Cos-

How is that sheet supposed to know how much time you spend running around mid fight. Do you spend the same amount of time running from Leo as from Solarian? etc.

You need people who can get out of the fire. Teron's ghosts, doomfire, flamestrike, fatal attraction, etc. All way more of a threat then someone who enjoys a terrible spec.

 

Iluat

Towelette I've read the first posts of this thread, but I haven't found what I'm seeking. I now that there's the way to have the EP from gear stats like agility, hit, crit, etc, but I'm speaking about the number of EP given by the weapon's dps.
If anyone knows the formula, something like "1 dps = 2EP" I happy to know it.
Thk anyway Towelette.

 

Khazrak

I join in on this question, was wondering the exact same thing..

 

Ricard

I believe Aldriana's spreadsheet will do this for you, if you open up the DamageCalcs sheet. It would be under the MHDamage or OHDamage columns.

 

Havenwood

Yes, check the DamageCalcs sheet on the Gear Spreadsheet if you want a meaningful number. Unhide the DamageCalcs sheet then look at the fields Q11 and S11 for the EP values of MH and OH. As a very rough estimate, a combat rogue may fall around 9EP for 1DPS MH, and 3EP for 1DPS OH. It really does vary with gear and buffs.

 

Ricard

Is it DPS? Or is it average damage? I'm referring to what's being calculated as the marginal change in those columns.

 

• Aldriana

Those columns are the EP value of 1 point of MH and OH DPS. Average damage turns out to be an inferior way of scoring items (particular on the OH) due to the fact that two weapons with the same DPS and different speed wind up with significantly different average damages.

Also, for those of you that were curious about the EP values of other things, row 11 of that sheet has EP values for just about everything under the sun, given that it's the entire basis of how the sheet makes gear recommendations.

 

drumbum

It's the EP value of adding 1 average damage.

Of course, if you are interested in the value of increasing 1 weapon DPS, you can simply multiply by the weapon speed. For example, adding 1 weapon DPS to a 2.6 speed weapon is the same as adding 2.6 to the average weapon damage.

Edit: Sorry, I misunderstood how it was labeled in the spreadsheet. (The row is labeled as "MH Damage" but is in fact DPS, not damage. I blame Aldriana. )

However, the point still stands about being able to infer one from the other. The EP value of adding one weapon damage would be equal to the EP value of adding 1 weapon DPS divided by the weapon speed.

 

antonied90

Hey.. I'm relatively new to the forums but straight to the point -> The dps spreadsheet shows that the optimal cycle for me would be 4s/5r snd cut. But so far I've always been using a 5s/5r rotation. Now from what I read so far here i understand that my dps will go higher with 4s/5r but what i cannot understand is why. I just dont see the logic behind it - the rupture uptime will be more but I won't always get the 25energy proc from 5s relentless. So can anyone please explain it to me?
Also if i start using this new cycle how should i start it in the very beginning - 1s/4s/5r or straight to 4s/5r?
And finally, since I've never used the cycle I have no idea how to best use my adrenaline rush on it, so any suggestion there, please? Thank you in advance!

 

chrisrichardson

I can't see a 2.4 Mutilate spec in the original post, or on the last few pages of this thread, so I'll attempt to contribute something useful.

41/20/0
[Obviously take what you want from the first line in the Combat tree. I personally prefer Gouge for PvP/grinding viability.]

Considering you poison applications should be brilliant at end-game with decent hit rating/haste procs/fast daggers, there's no real calling to improve those.

Which leaves the only real toss up between Murder and Master Poisoner.

I'm not sure if anybody has the numbers to deduce which of the two would be more beneficial, but I'm inclined to go with Master Poisoner.

Would be great if someone could provide a response/some data, thank you!

 

drumbum

Basically that's exactly what's happening. The spreadsheet believes it is worth gambling on the 25 energy return from Relentless Strikes to squeeze your Ruptures a little closer together. Now, in the general case, when the spreadsheets recommend a 4s/5r cycle, you can almost always go up to 5s/5r without even noticing a difference in your damage output. For example, I loaded your own gear into the spreadsheet with default buffs and it gives:

4s/5r SND cut = 1905.02 DPS
5s/5r SND cut = 1904.47 DPS

If you like the reliability of a 5s/5r cycle, then by all means, keep using it.

 

Gern

I often find myself with way too much energy, and nothing to spend it on. I run a 1s/5r cycle, but occasionally I'll get lucky with combat potency and I'll find myself with a full set of combo points mere moments after refreshing SnD. I can't refresh rupture because it's still got like 7 seconds on it, and dumping 5cps into a SnD that I just refreshed seems like the most counter productive thing I could do. I could pop it off with an eviscerate, but then I'd only have a couple of points to use when it came time to refresh rupture. So is it best to just continue SS'ing, with a full rack of cps, until rupture is ready to be refreshed?

 

• Feist-Mok

I think the ideal solution here, if you can see it coming, is to pop off a 3-ish point Evisc mid cycle so that you'll have 5 ready for your rupture again..

