Anaxie's Totally Revolutionary SoO Retribution T16 BiS & Theorycraft

[anaxie]

yes thats right

Thok may aswell be considered a 18-19% mastery multiplier because that is basically what it allows us to do with the excess haste..

That and openings will be quite potent again. 20 second duration statics are quite strong and unless I'm wrong should line up with guardian frequently and if you are lucky. Resync up when Wings and guardian meetup again. 115 ICD or something like the creche egg we used for 8 months of DS.

The trinket is excessively powerful.

[Deleted]

I think most people in here don't realise that there is no magic number of how much haste you should get. There is absolutly no reason to quit gemming haste when you hit 40.00% unless you see the value decrease. You need to check and do sims for your own gear, since it varies hugely.

Same goes for mastery and crit. Currently I am seeing an increase in dps by reforing crit over mastery, but we are litterally talking about reforing mastery from headpice into crit and then I have to go mastery on my next item again.

Also theres no point in debating if I get lucky or unlucky, its just plain retarded to argue 1 stat beats the other because "adds might die faster" or some other bs.

- - - Updated - - -

Are you guys taking into account thok's tail tip for the 40% haste "cap"? The haste we get through the trinket does affect our GCD/CD right?

If that's the case with a normal mode thok's tail tip 2/2 (7.5% amp) the 17,000 haste (40%) would roughly translate to 18,275 haste (43%).

If the 40% haste cap is already taking into account the haste we get from thok we would want something around 15,814 haste (37.21%).

If for some reason we wanted to achieve 21,250 (50%) we would want something around 19,767 haste (46.51%).

This would obviously change with higher ilvl's of thok's tip.

Sorry about the ninja edit

Thok's increase in haste is reflected on the number shown on your character sheet when you open it up. As far as I know having Thok's trinket increases the value of crit, and is why we see points where crit surpasses mastery.

[Maegor]

I think most people in here don't realise that there is no magic number of how much haste you should get. There is absolutly no reason to quit gemming haste when you hit 40.00% unless you see the value decrease. You need to check and do sims for your own gear, since it varies hugely.

Won't haste always sim higher until 50%? If you are going to test this isn't it better to do target dummy than simc, since what you are after is not having inactive periods in your rotation, which is what the t16 proc does for us?

[anaxie]

I think most people in here don't realise that there is no magic number of how much haste you should get. There is absolutly no reason to quit gemming haste when you hit 40.00% unless you see the value decrease. You need to check and do sims for your own gear, since it varies hugely.

Same goes for mastery and crit. Currently I am seeing an increase in dps by reforing crit over mastery, but we are litterally talking about reforing mastery from headpice into crit and then I have to go mastery on my next item again.

Also theres no point in debating if I get lucky or unlucky, its just plain retarded to argue 1 stat beats the other because "adds might die faster" or some other bs.

- - - Updated - - -



Thok's increase in haste is reflected on the number shown on your character sheet when you open it up. As far as I know having Thok's trinket increases the value of crit, and is why we see points where crit surpasses mastery.

Guess what though chuck. In the actual game it doesn't translate like that. Burst phases on the pull lust, downtime between boss burn phases allowing wings to come back off CD <thok, Garrosh somewhat> matters a huge deal. Thats just how the game works. And when the swing is as close as one item forge swings the stats a diff way, better to just keep stacking the one that equal and consistent.

The 40% haste was figured not by myself originally but Solsacra. And actually the initial number was 38%. 40% just seemed more rounded and better. Solsacra is arguably the best dps ret in this entire game atm CONSISTENTLY. Yes I am not infact the king shocking I know. He consulted with me the idea of it made alot of sense with how haste DOES devalue after a certain point.

Now if you can turn around and become the best player in the world like Solsa and then IN GAME prove him wrong then I might lend an ear.

Solsa even put up rank 1s towards the ass end of ToT on important fights just going a full mastery > haste build while still wearing T15. One would assume mastery is far more potent the sims suggest.

All this "sim your gear" stuff really make no sense almost everyone I see goes 40% haste > Mastery. So like. Wtf are people even simming at diff gear levels it honestly changes nothing. Who here can honestly say they're like FUCK YEAH time to forge into some juicy crit.

TLDR Consistency is best.

In b4 request about simulated target dummy numbers for proof!!!!

