My gear is 4/5 t6 and Cursed Vision, using Blade of Infamy/Blade of Savagery.
The spreasheet has been 95% accurate for me in terms of gems/enchants and estimated dps, however i have noticed that:
- Going from C.Swords to C.Daggers (Shard of Azzinoth MH - Messenger of Fate OH) costs me 64 dps which in my eyes is too big a dps loss for actually upgrading (?) to a dagger with ~6 dps more.
- Going C.Fist+Sword is actually worse than pure C.Sword despite the general belief. I used Vengeful Mainhand's and noticed that C.Sword will out dps C.Fist+Sword mostly because of the loss of Vile Poisons which talents are dumped to Fist Spec.
- Finally going from the Classic 20/41/0 spec to a C.Sword - Hemo dps increases my dps by ~18. I wonder how Hemo is modelled, are you consuming 100% of its charges? If so then probably my raid dps will be lower than estimated since the other melee will consuming the hemo charges as well.

Sounds about right, on the whole:

1) Combat Swords really does do significantly more damage than Combat Daggers. That 64 probably counts at least some contribution from Murder, but on the whole, Daggers really are just a weak spec right now

2) There's an extensive discussion about hybrid xxx/sword builds vs sword/sword builds in the Roguecraft 101 thread, and the conclusion seems to be that in terms of raw physical damage they are very competitive, but the loss of poison talents in hybrid builds gives sword/sword builds a narrow edge in most cases.

3) General consensus on Tri-spec Hemo is that it's total raid contribution assuming 100% debuff usage is better than Combat Swords at T4 itemization, comparable at T5 itemization, and slightly inferior at T6 itemization, which also seems to match your findings.

So, all in all: your findings with the spreadsheet match the conventional wisdom around here.

I was wondering how to tackle cycles that are based opon non 2xT4 gear. I got some upgrades recently that let me loose 2T4 and change cycle from 2/5 to 5/5 (at current stage testing raids with hemo 11/27/23).

5/5 cycle is a bit too long for most of the encounters. It works well in fights like naj'entus or rage/anetheron. However I would love to see an alternative for it in fights with lot of interuptions or shorter DPS periods with target changes (supremus, akama, some SSC/Tk fights).

I was trying to go with 1s/3s/5r. But was also thinking if hybrids like 3s/4r could work as well etc. Thats at least until I get some solid feelings about hemo swords, before i switch back to combat.

Any ideas?

1. The differences between swordspec and daggerspec are discussed in the Roguecraft101 and Changes for dagger viability threads. Without going into detail it boils down to 3-5 essential advantages of sword over dagger (in order of importance):
- OH-sword attacks proccing MH-extrahits through Swordspecialization (~50% of your extra hits are procced from your offhand)
- Mangle and Rupture synergy (in dagger builds your uptime of rupture is a lot shorter than with swords)
- Murder (2% dmg increase on murderable mobs)
- Vile Poisons (more Poison damage)
- 5/5 Lethality (swords) vs. 4/5 Lethality (daggers)

Yes ~65 DPS sounds a lot but i have the same gear 4/5 T6 + Vision + Infamy/Savagery and tried daggers. Its true^^

2. This has been the case a few patches ago but now the commonly acknowledged opinion is that Sword/Sword > Fist/Sword as you can read in the first post of Roguecraft101

3. You can tell the sheet how to model Hemo. In your case a ~18 dps most likely means that you have an older version of the spreadsheet and 'Include Hemo Debuff DPS Estimate?' checked.


edit:

On cycles: it really doesn't matter what cycle you use. There is no 100% accurate cycle because of combat potency. You can't do the same combination of styles over and over because combat potency means you will sometimes have more energy and sometimes less. Instead you have to adapt your playing style to the number of procs you are getting - you have to be flexible and react fast. (for example you won't have any problem keeping up Rupture and SnD 100% while Heroism/Bloodlust and Drums of Battle are up and your off-hand is fast enough - as it should be )
Never rely on cycles, just follow this rule:
Keep up SnD. Ever.
Try to keep it up using as few combo points as possible on each SnD (first one gives you 13 seconds while the second, third, forth and fifth only provide 4 more seconds each)
Use all of your other CP to maximize your Rupture uptime.

