• koaschten

The Sheet expects 1 Bloodlust per 10 minutes last time i checked. The increased number is to reflect shamans you rotate into the group only for bloodlust.

 

• songster

This is my experience. In fact, I find that a certain amount of variability in both the s and the r is forced by the presence/absence of various procs.

Generally I'm using 3s/5r, and it works fine. When I have lots of procs and extra energy to deal with, I can just let the bar fill up a bit and save it until I get a lean streak. However, if a lean streak goes on for too long, I'll wind up at 2 combo points on the slice and not enough time to get a third point in before it drops, whichmeans I use a 2s instead of 3s. Given that I'm still at very low average energy, I now know that I won't have time to go for a full 5r before the next turn of the cycle, so I drop that to 4r.

The upshot is that I use a pretty steady 3s/5r, but rarely I'll end up intercalating a 2s/4r to get back on track after a prolonged lean streak.

 

Alexsiss

99% sure this sheet assumes endless fights. No of heroisms should reflect the number of shaman you rotate in for hero. So if you have it set to 2 you would see the dps equal to you having 2 heroisms from now till the end of time(aka averaged haste) which is 0.04 haste passively(0.02 per).

 

• Latito

Alternatively, you can adjust the number of Bloodlusts you enter into the sheet to factor in the increased uptime due to a shorter fight. You can do this by multiplying the number of Bloodlusts you actually get by (10 / <Fight Length>).

So, for a 6-minute Brutallus, you would get 10 / 6 = 1.67. As such, a single BL would be worth 1.67 BL's in the spreadsheet. 2 would be worth 3.33, 3 would be worth 5. This works on shorter fights as well. Say your guild was capable of a 2-minute Teron. This would effectively make each BL worth 5. Rotate 2 BL's through a group and you have effectively 10 BL's. Consider that if your guild takes an extra 30 seconds to kill him, 2:30 total, you would be only getting an effective 4 BL's per actual BL.

It would be a bit easier if there was an actual "fight length" box, but that should then be used for trinket cooldowns as well and would likely get to be entirely way too much work. For now, a simple multiple and divide should suffice for those of us who actually care.

 

Towelette

This is correct. Aldrianna's spreadsheet assumes an endless, uninterrupted fight. This is also the primary reason that achieving the maximum DPS the sheet lists pretty much doesn't happen.

 

• Aldriana

Actually, the fact that the sheet assumes and endless fight is why you should be able to surpass the damage in the sheet in real uninterrupted fights. On an interrupted fight, you will of course do less DPS; but on fights like Teron, Brutallus, or no-graves Tidewalker, you should be able to do slightly more damage than the sheet suggests.

 

Towelette

"Pretty much" was probably the wrong choice of wording. "Usually" is more what I meant.

 

Auturgist

Recently I crafted a [Figurine - Shadowsong Panther], primarily for PvP, but I still wonder how it compares to my [Romulo's Poison Vial] for raid DPS. Of course, things are complicated by the fact that the trinket is bugged: the item description says that you gain 320 AP for 15 seconds when you use it, but the buff ends after 12 seconds (the duration of the old [Figurine - Nightseye Panther]). Needless to say, I'm hesitant to give up 35 Hit Rating and the poison proc from RPV for something that could be worse.

I don't know much about spreadsheets, but could someone maybe tell me how to find out how this item compares, with both a 15-second duration and a 12-second duration? (I'm not sure if they will fix the duration to last the full 15 seconds, or the change the description to reflect the current 12-second duration, so it would be nice to see how both versions compare...)

 

Towelette

If my math is wrong, please correct me. I'm only a mere English major.

Every 90s, you get 12s of 320ap.
12s\90s = ~13.33% uptime.
(0.1333) * (320ap) = an average of 42.67ap if used every cooldown.
80ap + 42.67ap = ~122.67ap total.

Every 90s, you get 15s of 320ap.
15s\90s = ~16.67% uptime.
(0.1667) * (320ap) = an average of 53.33ap if used every cooldown.
80ap + 53.33ap = ~133.33ap total.