[Tanyn]

Are you guys taking into account thok's tail tip for the 40% haste "cap"? The haste we get through the trinket does affect our GCD/CD right?

If that's the case with a normal mode thok's tail tip 2/2 (7.5% amp) the 17,000 haste (40%) would roughly translate to 18,275 haste (43%).

If the 40% haste cap is already taking into account the haste we get from thok we would want something around 15,814 haste (37.21%).

If for some reason we wanted to achieve 21,250 (50%) we would want something around 19,767 haste (46.51%).

This would obviously change with higher ilvl's of thok's tip.

Sorry about the ninja edit

I posted a table with the haste values needed for the 40% with each version of Thok's trinket somewhere in this thread. I will find it and relink it, I suppose.

- - - Updated - - -

Here is a small table to show the ilvls versus the actual % you get:


I will assume we have the VP needed to upgrade the items and do my math from there.

LFR(2/2): 5.9754% | 17000 / 1.059754 = 16041 Haste Rating
Flex(2/2): 6.6823% | 17000 / 1.066823 = 15935 Haste Rating
Normal(2/2): 7.5427% | 17000 / 1.075427 = 15808 Haste Rating
WF-Normal(2/2): 7.9764% | 17000 / 1.079764 = 15744 Haste Rating
Heroic(2/2): 8.514% | 17000 / 1.08514 = 15666 Haste Rating
WF-Heroic(2/2): 9.0035 | 17000 / 1.090035 = 15596 Haste Rating

[anaxie]

I posted a table with the haste values needed for the 40% with each version of Thok's trinket somewhere in this thread. I will find it and relink it, I suppose.

- - - Updated - - -

Here is a small table to show the ilvls versus the actual % you get:


I will assume we have the VP needed to upgrade the items and do my math from there.

LFR(2/2): 5.9754% | 17000 / 1.059754 = 16041 Haste Rating
Flex(2/2): 6.6823% | 17000 / 1.066823 = 15935 Haste Rating
Normal(2/2): 7.5427% | 17000 / 1.075427 = 15808 Haste Rating
WF-Normal(2/2): 7.9764% | 17000 / 1.079764 = 15744 Haste Rating
Heroic(2/2): 8.514% | 17000 / 1.08514 = 15666 Haste Rating
WF-Heroic(2/2): 9.0035 | 17000 / 1.090035 = 15596 Haste Rating

I dont believe this is right unless the Tooltip rounds up because a 574 says 9%. I don't have one myself to derive the exact % rating so maybe it just is rounded.

I do greatly appreciate the rating number needed with the H / HWf though very hand indeed. So essentially the trinket gives us about 3200 ish mastery <huge guesstimation> but 11.25% mastery is potent. love it

[Deleted]

Sometimes we have logic on one side and we have you on the other side Anaxie. Unless Solsacra can prove that his numbers are skewed or different from what Simcraft or Exemplars spreadsheet shows, I don't really care how much gear or what kind of logs he puts up. I also haven't Solsacra saying anything against what I am saying, unlike you which happens pretty much all the time.

What works for him depends a lot on his gear, and as you know gear scaling changes a lot depending on what offpieces and 2nd stats you have. Talking about a 40.00% cap which most of you seem to be doing is wrong. While it is true that at one point we reach a point where it not longer affects our GCD, it is not true that it becomes less important at exactly 40%. You need to do your own tests and simulations. With Thok crit also gains a larger DPS increase than mastery gains from the same trinket (flat % increase vs. % damage increase on critical hits) this is also why I believe I am seeing crit pass mastery untill I swap around 1000-1500 mastery into crit and then seeing mastery being slightly above again.

[anaxie]

Sometimes we have logic on one side and we have you on the other side Anaxie. Unless Solsacra can prove that his numbers are skewed or different from what Simcraft or Exemplars spreadsheet shows, I don't really care how much gear or what kind of logs he puts up. I also haven't Solsacra saying anything against what I am saying, unlike you which happens pretty much all the time.

