Siq

Thx for the detailed description of this problem.
Well, it was my fault, wrote "L" instead of "M" in one field.

http://www.savefile.com/files/669861

This is Zoro's version, just with this little bug fix.
Now the dps change should be as in bleedo's version.

p.s.: With 2nb & dagger combat, it might be better to use 5s/3r instead of 3s/5s/5r (5s/3r I have only in my personal 2.2.4c version, so its not in the actual spreadsheet).


Btw, does anyone has any infos about the actual proc rate of http://www.wowhead.com/?item=32410?
If the proc rate isn't too low, this gem could be quite nice for our class.

 

Dpsxpres

Long time lurker here. First time I've felt I had anything to actually add to the conversation.

Zoro (and Siq), thanks for keeping the spreadsheet alive. I think Cel is completly accurate that those who don't like the links are the decidedly minority. Since Pf has retired spreadsheet updates from the comunity are naturally going to be a little muddled and random. But this option is far preferred to just letting the project die. I think it might be for the overall best (even if it's more confusing) that now the comunity as a whole are responsible for updates, rather than one guru who we (as Zoro implies) expect to do the massive work needed to keep everything updated. Waiting for Bleedo to do everything might be expecting more from him than he has the capacity to give (due to rl time constraints, or whatever).

It would be nice if updates were less spammy and more centralized (like when Pf was the only one making updates), but you can't always have your cake and eat it too.

 

Melnor

Yea I can see where you're coming from as well.

 

Xizenta

Would an immune count as a fail to hit? I believe so, but wonder if anyone here thinks otherwise.

Just thinking about whether 2.1 will break 2pc abusers (assassination.)

 

pf

The only way for updates to become more centralized if one or two people decide to fully pick up the project. Anyway, the spreadsheet as it is now is in pretty good shape. I have very few problems with formulas and most of the stuff that needs to be done is mostly minor tweaking and adding new items/effects. Anyway, if Zoro/siq/whoever would be up to that job (as a collaborated or solo endeavor ), they should hook up with Bleedo and get the web info as well as the bugtracking info.

Anyway, the only real suggestion I have for improving the sheet now that slack cycle damage has been improved is that the windfury formula is still not spot on.
Windfury formula problems:
Does not include hidden cooldown (is there one for the totem?)
Only increases AP for extra attack, but actual buff lasts 1.5 seconds.
WF procs do not proc other procs (eg sword spec procs) in sheet (can this happen in game?)

Problems faced by fixing up the WF formula:
What is the best way to predict how many attacks are affected by WF AP buff?
How to make WF/Sword spec procs proc eachother without circular references?
How to best apply the changes to the next-stat sheet. Changing to extra AP calculations and implementing the hidden cooldown are not currently supported in next-stat. One would have to alter the WF formula in Next-stat to work with this change.

The above was never that much of a priority for me when I was running the sheet because I never had a clue how WF truely worked and so I did not even know the formula was bad. I do know that WF is very nice for raiding from my one experience with it on Thaddius as a 60 tagging along on a 70s naxx run. Thaddius was the only time during the naxx raid that I had anything other than battleshout and it showed as I ended up #5 in dps right behind the 2 other rogues in my group (consumables would have put me on par or ahead of them), gogo 37% unbuffed crit/20% hit + lvl 70 buffs + some luck! On patchwerk, I had a prot warrior who had BS up 50% of the time, and I finished with the alt 70 rogues who were undergeared and dont know what a cycle is.

http://thottbot.com/test/s39958 240 haste rating for 6 seconds.

No listed procrate, which could mean a PPM method. Also, depending on procrate, it might have an invisible cooldown. Not much info about it.

 

rj

Yes, I as well was comparing my buffed dps numbers. But even my unbuffed DPS is higher with the 4/4 gems as compared to +16 AP gems. Less than 1 dps higher per gem I swap, but still higher for 4/4.

It could be because my daggers are quite a bit better than Melnor's, I guess. I am not sure.

 

Triton

For what it's worth, I needed to socket a Girdle of Treachery and two +16 AP gems came out with higher DPS for me on the spreadsheet than two 4/4 or two +8 agility, both buffed and unbuffed. I only have around 200 +hit.

 

Gort

Can't speak to how to model, but I can say that to the best of my knowledge:

- There is no cooldown on totem procs.
- Procs will chain proc. (That is, yes, you can WF->SSpec->WF->SSpec. It HURTS when it happens on your first swing of the fight.)

 

rj

It could be the version of spreadsheet doing this. I was using 2.2.4b (no clue's who's revision that is.. heh).

 

Andeh

Pf, I can say from experience that WF & Sword Spec can chain proc off each other. I don't have a screenshot, but every now and then I get them, usually WF -> SS -> WF.

 

pf

are you sure its WF->SS->WF and not white->SS->WF with WF proccing from the first white? Without a detailed parsing of logs, there really isnt a good way to tell by experience because you have no clue whats proccing what.

 

Andeh

I typically run with my actual combat log hidden, and only notice this when all of the sudden SCT spams "Extra Attack!" three times right on top of each other. That's why I assume that it some combination of WF & SS procs. I'm not discounting the possibility that it could even be something like white procs SS procs WF, then a split second later offhand hits and procs SS. I can't remember seeing more than 3 back-to-back "Extra Attack" messages, however. My gut feeling is that since its a 20% chance to proc WF, and only a 5% chance to proc SS, the chain I theorized is most likely to occur, but still remain uncommon.

 

Melnor

Possibly. But what can I say, I've done enough Kara runs to probably be double exalted and haven't seen any daggers yet. Should have shanker in a couple weeks tho. And the feltooth isn't that bad

 

pf

I've had plenty of 3-5 chain procs occur which all started from one swing. This was with only sword spec. You cannot assume they proc off of one another without actually doing some statistics on it since there is no way to determine in the combat log what procced what.

