I actually thought about doing such a thing, but such a model requires a little sophistication beyond what's stated.

First, instead of tracking a straight disruption fraction, you need to explicitly track "n seconds of interruption every m seconds", as you don't actually start losing yellow damage the instant you are disrupted, while you do start losing white damage. For instance, if you experience a 2 second interruption every 10 seconds, you never lose any yellow damage; but if you experience a 10 second interruption every 50 seconds, your energy caps out at you do lose some yellow damage. Hence, while same have the same loss of white damage, the second will additionally lose yellow damage.

Second, you need to think about how this effects abilities with cooldowns, hidden or otherwise; As long as the interruption is (mostly) predictable and doesn't occur more than once every 20 seconds, you don't lose any Blade Flurry or Adrenaline Rush damage from it. Procs with internal cooldowns are a little trickier (since they do cool during interruptions, but you also risk losing uptime to the disruption). So you need to think a little bit about modeling that.

Additionally, since your ratio of white to yellow damage is changing, your cycle may change as well; hence, this needs to be factored in as well. It's actually this last point that's prevented me from adding such a feature to the Rogue Gear sheet; cycle computation becomes an absolute mess. If I ever get around to the programming-language implementation of this, I might put such a thing in; but in a spreadsheet I found it too messy to model with any accuracy.

Okay, so it's slightly more complicated than that, but it might be a nice future feature. It's not very high at all on my list (and thus may never happen), but I still think it would give a decent approximation for some fights. Like you said, you adjust to the circumstances. Given that most cycles give pretty similar DPS anyway, if one assumes that a good rogue is going to make smart choices with their adjustments (and thus achieve similar cycle DPS to their main cycle up to and immediately following a disruption), it still might work.

Of course the other way of dealing with it could just be a checkbox that's % DPS expected. If solid rogues are only realizing 80% of theoretical DPS because of the disruptions of a given fight, you could enter 80% and get an approximation without pulling out a calculator and dividing it yourself. I guess that could be the low-tech uncomplicated method of modeling it.

Very interesting, I think adopting cycle to boss is more being a good player than good rogue.

I also interested in mechanics how latency and fps disrupting normal cycles

Is Distract influenced by hit rating?
Currently distracting some bosses arrival with my fellow rogues in mount Hyjal. I am curious what mechanics are behind the bosses resistance to distract.

So recently, my guild and I killed Azgalor, and after the reset, as we were clearing through BT, the Syphon of the Nathrezim dropped. Being a mace rogue, I naturally bid and won, but was still unsure of how much of an upgrade this was from my current Dragonstrike. After doing testing on Archimonde, I've found that I usually do better with the Dragonstrike, but as my hit rating went up, my DPS went down, which brings me to my next point. I was doing some thinking, and wondering if there was a point of inflection in crit's favor over hit for Mace spec. So in general, which do you guys think would do me better for a weapon, and if you know, what would happen to be the more specific combat mechanics, like the roll system, and if there is, at which point crit becomes better than hit for a Mace rogue?

What do you mean by this? If you're adding straight hit rating and keeping everything else constant, there's absolutely no way your DPS is going down. So by that I have to assume you are sacrificing some other stats to get hit. In order to evaluate what's happening and address your theory about a point of inflection (which gut instinct I would say won't happen until the hit cap), you have to mention what kind of stats you're losing for every point of hit that's causing your dps to go down. For instance, if you're giving up 100 AP to get 5 hit, yeah your dps is going to go down, because that's a terrible trade.

My stats stayed fairly the same...maybe a small loss of AP, and crit...more a pure crit for hit trade though, statwise. The numbers exactly are: 1st Set:1867 or 1795 Ap, 250 hit, 26.49 or 25.04% crit (Depending on Hourglass of the Unraveller, or Bloodlust Brooch) and 12 expertise; and the 2nd set is: 1885 or 1813 AP, 222 hit, 28.52 or 27.07% crit (Again depending on HotU or BB), and 12 expertise. This is all while keeping the T4 2/5 bonus. And of course, my hit could've gone alot higher, but it would've been a gimp as you've stated.

EDIT: Forgot to mention both of those statistics are assuming Syphon, not Dragonstrike.

I've read the first post numerous times...but my question was more aimed at delving deeper into the actual combat mechanics: does the server calculate hit and crit at the same time, or hit then crit? I've used the spreadsheet from a post in this forum, and found that alot of the time, it's just what it says it is: theoretical DPS. As for the Dragonstrike v Syphon, I've noticed a very slight DPS using the Dragonstrike on Archimonde, but as has been stated before...terrible test environment.

Re: "How the server calculates stuff" - Attack table - WoWWiki - Your guide to the World of Warcraft

Re: "Archimonde fight shows this is better!" - Between Airburst and Fear.. this is probably one of the worst fights to judge damage on. At least Akama is always the same length of time you spend attacking him (within a few sec).

