Bolded the appropriate section under Eviscerate, and re-quoted the Envenom section to emphasize the fact that at essentially any level of gear, it makes no sense at all to compare Envenom against Rupture. I did not factor in Vile Poisons in comparing Eviscerate against Envenom, because I don't think it's a worthwhile comparison to make. Note that for each point you spend in Vile Poisons, the damage of your Envenom increases, but so does the damage you lose from having to refresh your Deadly Poison stacks afterwards. I'm sure there's a point where Envenom's damage increase over Eviscerate scales to the point where it's worthwhile to use it, but as it is Eviscerate/Envenom are highly situational finishers. The build that would primarily be concerned about this question rarely seems to be run in raids anymore, too.
So I'm a dying breed? Thanks for pointing out the points I missed in the first post (I thought I was paying more attention when reading :/ ). I'm 41/20/0 and I'm currently at T4/ZA level. Used to go Gruul, Maggy, SSC/TK (first bosses) on a weekly basis but only doing 10 mans now (with a 25man IF we're lucky to get ppl). Anyway, the reason I'm asking:
Like everyone else with this build, I find it easy to keep SnD up constantly during boss fights and rupture is on 80-90% of the time. What should the third finisher be? I use evisc most of the time and envenom when I got 5 poison charges ticking (sometimes it's just decided by what I feel like doing). Currently I have both Imp Evisc and Vile Poisons. Would switching imp evisc to murder make a bigger impact on DPS? Should envenom be used at all since it eats the poison ticks?
And the final question: Why is combat-mutilate considered so low on DPS for raiding? Does the -?% on the first post mean it's between -7% and -11% or is that just an estimation as to where mutilate MIGHT be but without any hard evidence on what the % reduction is?
I'm sure there's a point where Envenom's damage increase over Eviscerate scales to the point where it's worthwhile to use it, but as it is Eviscerate/Envenom are highly situational finishers.
Indeed. However, an interesting point is that the situation when you have to use Eviscerate (i.e. when rupture won't have time to tick out) is exactly the same situation that makes Envenom > Eviscerate (i.e. when the DP stack won't have time to tick). Thus my preference for the poison talents in a combat build (4 points spare) is to put them in Vile Poisons rather than Improved Poisons, even though the spreadsheet says the optimal choice is 2 points in each talent.
For those interested in heavy theorycraft number-crunching, check out Sage. It's a rather new computer algebra system available for free (it's GPLed) and about as powerful as MATLAB/MATHEMATICA, etc.
If you want to do statistics, R is an excellent package, also FOSS.
Indeed. However, an interesting point is that the situation when you have to use Eviscerate (i.e. when rupture won't have time to tick out) is exactly the same situation that makes Envenom > Eviscerate (i.e. when the DP stack won't have time to tick). Thus my preference for the poison talents in a combat build (4 points spare) is to put them in Vile Poisons rather than Improved Poisons, even though the spreadsheet says the optimal choice is 2 points in each talent.
Actually, I wasn't even considering that situation. I was considering the situation where the mob is immune to bleeds. Funny thing there, though, is that the two most notorious bleed-immune targets (Hydross and Void Reaver) are also poison-immune, and therefore Envenom isn't usable anyway.
Your reasoning for the situation you mentioned is definitely valid. Personally I spend the points in Vile Poisons as well, mostly due to educated guesswork, but even my own spreadsheet only puts them a very very small amount ahead of Improved Poisons.
Originally Posted by Achillion
Like everyone else with this build, I find it easy to keep SnD up constantly during boss fights and rupture is on 80-90% of the time. What should the third finisher be? I use evisc most of the time and envenom when I got 5 poison charges ticking (sometimes it's just decided by what I feel like doing). Currently I have both Imp Evisc and Vile Poisons. Would switching imp evisc to murder make a bigger impact on DPS? Should envenom be used at all since it eats the poison ticks?
You have two options: run dual DP and use Envenom, or run IP/DP and use Eviscerate. If you only have DP on one hand, even Mutilate can't grant you enough DP procs to make up for losing the DP stack on an Envenom. If you have DP on both hands, you're losing a pretty good chunk of potential DPS from IP just to get a questionable amount of gain from switching your third priority finisher from Eviscerate to Envenom. Either way I highly doubt it's actually worth it.
