With the change to glancing blows in yesterday's patch, I was rather anxious to hop on and benchmark it all. For the most part, I'd expected a fairly decent increase in white DPS across the board. However, I was really surprised at the numbers I found.

First off, I'm playing a human rogue, 20/41 combat fists with the BBW Paw/Glad Ripper, 22% hit, 28% crit, 1600ish AP. No additional weapon skill, though I'd not expect it to make a significant difference.

We ran through Karazhan last night, from a fresh instance through Curator, then back around to Nightbane. I managed to not save the event report from DamageMeters, but both DM and Recap were reporting that approximately:

- 51% of my damage was white damage
- 34% from Sinister Strike
- 6.8% from Deadly Poison
- 7% from Envenom

I benchmarked the entire run, as well as individual fights (Moroes, Maiden) and these numbers seemed to hold very consistent, never fluctuating more than 1% or so.

Now, I was a moron and didn't benchmark my white damage pre-patch, but that 51% number seems extraordinarily low. Before BC, our rogues, myself included, were tending to run at around 65% white damage. Obviously, Combat Potency and Surprise Attacks would be expected to change some of that, with the rest of the change attributable to the glancing change, but a 15% drop after a supposed glancing fix seems rather severe.

What I'm mostly looking for are some before-and-after numbers to try to evaluate the scope and impact of the glancing changes. The Blues have made it pretty clear that they feel that the glancing change brings melee DPS back in line to where they're supposed to be, but I wasn't feeling that last night. I ended up behind an elemental shaman and a frost mage for the run, with the shaman at about 20% of the raid's damage, and myself at about 18%. I didn't have any offensive buffs beyond Might/Kings/Mark (no BS, melee totems, unleashed rage, TSA, LotP, etc), so that might have been it, but last night felt very atypical - before, I would tend to end up a slight bit above the mage, and a good bit above the shaman.

Does anyone have white damage ratios pre-2.1 for a similarly-geared rogue? Alternately, does anyone have pre- and post-2.1 white damage ratios for comparison? Do we have any kind of hard data indicating what the extent of the glancing changes are, and to what degree they've affected sustained melee damage output?

Individual boss fights would be more useful in that lower level trash (71, 72) do not cause glancing like bosses.

(On a side note, why envenom as combat fist?)

Conventional wisdom is that glancing blows will happen 40% of the time on white damage against a +3 mob/boss.

The patch notes describe the change as follows:

This implies that they are changing the percentage at which glancing blows happen, not the damage done reduction component (as weapon skill used to do).

Does your logging show what your glancing blow percentage was on white damage vs. bosses?

If they really did reduce the percentage of glancing blows then it would also have some secondary effects, like raising the soft crit cap on white damage, so I'm very interested in seeing the numbers.

Well, I specifically benchmarked Maiden, which is about as close to a rogue punching bag fight as you get in Kara. The numbers for that one fight were virtually identical to my overall numbers for the evening.

I dropped my last three points into Vile Poisons, mostly for the dispel resist for PVP, but it tends to give slightly higher damage output than Eviscerate, and the 30%ish buffed crit tends to push Envenom a decent bit ahead of Rupture in the tests I've run - enough to justify the 10 energy cost based on "gut feeling", though I haven't sat down and actually crunched the numbers out yet. Mob mitigation levels tend to make Evis pretty wimpy on anything but Aran.

Unfortuately, no, it doesn't. If anyone knows of a good glancing logger mod, I'd be happy to run with it. Alternately, I guess I could just record and parse a raw combat log.

I'll help in the testing by beating up a bunch of L70 mobs if someone can recommend a good logger for this purpose.

Type /combatlog then use the log file that it spits out into your WoW folder with WWS at lossendil.fr.

And yeah, somehow I missed that you used specific boss fights.. My fault.

The best pre-post testament I can provide is from our Morogrim fight last night. Our rogue that typical hovers around 1000-1050 dps on that fight was pumping out 1250-1300 dps, sustained over several minutes. Quite an increase in DPS I would say.

I took Imp Evis - again - for PVP purposes, primarily. I'm primarily a PVEer, but I've been doing some arena play, as well. I've considered dropping it, but that leaves me with fairly gimped PVP burst, since Envenom isn't an option. Rupture is a joke (rogue anti-vanish aside) anytime healers are involved, and XA isn't worth the investment since the rest of my team's DPS is magical.

We've been using a prot warrior tank, so sunder's up as expected. I've got SnD perma-up (which is a lot easier with the Netherblade bonus!), and I think I used Shiv all of twice the entire night. For the most part, I'm following conventional combat techniques pretty closely, poison choices excepted.