 

drumbum

I personally don't think popping a low rank Eviscerate is *ever* a good idea. I think you're better off spamming SS past 5 combo points until Rupture is finished, but I'd have to spend more time thinking about it.

 

Aszune

I had a question, i didn't see it anywhere else, but i may have not looked hard enough

Is there a point on a rogue, where you can have too much haste (almost a haste cap i suppose)? Reason i ask this is cause more haste means you swing faster, which means more offhand attacks land, more combat potency procs, which would lead to you not being able to smash Sinster strike fast enough with GCD to actually use all your energy

 

drumbum

That amount of haste is not really attainable. You can spend roughly 40 energy per second if you are not energy starved (one SS per 1 second GCD) which means you need about 2 combat potency procs per second to generate that much energy (10 from natural regen + 30 from two procs). With combat potency occurring on only 1/5 of offhand swings, you'd have to get your offhand swing speed to about 0.1 to generate that much energy.

With 2pc T6 SND (35%) and a 1.4 offhand you'd need roughly 937% haste to get there. Good luck.

 

Killaq

Drum I would like you to review my rotations and see if you can spot any flaws with it (I'm attempting to become as efficient as possible).

I run a 3s/5r Rotation which rarely ever drops SnD, and if it does I have it back up within 1-2 seconds.

But where I often think I have flaws is on my CD rotation.
Once I have my 3pt SnD up I usually get 2-3 pts and blow AR->BF->Trink, and if the fight isn't healing intensive Blood Fury. (Yes, I am an orc rogue)

Once thats up I quickly finish my 5 pt rupture move into another 3 pt SnD and then work towards a 5 pt evis where my CD's begin to wear out, this is followed by a 3pt SnD and then a return to Rupture/SnD. By doing the evis I avoid the downtime of waiting for my rupture to wear off, and usually seems to put me above my rogue counterparts.

See anyway I could improve this? I realize rupture/evis is but a small percentage of my damage, but with the guild I raid with, it often means above or below 1-2 people.

 

Towelette

The increased damage from Rupture uptime will theoretically outweigh the times that you do not get the Relentless proc. However, as has been stated a couple of times in the past several pages, the total DPS difference between 4s/5r and 5s/5r is really not all that much. If you're more comfortable with the 5s/5r cycle, you can stick with it and not see much of a drop in DPS at all. It may not even be noticeable to you.

The way I've read it in this thread, and the way it makes the most sense to me logically, is to start with 1s and then build into your normal cycle. The reasoning is that SnD is extremely important to Rogue DPS, so getting it up as early as possible if desired. I do know Rogues that begin with 2 SSs for a 2CP SnD to build into their cycle, but I wouldn't go any further than that with my CPs, as you have to regen energy and extend SnD downtime to go beyond an intitial 2CPs. Going for 4CP immediately is not a good idea because you leave yourself with several seconds of time that SnD is not up. My recommendation is to just start with a 1s and then work into your cycle from there if possible.

Seconding this. In recent memory, I can't recall any Evis of less than 4CP even outdamaging my SS.

This question has been addressed in the past 10 pages or so, but I believe the answer was along these lines:
-Keeping SnD up is always the primary objective. As a rule, anything that would decrease it's uptime is generally a bad move. Have you checked the difference in damage on the spreadsheet between a 3s/5r and a 4s/5r cycle? You may actually do better switching to 4s/5r and having SnD never drop off.
-Dropping a 5CP Evis because Rupture is still ticking is not necessarily an increase in damage. Using the Evis and then going back into your cycle is going to make your next Rupture later than it normally would be, thus decreasing its uptime. You're trading X amount of Rupture uptime for an Evis, which does less damage.

Rupture being up doesn't count as its downtime. The time it takes from that Rupture to complete until the next Rupture you use is the downtime; what you're looking at is rather how best to spend your CPs and energy while it's up. Is your using Evis increasing the time between Ruptures finishing and being refreshed?

Veterans correct me if I'm wrong, but I think what it would come down to is if the damage difference of using an Evis over spamming SS at 5CP would outweigh the increased Rupture downtime. Does that sound about right?

 

wowmarx

There are many threads / posts about standard rogue cycles, but I haven't seen much formal theorycraft about a cycle that maximises Ashtongue Talisman uptime as part of a modified Snd/Rupture cycle.

Ashtongue:
Equip: 20% chance per combo point for your finishing moves to grant 145 critical strike rating for 10 sec.

I've often found myself following a standard t6 3snd-cut/5r cycle, but modifying it on the fly to maximise Ashtongue talisman uptime, usually doing quicker finishers if the trinket hasn't procced, or hitting a 4scut/4r cycle intermittently to give 2x80% proc chances on the trinket.

Is there any formal theorycraft about the use of this trinket and how to modify a standard 3s/5r cycle to maximise trinket uptime - and especially modifications to the standard cycle to "get the trinket back up and running" if bad luck means it hasn't procced in a few cycles

 

koaschten

You should check out the dmgcalc sheet of the rogue gear spreadsheet and notice the minor differences between 3/5 and 5/5 cycles. With energy queueing you can in theory even do more dps than any sheet with 5/5 cycle and AToL suggests. A 4/4 cycle is ALWAYS inferior because your rupture is not free anymore.