Thats really all you need to know about cycles. They provide an idea but modeling combat potency will never be 100% accurate with it beeing dependant on your weaponspeed and haste.

Hello, I have a couple questions !

1) Why don't you all speak french !!! Ok ok, I'll do my best...

2) I saw on the Buffed dps sheet ( Cell Q7 ) : 18.88 for Expertise.
It seemed to be clear in an old thread that the value displayed in game for Weapon Expertise (The old stat) was in fact the value used for calculations by the game : 18 in my exemple, not 18.88
What about the new stat and the spreadsheet ?

You are correct in that being a more accurate modeling. There is one complication which is the primary reason this has not yet been changed. It wouldn't be very hard to floor this value, but doing so would mess up the AEP calculation for Expertise. Because the NextStat Macro adds 1 to determine increased DPS value for each stat, the result would become either 0 AEP or a higher than expected result (if it breaks into the next whole value). So, until I find the time to alter the 2 NextStat functions to work correctly with a floored Expertise, I won't be changing that. It's something I want to address soon and it's been on my "to do" list as part of the calculation run through. So far, I haven't made many changes in the base calculations besides corrections that have no impact beyond themselves or, in the case of Seal Fate or Hemo Debuff vs Debuff Estimate, were incorrect enough that correcting them made the spreadsheet clearly more accurate than it was.

Other posts...
Batory ---
I think you are on the right track. Those seem like the correct shorter cycles to consider. I think Todesbote explained the issue with optimum cycles pretty clearly. The only thing I'd add is that since the spreadsheet models a fight in terms of constant DPS on one target, the suggested cycle may not be optimum in some circumstances (like fights with movement at set intervals) meaning one may have to adjust one's cycles to accommodate situations.

Stokos ---
Aldriana and Todesbote pretty much said all that needs to be said. I'll just add that the reason the Hemo Debuff Estimate is on by default is because people already realize that Hemo is inferior in personal DPS but the question is whether speccing Hemo might be a benefit to the raid (just as other DPS classes might shift a few talent points to improve buffs for the rest of the raid or Warriors use specific trinkets to give the rest of their party an increase to AP). In order to get a more correct estimate of personal DPS, one would turn the Hemo Debuff Estimate to No on the Talent sheet (options to the right) and turn the Hemo Debuff on in the buff section (below the equipment on the Gear page).

Now, about the Debuff Estimate itself. That is the overall benefit of the debuff. I think it's in the correct ballpark, but it is very difficult to model as it's a complicated effect that depends on raid stats and composition beyond what is modeled on the spreadsheet itself. There might be some room to improve it, but I haven't seen a very comprehensive analysis of it's true value yet. One might have to go line-by-line through a Combat Log to see who is using up charges and which skills these are. Then figure out based on talents what multipliers further modify the debuff, which classes have enhanced crit chance, etc in order to narrow down it's value.

I believe the general consensus is that as soon as Combat Swords gets close to Hemo with the debuff estimate, it's time to switch. It becomes harder to justify taking what might be a 100 dps hit for a hard to quantify and non-obvious raid buff.

There is one minor change that I would like to see -- instead of just showing the best DPS cycle, maybe show the top 3 or top 5 cycles. Some cycles are only slightly worse DPS than the optimum one, but are much easier to pull off. This may help people vary their optimum cycle for the current fight -- i.e. one that requires a lot of getting in/out you'll be using a short cycle and it would be nice to see exactly what that cycle would be.

You can go and look at the Buffed DPS sheet, of course, but it's not pretty and many people would get confused. Having a Top 3 or Top 5 cycles table instead of just the Top DPS cycle would be beneficial for everyone, I think.