Originally Posted by Achillion
And the final question: Why is combat-mutilate considered so low on DPS for raiding? Does the -?% on the first post mean it's between -7% and -11% or is that just an estimation as to where mutilate MIGHT be but without any hard evidence on what the % reduction is?
Essentially, empirical testing has shown it's inferior to combat, and that's as good as we know, because no solid model of Seal Fate currently exists. The DPS spreadsheet's model is bugged and even if it were fixed, Seal Fate is still a touchy thing to model. So to answer your second question, yes, the -?% indicates that I have no clue where it actually ranks and just took a guess that it's probably inferior to those other builds, but superior to Shadowstep.
...
Oddly, this is somewhat reversed now. In purely sustained fights where swords can maintain cycles like 3s5r and keep very high rupture uptime, they blow past daggers. It is in interrupted fights where swords have a hard time maintaining tight cycles that daggers have the ability to compete, as dagger builds are much less reliant on rupture and much more reliant on simply keeping SnD up and backstabbing. Hence, to the extent a dagger rogue is going to be able to keep up at all, it's going to be on trash and interrupted fights, not on purely sustained fights. Which is exactly reversed from what it was pre-BC, but that's the way it goes.
In theory, I agree here. However, in practice it seems that a lot of fights which are interrupted/mobile hit the second problem with daggers: positioning. (At least, this is true in the 10 man content I have seen.) For example, I had a hell of a time backstabbing the sparks on Curator, or backstabbing Aran when he turned to cast. I've also noticed that totems don't have a "back", making any positional sort of build inferior to a non-positional one on Lynx Lord Halazzi in ZA. I actually have a macro right now that switches me to a sword so I can sinister strike the totems, then switch back to daggers for Mutilate on Halazzi.
What am I going to do with this thread is over time establish the effectiveness of both the mongoose and executioner enchants for rogues. First will be the individual comparisons that will eventually include my own studied PPM, then when combining them.
As a rogue, it takes ~14 AP in order to increase our DPS by 1. That being said at AP of (see below) you get an increase to DPS of (see second number
AP: DPS:
1400 100
1500 107
1600 114
1700 121
1800 129
1900 136
2000 143
You can kinda see a pattern here, 100 AP = ~7 DPS increase, now with that being said we all know that mongoose proccing gives a 120 agility increase which is 3% to crit and ~9 DPS increase.
Now let us calculate the % of damage increased when mongoose procs
AP: DPS:
1400 9%
1500 8.4%
1600 7.8%
1700 7.4%
1800 7%
1900 6.6%
2000 6.3%
As you can see, mongoose gives diminishing returns the higher your AP gets.
Let us now look at executioner. What this gives you is 840 armor penetration. A fully debuffed boss will give you -4010 armor reduction (2600 for 5 sunders, 610 for faerie fire, 800 for Curse of Recklessness) I do not believe that we use CoR so our fully debuffed mob will be at -3210 armor. Here is the armor value for most high end bosses.
Serpentshrine Cavern:
Hydross the Unstable: 7700
The Lurker Below: 7700
Leotheras the Blind: 7700
Fathom-Lord Karathress 6200
Morogrim Tidewalker: 7700
Lady Vashj: 6200
Black Temple:
High Warlord Naj'entus: 7700
Supremus: 7700
Shade of Akama: 7700
Teron Gorefiend: 6200
Gurtogg Bloodboil: 7700
Reliquary of Souls:
- Essence of Suffering: 0
- Essence of Desire: 7700
- Essence of Anger: 7700
Mother Shahraz: 6200
Illidari Council:
- Gathios the Shatterer: 6200
Illidan Stormrage: 7700
As we can see, on average most bosses have either 6200/7700 armor, minus VR who is at 8800. What this means is that the mitigation done by a lvl 70 attacker is going to be R = (Ar)/(Ar + B)*100
where
R=damage reduction
Ar=armor
B=(467.5*Enemy Level - 22167.5) = 11960 for a 73 mob or 10557.5 for a 70 mob
So the damage reduction for a mob with x armor is equal to y reduction
X: Y:
6200 36.34%
7700 39.16%
8800 42.38%
So lets take out the -3210 armor that we should have from sunders and FF
X: Y:
2900 19.51%
4490 27.29%
5590 31.85%
NOW for the part that we've all been waiting for! the damage increase when executioner procs on a fully debuffed boss mob.
