I agree that Judging to stack Censure at range would be counter-productive, but if we only worry about stacking it during a ground phase, we only have to Judge once every so often to maintain Censure and can spam Wrath during Judgement's cooldown. Such a fight, though, would likely have something for us to be doing on the ground, such as Nefarion's air phase requiring us to DPS adds on platforms - unless, of course, Blizz makes another Atramedes-stlye fight where air phase is simply dodging fire.

1) It's 20s total

2) Probably a mistake with the buffing. Should've started the fight with it.

3) It's on pre-combat cast, it should've been okay.

4) The reforging is the most optimal, and I should defend that. It's easily achievable to have exact numbers on the cap of hit and exp, but you get much more Haste (and/or Mastery) by not doing so. Yes it has wasted stats (a little bit of crit), but it also has maximized potential (as much haste as possible). Doing so has been a DPS increase in all my testings, you can test yourself.

5) Probably a mistake when I was changing between sheets to fix some other stuff!


As I said, the sims are not 100% perfect! Feel free to try your owns!

It's not hard, you just have to be patient.

1) Always remind the others involved in the raid that you're running tests and they should be careful not to interfere.

2) Do some runs until you achieve a close enough buff up-time in both situations.

3) Remove the runs you felt have been interfered in some way. Remove the ones that missed an important RNG (landslide),

4) Clean up all the crit damage from the raw results.

5) Do some math and re-add the crit damage based on your crit rating (do it correctly and remember there's hand of light involved).

6) You shall have now some fairly clean numbers. You can clean more data if you want to do further mathing.

7) Remember these are estimates and not hard results. Do further testing to confirm the vality of your observations. When doing observations, remember there's still a fairly big threshold for error in the data.

I think that may be the problem people are having with your work. You seem to be contradicting yourself. Here you are saying that what you are getting are just estimates and not hard results which is exactly the opposite of what Exemplar and the others are after.

You also contradict yourself by saying that you need further testing to confirm your own observations, when earlier you clearly made the argument that you did not need large sample pools of data to confirm your suspicions. Whereas in this quote you admit to needing a more testing, hence a larger pool of data to validate your claims.

And the last point of "remember there's still a fairly big threshold for error in the data" is exactly what these guys are working hard to remove.

But still all of your own solo work only confirmed in the end what has already been agreed upon through analysis of large amounts of data by these guys that haste>mastery. And though the hybrid build you speak of may be slightly beneficial in one instance on an encounter like Spine, I think its safe to say that there are few encounters that follow that exact model.

Furthermore, consider this quote from a developer: "So there definitely were lessons learned from Spine: specifically the importance of stressing varied and conflicting mechanics at the same time, instead of just one." - in reference to the burst potential of specific classes being stacked for that particular type of mechanic; it would seem that type of encounter is not something they want to purposefully develop in the future which would make any type of build made for such a specific fight rather pointless.

I did provide hard data, the estimatives were merely a setback! Jokes aside, the "soft" data was just the beginning of the research. In the finishing of it, I did 25k iterations tests on SimC - which is basically as "hard" as it gets!

There, you answered the question of it all by yourself.

-> Do not need large sample pools of data to confirm SUSPICIONS.
-> Need further testing to VALIDATE the claims.

I provided date with little threshold for error as the final result of the research! Only the initial data wasn't accurate.


Well, but that's the fun of it! Proving it wrong or right, it's knowledge we all obtain - until proven wrong!
Be it useful or not, it's knowledge nonetheless!

That's weird, as it says "TV count: 8" "Interval: 2.23s" -> totalling 17.84s.
And the graphs clearly show "time=20s" as the end of it.

Care to elaborate why is it definitely 10s? I seem to be missing on that.

That is something I did not know, thanks! Will keep an eye on that for further testings.

Please do! I'm eager to hear about your results.

What was Septicret doing with all their HP? I don't see any usage of Harsh Words, which could push ranged Ret even higher.

Also, I see things like you having only 83% uptime on Inq, only have two AW in a 7.5 minute fight, only have 3 HA in a 7.5 minute fight, only five execution sentences in a 7.5 minute fight.

Meanwhile, Seprticret has 3 AW uses, 7 Execution Sentences. He also has higher crit rates on most of his abilities than you.

@Rafaello

That's my mistake, sorry. Fight length is listed at top as 10 seconds, I didn't double check graph times.

Step two is where you're going wrong. Instead of removing all your reforging, use tools like wowreforge.com or any of the reforging addons to reforge your new gearset as optimally as possible using your old stat weights. It's extremely unlikely that changing a single piece of gear will change your stat weights dramatically enough that your entire reforging priority changes with it.

@stRafaello:

Statistical significance - Wikipedia, the free encyclopedia
Statistical significance - Wikipedia, the free encyclopedia

By design, damage in WoW is largely RNG based, so there will never be empiric "proof" or "final" results. You can only try to maximize the confidence in your results by maximizing your sample size. There is no real "noise" in WoW given that you repeat tries (samples) with identical prerequisites under controlled conditions (which SimC or Excel sheets are able of doing).

SimC makes sure of all that while your individual runs are prone to error (buffs not present / forgotten, etc., errors in rotation) and actually have noise.
Regarding sample size, SimC does 50000 noise-free runs in about 1 minute... you´ll never even get close by doing this real-time.

Only advantage of doing it real-time in WoW would be, that you are using the actual engine vs. a "reverse-engineered" one. But this does not make up for the huge gap in significance.

What real-time testing IS good for is for double-checking mechanics, behaviour, etc. to validate your engine and base numbers or to double-check sim results whether they are within the predicted range of error.

If you are trying to compare your results to the simulated ones from SimC or Excels as equal, you are wasting your time.

That was not the question. And changing a piece without hit for one with hit on it i.e. quite drastically changes your reforging options.

I am asking for the academic/theoretical optimum that using sims/tools can achieve, not for the 80/20 approach.

Zephur, I get your point and that might enhance the result, but I am not yet convinced this is the optimum.

It depends on your old stat weights being "right" and optimal which I am generally challenging with my question. Or to put it differently: What do you do the first time when you do not have any prior weights/reforges/enchants/gems?

In my opinion simc would need to do its 50.000 iterations for each possible combination of gem/enchant/reforge and then compare the results. This would take ages of course to compute, but in my opinion is at present the only way of getting the optimum result with maximum confidence. And I believe this is not the way the sims work presently.

You would of course try to reduce the number of combinations possible by ruling out obvious nonsense like i.e. "spirit", but if you do not start off clean every time you will get a biased result. You are after all using results from a different gear setup which are not accurate anymore to generate the new one. At least that´s still my opinion at this point.