I got two quick questions

I'm a bit confused.

I remember having my recount running (latest version) at our latest Patchwerk kill and at the end of the fight I was sitting at 5300'ish dps.

However looking at the WWS stats I'm clocking in at 5900. What causes this large discrepancy? Is recount really THAT inaccurate that it can put me over 600 dps off? What could be the reason for that difference? Is WWS counting something that recount doesn't? :huh:

/boggle

(link to wws: Wow Web Stats)

And which of these belts are the better option:

[Girdle of Razuvious]

[Girdle of Chivalry]

Assuming I'm under the hit and expertise cap with both.

Armory link here: The World of Warcraft Armory

Cheers

Short answer: yes. Longer answer: recount has some different rules about calculating DPS, so a very specific set of circumstances could lead to very different results (recount is more likely to show higher dps if your RNG is better at the end of a fight). Also, as mentioned above, websites like WWS and WMO filter damage to the boss only.

[Girdle of Chivalry] IMO (especially in 3.1), but really these kinds of questions can be more easily solved using Rawr or the spreadsheets, and you should really consider getting one or the other.

Does weapon speed make a difference for prot anymore?

Specifically, I'm wondering if Broken Promise is worth getting (Last Laugh is not an option).

Weapon speed isn't nearly as important as it used to be. If you need more mitigation get a "tanking" weapon. If you need more threat, go for a dps weapon.

Most DPS weapons will actually be relatively low in TPS for a paladin as I'm pretty sure none of them have str on them. If you are looking for a threat weapon, look for the highest Base Weapon DPS and highest STR. Last Laugh will be much more threat than any other 1 hander.

LL is definitely best, but I think Split Greathammer and perhaps one or two others ever so slightly edge out BP. Not by enough to make them worth using though imo.

Frankly, the biggest difference weapon speed makes for prot, is that slow weapons will cause less parry haste from bosses. Personally I rate that quite highly and tend to prefer a slow weapon, but even with an over-heavy weighting towards slowness (my spreadsheet's currently set up for an almost pure swing-damage boss, and isn't rating TPS highly right now), I've got Last Laugh as #2 in slot (behind Broken Promise).

Parry haste is an increase in damage taken, if you ignore it then you're undervaluing certain stats, and you're potentially making yourself worse. As a paladin parry haste accounts anything from a 2-4.5% increase in swing damage taken (2% with a decent whack of expertise and 2.6 speed, 4.5% with no exp and 1.6). This is significant, though not devastating in any encounter which currently exists, and even taking this into account avoidance stats are a much better way to increase your mitigation against swing damage. As I stated, My current high valuation of slowness is based on a setup which is highly biased towards swing damage (It's assuming 95% damage taken from 10k swings every 2s, that's a pretty dull encounter for a tank and not something that's going to stress one) and with very little attention paid towards TPS (which scales up significantly with ilvl). This is based on my current prot gear, which is probably not quite good enough for naxx and just needs some straight-up mitigation before I can take it there. In this setup, I rate the difference between a 1.6 and 1.7 speed at about 10 dodge rating, this is about halved if you compare 2.5 to 2.6. When my gear is better, it'll be less.

Anyhow, the topic has drifted somewhat. The context of my bringing up parry haste was "how important is weapon speed" and the answer is something along the lines of "as important as mitigating parry haste", a logical corollary of that is if parry haste doesn't matter then weapon speed doesn't matter. Two exceptions, (1) Faster weapon speed = SoV up to 5 faster, (2) If you care about DPS and have reckoning, slower weapons => more reckoning damage up to a hasted swing speed of 2.0.

Thanks for the reply Nobby. I've logged out in my Ret gear again today, been forgetting to change out of the prot set.

I've done some more testing lately, including "cheating" by spamming HoW on the 1 HP dummy. I can peak out at close to 3k (not using the HoW glyph atm), but when i go back to using the normal Judge/Cons/CS/DS rotation I'm back in the 2300 range. What's odd is that another Ret in my guild gets about the same as I do on the dummy (slightly better gear), maybe a bit lower, but almost doubles that number in Naxx 25.

Is there something about testing on the target dummy that may mistakenly give a lower-than-expected number? I know with raid buffs we can expect a pretty decent increase, but am beginning to think that the static 2500dps target may treat Ret unfairly, just as it would a HaT rogue.

The World of Warcraft Armory

Thanks.

Unless they corrected it, test dummies do not display DoT damage properly when they have 1hp. They only display taking 1 damage, rather than the full DoT.

RV (Righteous Vengeance) is a DoT which does 40% (on live, 30% in PTR) of a crit Judgement or Divine Storm over 8 seconds time.

If that does not display properly, Recount or WWS will not record it properly. Therefore your DPS is lower than true.

Try to hit the dummy at an unusual hour when it (hopefully) has more than 1 hp in order to get more accurate statistics.

Exemplar - I'm pretty sure Recount is picking up RV damage, but I'll double-check. IIRC, it was in the range of about 6-8%.