X: Y:
2060 14.69%
3650 23.38%
4750 28.42%
Thus, the damage increase for a mob with x starting armor will be y when fully debuffed and executioner proccing
X: Y:
6200 4.82%
7700 3.91%
8800 3.43%
As you can see, mongoose (at least with these preliminary numbers) does do more for a rogue when looking at damage % increase.
Over the next couple of weeks, I will be tracking the PPM, uptime and downtime of both my mongoose and executioner so expect more data as it develops.
I tested Executioner's procs using Procwatch mod. During the Fathom-Lord Karathress fight this recorded how often it procced per hit and every second of combat.
Recall:
803 Hits
27 Procs
11:11 of Combat time
which equates to ~29.7 Hits/Proc
total of 2.4 PPM which translates this:
Executioner on a boss fight, when consistently using SnD and attempting to hitting the target at all times, this enchant is up around 36 total seconds out of 60 seconds in a minute. These hits btw are counting my TOTAL hits, MH and OH
Trash, this proc is MUCH lower, around 1.3 or so.
My OH speed is at 1.15 with SnD up
Mongoose is currently only on my MH, with SnD up, i run 2.00 speed
On trash clearing to Fathom-Lord, here are the numbers
1451 Hits
58 Procs
21:21 Mins of Combat Time
which equates to ~25 Hits/Proc
total of 2.7 PPM which translates (on trash) to:
Mongoose on trash, when consistenly used with SnD, will run about 40.5 seconds out of a total of 60 seconds per minute.
On Fathom-Lord Karathress, here are the stats (remember that there is some running + wipes, etc etc STFU already!!!!)
692 Hits
27 Procs
10:13 Mins of Combat
which is equal to ~ 25.6 Hits/Proc
total of 2.6 PPM which translates (onto the boss) to this:
Mongoose on Fathom-Lord Karathress (remember that we did have to run around A LOT), when consistently used with SnD will run about 39 seconds out of a total of 60 seconds per minute.
One thing to remember about the proccing of mongoose is it is on my MH only, OH had executioner at the time I was testing.
The DPS spreadsheet's model is bugged and even if it were fixed, Seal Fate is still a touchy thing to model. So to answer your second question, yes, the -?% indicates that I have no clue where it actually ranks and just took a guess that it's probably inferior to those other builds, but superior to Shadowstep.
This may seem like a completely retarded question, one that has probably been answered numerous times to supposed-do-gooders......
but, why is it not possible to take the idea of seal fate to a math-guru/theorist and have them try and make an accurate model? I know we have some computer/programming experts and even some calc stars (I was fairly good at calc) but as far as modeling something in a mathematical approach, seems like we might have to get someone who does this for a living.
I'd be more than willing to look into it, if someone would lay out WHY the previous attempts didn't work. As far as the talent goes, it seems straight forward. When maxed out, 100% of your critical strikes, per your crit strike %, will add an extra combo point. Couldn't we take a a general concensus of attacks total per average bossfight (I'm guessing around 750-1000) and then multiply in the crit chance (Higher for BS's and such) and then take the number of added combo points over an average fight and then factor in the gains in DPS? OR on a totally different avenue, possibly look at a normal 2-3 finisher cycle and model based on that (allowing more specificity)?
*As I type this I'm seeing the depth of error we could get in to and why most people have steered clear of trying to model this...*
There was some discussion of the issue in the Hemo vs combat thread, and it really boils down to the fact that Seal Fate is a special case where the normal methods of long-run average cycle performance measures won't be accurate enough to represent the actual performance of the thing.
