View Full Version : DW Fury Guide - 5th Edition

04-14-2008, 10:28 PM
I'm currently working on an update to Meeks' fury guide and wanted as many people as possible to give their input/ask questions/do some testing (so I don't have to because I am lazy).

See the (very) rough outline so far at:
WoW Forums -> Corb's Fury Guide - Kinda (http://forums.worldofwarcraft.com/thread.html?topicId=5784525275&sid=1)

04-15-2008, 01:47 PM
"A real analysis of the differences between fast & slow offhands. We all know the standard arguments, but I want to put this to bed for once and for all."

The difference is very minor. The same speed thing is almost a non-factor, as its not really clear that it causes any (or much) more extra flurry procs than just seem to randomly occur. Non-matched speed weapons can also give extra flurried attacks, but it seems that as the speeds differe more and more you just get less of them. That is, 2.7+2.7 will give the most 'extra' flurry charges, but 2.7+2.6 will give only slightly less, and 2.7+1.6 will give somewhat less than that.

The fast weapon is slightly worse in that you do less WW damage, however this is with your OH weapon so its already a smallish part of the WW damage. This isnt a big factor in single target fights (though a slow OH for AoE is probably good to have)

Additionally, fast weapons give smoother rage generation which will tend to allow better rage management and probably more slightly heroic strikes (or less missed/delayed cooldowns due to lack of rage). Also, fast is better during execute time.

Overall I think the difference is so minor that it should be used only as a tiebreaker between equal weapons. definitely 2.7 vs 2.6 is not worth thinking about. 2.7 vs 1.6 is worth thinking about, however if the 1.6 speed item is higher ilvl and with better stats, it should be used.

"A true test to evaluate the possibility of dual Executioner as an alternative to Exec MH potency/goose OH. (YES I KNOW THEY DON'T STACK)"

Executioner uptime is around 30%ish. The loss from non-stacking would be when both are procced at the same time. This would be somewhere in the area of 30% * 30% = 9%. Thus the OH executioner would give about (30-9 = 21%) extra uptime. Say 20%. Its basically worth 2/3 what executioner is.

It is a fact that for good gear levels, OH goose is better than Potency. For a T5+ gear level this is definitely true.

Executioner and Goose have the same proc rate, so we just have to compare the proc benefits.

Executioner is 840 ArP. Thats somewhere around 7% more dps.

Mongoose is 120 Agi (4% crit, with kings up) and 2% haste. Thats around 5-6% more dps.

However, 2/3 of Executioner (what you get from OH, if its also on MH, since they dont stack), is 2/3 * 7% or about 4-5% more dps. Which is worse than Mongoose.

Basically, executioner is better than mongoose, but not by such a large amount that its still better when youre losing 1/3 of the OH procs due to it not stacking with the MH proc.

Executioner/Mongoose is the way to go, definitely, once above a reasonable gear level. (Below that gear level, say when one is still in blues or has only some epics, Potency/Potency is probably best, due to cost if nothing else, given that gear at those levels will hopefully be replaced soon).

04-15-2008, 02:03 PM
"A caveat that acknowledges that the highest i-level item is typically the highest DPS item, no matter if it has ideal stats or not (i.e. the badge shoulders having higher DPS than t4/ragesteel)."

This is not always the case. Itemization is critically important, and can even overcome ilvl differences of 15 levels. Examples:

Red Belt of Battle (ilvl 128, very well itemized) is listed as 63 dps for me on maxdps.com
Belt of Seething Fury (ilvl141, Str/Haste) is 51.

Netherforce Chestplate (ilvl 125 from Mag.Terrace) is 79
Breastplate of Ire (ilvl 141 badge chest) is 67.

The Gruul gauntlets beat out the higher level T5 and even the much higher level Pillager's Gauntlets.

Specifically: Items with only two itemization stats that have haste as one of them are universally terrible, and are worse than items of 15 levels lower that were well itemized.

Items with large quantities of Hit rating are terrible unless they are well itemized. Hit rating is best gained through items with (Str or AP)/(Crit or ArP)/Hit, with less hit that either of the others. Like Red belt of battle for example.

Hit rating beyond 9% is poor, not only because it only affects about 40% of our damage (non heroic strike white swings), but also because getting much more of it than that generally means getting items which arent as well itemizaed as others. If you get much over 9% hit and its not all from a bunch of well itemized pieces that all have 15ish hit rating, thats terrible. I personally am sitting at 6-7% hit (plus precision).

04-15-2008, 02:04 PM
highest i-level item is typically the highest DPS item

04-15-2008, 02:09 PM
"A realistic model for determining how extra rage increases DPS (something besides extra rage/12 = heroic strikes, which is how the spreadsheets currently do it, I can go more into this if needed, as it is my current biggest sticking point with the spreadsheets that are out)"

More like rage/20 = heroic strikes, or so, given that you lose the white attack rage. I think rage/X is the best way to model this, just have to determine X.

"Testing to show whether yellow attacks are based on a single-roll system like we all believe to be true, or on a two-roll system, as some limited testing has shown."

I am 100% certain that special attacks are on a two roll system. This is the only way to explain how my crit is much, much lower than it should be when I am tanking, on specials, but it accurate for white attacks.

For example, I tend to have about 20% crit when tanking with a feral in group(about 12-13% base + buffs + mongoose + feral). Before there was expertise, I was missing about 5% miss + 6% dodge + 12% parry = 20-25%). I would routinely see about 20% crit on white attacks over the raid but 15% on specials. I was losing a quarter of my crit, and it made no sense to me. I saw this raid after raid for MONTHS, every time.

Now that I have expertise (above dodge cap), and more hit rating, guess what, my crit on specials is up! Its like 18% now, it is still less than white attacks but not by as much. This is fully explained by the two-roll system on specials.

There is a tremendous amount of evidence that Specials are two-roll and white attacks are one-roll. I have no doubt that this is the case, based on months and months of evidence from my WWS.

04-15-2008, 02:12 PM
highest i-level item is typically the highest DPS item

Yes, if we're talking about 10+ level differences.

I think its good to add that MOST of the times when this is not the case, it is because the high level item has only 2 stats and one of them is haste. (Many BT gears and badge gear items are like this).

Also for equal level items, 3 different stats is nearly always superior to 2 stats, even if the 2 are both good, and one of the 3 is hit or haste.

04-15-2008, 02:38 PM
uh oh. someone poked crimson-bear out of slumber. Expect good info to ensue!

04-15-2008, 03:51 PM

04-17-2008, 08:18 AM
Corb - I'm gonna add this on WOW.com too, if you're thread's still up ...

1. Execute (didn't see it in the inital list) - Process by which Execute can be managed, either with fast weapons at the ready (a macro button is easy enough) and an understanding of how to judge Execute range. Latency, to me, is an issue - alot of times I sit with a full rage bar and the mob goes down before I can get Execute off. Maybe an interesting topic?

2. "Field Testing" - alot of us use Spreadsheets like EJ and Maxdps, etc. But getting into the field with a setup, and checking data live is often most helpful. We all probably think we're faster on the rotation than we are in real life! A live test, with reduced buffage, and gear comparisons, etc. is helpful. I wonder if noting the key elements to a live test in the field would be helpful. Where to go, mobs to target, stat counters to use, buffs to get or not get (ie Kings? no pots? does it matter?, etc.)