 | 72.7% |  7.9% |
|       CSw>J>AS>HPr+ex>HW>HoW>HPr>Cons | 395447 | 158380 | 158683 | 691 |   3 | 72.7% |  7.9% |
| CSw>J>AS>HW+FW>HPr+ex>HW>HoW>HPr>Cons | 395841 | 158405 | 158713 | 692 |   3 | 72.7% |  7.9% |
"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 - MoP/5.x

Postby Thels » Mon Jan 27, 2014 5:03 am

CSw>J>AS>HW>ES+ex>HoW>ES>Cons seems the highest DPS option from that list, but is there any chance you could compare the following two options?

CSw>J>AS+GC>HW>AS>ES+ex>HoW>ES>Cons
CSw>J>AS+GC>HW+FW>AS>HW>ES+ex>HoW>ES>Cons

Surprisingly, CSw>J>AS>LH>HW>HoW>Cons is the best DPS option with LH while optimizing HoPo gain. That doesn't feel right. ES deals more damage single target than LH, still, we prio HW over ES, but we prio LH over HW?
ImageImage
User avatar
Thels
 
Posts: 1233
Joined: Tue May 15, 2007 7:30 am
Location: The Netherlands

Re: Theck's MATLAB thread - MoP/5.x

Postby theckhd » Mon Jan 27, 2014 9:53 am

I'll add those two and re-run the sims to see if that oddity is consistent. My guess is that it's RNG - even though SimC Reports "DPS Error" as ~80 DPS, I don't know offhand what that metric represents (i.e. is it standard deviation, or st. dev. of the mean? Or something else?). My guess is that these results are only accurate to +/- 200-300 DPS. This is with 50k iterations; for the final data I might bump it up to 100k to see if that helps.
"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 - MoP/5.x

Postby theckhd » Mon Jan 27, 2014 2:51 pm

Here are the results, I think comparing this to the previous set makes it clear that the DPS of any individual sim can vary by +/-200 DPS, at least. Our confidence interval is probably twice the "Max DPS Error" metric. I'll try to look into what exactly SimC is doing to calculate that metric so I can determine what it is, statistically speaking.