What works for him depends a lot on his gear, and as you know gear scaling changes a lot depending on what offpieces and 2nd stats you have. Talking about a 40.00% cap which most of you seem to be doing is wrong. While it is true that at one point we reach a point where it not longer affects our GCD, it is not true that it becomes less important at exactly 40%. You need to do your own tests and simulations. With Thok crit also gains a larger DPS increase than mastery gains from the same trinket (flat % increase vs. % damage increase on critical hits) this is also why I believe I am seeing crit pass mastery untill I swap around 1000-1500 mastery into crit and then seeing mastery being slightly above again.

Solsacra can't remember his MMO champ info lol...

Also of course I talk all the time. My thread / personal blog yo.
One would assume that in Anaxie's Totally best-of-all-time thread would include lots of in your face Anaxie.

You cling to target dummy sims too much. Might i suggest opening your eyes in a H25 man raid once in awhile. You know the part with actual gameplay and class / strat variance and downtimes / burst phases.

[Deleted]

I dont believe this is right unless the Tooltip rounds up because a 574 says 9%. I don't have one myself to derive the exact % rating so maybe it just is rounded.

It is rounded up on the tooltip. If you go to Wowhead you can see the exact percentage with decimals.

Edit: Hmm looks like it wasn't Wowhead, I can't remember where I found it. The regular WF version 559 gives around 7.5 where as the 2/2 regular WF version gives around 7.98. Let me find out where I saw it.

- - - Updated - - -

If you equip and unequip the trinket you will also find that the haste rating that it gives does not match a rounded number.

[anaxie]

It is rounded up on the tooltip. If you go to Wowhead you can see the exact percentage with decimals.

Edit: Hmm looks like it wasn't Wowhead, I can't remember where I found it. The regular WF version 559 gives around 7.5 where as the 2/2 regular WF version gives around 7.98. Let me find out where I saw it.

- - - Updated - - -

If you equip and unequip the trinket you will also find that the haste rating that it gives does not match a rounded number.

Can't equip / unequip it. Thok gives me cancer healer trinkets only.

But tooltips have always been wonky. Feather was incorrect, Apparatus was incorrect.

Best to just equip and find the exact amount this way ALWAYS.

[Deleted]

Won't haste always sim higher until 50%? If you are going to test this isn't it better to do target dummy than simc, since what you are after is not having inactive periods in your rotation, which is what the t16 proc does for us?

It depends, if you mean buffed 50% then I think you will see haste being better for various gear setups untill around 52-55%. It would be best to do a personal dps test on a dummy, but you need to be raidbuffed and have the dummy debuffed too. Further more what works for you can not be generalized to always be the better choice for others. The mistakes you make in your rotation and how RNG can affect your results are also hard to compare with others.

Basicly 1 stat could show a better DPS increase for you because your rotation is different. Also to reduce RNG (meaning your crit chance is your crit chance) will take a long time of testing, you can't do that.

[Tanyn]

You're contradicting yourself. You're saying that it's gear and person dependent and then in the same breath you're saying we're wrong for using the 40% numbers. Isn't it possible that both could be true? 40% Haste could work for people.

It was also said in this thread a ton of times that the 40% was not a hardcap. It'll fluctuate per person depending on many things. Latency, other gear, etc. It was also suggested that people sim their own character or try the changes in-game.

How do you expect Solsacra to back up this idea? You won't accept logs as proof, so I'm not sure what you expect him to provide as evidence.

Have you tried stopping at 40% haste and testing your numbers and then going straight haste and then trying it again? Or have you simmed it at these levels?

There's a major flaw with SimCraft. It's made by people who are doing a similar thing that Anaxie is and using their own "data" and making a program to generate numbers. There's no verification that the ways they're coming up with numbers is the correct way. At least using actual data from gameplay seems a little more solid than generating numbers from possible outcomes.

All in all, this is my opinion. So, don't freak out and nerdrage.

[anaxie]

Tanyn said it better than I could had. I'm a man of many words but I convey things in.... less desired fashion.

[Deleted]

For progress raids I would still go with the gems that gives me the most single target DPS. If crit shows higher damage than mastery for single target, it means that it shows higher damage overall. I'd never gem for what gives me most damage if I can hit 5 adds for 30 seconds at at time, just like I will always try to cleave off a single target unless I'm stacking censure on different targets.

Mechanics will override any of these small differences though. The 500 dps gain would hardly matter over playing correctly.

