New patch notes update:

Guess we can knock this spec on the head and call it done.

The coefficients used for Daggers are different than Swords and Maces and Fists for Shiv.

Not that it matters anyways, since they basically nerfed Shiv into a hole.

But, until 3.1 hits Live, CombatShiv and AssShiv are still the ways to go if you can't find a good HaT group.

For anyone who is still interested in this spec and are looking for a solid WWS to compare off of:

Wow Web Stats

With the news to shiv, this isnt worth as much as maybe a couple weeks ago but none the less thought it might help someone. I was RNG shafted on berserking buff but none the less was pretty happy with the dps.

There's a good chance you aren't getting "RNG shafted" but "mechanic shafted" as I pointed out in my 2 runs on the last page. Shiv is going to proc PPM mechanics far less often than SS (assuming you're using the appropriate weapons). I'm interested to know how you did as SS combat before this run. Your kill time was substantially faster so comparing direct numbers to mine isn't terribly useful, but did you find an overall increase going to combat shiv? WWS reports of "all bosses" for both specs would be my first and probably most accurate place to check this. I know everyone just assumes the sheet gets it right and that shiv is better, but real-world results are worth a heck of a lot more, imo.

With the 0.4.7 sheet? Perhaps it's because I compare with BiS (although for my own gear I have similar results).
BiS IIV gives 6117 vs Assassination Shiv's 5909.

Not sure if this is helpful or only a gimmick presentation of data, but I have made a graphical representation of Deadly Poison procs.
As far as I know this is kind of new, at least I cannot remember having seen this before. Although I admit I wasn't following the more theoretical and mathemetical discussions on poison proc chances / envenom cycles at the time they were around, so if this is redundant data, please bear with me.

Also, I wasn't sure which thread would be the most adequate for this. I chose this one, just because 'PvE Poison DPS' had the most correlation with what I was doing. Didn't want to open a new thread for this, and since the Shiv part of this thread may die out now sooner or later, focussing on the other part of the title seemed reasonable.


So what I did, I wrote an addon (or better, tried to with my very limited knowledge of LUA) that catches my Deadly Poison procs on a target dummy. I made use of the UnitDebuff API, which also returns a value for expiration time. This approach let me catch refreshes on 5 stacks, as the expiration timer of the debuff then is greater than it was before.

For the data I was attacking a boss level training dummy in Dalaran, completely naked with only two [Webbed Death] equipped, and with my talents being 0/0/0, for an duration of one hour. Deadly Poison was only applied to my off hand.
After I collected the data, I wrote a PHP script that would parse the LUA code and generate an HTML file with the evaluated data and a graphical representation of both only events (procs, drops), as well as a time line with a resolution of 1 second which shows procs, drops, times when there is no poison present and time when poison is still on the target.


Caveats and Notes:

• As I was completely naked besides my two daggers, and without any talents, I had only 42 hit rating and thus encountered reduced poison application through reduced white hits (cannot proc if there's no hit) and spell misses.
• I didn't use any special attacks. My opener was a Garotte, which shouldn't proc the poison because it was only on my off hand.
• As I did not use any specials, I also didn't use Slice and Dice. So my weapon speed was exactly 1.4.
• I did this test with the last PTR build (not the one currently on the server). Things may have changed since then, and also I am not sure how the latency played into the results as I had no addons enabled besides my data collector.
• There may have been other people attacking the target dummy. I wasn't watching for most of the time. Shouldn't skew the results by too much though.
• Not sure why it shows 3613 seconds instead of 3600. I stopped attacking right when my poison went to 0s. You may want to ignore the first 13 seconds for further analysis.

Following are the statistics for my data, and the full HTML file can be found here:

DeadlyStacks (beware, 625kb HTML file)

Time in seconds:	3613
Number of Events:	543
Number of Procs:	449
Number of Drops:	94
Uptime seconds:	2484.992
Downtime seconds:	1128
Uptime %:	68.78%
Min Time between Procs:	0.095
Max Time between Procs:	31.382

I also included a copy & paste section with the data, so that one can put the data into an Excel file if you don't liky my presentation (not sure if I like it myself) or just want to play around a bit.
I'm not the very best at analyzing statistical data, so I'll leave that to other people, if they can come up with something useful from that data.

I didn't include a legend however, so here is what the graphs mean:

• Dark green means it is a poison proc
• Light green means the poison is currently on target and ticking, but no proc occured.
• Red means the Deadly Poison stack has dropped.
• Light red means there is currently no poison on the target.
• The X-axis represents either one event each (first graph), or 1 second on the time line (second graph)
• The Y-axis represents the number of stacks currently on the target. I chose a value of 0.5 for no stack, so that you can see that pretty color.

Will be trying out Combat Shiv tomorrow in Naxx with SR / Hailstorm. Spreadsheet says I can hit 5k dps with my gear, which is ~5% higher than IP/IP Muti on Undead.

I was not aware that Shiv had reduced PPM for enchants, is this modeled by Vulajin's spreadsheet?

Why would it have a lower chance to proc weapon enchants? It wouldn't.

I am referring to post #204 in this thread.

So you're going to try to test the relationship between PPM weapon enchants(mongoose/berserking) and PPM poison enchants(IP/WP/DP)?

You know that is going to take forever and the results will still be less than definitive?

I think the reason Shiv would have a lower chance than SS to procc weapon enchants comes from the fact that the weapon will be faster, which in turn would lead to a lower % chance per hit to procc the weapon enchant.

But not when you account for instants, which was the point he was getting at. A slower weapon will have better enchant proc uptime given a reasonably similar number of specials coming from it.

Without doing the exact math to pinpoint the offset, remember that you will be performing more shivs due to both the lower energy cost and the lack of the possiblity to proc extra combo points.

Shiv (with a 1.5 speed weapon) is only 5 energy cheaper than talented SS. That's not good enough to offset the difference in proc chance.

Also, finishers (except SND) can proc enchants as well, so the fact that glyphed SS leads to faster combo point generation is not in any way detrimental to weapon enchant proc rates (in fact, it's the opposite because finishers cost less energy).

Tried out Combat Shiv with WD/WP + Widows Fury/IP in Naxx this week. Easily seemed to outperform WD/WD Mut for me, and with my guilds lack of hunters or anything that would build a decent HAT group, outperformed that as well.

Combat Shiv Patchwerk (5253 DPS)
HAT Patchwerk (5019 DPS, Shitty group comp!)

I am a little disappointed with the DPS, as I've seen plenty of numbers from 5500-5900 (such as in post #203). Of course, I wasn't using Hailstorm for my OH, which would give me a slight boost, but other than that, the buffers and debuffers don't seem too far off from each other without seeing his correct armory for the build, aside from me forgetting to pop a haste potion, and him having Mirror of Truth, VS my FOTFF. Anyone else notice something I haven't been doing right?

will this built work with MH [Calamity's Grasp] and OH [Hailstorm]. I do have [Webbed Death] x 2 as a Mutilate. Will [Webbed Death] as a MH will be better than MH [Calamity's Grasp]?