As far as how that compares to Romulo's Poison Vial, that will depend on your gear and you'll probably need to use the spreadsheet to figure that out precisely. If I remember correctly, something like 14ap equals 1dps, though I'm not positive and can't remember if anything else factors into that for damage calculation purposes.

EDIT: I see that you're Sub-specced. If that's how you raid, the 12s duration would increase to ~134.94ap and the 15s duration would increase to ~146.66ap with 5\5 Deadliness.

 

Auturgist

Hahaha! Jesus no, man... I just haven't respecced since scrambling for Arena points late last night! I raid 19/42/0 Fist Spec., with 3/5 Vile Poisons.

But thanks for the math -- I'm an English type too, and always have a hard time trusting my own calculations even when I have the patience to try to crunch numbers.

 

drumbum

If you edit the "AP" column box in the Trinkets sheet for Berserker's Call (or Bloodlust Brooch) to either =80+320*12/90 (for 12 seconds) or =80+320*15/90 (for 15 seconds) you should be able to just use that to compare it.

 

Alexsiss

15 second duration should be better in almost every case(exceptions of course). the 12 sec is more on par with poison vial and will depend on what other procs you have increasing the value of hit.

 

Auturgist

What you are saying doesn't seem to make any sense. I'd think 15 seconds would be categorically better than 12 seconds, and I don't understand why either version would depend on what other procs I have that increase the value of Hit Rating, as the trinket doesn't offer any Hit Rating. Unless you mean to say that the value of Romulo's Poison Vial will depend on what other procs I have that occur "on hit" -- in which case the answers are: Mongoose x2, Shattered Sun Pendant of Might (Aldor/Light), and Shard of Contempt's Heroism proc.

 

Towelette

What situations do you think it would not be better? I can't see how the trinket would lose any value at all with a longer duration. The cooldown starts when you activate the trinket, not when the on-use effect has completed.

 

• Aldriana

What he means is that 15 second Panther should be better than RPV in almost every case (with exceptions), whereas the 12 second Panther is more comparable to RPV. Obviously 15 second Panther is always better than 12 second Panther.

It also might be noted that, at equal EP, an activated trinket - particularly one with a relatively short duration buff - tends to be better than a proc trinket, as you have more control over when it goes off. This doesn't matter for all fights, of course, but there are certainly fights where it does. On the other hand, it does take a little more attention to remember to use, and if you don't use it close to every cooldown in boss fights the advantage disappears pretty quickly.

 

Towelette

Ah ok, I misunderstood.

Auturgist: As for RPV's value and hit rating, what he means is that certain other trinkets increase the value of +hit rating when equipped. For example, equipping a DST increases the value of +hit, which should make RPV worth more with DST equipped in the second trinket slot than it would be worth with, say, the Panther in the second slot.

 

Auturgist

No, no, I get that, but that wasn't what was actually written; the way his post was phrased implied something other than what he meant to say, which is why I commented for correction/clarification, and also listed my "on hit" proc effects anyway to give an idea what could possibly contribute to the value of RPV.

Basically, I couldn't think of a reason 15 seconds wouldn't be absolutely better than 12 seconds. And then I also didn't see how either version of the Panther somehow made Hit more valuable, being as it is an activated trinket that gives pure attack power. I think the ambiguity occurred with the use of the conjunction "and" in the line: "the 12 sec is more on par with poison vial and will depend on what other procs you have increasing the value of hit."

 

Alexsiss

The reason i noted hit rating is because anything that can proc from your oh will be valued more the more hit rating you have as you are missing less attacks

 

• Aldriana

Quick thoughts on Righteous Weapon Coating, which I used for Brutallus tonight:

* Spacing on procs was universally between 45 and 50 seconds; thus, it seems likely that, like the SSO Exalted neck and Shard of Contempt, it has a 45 second internal cooldown and a high proc rate (not sure how high we're talking, but definitely at least 10%, and it could easily be twice that.

* Thus, in a sustained DPS environment, it gives roughly 21% uptime, I would estimate; this would make it equivalent to roughly 63 static AP.

* With typical raid buffs, one gains about 1 DPS for every 2.5 AP; hence, the damage bonus of 63 AP is something like 25 DPS. Compare to Instant Poison, which does about 35 DPS on either hand.