[Neldarie]

Anything above 40% is perfect world, mathworld (any fraction of a second: be it encounter mechanics or just breathers in between you aren't hammering gcds haste "value" is gone). I tend to just use AMR with 17k haste softcap reforges. After that I use sim-C(10k+ its) to run sims and adjust/juggle crit vs mastery until both are equal in value. This so far gave me best overall results ST + AoE/cleave. But overall I found that Crit pulls ahead in AE and mastery(very wep dmg dependant) slightly in ST.

[anaxie]

Anything above 40% is perfect world, mathworld (any fraction of a second: be it encounter mechanics or just breathers in between you aren't hammering gcds haste "value" is gone). I tend to just use AMR with 17k haste softcap reforges. After that I use sim-C(10k+ its) to run sims and adjust/juggle crit vs mastery until both are equal in value. This so far gave me best overall results ST + AoE/cleave. But overall I found that Crit pulls ahead in AE and mastery(very wep dmg dependant) slightly in ST.

I can get on board with this. No doubt part of the effects of Crit in AE would be the flurry of xuen. However as gear scales up adds get smoked faster so usually ST mastery preference should still end up superior. Then again raids have diff comps :> Yours could be different.

Neldarie can I touch your mass WF gear. please?

My trinkets make me want to cry.

[Deleted]

You're contradicting yourself. You're saying that it's gear and person dependent and then in the same breath you're saying we're wrong for using the 40% numbers. Isn't it possible that both could be true? 40% Haste could work for people.

It was also said in this thread a ton of times that the 40% was not a hardcap. It'll fluctuate per person depending on many things. Latency, other gear, etc. It was also suggested that people sim their own character or try the changes in-game.

How do you expect Solsacra to back up this idea? You won't accept logs as proof, so I'm not sure what you expect him to provide as evidence.

Have you tried stopping at 40% haste and testing your numbers and then going straight haste and then trying it again? Or have you simmed it at these levels?

There's a major flaw with SimCraft. It's made by people who are doing a similar thing that Anaxie is and using their own "data" and making a program to generate numbers. There's no verification that the ways they're coming up with numbers is the correct way. At least using actual data from gameplay seems a little more solid than generating numbers from possible outcomes.

All in all, this is my opinion. So, don't freak out and nerdrage.

Im trying to dismantle the idea of a 40% cap, it's just a number to aim for, but this thread hypes it too much. It fluctuates a lot, the reasoning behind why we should stop at it makes no sense. I haven't seen Solsacra suggest anything or haven't even mentioned him so I don't see why he is even brought into the discussion, it is Anaxie who thinks it makes his own post more valid if he namedrops Sols name or Neldra, newsflash it doens't.

I don't accept logs as proof because it is hugely different from player to player and it depends on your tactic. Logs showing high dps could mean you were being prioritized in the raid or piggybacked to meters.

- - - Updated - - -

The thing with you Anaxie is you go along with anyone who is willing to pat you on the back, and sadly that is not how the real world works. If you want to prove something that you can generalize you have to be able to provide evidence that it works for everyone.

You're a good ret but you say a lot of things :P

[Deleted]

Logs: Sample size = 1
The argument, that haste decreases in value if you aren't pushing buttons hard all the time or if you are out of range of the boss, is valid. But you can't do enough tests to be able to see such a little difference out of the RNG.
You can go for mastery over haste at 40% if you think it's better. But you can't say that it's "the way to go", because you can't prove it.

[Zephur]

Not sure why people insist on being so contrarian in this thread. It should be self-evident to every ret that haste is amazing only up to that point where you become GCD locked, at which point it becomes our worst secondary. For most people, based on the experience and testing of exceptional players, that number is going to be somewhere around the 40% mark, +/- a couple points to account for latency and player skill. Simcraft shows that the value of haste diminishes past a certain point, too; just look at the chart on page 269.

There's really no point in making a fuss about whether we should stop at 38%, 40%, 42%, or any other number. Once you notice that you're GCD locked 99.9% of the time, you're obviously losing dps by sticking with haste. Swap ~500 points into mastery/crit and try again. Everyone will find their own sweet spot, again based on latency and skill.

[Deleted]

based on the experience and testing of exceptional players

People who are the best at doing things in practice (who you refer to as exceptional players), aren't always the best at theorising. This is the same reason I think discarding Exemplar's advice based on progress is foolish.