Haste>Master>Crit??

[Taddeous]

According to Exemplars spreadsheet he has accurately tested that haste has moved up. What is your guys input on this? Is Simulation Craft correct after all?



Important update! Emergency spreadsheet update: here. Read below for reasons.

Warning that Autoreforge is presently tentative. I haven't had time to completely rebuild from the ground up and validate all comparatives. With Autoreforge flaky it will impact the automated Item Calc routine, but directly entering different gear/gems/reforge/etc will function properly.

On with the reason:
Yesterday I got a PM from Cayse who was wondering why the value of Haste changed wildly when the Fight Length setting in my spreadsheet was changed.

I checked and Cayse was quite correct. I looked deeper and found a huge bug in my spreadsheet. Back when I changed everything to probability it seems I failed to update autoattack and Censure ticks appropriately (I think I did, then ran one more test of the old to compare/verify and the old overwrote the formula in the cell). As such they were static values. Just watch what happens to your DPS and you increase the Fight Length to 6000 seconds and still make only 110 autoattacks. I'm amazed no one has caught this earlier - it appears to have been present since the February update.

Since the quantity of autoattack and Censure were not being updated, they were only being impacted by Hit, Expertise, and Crit (Miss, Dodge, and damage increase) - not Haste. Now that the spreadsheet properly calculates the # of autoattacks which occur and the # of Censure ticks which occur - Haste increases in value.

Haste has increased significantly enough that in the gear I've tested, it's now our #1 stat. Which is almost turning our present worldview on its head. I'm now seeing Haste > Mastery > Crit.

I'd somewhat taken this for granted, but will explicitly state it. I don't just permit people to view all the formula and calculations in my spreadsheet (barring the macros), but I actively encourage people to review how the sheet functions. Did I miss a buff that affects an ability? Add an inappropriate one? Do something entirely hamhanded and stupid (like leave # of autoattacks a static value)? Many of my spreadsheet updates are due to missing/incorrect items reported to me via PM and I welcome bug reports or suggestions to fix/streamline calculations.

P.S. For those who had reported Simcraft showing Haste more valuable - you appear to be correct and Simcraft was the more accurate on this detail. Why my values agreed very closely with Redcape, which should not have had this flaw, I do not understand.

[Requital]

According to Exemplars spreadsheet he has accurately tested that haste has moved up. What is your guys input on this? Is Simulation Craft correct after all?



Important update! Emergency spreadsheet update: here. Read below for reasons.

Warning that Autoreforge is presently tentative. I haven't had time to completely rebuild from the ground up and validate all comparatives. With Autoreforge flaky it will impact the automated Item Calc routine, but directly entering different gear/gems/reforge/etc will function properly.

On with the reason:
Yesterday I got a PM from Cayse who was wondering why the value of Haste changed wildly when the Fight Length setting in my spreadsheet was changed.

I checked and Cayse was quite correct. I looked deeper and found a huge bug in my spreadsheet. Back when I changed everything to probability it seems I failed to update autoattack and Censure ticks appropriately (I think I did, then ran one more test of the old to compare/verify and the old overwrote the formula in the cell). As such they were static values. Just watch what happens to your DPS and you increase the Fight Length to 6000 seconds and still make only 110 autoattacks. I'm amazed no one has caught this earlier - it appears to have been present since the February update.

Since the quantity of autoattack and Censure were not being updated, they were only being impacted by Hit, Expertise, and Crit (Miss, Dodge, and damage increase) - not Haste. Now that the spreadsheet properly calculates the # of autoattacks which occur and the # of Censure ticks which occur - Haste increases in value.

Haste has increased significantly enough that in the gear I've tested, it's now our #1 stat. Which is almost turning our present worldview on its head. I'm now seeing Haste > Mastery > Crit.

