Fury Warrior 5.4 BiS list!

[Deleted]

Ok about the 2 piece Bonus in Simcraft :

Iterations: 10000Fight Length: 360 - 540

tier16_2pc_melee Rage 330.90
colossus_smash 16924 5.0% 42.6

so its 330 Rage in 42 CS ? 330/42 = 7,8

does that mean the Simulation is doing 1,5 Special Attacks in Every CS ? that would be ..... BAD ?

thats 330 rage in 450 secounds.

If u do 3 CS in 1 minute u get like atleast 60 Rage ? 450/60 = 7,5 .

So if u take the simulation he gets (330/7,5) 44 Rage a Minute WITH PROCS.

But on PTR u can get 60 Rage WITHOUT PROCS whats wrong there?

for the haters 1 Minute fight 3 cs = No procs
http://www.worldoflogs.com/reports/1...uzl/details/5/

dont look at DPS go to buffs gained power gaines : 2 piece boni = 125 Rage in 1 minute. WITHOUTPROCS

Simulation gets 330 in 7,5 Minutes WITH PROCS?

Did i get something wrong here?

[CollisionTD]

Thats a display bug. It means that the bonus proceed 330 times, for 1600 rage. Thats roughly 38 rage gained every cs or an average of 8 specials. You can verify this under "resources gained"

I'll fix the display bug tonight

[Oliria]

Thats a display bug. It means that the bonus proceed 330 times, for 1600 rage. Thats roughly 38 rage gained every cs or an average of 8 specials. You can verify this under "resources gained"

I'll fix the display bug tonight

I do admire your patience with this one Collision

[Deleted]

And how can it have Overflow ?

only way i see is u go in CS with 120 or 100 Rage and get a proc from ur Mastery.

[CollisionTD]

And how can it have Overflow ?

only way i see is u go in CS with 120 or 100 Rage and get a proc from ur Mastery.

Most overflow happens in the first 20 seconds or is a result of multiple sudden death procs. The current action list doesn't use heroic strike at all, but I probably need to add a last resort heroic strike action to prevent it.

The current logic is that its better to cap rage for a gcd than use heroic strike, but with a string of cs procs rage income will be too high for slam alone to keep down.

- - - Updated - - -

well u need acces to all simcraft data then if u need to fix everything.

What did i get wrong ? I mean DW doesnt go with REck on PTR i tested it 100 times.

Btw in a 450 secounds fight lenght simulation there are 138Ticks of DW hows that? 450/3 = 150 ticks. if the first starts after 3secs its 149 if iam right ?

and yeah i told warriorsarri about that skullbanner thing, most of ppl think that it gets snapshotted.
But that was a HUGE BUG, every sim in 5.3 was wrong there, most of all the Warlock simulations.
I mean think about a crit doom or something like that with 20% more damage for 2 minutes?

All simcraft code is public on the website, anyone can look at it.

I understand the crit damage bug is huge for classes that use a lot of snapshot dots. However... Warriors only have one dot and we refresh it every time we use mortal strike or bloodthirst, so the bug only lasts one or two ticks of deepwounds every 3 minutes. That is very minor.

I'll check deep wounds ticking. Can you link me the logs where you tested all of this ?

- - - Updated - - -

We could see anything, I expect a decent sized number pass addressing each spec soon enough, if not then 1 week into the tier.

It's fairly obvious that I want to see the math that makes you believe Storm Bolt is superior, where are the superior numbers other than simcraft results.
My blank statement was a counter to your blank statement encouraging you to show the math and prove me wrong, I then atleast put some "fake" (or theoretical) math forward (due to not knowing the storm bolt damage).

As to the blankness of my statement, it's words that I've heard from Landsoul himself saying Bloodbath is "probably better" whether you want to believe it or not, I've heard it so I'll stick to it.

I feel like I sound hostile but I'm trying not to .

I understand. The only reason I haven't posted math/proof is because I'm at work posting on my phone.

But seriously, what bugs/discrepancies do you see with simcraft ? Or what can be added to make it better.

[Deleted]

i dont have loggs for the Reck DW testing but i can stream and post u Vid after.

[Deleted]

But seriously, what bugs/discrepancies do you see with simcraft ? Or what can be added to make it better.

There's never a strictly patchwerk fight in the game, and yet everytime I see someone putting a point forward it's based off what they've found on simcraft.