Thus the conclusion is that Righteous Weapon Coating is inferior to Instant Poison, right? Well... mostly. In a vacuum this would be true; the caveat is that if there's an enhancement shaman in the raid, then the fact that Instant Poison eats Stormstrike charges is a liability - the Stormstrike charges are better spent on higher-damage nature attacks, and the fact that Stormstrike gets taken down faster means (as I understand it) that all the Deadly Poison in the raid does less damage as well.

Thus, the *real* conclusion is that if you're stuck without WF and have to use something other than Deadly Poison, and you have an Enhancement Shaman in your raid, you're probably best off using Righteous Weapon Coating. Other than that, it's not worth using as a rogue.

 

x1tiger1x

I have a question on how Executioner and Mongoose are modeled. I'm not trying to disprove the spreadsheet, I just want a better understanding of how it works.

Executioner and Mongoose are both 1.2PPM with no internal cooldown. Therefore, they are up about 18s out of every minute for an uptime of 30%. Now, if I remember right, the WSC was (or is?) modeled at about a -430 passive armor penetration. Is executioner modeled at a passive 252 (840 x 30%) Armor penetration? And is mongoose modeled at a passive 36 Agility and .6 haste?

Unless I am wrong.

 

Left

No, the modeling is actually a good bit more sophisticated than that. Without getting into it too much, haste, +hit, weapon speeds, WF, and number of special attacks affect the number of procs for each ability. The spreadsheet takes all that into account and adjusts uptime accordingly. Aldriana can address the math in more detail than I can, however.

 

• Aldriana

Roughly speaking, you're missing the following points:

1) 1.2 PPM assumes you're only autoattacking and never miss. The fact that your autoattacks miss sometimes thus lowers the number of procs you get; the fact that you have instant attacks increases the number. In practice, your MH weapon will proc more than 1.2 times per minute, and your OH slightly less.

2) Since procs can overlap, proccing 1.2 tiems per minute does not mean you'll get 18 seconds of uptime out of it or 30% uptime. The actual formula is rather more complicated.

For a more detailed discussion of these issues, I'd refer you to the Proc Mechanics article in the TTT.

In the spreadsheet, then, rather than modeling it as some static amount of agi and haste or ArPen, it's actually modeled as a proc, and thus correctly changes value based on the first-order effects of hit, expertise, sword spec, and windfury, and the second order effects of haste.

It also models Expertise as 840 ArPen with uptime of U, rather than U*840 ArPen all the time - which makes a difference.

As a historical note: the ability to model procs as procs was actually one of the major motivators in the creation of this sheet; at the time, the DPS sheet was attempting the sort of passive conversion that you mention, which was causing accuracy issues at times.

 

• Aldriana

Public service announcement:

I just found a bug in the 5sXr3s sheet. If you are:

a) Aldor
b) Using the SSO exalted neck, and
c) 5sXr3s is your optimal cycle

the results you get out of the spreadsheet will be utterly bogus. If you satisfy a) and c) but not b), everything should more or less work, except the SSO neck will be undervalued.

I'll be fixing this in the 0.10.1 release, but, in the meantime, if you want to fix it for yourself, go to the 5sXr3s page and fill the formula in B337 across into all cells in that row.

 

Rivalry

Excuse me if this is completely obvious and I missed it and/or it's been posted before. With my gear setup, when I select a boss armor of 7685 it displays an optimal cycle of 2.8s5r, and if I change the armor to 6200 it changes to 5s5r. Would like to know what the logic is behind this.

 

• Aldriana

Compressed cycles of the form Xs5r sacrifice SS damage (as you're spending energy on finishers) to get higher rupture uptime. Thus, they're better than 5s5r provided the loss of SS damage is smaller than the rupture damage gain. At 7685 armor, it is. However, when you drop to 6200 armor, that increases the damage of the SS but not the rupture, which thus causes the tradeoff to no longer be worthwhile; hence 5s5r takes over.

Note that we're talking differences on the order of a tenth of a percent here. Open up the DamageCalcs page if you want to see what the actual damage for the 2 cycles are.