It's not that it's impossible to model, it's just that we're already pretty confident that it's inferior to combat (from pretty good amounts of empirical data), and as such, going to the amount of trouble it seems like it will take to model it with the degree of accuracy to which we've modeled combat cycles and DPS doesn't seem worth the trouble. If you'd like to try for it, be our guest, as who knows whether Blizzard will suddenly make builds involving Seal Fate viable with Wrath, sending all us theorycrafters scurrying about to model it properly.
There was some discussion of the issue in the Hemo vs combat thread, and it really boils down to the fact that Seal Fate is a special case where the normal methods of long-run average cycle performance measures won't be accurate enough to represent the actual performance of the thing.
Wouldn't it be possible to do a short cycle build model and then extrapolate over a larger one, of course accuracy would be diminshed the larger the cycle you extrapolated it over, but it seems that looking at a "boss fight" would be impossible for a seal fate considereing all the possibilities, but looking at a much shorter encounter and trying to model the possibilities would seem more logical...
I think what it fundamentally comes down to is that there are some people here (myself included) who have a pretty good idea *how* to make a decent SF model; it's just that doing so is a heck of a lot of work, and it hasn't been deemed worthwhile. In particular, the major problem is that one needs to consider variance in addition to the average, as combo points are a lot more granular and a lot less able to smooth out burstiness of cycle. That is, it's not enough to say "on average, each attack genrerates 1.4 combo points" - you have to model it as "40% of the time you get 2 combo points, and 60% of the time you get 1" - which means the problem pretty quickly becomes lots of case work that you ultimately average together. There's nothing fundamentally hard about doing so: it's just a very long, messy calculation that no one has felt like doing, because no one expects to demonstrate anything other than "SF builds are clearly inferior" - which we already know.
Also, I might note that a spreadsheet format is singularly ill-suited for the task. If the aforementioned programming-language based approach ever turns in to something, I'd probably throw a SF model in because it would be slightly less painful to do in that format; but in terms of doing it in a spreadsheet, it's just really not worth the effort in my estimation.
Well thats good to know, though sad to hear. I figured that a spreadsheet model was going to be fairly difficult with the restrictions that a spreadsheet inherently has, I guess I just hoped that it was still possible to deem that valuable enough to spend time, but it's not my time, so I def understand not wanting to work on something we are pretty sure is inferior.
Hey I was wondering if you could help me choose my offhand weapon. I recently respecced to combat, the plan is to go sword when i get a good main hand sword. But for now I am the typical 20/41/5 combat dagger spec. I am finding the dps is great but I was wondering about my weapons. I am currently using Malchazeen and Blade of the Unrequited (with three+8 hit gems in) both with mongoose. In my bags I have the season 2 offhand sword. It is a fast weapon than my current offhand with alot more dps aswell but obviously my weapon spec is dagger at the moment, which would be better?
Also there is the option of buying the mainhand sword from the aldor and going sword spec now. But obviously the dps on the aldor sword is far below that of Malchazeen.
So basically should I change anything or just stick with the dagger until I get Spiteblade/season 2 MH/ZA MH?
have you checked out hte gear/dps spreadsheets? Plugging in your information and both specs (20/41 or 15/41/5) and appropiate gear, you'll be able to see a rough estimate of dps based on a normal boss fight.
I'm going to guess that aldor sword + s2 oh will be more dps than you think, and depending on the fight where BS's might be an issue, I'd say it pulls ahead quite a bit. But again, try the spreadsheet and see what happens..you'd be amazed at the wonders it can do.
Mongoose is currently only on my MH, with SnD up, i run 2.00 speed
On trash clearing to Fathom-Lord, here are the numbers
1451 Hits
58 Procs
21:21 Mins of Combat Time
which equates to ~25 Hits/Proc
total of 2.7 PPM which translates (on trash) to:
Mongoose on trash, when consistenly used with SnD, will run about 40.5 seconds out of a total of 60 seconds per minute.
On Fathom-Lord Karathress, here are the stats (remember that there is some running + wipes, etc etc STFU already!!!!)
692 Hits
27 Procs
10:13 Mins of Combat
which is equal to ~ 25.6 Hits/Proc
total of 2.6 PPM which translates (onto the boss) to this:
Mongoose on Fathom-Lord Karathress (remember that we did have to run around A LOT), when consistently used with SnD will run about 39 seconds out of a total of 60 seconds per minute.