If this class were as easy to "sim" as people make out, Landsoul wouldn't have felt the need to make an overly ambitious spreadsheet to the point where it's too ambitious for him to finish.

I just don't think it has every variable correct in it's calculations. Not to mention there's the occasional rotational error. In the past when I've spoken to good players and mentioned Simulationcraft they either laugh or disregard what I'm saying that relates to simcraft. I don't trust it and hate when people use it as a defence.

[Deleted]

Colusion can u tell me when u have time ? Cant save the stream somehow have to show u "live" in stream !

- - - Updated - - -

http://www.twitch.tv/chrizzzy1/

[CollisionTD]

Colusion can u tell me when u have time ? Cant save the stream somehow have to show u "live" in stream !

- - - Updated - - -

http://www.twitch.tv/chrizzzy1/

My time at the computer is erratic. Logs would be best.

[Deleted]

Got 6,7% Crit + 30% Reck sould be about 35.

But only 1 dot crit out of 20 thats 5% crit.

I know its just 1 Try but its 20 Ticks DONT TELL ME I NEED TO TEST MORE!


http://www.worldoflogs.com/reports/r...gon/details/5/

[CollisionTD]

There's never a strictly patchwerk fight in the game, and yet everytime I see someone putting a point forward it's based off what they've found on simcraft.

If this class were as easy to "sim" as people make out, Landsoul wouldn't have felt the need to make an overly ambitious spreadsheet to the point where it's too ambitious for him to finish.

I just don't think it has every variable correct in it's calculations. Not to mention there's the occasional rotational error. In the past when I've spoken to good players and mentioned Simulationcraft they either laugh or disregard what I'm saying that relates to simcraft. I don't trust it and hate when people use it as a defence.

I have top 20-30+ ranks on nearly every encounter. I find it hard to believe that simcraft is that far off when i base a lot of my rotation choices off it, as the results speak for themselves. Back when landsoul released his spreadsheet in dragon soul, the conclusions drawn from his spreadsheet and the SMF guide I made back then were the same. There were some minor disagreements on the damage numbers, but not significant enough to change how we played.

Did any of these people give you legitimate reasons? Or did they just scoff and move on? There's a huge difference.

Someone else mentioned that its how you use the tool that matters. So I'll explain what I do with it.

I use it to find relative differences.

Example.
I want to know if its best to use heroic strike inside of cs or use wild strike every gcd outside to bleed rage.

I set up the normal simulation, and disable colossus smash. Note the damage per rage numbers for both, and then re run it with a rotation that only uses those abilities during cs.

The program may have some bugs, but those bugs affect heroic strike and wild strike the same.

Its an excellent sand box. The default profiles are simply guides for newer players.

[Deleted]

I have top 20-30+ ranks on nearly every encounter.

Because ranks actually mean something? All that shows is you've managed to whore better than most others in your BiS gear.
Why aren't those all rank 1's if simcraft is steering you in the right direction and you have best in slot gear

Did any of these people give you legitimate reasons? Or did they just scoff and move on? There's a huge difference.

Pretty much the same reasoning I've given you.

Someone else mentioned that its how you use the tool that matters. So I'll explain what I do with it.

Maybe I'm old fashioned, I'd rather hit a dummy for a while than trust a tool that could potentially give me an incorrect value for something like that.

[baconjesus]

Because ranks actually mean something? All that shows is you've managed to whore better than most others in your BiS gear.
Why aren't those all rank 1's if simcraft is steering you in the right direction and you have best in slot gear

Pretty much the same reasoning I've given you.

Maybe I'm old fashioned, I'd rather hit a dummy for a while than trust a tool that could potentially give me an incorrect value for something like that.

Ranks do mean something, they indicate performance.

There is a bit of RNG and cheesing to get some the #1 ranks. Getting feathers to keep buff up on jikun without having to dps adds, etc.

Old fashioned? I'd go with stupid. Simcraft is a tool, if you lack the ability to use the tool properly, it's your fault. No one said only use Simcraft, it's just another way to have information consolidated to analyze.

[Deleted]

Ranks do mean something, they indicate performance.

The problem is, top players generally don't cheese to get ranks, so they don't indicate performance. For example method runs with 1/2 alts 1/2 mains, so any boss that requires a fast kill time for a decent rank, they don't rank on. The best guild in the world is out of the running.
Ranks are entirely arbitrary these days.

Old fashioned? I'd go with stupid.