talents Rotations
Max DPS Error: 91
Talents: 312232+custom
Glyphs: focused_shield/word_of_glory/final_wrath
Code: Select all
|                                    Rotation |    DPS |    HPS |   DTPS | TMI | Var |  SotR |  Wait |
|                     CSw>J>AS>HW>HoW>Cons>ES | 404804 | 157844 | 158165 | 492 | 200 | 73.1% |  9.6% |
|                     CSw>J>AS>HW>HoW>ES>Cons | 406976 | 157856 | 158174 | 357 |  63 | 73.1% |  9.8% |
|                     CSw>J>AS>HW>ES>HoW>Cons | 407139 | 157885 | 158207 | 383 |  70 | 73.0% |  9.7% |
|                     CSw>J>AS>ES>HW>HoW>Cons | 406760 | 157737 | 158052 | 502 | 189 | 73.2% |  9.9% |
|                     CSw>J>ES>AS>HW>HoW>Cons | 406476 | 157790 | 158110 | 332 |  25 | 73.1% |  9.9% |
|                     CSw>ES>J>AS>HW>HoW>Cons | 405132 | 157812 | 158132 | 366 |  35 | 73.1% | 10.0% |
|                     ES>CSw>J>AS>HW>HoW>Cons | 405262 | 158135 | 158459 | 575 | 213 | 72.7% | 10.0% |
|               CSw>J>AS>HW>ES+ex>HoW>ES>Cons | 407061 | 157858 | 158177 | 418 |  99 | 73.0% |  9.8% |
|               CSw>J>AS>ES+ex>HW>HoW>ES>Cons | 406700 | 157815 | 158135 | 480 | 259 | 73.1% |  9.8% |
|         CSw>J>AS>HW+FW>ES+ex>HW>HoW>ES>Cons | 407187 | 157787 | 158109 | 416 |  67 | 73.1% |  9.7% |
|         CSw>J>AS+GC>HW>AS>ES+ex>HoW>ES>Cons | 405762 | 157755 | 158077 | 570 | 365 | 73.1% |  9.7% |
|   CSw>J>AS+GC>HW+FW>AS>HW>ES+ex>HoW>ES>Cons | 407219 | 157829 | 158148 | 373 |  54 | 73.1% |  9.7% |
|                     CSw>J>AS>HW>HoW>Cons>LH | 393282 | 157995 | 158313 | 379 | 165 | 73.1% |  9.7% |
|                     CSw>J>AS>HW>HoW>LH>Cons | 393939 | 157990 | 158296 | 293 |  40 | 73.1% |  9.8% |
|                     CSw>J>AS>HW>LH>HoW>Cons | 394072 | 157941 | 158251 | 283 |  58 | 73.1% |  9.8% |
|                     CSw>J>AS>LH>HW>HoW>Cons | 393892 | 157921 | 158222 | 307 |  64 | 73.2% | 10.0% |
|                     CSw>J>LH>AS>HW>HoW>Cons | 393699 | 158053 | 158357 | 414 | 304 | 73.1% | 10.0% |
|                     CSw>LH>J>AS>HW>HoW>Cons | 394312 | 158082 | 158388 | 392 | 104 | 73.0% | 10.0% |
|                     LH>CSw>J>AS>HW>HoW>Cons | 395052 | 158316 | 158627 | 407 | 140 | 72.7% | 10.0% |
|               CSw>J>AS>HW>LH+ex>HoW>LH>Cons | 393794 | 158053 | 158358 | 324 |  72 | 73.1% |  9.8% |
|               CSw>J>AS>LH+ex>HW>HoW>LH>Cons | 393790 | 157980 | 158285 | 318 |  64 | 73.1% |  9.8% |
|         CSw>J>AS>HW+FW>LH+ex>HW>HoW>LH>Cons | 393965 | 158031 | 158338 | 428 | 256 | 73.1% |  9.8% |
|         CSw>J>AS+GC>HW>AS>LH+ex>HoW>LH>Cons | 393028 | 157967 | 158273 | 494 | 387 | 73.1% |  9.7% |
|   CSw>J>AS+GC>HW+FW>AS>HW>LH+ex>HoW>LH>Cons | 393717 | 158015 | 158325 | 364 | 159 | 73.0% |  9.8% |
|                    CSw>J>AS>HW>HoW>Cons>HPr | 396650 | 158260 | 158579 | 338 |  37 | 72.9% |  7.6% |
|                    CSw>J>AS>HW>HoW>HPr>Cons | 395982 | 158417 | 158726 | 331 |  55 | 72.7% |  7.8% |
|                    CSw>J>AS>HW>HPr>HoW>Cons | 395922 | 158348 | 158658 | 350 | 112 | 72.7% |  7.9% |
|                    CSw>J>AS>HPr>HW>HoW>Cons | 394881 | 158167 | 158475 | 310 |  46 | 72.9% |  8.0% |
|                    CSw>J>HPr>AS>HW>HoW>Cons | 394627 | 158172 | 158482 | 324 |  59 | 72.9% |  8.0% |
|                    CSw>HPr>J>AS>HW>HoW>Cons | 393760 | 158422 | 158730 | 324 |  62 | 72.7% |  8.1% |
|                    HPr>CSw>J>AS>HW>HoW>Cons | 395560 | 159531 | 159856 | 515 | 104 | 71.6% |  8.3% |
|             CSw>J>AS>HW>HPr+ex>HoW>HPr>Cons | 395869 | 158422 | 158732 | 353 |  84 | 72.7% |  7.9% |
|             CSw>J>AS>HPr+ex>HW>HoW>HPr>Cons | 395607 | 158406 | 158711 | 898 | 905 | 72.8% |  7.9% |
|       CSw>J>AS>HW+FW>HPr+ex>HW>HoW>HPr>Cons | 396031 | 158397 | 158705 | 312 |  30 | 72.7% |  7.8% |
|       CSw>J>AS+GC>HW>AS>HPr+ex>HoW>HPr>Cons | 395034 | 158196 | 158502 | 311 |  49 | 72.9% |  7.7% |
| CSw>J>AS+GC>HW+FW>AS>HW>HPr+ex>HoW>HPr>Cons | 395909 | 158406 | 158716 | 348 |  47 | 72.7% |  7.9% |
"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

PreviousNext

Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: Yahoo [Bot] and 1 guest

Who is online

In total there are 2 users online :: 1 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: Yahoo [Bot] and 1 guest