One thing to remember about the proccing of mongoose is it is on my MH only, OH had executioner at the time I was testing.
I see most rogues on my server running around w/ Ex on their MH (Myself included) Mongoose OH. Is there going to be a significant proc difference when Ex is applied to a slower weapon?
If Mongoose looks to be more beneficial to a rogue would it be better to put it on the weapon that would proc it most often (fast OH)? Or does weapon speed not factor into the proc rate of either?
I see most rogues on my server running around w/ Ex on their MH (Myself included) Mongoose OH. Is there going to be a significant proc difference when Ex is applied to a slower weapon?
If Mongoose looks to be more beneficial to a rogue would it be better to put it on the weapon that would proc it most often (fast OH)? Or does weapon speed not factor into the proc rate of either?
To the best of my knowledge (and let me be VERY honest by saying that we are still VERY unsure about Executioner except in a broad "it's about as good as mongoose" sense), this is how I personally break it down.
That is - Executioner and Mongoose are both strong, and both strong on the mainhand, so it's more personal preference at this time. They're both (as far as we can tell) PPM procs, so mainhand will proc more, regardless of speed (due to instant attacks with the mainhand).
Please note that Ex/Ex is fairly weak due to the funky mechanics it has with overwriting/renewing itself instead of stacking. Unless this changes, my opinion is that Ex/Ex is very inferior to the other options.
It's my personal guess that as you move into more Armor Penetration gear (ZA and T6) that Executioner on the MH will start to outperform Mongoose. This is due to Armor Penetration improving all rogue stats, including itself, so as you get more ArmorPen, it's more powerful (up till you're putting a target's armor at 0). This makes the final graph for T6 rogues look like:
Again this is all "educated speculation" at this point, but (to my knowledge) it's the best we have. If you have concrete information about Executioner other than "Wow I feel like it's up a lot!" please post in the Executioner thread on these boards.
Thanks for sending me to the spreadsheet. Useful although it doesn't have the blade of the unrequited in the offhand dagger list for some reason which makes it hard to get exact numbers.
Well, two things on that. 1) If the gear spreadsheet doesn't have it as an option, it's probably very non-optimal, answering your question, and 2) The DPS spreadsheet should give you the answers, as it has a larger range of equipment supported.
To be perfectly accurate: Blade of Unrequited was excluded not so much because it's clearly inferior (although, by the metric of .1 speed ~= 10 OH DPS ~= 35 AP, it most likely is) but because weapon socketing isn't wholly hooked up and it didn't seem worth adding purely for the one item that isn't notably good anyway. However, with more socketed weapons of reasonable quality in ZA, it's becoming somewhat more important to handle socketing weapons correctly, so that may show up at some point.
I see most rogues on my server running around w/ Ex on their MH (Myself included) Mongoose OH. Is there going to be a significant proc difference when Ex is applied to a slower weapon?
If Mongoose looks to be more beneficial to a rogue would it be better to put it on the weapon that would proc it most often (fast OH)? Or does weapon speed not factor into the proc rate of either?
All the testing I did purely involved Ex in the OH, so unfortunately I can't tell you. As stated, we believe that it is a PPM, but looking into it, I'm not 100% sure about it, just based on the numbers. The thing with Ex is that if you check out the debuffed armor of bosses, it increases your overall damage by up to ~5% at any range of AP. When proccing Goose you have to remember that the higher your AP, the less increase you will get. The thing that I like about Goose is the fact that it also increases my rupture damage, Ex only helps your Evis finisher.
Of course, it should be remembered that finisher damage is ~10% of your total damage output, so the fact that Executioner doesn't help it is not a show-stopper.
When Executioner was first announced I did a rough estimate in the Rogue Gear sheet and concluded that if Exec and Mong had the same proc rate, Exec would be slightly more damage, but on a small enough scale that you could argue for Mongoose based on the extra dodge it gives (yes, they're that close together). If this assumption is valid, then, it's easy to show that Exec MH Mong OH maximizes DPS.
That said: it's not at all clear that the stated assumption is valid, so take that conclusion with a grain of salt.