So not trusting something that could potentially steer you in the wrong direction is stupid?
It's nothing to do with not being able to use it, not at any point did I mention an inability to use it, I mentioned a lack of trust.

The better players I know prefer to do physical testing and napkin math. Are they stupid for not using Simcraft? no.

[Archimtiros]

Because ranks actually mean something? All that shows is you've managed to whore better than most others in your BiS gear.
Why aren't those all rank 1's if simcraft is steering you in the right direction and you have best in slot gear

Pretty much the same reasoning I've given you.

Maybe I'm old fashioned, I'd rather hit a dummy for a while than trust a tool that could potentially give me an incorrect value for something like that.

Do we even play the same game?
Firstly, Collision has very good gear, but you really aren't very far behind him at all; secondly, rank 20-30 doesn't mean you "whore" with BiS gear. By whore I assume you mean cheese the fight to make things work for you. Some people do that, but there actually isn't a whole hell of alot of room to do that as a Warrior on most fights. Regardless if you WERE doing that, in very good gear, you'd be seeing rank 1-10's, not rank 20+. Really I would put anyone who ranks in the top 50's consistently amongst the better/best players in the world. After that, especially with RPPM, a large chunk of it comes down to procs, crits, luck and how you do the fight.

Simcraft is a very good tool, I have had many of the same discussions where people laugh at using simcraft as a reference and move on. There is good reasoning for that, because the tool is only as good as the people who maintain it, a job I think Collision has done very admirably. Also, Simcrafts results are constantly evolving based on a variety of factors, changing one piece of gear will often change your stat weights for example, many people do not understand it takes more than running the sim once to give you all the information you need. They run it once and never touch it again.
But just because Simcraft is "steering you" doesn't mean that you will be able to pull off a rotation at 100% effectiveness, or that lady luck won't kick you in the ass.
Nobody is perfect, even those that practice constantly.
Lastly, the dummy is useful for some basic things but it misses alot of variables, for one raid buffs and debuffs. Unless your trying to figure out static things like ICD's, and you practice on it for hours and hours, it's not a very good reference point at all.

[baconjesus]

The problem is, top players generally don't cheese to get ranks, so they don't indicate performance. For example method runs with 1/2 alts 1/2 mains, so any boss that requires a fast kill time for a decent rank, they don't rank on. The best guild in the world is out of the running.
Ranks are entirely arbitrary these days.





So not trusting something that could potentially steer you in the wrong direction is stupid?
It's nothing to do with not being able to use it, not at any point did I mention an inability to use it, I mentioned a lack of trust.

The better players I know prefer to do physical testing and napkin math. Are they stupid for not using Simcraft? no.

Simcraft is nothing but math.

If you can't make the changes, adjustments, modify the profiles or the source, then yes, that's an inability to use a tool to it's full potential. Attempting to discredit something because you don't "trust" it when it is transparent and modifiable by you is nothing but unapologetic stupidity or sincere laziness on your part.

It is a tool, how you use it is important. If you don't want to make the effort, that's fine, that's on you. There's nothing inherently distrustful about it when it's transparent.

[Archimtiros]

The problem is, top players generally don't cheese to get ranks, so they don't indicate performance. For example method runs with 1/2 alts 1/2 mains, so any boss that requires a fast kill time for a decent rank, they don't rank on. The best guild in the world is out of the running.
Ranks are entirely arbitrary these days.


So not trusting something that could potentially steer you in the wrong direction is stupid?
It's nothing to do with not being able to use it, not at any point did I mention an inability to use it, I mentioned a lack of trust.

The better players I know prefer to do physical testing and napkin math. Are they stupid for not using Simcraft? no.

No offense intended to them; but you realize that the people in the top ranking guilds in the world are not necessarily the best players in the world right? Not by a long shot. Not going to mention names, but there are some down right mediocre players in alot of those guilds. They make up for it by having the time, dedication and willingness to persevere. That alone is as important as good play for a World Ranking guild.

A lot of top players cheese to get ranks. But they don't all. Before he quit, I raided with one of the best DK in the world; I've seen him cheese a few times, but not at the expense of the raid group, only on farm, and to beat his own rank ones that he already had. There also isn't opportunity for a Warrior to cheese in many fights in ToT. If it was strictly luck or cheesing there would not be many players out there with multiple orange parses across a variety of bosses.
Some of it comes down to how you do the fights, for example, my guild kites Tortos bats instead of killing them = no aoe damage. I also am on the first nest rotation due to reliable BStorm aoe, and so I never get nutriment buffs. Or your raids DPS on Jin'rokh for example is very important for how many pools you get/when you get them.