Between your two sets of gear, net difference is:
+18 AP
-28 Hit
+2.03% crit (~45 crit rating)

So um.. yea, 45 crit rating and 18 AP is definately better than 28 hit. Hell, the 2% crit > 2% hit regardless of the AP (and you get slightly more than 2% crit, slightly less than 2% hit).

Either way - if you're on Archimonde.. you'll likely be upgrading your T4-2pc and Hourglass/Brooch soon.




In a more general sense, the difference between hit and crit is a bit.. strange for rogues. Crit has the noted advantage of applying to both autoattacks and specials. For specials it generally means 2.3x as much dmg (Lethality).. possiblt an extra 3% more for RED meta. However, more crits mean less hits, as described in the Attack Table linked above. This means no extra landed attacks. Hit on the other hand converts a miss to a hit. This means a smaller "per-attack" dmg increase (going from 0 to 100 dmg for a hit, 100 to 203 for a crit autoattack, 100 to 236 for a special), but you have more landed attacks. This means more chances to proc: Mongoose, Combat Potency, Sword Spec, Windfury (proc'ing an extra attack), Windfury attack (the actual extra attack, which can miss), Dragonspine, etc. Now.. as a mace rogue, you don't have sword spec.. but the rest should still apply.

In terms of "point of inflection".. its actually the reverse of what you may think. Hit scales with it self in a positive way. The more hit you have, the better it is to get more hit. This is largely due to WF. The more hit you have, the more likely you will proc WF attacks.. and thus want even more hit.

Lets look at a standard attack table for roughly your gear: (played with the numbers a bit to make things easier)
5% miss
3.25% dodge
25% glance
26.75% crit
40% hit
This works out to: (for maces)
(5 * 0) + (3.25 * 0) + (25 * 0.75) + (26.75 * 2.05) + (40 * 1) = 113.6
Adding 1% hit or 1% crit both increase your total dmg nearly the same.. slightly more for crit due to the modifiers but nothing significant. Adding 1% crit turns a hit into a crit.. so you turn a 1.00 into a 2.05. Adding 1% hit turns a 0 into a 1.00.

Say you had some really crit-heavy gear.. (note.. this is not *too* far off from my gear)
5% miss
4% dodge
25% glance
50% crit
16% hit

(5 * 0) + (4 * 0) + (25 * 0.75) + (50 * 2.05) + (16 * 1) = 137.25
Adding 1% crit moves you to 138.3. This is a 0.76% increase in damage. Adding hit will increase your damage similarily. The added bonus of hit, is that you gain more procs too.


When talking about sword spec and WF..
Say you had:
15% miss
6% dodge
25% glance
40% crit
14% hit
You only are landing 79% of your attacks. This means 79% of your attacks have a chance to proc WF/SS.. which in turn has only a 79% chance of landing itself. This results in a 0.79 ^ 2 chance (times the 5 or 20% for WF/SS to proc).. roughly a 62% chance of landing a WF or SS attack.

If you increase your hit and expertise a bit:
2% miss
3% dodge
25% glance
70% crit + hit
You end up with 95% of your attacks landing (16% higher than the above example). This turns into 0.95 ^ 2 landed WF/SS attacks.. or a 90% chance to land WF/SS. Thats 28% better than before, with an increase of only 16% chance to land the attack.

Hit scales very nicely with WF/SS.. just always adds the same damage increase point after point. Going from 1% to 2% crit nets you the same raw damage increase as going from 50 to 51%. This is a similar effect to AP, which does not scale with itself at all.



All this math an theorycraft aside - you can pretty much just trust the spreadsheet :P. In your gear tradeoff.. you got a lot more crit for a small hit loss. Use the APEP numbers provided to figure out what is an "acceptable" tradeoff in terms of hit vs crit. In general, 1 crit RATING < 1 hit RATING. However, 1% hit < 1% crit. You tried to compare less than 2% hit to more than 2% crit.. and added in some AP to the crit side.

Thank you very much for the more in depth explanation...as you pointed out I can now see that it was also not an even comparison. Off on a bit of a tangent...is it just me or has anyone else had horrible luck with trinket drops? =P. Took me almost half a year to get a Dragonspine.

Beyond that though, it seems as though you've traded 18 AP and ~2 % crit (lets call it 44 crit rating because it's close enough), for 28 hit rating.

Using the EP values from post 1 for swords (which will bias hit to be even better than it is for you), we can see that you traded 95 EP in AP and Crit for 61.88 in hit.

Hit is a good stat, and better point for point than anything else - but other stats are still worth something too, and trading away too much crit or AP or DPS or Stam (Glass cannons die. Dying sucks.) for hit is a bad deal.