I'd somewhat taken this for granted, but will explicitly state it. I don't just permit people to view all the formula and calculations in my spreadsheet (barring the macros), but I actively encourage people to review how the sheet functions. Did I miss a buff that affects an ability? Add an inappropriate one? Do something entirely hamhanded and stupid (like leave # of autoattacks a static value)? Many of my spreadsheet updates are due to missing/incorrect items reported to me via PM and I welcome bug reports or suggestions to fix/streamline calculations.

P.S. For those who had reported Simcraft showing Haste more valuable - you appear to be correct and Simcraft was the more accurate on this detail. Why my values agreed very closely with Redcape, which should not have had this flaw, I do not understand.

I find it hard to believe haste is our #1 stat we can't hit 3.0 CS CD so there just is no reason to be pushing haste.

[Naididae]

My sims and Redcape's still show haste as the least valued.
You can see my comments on the post on EJ itself, but my guess is that there might be something wrong with exemplar's spreadsheet on this one. I can't find reasons for why haste would jump so much from 4.1 to 4.2. Exemplar's spreadsheet can have haste worth nearly 70% of strength!
Either way, we have to see, but simulationcraft, at least from the various tests with multiple armory people, show haste as the least valued and mastery >= crit for tier 12 geared people.

So, for everyone out there, try and help out to figure out if haste is really that far ahead now, but I'm taking it with a grain of salt for now.

---------- Post added 2011-07-21 at 11:11 PM ----------

Also, I noticed simulationcraft and redcape's to be more accurate in terms of finding out what my average dps will be, so I have a feeling Exemplar might be overvaluing something right now, like Censure as I noticed earlier today, although I don't think that is enough to push haste so high.

[Cayse]

Wait for another day or two, just after he updated his sheet Naididae pointed out the censure formula was wrong and he hasn't updated it again.

A few random rets at different gear levels I've through simcraft on 25k+ iterations have haste almost 1:1 with crit, but both below mastery. Mostly on people who have a weapon falling behind their other gear.

I had one sim on a paladin with a 391 weapon and mostly heroic t11 gear where crit was valued very slightly over mastery, and haste well behind.

[Kisko]

In these uncertain times, when we can't really say which sim is right and which is wrong, just go by ilvl and amount of STR. There is a good chance that Haste is close enough to Mastery and Crit to not be a factor when choosing gear.

Should we change our reforging strat? I wouldn't do that now, until we will get confirmation.

[Badpaladin]

Given the general wonkiness haste brings to the table anyways, I'd reserve any kneejerk gear requests until later.

[Butosai]

I almost fell out of my chair reading this. What a turn of events!

[Deleted]

I've not deep inside maths of stat weights, but i've often played retribution reforging all to haste and I always feel that my dps was improved (hit-expertise cap, haste > mastery > crit).

[Oneirophobia]

I almost fell out of my chair reading this. What a turn of events!

As a shadow priest, I find this shocking. ;P

[Naididae]

Just to bring this up again cause I think this is important and being left to the sidelines.
Some new info was gathered as Exemplar fixes his spreadsheet, having Mastery >= haste > crit.

It would be great if someone with more direct contact with simulationcraft devs could look and see if there is anything suspiciously wrong with ret simulation.

[Taddeous]

There still isn't enough information to confirm Mastery>= Haste > Crit. Everyone is kind of just waiting to see what happens. This may be true when you have the T12 2 set b/c of the extra damage to Crusader Strike. I'm still confused on what's going on here. I wish someone would come with a solid answer to provide the evidence.

[Badpaladin]

There still isn't enough information to confirm Mastery>= Haste > Crit. Everyone is kind of just waiting to see what happens. This may be true when you have the T12 2 set b/c of the extra damage to Crusader Strike. I'm still confused on what's going on here. I wish someone would come with a solid answer to provide the evidence.

Would the DoT ticking scale enough such that you won't get the thing over-writing itself? If that's not the case, then haste might have spots of undesirability in-between DoT ticks.

Fuck I wish I could test this stuff. Requital, get your ass in here and start hammering away at dummies.