 

Xizenta

I would assume that a sword spec proc would be able to proc a windfury, since it is in every way a normal attack. All it does is break your current main hand delay. I'm also led to believe that windfury has at least some form of hidden cooldown, since it never has procced off of itself (from what I have seen) even though it is also noted normally like any other swing in the combat log.

 

Cel

I've had SCT kick back multiple windfury EXTRA ATTACK messages simultaneously before, and I'm combat daggers. This leads me to believe that if you land a Backstab or some instant at nearly the same time an autoattack lands, they both can fire off a Windfury, thereby showing there is no cooldown, unless I am remembering poorly. I have no real support other than observation, though.

"There is much pleasure to be gained from useless knowledge." - Bertrand Russell

Alpha is recruiting... go go.

 

Kytrarewn

OK, fucked around with the spreadsheet because I had some free time.

Your "non-consumable, but raid-buffed" DPS is 1243.95

With Talon of Azshara (Still the best raid-dropped one-hander, I believe?) and Latro's (same enchants elswehere), and your buddy's spec, your "non-consumable but raid-buffed" DPS would be 1243.97.

Your buddy, with his gear (Going off memory on the weapons since he has the big iron fishing pole equipped now, I think it was Spiteblade(mongoose)/Fireguard(15agi)) has a raid-buffed, but non-consumable, DPS of 1042.61. With Blazeguard(15agi), which I think it also might have been given that I remember his offhand having a significantly better DPS than yours, it would be 1049.95. If he were to get Latro's OH, it would be 1053.54.

The one place that a bit of error crept in was that the spreadsheet doesn't have Bloodfang Belt any more, so I tossed in Shifting Sash of Midnight instead, which, off the top of my head, is likely a little bit better, but not a game-breaker.

As a side note: While the spreadsheet doesn't calculate Ruthlessness, and, as such, Murder comes out ahead by all of 20DPS, if you were to respec, I'd look into picking that up instead.

So neither of you were quite living up to your overall "Patchwerk Potential", but given the nature of the Mag fight add-wise, you were doing pretty respectably, and the overall difference between your DPSes was about what would be expected.

And, were you to get the Morogrim sword and respec, your overall DPS would go up by 0.2pts.

I'd probably save DKP for the Vashj MH dagger and/or arena points for the Gladiator's OH. It would provide a significantly bigger DPS boost than respeccing.

 

Cloak

Sword's extra hit procc'ing off itself got done away with like 7 patches ago. So if you're getting 3 extra swings instantly, there needs to be another instant attack to propagate it.

The more you know, the less you understand.

 

Melnor

http://forums.worldofwarcraft.com/th...eNo=2&sid=1#27

In light of this new information, how much of an impact is going to have on the spreadsheet formulas?

 

• Aldriana

I'm not sure what's actually in the spreadsheet, but assuming it's a simple ratio - not much. If I recall my earlier calculations correctly (I don't have them in front of me), the intercept for rogues is pretty close to 0 - I think I came up with an agi-to-crit formula of something like crit = agility/40 - .29% A fudge factor of .29% shouldn't change a whole lot.

 

pf

This change never made live, it didn't even last a full week on the PTR that patch. Sword specs most definitely still do proc off of eachother and always have. If you do not believe me, try to find the patch notes that say anything to that effect.

I determined this to be a y=mx+b formula a long time ago(I'm sure others came to this conclusion independent of me as well), and unlike hunters, I figured it was done on purpose. There is very little chance a formula like that could be going wrong in the way that it was and it be accidental. Anyway, taken right from the spreadsheet, the agility portion of crit is: (E77-11.5)/40 where E77 is agility, so rogues base crit rate is: -.2875%

 

Andeh

Windfury hasn't been able to proc itself, totem or weapon, for probably close to a year now.

I understand you're trying to be scientific about it, Pf. I am certain at least once or twice when I got chain procs I actually looked at the combat log, and saw that they were of mixed type. Again, I didn't think to screenshot it. I will edit my SCT so that WF & SS have different proc messages, and then throw a tantrum to get WF this weekend's raids. And maybe create a new combat log tab that's a bit cleaner for watching this.

BTW, according to WorldofRaids, the Consortium vendor now sells the Major Striking enchanting recipe.

 

pf

You cannot watch for it, thats why I am trying to tell you. In the log you might see:
SSproc
WFproc
SSproc
SSproc

This does not mean that WF procced SS, as we know SS can proc itself and all the SS procs could be from prior SS procs or white attacks. The only thing that can be determined from simply looking at logs is if WF has an invis cooldown keeping it from proccing itself. This would not be hard as you would just need to find a case where your windfury buff lasts more than 1.5 seconds and you have 2 procs in that time. To determine if WF can proc SS would be much more difficult, I assume that if WF uses an invis cooldown that WF can indeed proc SS, but without a statistical analysis, there is no way to prove it.

 

Cloak

I always just assumed it went live.

The more you know, the less you understand.

 

Gogge

I checked the last Gruul run and the lowest time between any "gain 1 extra attack from WF"-messages was at 1516 ms (for reference, the lowest Sword Specialization was at 875 ms), with the supposed 20% proc rate one would expect a few back-to-back procs, so it seems like there is a cooldown on the totem version too. I usually play with 200-500 ms delay so I wouldn't put much value in the time stamps in the combat log (link).

Edit: I can't remember getting any of the chain procs from Sword Spec since a long while back (1.7?), it was rather usual before (nostalgia screenshot) that but it's not happening to me anymore.