hi leica,

its possible to know your shot rotation or your castsequence?
i've seen your damage meter as bm here http://www.lossendil.com/wws/?report...&s=20683-21360

and i admit im "Oo" to see this, its awesome :o need some advice plz ^^

i see you dont use multi and arcane often, you dont have them in your rotation? or downranking?

thx in advance

Never.

I suppose you could do that, essentially it would just be weighted averaging. It would just be %uptime * speed for all the respective speeds, then sum those weighted values. So using my example from before, with frenzy uptime at 73%, the average attack speed would be 0.99*0.73+1.28*(1-0.73)=1.0683. That looks about right, but honestly you may want to double check that. You could pretty easily check this with WWS: time dps presence times time of fight (so percentage * total time) then divide by the number of auto attacks to get a rough idea if this is correct. Honestly, because the variation from 0.99 to 1.28 is relatively small, you may not get good verification (in practice there are too many extraneous variables arising from time spent moving, etc), but in theory that should work.

Get the cast sequence directly from WWS : browse the log and filter. Example :

http://www.lossendil.com/wws/?report...8&fiaa=a&fit=0

Or if you also want the buffs & gains :

http://www.lossendil.com/wws/?report...=2&fit=3&fit=4

Leica's secret seems to be a perfect pattern of "Auto Shot/Steady Shot" for 66% of total damage. His pet does 24% of the total, and other shots are used for a mere 10% of total, mostly on adds.

This is pretty much accurate. I'll fire off arcane shots if I'm on the move, ditto on multi-shots if I need to travel a fair distance. I always tend to 'stutter-step' when running to make sure I don't lose autos as well, so my pattern is like... auto arcane on the run, autoshot time says new auto coming, stop multi-auto.

Fights where it's just stand and fire, it's pure steady/auto, typically during quickshots even, I could probably improve there a bit.

An interesting note, when I went from Sunfury to Serpent Spine Longbow my dps stays relatively static, this was because I gained deadtime between autoshots and steady shots. I fixed this by going from Vengeance Wrap to Crystalweave Cape. I lost a bit of AP and Crit, but the haste on the cape actually buffed my dps even with that loss. I'm looking foward to when I can pick up some haste gear from BT and go back to a more appropriate cape.

My guild is recently working on Vashj, and has requested that I spec into Hawk Eye to kill spore bats. What I am wondering is if The Beast Within is better than 1 point in Mortal Shots or vice versa, for damage?

Any insight would be appreciated.

You can still reach spore bats with 35 yard range. You just don't have as much freedom in where you can stand. As a BM hunter stand directly below the flight path, and you'll still be able to get 2-3 shots off. It's much easier if you split the damage among a few people along the flightpath. None will have to take their attention off Vashj for too long, and since the approach is predictable you can knock 'em out pretty reliably with just a few hunters/mages along the flight path paying attention for them.

I've hit a brick wall with my DPS as of late --

Specc'ing to 41/20/0 has definitely increased by DPS by a substantial amount (along the lines of 150+ more dps). However, I'm not currently happy with where it's at, especially after going through this thread. I don't have WWS, but here are some numbers I've put out in recent raids using SWS with pets merged:

Magtheridon (we don't make use of the banish): 688, 591, 564 - Both of those were at a side cube which put me in range to dps him while waiting click. Both those attempts ended right around 30%. A few nights later, on an attempt where we downed him for the first time I was using a scorpid and was at the cube next to the door (meaning I had to be out of range of mag for about 15+ seconds). Here's what makes me sad: I died at about 25%, I was at 280 dps and my scorpid was at 282. He out damaged me by 1200. Recap had a max scorpid tick at 1500 (it must have been refreshed during the 1 second or so Mag was banished).

Gruul: 623, 680, 583.

I'm not happy with these numbers and I'm not sure what to do. Here is what I have tried

1. Shot Rotation: I use a cast-sequence to do a pure auto/steady rotation. I've installed bigtrouble and have done a manual rotation for +/- 15 dps difference than the cast sequence. I've tried replacing steadies with arcane/multi-shot, but that resulted in less DPS and much greater mana consumption. I've tried clipping my autoshot to squeeze in a steady+arcane/multi, but my 2.03s attack speed combined with 277ms ping resulted in too much clipping. So I use an auto-steady castsequence.