[Jonuts]

Which raises a really silly idea. Stack enough haste for Apparatus to proc haste, and you have a 15 second bloodlust on a 2m CD. Might turn out to be viable.

Either way, *SOMEONE* has a bug in their math. Save your panic until they figure it out properly. Or figure it out yourself. Either way, I'd be quite happy to see haste more valuable. If only it lowered the GCD for our non-spell abilities too

[dunst306]

This seems rather shocking because yesterday in a trash run for FL, I told a paladin to stop reforging for haste and to reforge for mastery. Maybe he might have been ahead O_o

[Ulmita]

I personally believe that something has changed since 4.1 or it was always like that. I went and tested some gear set ups on hc dummies testing out the t.v crits (was paranoid about crit suppression). One of my particular gear set up had me on 21.12 Mastery (no mastery trinket), 9.2% crit and 5.01% haste which one would think it would be the dream set for rets with redcapes stat weights in mind.

The fight lasted for about 142 Millions worth of damage, pulling full rotations etc and in that time frame i ended up with a bit over 1k T.Vs (1006 to be exact). I was surprised to see that my sustainable damage was 1- 1.5k dps lower than a different gear set up i had used the other day to test out stuff. That specific gear set had around 2.5% more haste and lower mastery; crit %was a bit lower too.

I don't know what to think about this one, rather than its time for some serious dummy smacking while in a reforging nirvana. Btw i have also noticed this: When i plug my gear to exemplar's spreadsheet, i am getting significantly higher crit % on my attacks from what i am getting in reality. And by significant i mean 9-10% more crit (at least to the T.V) which just doesn't happen in game. I might be miss-reasing stuff or the sheet might not calculate the crit suppression of bosses (4.8%) not sure what to make yet.

[Bassnectar]

o.O haste has always been #1 (since cata) then crit then mastery....

[Badpaladin]

o.O haste has always been #1 (since cata) then crit then mastery....

Ret thread, not a holy thread.

[Ulmita]

you might be deluded by Sanc titty of Battle talent.
but, due to current itemizaton it is impossible to reach haste hard cap (3 sec CS cd), thus, its one of the main reasons you should avoid haste like a plague as Ret.

Unless you can come up with some kind of proof, please refrain your self from making comments like that. At this very moment the community is searching why the haste weight stat is inflated so much (at least to the only spreadsheet we got available), to the point where, theoretically speaking, you would want to reforge into it.

[Deleted]

Haste cap is reachable only under effect of some trinket (i've reached with 363 ilvl using apparatus effect), so forget to reach 3 sec CD.

I feel that the 3 secondarsy stats (mastery, haste and crit) are at the moment balanced, with mastery slightly better than other two

More haste = more cs and Tv and then more mastery damage (also more art of war proc)

more crit = bigger hit, so more mastery damage

So just use gear with highster str and mastery, other stats are not influencing dps in a relevant way as str and mastery can do.

[Ulmita]

right back at you, good sir.
once upon a time somehow Exemplar's spreadsheet has shown that haste is not that bad, and you are ready to teach others how to reforge for all-out haste?
wait a bit.
its in discussion, its being tested.


edit: and for duck's sake, it IS impossible to reach haste cap in current gear.

Firstly, i did not teach or erged anyone to reforge to anything. If you read again what you quoted you'll see that i mentioned that the community is trying to figure out wth.

Secondly in today's news,

You want about 20% more Mastery than Haste (1.2:1 Mastery/Haste). Too much more than that and Haste becomes superior. The rate changes so slowly, though, that you could be as far as 1.4:1 before you would even get a few DPS variance. Thus it should be pretty safe to just reforge for Mastery without worry that you're going too far.

Crit is a reasonably close third. In most reasonable gear/reforge setups their values are about 1.25 (M/H) to 1.1 (C)

This is were we stand atm. Sim vs Spreadsheet. All i am saying is that we do not yet know for sure, if haste becomes our best secondary stat, if and only if, we are able to hit the 3 sec cs. According to the latest finds, it doesn't need to.