When it comes to group dps, having higher personal dps than the rest of the group means that you will do better in comparison right? If your killing Maegera and the raid dps is trash but yours is very good, its much more likely that the bats will stay alive longer for you to do a higher portion of DPS than the other players, instead of them dying 2 seconds into your bladestorm, and that the head will survive long enough for you to get a second/third Colossus Smash on it, instead of dying before the adds even pop.
Not only that, but longer is actually better for DPS to get a good rank because it is more opportunity to line up and use cooldowns effectively. This is another reason some Warriors were not using 4p on Farm, because they might not be able to use Banner effectively.
Maegera for example is a fight I haven't gotten a good rank on in months because we kill it too fast. Likewise I haven't been able to beat my rank 1 on Ra-den because we now kill it about 10-20 seconds before I can get 3 sets of CDs.

I will agree that ranks are mostly arbitrary however. With RPPM alone, it is way too hard to control when and how you perform. Which is why I said that anything around top 50 and below I recognize at being the better/best players. Out of 150 non subscriber ranks, and the many many players who do not rank.

[Draconess]

My personal preference to the 4 piece this coming tier is that its utter garbage with a terrible proc chance and there are better subs for the legs, chest, and shoulders all with crit/mastery and better itemized sockets.

As for the Malkorok axes I don't think will be BiS as there is a lack of expertise on gear unless you use the tier shoulders, the weapons for smf would be the sword x2 off of the Klaxxi paragons as it has exp/crit. http://ptr.wowhead.com/item=103972

Also the manoroth shoulders i don't think will be bis either as garrosh drops another pair of strength shoulders with crit/mastery and 2 yellow sockets which means for much more crit and from the math I've done if you don't use the 4 piece you will have roughly 18,000 crit rating from gems/chants/gear.

There is also a crafted hit/crit belt that only takes maximum 21 days to make if you get the recipe early which means less reforging for hit also and its got crit so that's automatically better than a haste/mastery belt unless your pushing 53% crit with buffs which you wont til heroic gear. http://ptr.wowhead.com/item=98616

[CollisionTD]

Because ranks actually mean something? All that shows is you've managed to whore better than most others in your BiS gear.
Why aren't those all rank 1's if simcraft is steering you in the right direction and you have best in slot gear

Pretty much the same reasoning I've given you.

Maybe I'm old fashioned, I'd rather hit a dummy for a while than trust a tool that could potentially give me an incorrect value for something like that.

As people have noted already, rank 1s are 50% luck and 50% padding. Consistent performance is what matters, and what I do consistently does very well on all fights. I'm sincerely not bragging, only trying to use that as evidence that simcraft does work as a tool.

Your reasoning so far:

1: No patchwerk fight. That's fair, however there are also 2 other fights that you can change it to that aren't patchwerk. Helter skelter, hectic add cleave (based on Horridon.), and you can fairly easily put in commands that will do different fights. Admittedly, movement isn't well modeled, but warriors tend to not have issues with movement fights.

2: The variables aren't calculated correctly. The variables that simcraft uses are literally stripped from game .mpq files. Hotfixes may not be in there, but they're generally updated fairly quickly. It's not hard to multiply a couple of numbers together.

3: Rotations: I can't speak for all specs, but the fury warrior one is definitely well done. If you think there's a better way to play, I'll try it out and see.

4: It's not nearly complex enough: I'll need more information on what exactly you mean here.

5: It's just not correct according to my sources: Have your sources come here and cite their argument, I'm interested in hearing it because maybe it's something that can be fixed, or maybe they have a fundamental misunderstanding of the output that the program gives. A lot of people have said "It doesn't attempt to line up cooldowns" as their argument for a year, even though it has 3-4 lines that specifically say to wait for certain cooldowns or boss health percentages.

Hitting a dummy will tell you very little unless you can play exactly the same way for 10-20 minutes. Any mistakes in your rotation will throw off your results.