2. Gear (armory profile to the left): Obvious shortcomings are my trinkets. I'm *still* trying to get the hourglass from BM, and I'm working towards the bloodlust brooch. Other than that, I'm not sure if there are any other obtainable upgrades outside of raids. I've gone through hundreds of items in Cheeky's, and still cannot find a viable upgrade for my Fel Leather Boots. I plan on replacing my belt with a General's PVP belt come the pvp patch. Cheeky's tells me my gems are currently the best I can do at the moment (which surprised me, as I gemmed most everything to support a survival build for 2.1). Anyone see something I'm not?

3. Pet: I use a Scorpid with Rank 4 poison and apply it properly for Magtheridon and Gruul's. In Kara and SSC I take my Ravager. Both pets have CR and Avoidance. Scorpid uses only Scorpid Poison and auto attack, Ravager uses Gore and Bite (both rank 9). I can generally reposition my pet during cave-ins in no longer than 5 seconds, while I'm moving I use Arcane and serpent sting, by then I've normally stopped moving and begun to resume my steady-auto rotation.

4. Consumables: I use Warp Burgers, Major Agility Elixir, Major Mageblood Elixir, Superior Mana Oil, Fel Mana Potions.

5. Spec: I am fairly certain my BM build is as optimized for raid dps as I could get it. Anyone disagree?

This is what I do and what I've tried. Any suggestions?

In my experience using SWStats to calculate "DPS" (as opposed to just Damage) gives lower than actual results.
If most people on this thread are reporting WWS DPS numbers and you're using SWStats numbers that could be one thing that would lower the DPS you *think* you're doing without affecting actual DPS.


You reposition your pet for CAveins? I generally just pull mine out, wait, and send it back in so that I don't have to interrupt auto/steady. I wonder which is better. Losing FI is bad, but so is losing shot rotation.

The last thing is my DPS depends a lot on where I'm standing. If I'm able to stand on a rock for gruul (or get there in just 2-3 steps) I do a lot more than if I'm having to get flung around and reposition myself etc. (I'm not running to a safe zone when I'm far away - this is just the shatter + bandage time)

It sounds like you know what you're doing and your gear is good enough that you should be seeing higher numbers than you are. I can't really diagnose any problem with your technique without actually seeing a WWS log and browsing through some of the data.

Here is our latest Mag kill if you feel like looking at any of my numbers or stats. I'm using a steady/auto shot macro. I was clicking an orb that was about 5 yards out of range, so I had to time it as closely as I could before running to the cube. This was with a lvl 69 scorpid with rank 4 poison- popping my trinket and BW before applying poison on the first stack.

http://www.lossendil.com/wws/?report...ham&s=991-1673

Try going into the log and filtering the data if you're looking for shot rotations or specific trends.

I was curious to see if people using SWS are still receiving higher numbers than my SWS reports. I know that WWS stops the dps count during cave-ins, moving, and shatters, but to the tune of a 400+ difference? I'm curious to find out.

That is something I have not thought of. Can we simulate this mathematically? I may try just pulling my pet out tonight on Gruul and maintaining my rotation barring some math to show otherwirse.

As MM with hawkeye I could stay very close to the wall if not on the wall and still DPS. As BM, I generally can't, but I'll try playing with different positioning around the room. Up until grow 6 or so our healers can afford to spot heal so the DPS does not have to bandage, after that it's on ourselves to bandage/healthstone. Beyond that, it's my goal to never take damage anyways, but I can't always help mages blinking near me or people taking a nice walk in the woods towards me. As an aside, the numbers I posted were kills around the 11 growth range.

I'm going to assume you use TBW as soon as it's up each time (just for the sake of making calculating easier). For 18 seconds out of every 120 seconds, you do 10% more damage. Distributed for time, that's 1.5% additional damage. A 5th point in Mortal shots means that x% (x is your crit chance) of your shots do 100*(2.30-2.24)/2.30=2.6% additional damage. Using the equality 2.6*x>=1.5, the break-even point is x~0.58, or 58% crit. I'm gonna make a stretch and assume you have less than 58% crit - for sustained dps, 5/5 Mortal Shots is probably not superior to TBW (keep in mind the assumption that you're using TBW every single time it's up without delay). Also, keep in mind that stacking trinkets + TBW will yield far superior damage to trinkets + 6% additional crit damage.

What do you mean? I thought everyone has 63 talent points. Yeah, I occasionally make silly mistakes, but you get the picture. Hope that helps.

After some abyssmal Vashj attempts leading up to our kill last weekend you can do the bats without Hawk Eye. Just position yourself to the west and be sure to catch them as they come in. Have all of your hunters just toss their shots at them as they come in and they don't get far at all.