So I'm a dying breed? Thanks for pointing out the points I missed in the first post (I thought I was paying more attention when reading :/ ). I'm 41/20/0 and I'm currently at T4/ZA level. Used to go Gruul, Maggy, SSC/TK (first bosses) on a weekly basis but only doing 10 mans now (with a 25man IF we're lucky to get ppl). Anyway, the reason I'm asking:
Like everyone else with this build, I find it easy to keep SnD up constantly during boss fights and rupture is on 80-90% of the time. What should the third finisher be? I use evisc most of the time and envenom when I got 5 poison charges ticking (sometimes it's just decided by what I feel like doing). Currently I have both Imp Evisc and Vile Poisons. Would switching imp evisc to murder make a bigger impact on DPS? Should envenom be used at all since it eats the poison ticks?
And the final question: Why is combat-mutilate considered so low on DPS for raiding? Does the -?% on the first post mean it's between -7% and -11% or is that just an estimation as to where mutilate MIGHT be but without any hard evidence on what the % reduction is?
I've been playing mutilate throughout most of TBC and the only time you should be Envenoming on a boss is when you want to see "Big Numbers" or have a pvp spec (53/0/7 or a variant) or as the last finisher at 1%.
Finisher preference has always been SnD > Rupture > Eviscerate for me.
Poison choice should be WF/DP or IP/DP.
I haven't experimented with DP/IP, which, presuming that you have enough hit to keep Deadly rolling, might be better.
As for the "rough" DPS estimate, I think the reason why mutilate is so low there is that there hasn't been much data gathering on it. From personal experience I can say that it's not as low as 7. But maybe that's just me.
Fans glory to the Gladiators,
Gods glory to the Heroes.
I've also noticed that totems don't have a "back", making any positional sort of build inferior to a non-positional one on Lynx Lord Halazzi in ZA. I actually have a macro right now that switches me to a sword so I can sinister strike the totems, then switch back to daggers for Mutilate on Halazzi.
Is this indeed the case? I have never had a problem with backstabbing Spitfire Totems; I've done FLK as least as recently as two weeks ago. Is Halazzi a special case? My guild does not run ZA regularly, so I have no experience with that particular boss.
Is this indeed the case? I have never had a problem with backstabbing Spitfire Totems; I've done FLK as least as recently as two weeks ago. Is Halazzi a special case? My guild does not run ZA regularly, so I have no experience with that particular boss.
No, totems definitely do have a unique 'back' even if there is no graphical reason for it. Kind of like finding C'thun's back. It doesn't make sense, but it's there.
I've been playing mutilate throughout most of TBC and the only time you should be Envenoming on a boss is when you want to see "Big Numbers" or have a pvp spec (53/0/7 or a variant) or as the last finisher at 1%.
Finisher preference has always been SnD > Rupture > Eviscerate for me.
Poison choice should be WF/DP or IP/DP.
I haven't experimented with DP/IP, which, presuming that you have enough hit to keep Deadly rolling, might be better.
As for the "rough" DPS estimate, I think the reason why mutilate is so low there is that there hasn't been much data gathering on it. From personal experience I can say that it's not as low as 7. But maybe that's just me.
On the IP/DP Vs DP/IP issue: I always use 1.8 speed daggers on both hands and I must say I haven't seen much of a difference on either setup. I currently have 245 HR. The only difference is that with DP on OH I can sometimes shiv to reset the DP timer. For instance at some fights were you have to stop DPSing and DP ticks away and you start hitting again when your 5 poison stacks are about to go out. I've read on forums and guides that shiv should almost never be used in a raid environment and I suppose a mutilate would reset the timer at that point, but it's nice to know you have control and you're not crossing your fingers for a poison proc.
On WF Vs GoA: I've read somewhere around here that WF isn't THAT much better anymore (since a recent downgrade I think). Personally, I always asked for WF (cos the guides said so :/ ) when it was up to me but when I was in a group that dropped GoA I was rarely getting non crits with Mutilate (stick a druid in that equation and it's to die for). I always wondered, since seal fate is our thing, isn't GoA at least on par with WF for assassination rogues?
12/11/07, 6:42 AM
Similar Threads