- - - Updated - - -

Storm bolt math:


Storm bolt: Does 500% weapon damage, hits with main hand and off hand. 30 second cooldown
Bloodbath: Basically +30% damage to specials for 12 seconds. 60 second cooldown

500% weapon damage for TG: Fully buffed the character sheet shows an average weapon damage of 50k. All calculations will be done assuming 50k weapon damage, and all other factors will be equal (No calculation done for crit/enrage as both spells benefit equally.)
50k*5= 250k damage main hand.
0.625*50k*5= 156k damage main hand.

Used on cooldown, one storm bolt will be inside CS every minute.
250k*1.5 = 375k
156*1.5 = 234k

So every minute, storm bolt will hit for (250k +156k + 375k + 234k) = 1.015 million damage.

Now for bloodbath. How much damage do we need to do in 12 seconds to equate to that?

1.015/0.3 = 3.383 million damage.

I want to absolutely prove to you that storm bolt is better than bloodbath, so I'm going to throw the kitchen sink at bloodbath and say that we have infinite rage and the only restrictions are cooldowns on abilities and the gcd. Additionally, every non-BT gcd will be RB, even though that's highly unlikely to happen.

BT-CS-(BLOODBATH)-RB-BT-RB-RB-BT-RB-RB-Dragon Roar
+ 8 heroic strikes.

RB: 190% weapon damage.
MH: 50 * 1.9 = 95
OH: 50 * 1.9 * .625 = 59.375
3 rbs are inside of CS, so multiply by 1.5
(95 * 2) + (95 * 1.5 * 3) = 617.5k total MH damage.
(59.375 * 2) + (59.375 * 1.5 * 3) = 385.9375 OH

BT does 90% + 1000 damage.
50 * 0.9 = 45k + 1k = 46k.
One of them is inside CS
46k *1.5 + 46k = 115k

Heroic Strike:110% + 741
50 * 1.1 = 55k + 741 = 55.741k
4 of them are slammed inside CS
4*55.741*1.5 + 4*55.741 = 557.41k

Add it all up:
617.5 + 385.9375 + 115 + 557.41k = 1.675 million

Wait! We haven't calculated Dragon Roar yet! How much damage will dragon roar need to do in order to tip the scales in bloodbaths favor?

3.383 - 1.675 million = 1.7 million damage.

I'm not even going to do that calculation, because both you and I know that dragon roar never goes over 1 million damage even if the stars align.

So not only is stormbolt better than bloodbath on single target, it absolutely kicks the crap out of bloodbath.

Bonus math:

What about the opportunity cost?

Well, one of the storm bolts doesn't have any cost because it happens outside of CS, where we would use heroic strike instead of wild strike, and the rage efficiency difference is fairly minimum.

Inside of CS:

BT-CS-Storm Bolt-RB-BT-RB

We're trading out the additional CS damage on raging blow and pushing BT back one GCD if we get "unlucky" and CS procs RB.

So we're losing 0.6667*RB Damage and 0.333*BT Damage + 3 rage. There is no enrage loss uptime as enrage covers 4 gcds.

0.667*RB = 102.96k damage lost
0.333*BT = 30.66k
The 3 rage will be lost from heroic strike, which has a rage efficiency somewhere around 4k inside of CS, so 12k damage.

102.96 + 30.66 + 12k = 145.62k damage

1.015 mil - 145.62k = 869.38k
Divided by 0.3

So that's only 2.89 million damage that you need to do in 12 seconds. Even if dragon roar hit for 1 million, bloodbath still falls 200k short of that.

- - - Updated - - -

I remember Simcraft not even Using shout prefight so warrior didnt start with any rage did that get fixed ?

Yep. I implemented that 8-11-2013:
https://code.google.com/p/simulation...detail?r=17219

Tweaked it to use game-data 8-29-2013, and also have the rage show up in the "Resources gained" section.
https://code.google.com/p/simulation...detail?r=17412

resources.current[ RESOURCE_RAGE ] = initial_rage; // User specified rage.
 else
   player_t::resource_gain( RESOURCE_RAGE, ( find_class_spell( "Charge" ) -> effectN( 2 ).resource( RESOURCE_RAGE ) ), gain.charge );
     if ( glyphs.bull_rush -> ok() )
        player_t::resource_gain( RESOURCE_RAGE, ( glyphs.bull_rush -> effectN( 2 ).resource( RESOURCE_RAGE ) ), gain.charge );

[Adsertif]

You nerds just got told.

But is there a threshold that Bloodbath would beat out SB regarding an AE fight, where uptime would be high for cleaving?