I've been debating between improved and vile, but what really makes vile attractive to me is the dispel resist. It's a lot of fun getting a paladin to sit there spamming Cleanse trying to get rid of that Wounding stack on his warrior, rather than actually healing said warrior.

To be honest, I'd be quicker to attribute this to gear buffs before the glancing change. Raw DPS numbers don't mean a lot due to other factors that changed, which is why I'm mostly interested in white:yellow damage ratios. A 25% damage upgrade from glancing changes alone is a bit suspect.

Glancing blows don't happen against same level mobs.

Really, what needs to be done is a large data sample of white damage attacks on something flagged as a boss. Gruul and Aran wouldn't be bad for single fight samples. Patchwerk would be pretty decent as well, if you happen to be in Naxx for some reason. A comprehensive log of white damage vs all the bosses in a full clear of Karazhan would also work. I was planning on having the melee in my guild export their combat logs from our next gruul fight; parsing those for the relevant data would be quite easy.

We have Gruul tonight, assuming we don't decide to do Mag instead due to the Krosh bug. I'll do what I can to capture and upload my combatlog. If we can just start collecting data points, I think we could build a decent overall picture.

http://www.juggernautguild.com/wws/2...ory/index.html

From the 3 logs last night, it's pretty obvious that the new glancing blow rate is around 25%, and the reduction in damage also hovers around 25%.
The deviance on Lurker is due to his adds.

I just beat up L70 Demon Hunter Initiates on the top of the BE pyramid in SMV for an hour or so. Parsed with Kombat Stats. 350 skill, two axes.

Glancing rate is right at 5% (35/668 = 0.0524).

Glancing damage reduction is more tricky to get a hold on because I was dual wielding. The best estimate I have is Max/Max, because I'm assured that they're both from main hand and with max AP (UR esp). Max Glance/Max Hit is 419/480 = 0.873. (This calc checks when using crits: Max Crit/Max hit = 954/480 = 1.99.)

So unless there's something funny about these mobs*, an estimate for equal level mobs is 5% glance rate for 13% damage reduction.




* High Defense skill or high armor might throw this off, I guess. These mobs dual wield, so I doubt they have high armor, at least.

Our top dps rogue basically destroyed everyone last night. Did something like 1100-1200 dps on tidewalker. Said it felt like his dps increased 200-300.

I'll echo what somebody said above -- last night's increase is attributable to multiple sources. Our weapon DPS got buffed, as did our stats, as well as the glancing blows reduction.

Ok, so assuming a change from a 40% glance rate @ 70% damage (88% of optimal white DPS efficiency) to a 25% glance rate @ 75% damage (93.75% of optimal white DPS), we can expect to see a 5.75% increase in white DPS.

Grimx's numbers seem to show melee accounting for 50-60% of each rogue's overall damage (and pegs me at the low end of that scale - ouch), so does it sound reasonable to assume that rogues got a 2.9%-3.5% overall damage increase, not considering things like the crit cap (which isn't an issue from behind) or rage generation (not applicable for rogues)?

Today in Blood Furnace heroic I did around 30% of the damage while a frostmage, that usually did a bit more than me came to 24%. Item changes were Deathblow Googles instead of Gnomish Battle googles and Romulos Poison Vial instead of Drake Fang Talisman.

In Karazhan afterwords me and another rogue were also on 1 and 2.

So no numbers, but from the feeling I got a huge damage boost.

As a DPS warrior I went up around 250-300 DPS on Void Reaver.

Maintained a wicked 2600dps on Solarian untill she re-emerged and killed me(Still learning the fight), a hefty 600dps increase. The flurry nerf is still a pain though...

One of our rogues actually pulled aggro on Magtheridon last night, on both a wipe attempt due to half the raid crashing due to a spike, and then on the kill. The tank had a good ~20-30 second head start as we finished off the last add and then dealt with the knockback. For the kill attempt, the rogue got a battle rez and topped the DPS meters, the first time a rogue has done so on that fight for us.

Our enhancement shammy also seemed to have gained a solid DPS improvement, as did our mages, to a lesser extent. Warlock/Spriest DPS stayed roughly where it was from what I could tell, with the 5% damage buff and the gear improvements canceling each other out.

Looks like the DPS hierarchy has been shifted a bit, and rightly so.

BM pet DPS seems up 30 dps or so, MM about 18, before new KC damage is added in

Our class positions mostly stayed the same, things just died faster.

Gruul tonight: http://www.lossendil.com/wws/?report=ohofxssttk6wi&m

Looks to support the 25%/75% argument.