I'd personally drop the point in Beastial Swiftness and 1 point from Frenzy and put them into Animal handler, it's the only way to increase pet +hit and gaining 4% hit against level 73's will be a nice DPS increase. 4 Points in Frenzy will be enough to keep it up almost 100% of the time.

It looks like you have your damage model from the post above.

I understand wanting to spec in a way that's best for the guild, but wouldn't it be remiss not to suggest to them that Hawk Eye is not necessary to kill the bats, and in fact, skipping it allows you a higher dps spec for not only the bats, but also Vashj and whichever adds you handle?

As for them not budging, I suspect that if they've got the best interest of the raid in mind, seeing posts from multiple hunters here on these forums that show the bats can be killed without hawk eye will likely change their minds.

The problem with pet crit is Kill Command. KC triggers FI/frezy as well when it crits and the crit coming from KC depends on your crit rate as a hunter so it makes the whole calculation very messy. I'm working on a FI/frenzy uptime calculator which includes KC crits as well, don't know when I'll finish it. Tonight I just tested the FI uptime on blasted lands mob. In 2 minutes of shooting FI was up 117 seconds. Pet crit chances were 33% KC crit, 8% LB crit, 17.9% melee crit.

Assume pet crits once and enters FI for 10 seconds and frenzy for 8 seconds. In 10 second time the first 8 seconds pet has 0.98 attack speed with frenzy, that means he outputs 8 white attacks in 8 second time and then at least one more until FI fades away, that's 9 white attacks. Pets white crit chance is exactly 15%, so out of those 9 attacks, the chance of none of them critting is 0.85^9 = 0.23. Then there's lightning breaths. In 10 seconds, pet outputs 4.34 many LB's (2.3 average time between LB). LB crit chance is around 11%, so the chance none of those 4 LB's crit is 0.89^4 = 0.627. Then there's kill command. Assuming you use 2 times KC in 10 seconds, KC's crit chance is 35%, so the chance of 2 KC's not critting is 0.65^2=0.4225. When you take the product of all of them

0.23*0.627*0.4225=0.06. That's less than 1 in a thousand. So once the pet goes into FI, the chance of ever getting out is extremely low.

Emphasis mine. Do you have a source for this? I've seen nothing to indicate that KC uses the hunter's crit rate. Are you forgetting the talent that increases KC's crit rate by 20%?

When I read it I thought:
For a Kill Command to crit, it must first be used.
For a Kill Command to be used, the Hunter must crit.
Thus, how often Kill Command crits (and is used) depends on the Hunter's crit rate.
Reading it again, however, does leave me with doubts, though...

hmm steady-auto sequence + scorpid = win as bm :o

I usually have 300-500 ms, which wasn't that bad when I was MM. Now as BM I find my latency a lot more noticable. The biggest problem I have seems to be KC. I tried to macro it with steady, then arcane and multi, and got pretty bad delays. Now I manually do it, but an extra button to press + lag really messes up my rotations.

For the people who have high latency, how do you work around it as BM?




Also as a bonus question, do you think Wolfslayer gun or even Don Santo's gun is better than Sunfury bow? The guns have a lot less dead time, not to mention DS gun proc adds another level to scorpid poison madness.

That's the line of thinking you have to use to find the average time between kill commands. This work isn't mine (you can find it in this thread on tkasomething http://forums.tkasomething.com/viewtopic.php?t=4231) but I'll put it here, since we're discussing it now:
Another thing to keep in mind is that the number of KCs inside a single FI/Frenzy buff duration (10 or 8 seconds, respectively) can be either 0, 1, or 2. You either get no hunter crits to allow for KCs or you get at least one crit; depending on the cooldown state of KC and if you get another crit, you may or may not get a second KC in. This greatly simplifies calculating uptime since you can't do partial KCs - the number of KCs must be an integer; because of this you can establish upper and lower bounds. Furthermore, it is very difficult (nearly impossible without simulation, I'd estimate) to know what the cooldown state of KC is, on average, at the start of every FI/Frenzy duration.

Similarly, you can estimate how many non-kc specials by estimating focus regen (which we know how to do) and assuming that your pet will be spamming abilities. A simple focus-accounting approach should give you the number of special attacks that can crit in an average span of time with an arbitrary length.

I realize that using time-averaged methods isn't exactly everyone's favorite approach, but distributing for time greatly simplifies the math and, over sufficiently long periods of time, time-averaged models agree with reality fairly well.

Ahh never mind, it was edited and fixed allready.