 

wowmarx

Indeed, having also tried the 5s/5r cycle, it works fine for some reliable trinket uptime, but again, at 10 secs trinket uptime per 5cp cycle, it strikes me there must be a way to maximise uptime further.

The problem is that I'm basically experimenting with uptime in practice, including 4scut/5r cycles, but there's a lot of improvisation here, depending on relentless strikes/potency procs as well as haste etc - so what I was wondering is whether anyone had done some formalised theorycraft on a specific ashtongue cycle that maximises that 10 second proc uptime while still maintaining a consistent cycle over time.

 

• Aldriana

As it turns out, there's no real good way to increase Ashtongue Uptime without simply having more energy to spend. To get a cycle shorter than 5s5r involves using smaller finishers; and smaller finishers don't always proc. When you work out the loss of proc uptime from nonprocs and compare it to the increased uptime from a shorter cycle, it works out to be mostly a wash. As such, it's generally believed that the best cycle to make use of AToL is 5s5r, because it has uptime as good as any and allows you to do energy queuing and squeeze in a few extra SS during the buff uptime.

It also might be noted, of course, that the effects of cycle optimization on trinket uptime are small on the scale of total DPS: that is, don't totally redo your cycle around one trinket. It's not a bad trinket, but it's not good enough to change the usual rules.

 

• Vulajin

Well, the issue is that uptime has specific limits since you're always making a tradeoff. If you run 5/5, then you're guaranteed to have the trinket proc on every finisher, and therefore have 20 seconds of uptime per cycle. On the other hand, if you run 1/1, you could have 20 seconds of uptime, but only 4% of the time. (In actuality, it doesn't work exactly like that for 1/1; actually, the performance of 1/1 for the trinket will be pretty good thanks to the sheer number of finishers being used.)

Obviously, for each combo point we chop off a finisher, we lose 20% chance of having the trinket proc on that finisher. Since the proc duration is 10 seconds, we can roughly estimate that each combo point reduced means an average uptime reduction of 2 seconds. However, each combo point reduced also means a shortening of the cycle. Let's assume that, for a given cycle, the uptime of the trinket is U = T/C, where T is the length of time the trinket is active, and C is the cycle duration. Now assume that reducing either finisher by one CP reduces the length of the cycle by X. In order to achieve an uptime increase by chopping off a combo point, the following condition must be met:

(T-2)/(C-X) > U
T-2 > U(C-X)
T - 2 > T - UX
UX > 2
X > 2/U

This is actually a reasonably easy condition to meet for most cycles, if you take as a baseline the cycle with the most CP invested in finishers: 5/5. This cycle has uptime of roughly 67% (determined via spreadsheet), so X must be greater than 3. There is an interesting breakpoint involved here. If X > 3, then decreasing 1 CP from 5/5 is an uptime gain; and since U is higher for 4/5 or 5/4, then 2/U is lower, therefore decreasing 1 CP from those must also be an uptime gain (since X is static); and so on, to the conclusion that 1/1 is the best trinket uptime cycle. On the other hand, if X <= 3, then decreasing 1 CP from 5/5 is not an uptime gain; and since U is lower, then 2/U is higher, therefore decreasing 1 CP from those must also not be an uptime gain; and so on, so that 5/5 is the best trinket uptime cycle. Of course, for most builds other than Mutilate, X is greater than 3, so it stands to reason that 1/1 is the greatest trinket uptime cycle.

So the real question is, how much of a gain is this, really? Well, you're not going to get 100% uptime, period. And even if you go all the way down to 1/1 for maximum uptime, you're losing a lot on your actual cycle: energy efficiency (read as Sinister Strike DPS) and Rupture DPS, both in large amounts. Between 5/5 and, say, 4/5 or 5/4, you don't lose as much DPS from the cycle itself, but you also don't gain as much uptime, so it's basically a wash. In the end, just trust the gear spreadsheet, its model is perfectly sufficient. Yes, it feels like you can "game" the trinket for better uptime, but in reality, no you can't, other than some simple energy queueing between a trinket proc expiry and your next finisher.

 

• Aldriana

You need to be a bit careful with carrying that analysis through to it's logical conclusion; if your cycle gets too short, you start losing uptime to overlapping procs. For instance, in the case of 1s1r, you're doing a finisher, on average, every 3 seconds. Thus, if you get procs on consecutive finishers, the first one doesn't give a full 10 seconds of uptime - it only gives 3. So a slightly compressed cycle - 3s5r or 4s5r - may experience slightly higher uptime than 5s5r, but if your cycle gets too short, your uptime actually starts to drop off as your cycle just isn't long enough to get the full benefit from procs.

 

• Vulajin

Well, I ignored that consideration mostly because it makes the math overly complex. In addition, I'd say there's about as much loss to overlapping procs as there is gain from a proc from a previous finisher carrying through extra future finishers simply due to the brevity of the cycle. All in all, I'd say it's obvious that the important consideration isn't trinket uptime anyway, and a 1/1 cycle obviously has major negative consequences such that any considerations about trinket uptime are moot.