Either that, or have two cycles -- Top DPS cycle and Top DPS Short Cycle, which would show the highest DPS cycle under a certain time threshold.

Any comments?

Songster---
Expertise
Yes, that is the plan, adding a whole value and dividing, actually 2.5*82/52 to be exact. But again, requires time and testing. Should be soon. As an FYI, the currently encoded rating conversion in the game for levels 70-100 is supposedly (Level + 12) / 52 or, in other words the conversion will be 2.5*92/52 for an 80. Same applies to other conversions.

Stabmaster---
Actually, I've discussed something very similar with Todemax who has been helping with the cycle sheet. First step is adding more cycles, especially the missing short dagger cycles and 4+ cycles for non-mutilate Seal Fate builds. After that, the plan is to do something along the line of what you suggest. It would be very nice to also know your fallback cycles. An additional recommendation of what cycle to switch to when you see you are running short on energy and what cycle you should think about jumping to is you are capping out. A top-5 list with estimated energy gain/loss might be enough leaving the decision up to the user.

I'm currently hemo as well, but considering switching.
Kael is dead, and we're working on Hyjal and BT, so T6 is around the corner.

While hemo / combat are on the same level, i'd consider hemo to be better because it (however slightly) increases the tank's threat which would allow the mages, warlocks etc to do more damage without overaggoing. And gives me a reduced cd on vanish.
As soon as combat shows up above combat again though, i'll definitely switch.

Perhaps the next version of the sheet will give me some more clarity.

Just wanted to say thanks for the latest update on spreadsheet, talent/gear profiles really save my time and make life easier, thanks.

They may well change this when the the level cap actually increases, though. For levels less than 60, it's (L-8)/52. For 60-70, it's (52+3(L-60))/52. I would guess that the scaling from 70-80 will wind up being even stronger. Something of the form (82+N(L-70))/52, for N > 3. They'll probably choose a value once they've seen how far gear has progressed when the expansion releases.

I'm going to take a guess and assume you're looking at Unbuffed DPS (which does not count WF at all).

It turns out that Flooring Expertise remarkably didn't break anything, so it will be included in next update.

The Windfury effect on the sheet is a little strange, but it makes sense. Since Windfury is coming from a party member and not you, it doesn't count for unbuffed DPS. Your poisons, on the other hand do count, as those you get running solo. This is why looking at unbuffed Windfury is worse...in fact it's 0 DPS as you cannot gain it's benefit solo.

2.3.2.10 & 2.3.2.10NA are now available

This was primarily a Mac fix for various Macros. Hopefully all Macros in the NA version should now work on a Mac (with Excel versions that support Visual Basic - in short, not the latest Office)

2.3.2.10 dmm

Added
Gladiator 4-piece bonus to Set bonus list.

Fixed
Ranged, Mainhand and Offhand socket bonuses.
Tuskbreaker now correctly attributed to Zul Aman.
All Macros now coded to work with Macs (except Armory Import which can't).
Fixed issue where Buffed DPS on the Gear sheet always said Sinister Strike, now correctly allows other options.
Expertise is now properly floored.
Added Resilience to various Arena weapons.
Ghostly Strike option now only applying to Hemo.

Nice work on that last upgrade, allthough, I found a little bug.

Some gems are not listed as gems from BT or Hyjal. They appear in my list of upgrades but I unchecked both instances. In this case the gems are Crimson Sun and Delicate Fire Ruby.

Crimson Sun is a jewelcrafting BoP gem. Delicate Fire Ruby is not currently in game, but is also believed to be a JC gem, seeing as it's +12 to a stat and not +10.

Ohw, I'm sorry for that. I thought Jewelcrafting wasn't enabled because I enabled the sheet to import all settings as well from the previous version. After I checked it was on, so I disabled it. Sorry for the trouble.

Unfortunately, I have not gotten to adding the other buttons to the import routines yet. That does remind me to add that to the to do list now that I've got the Macros working (knock on wood) for the two major platforms.

Back to AR: I'm assuming here that you're going to be calculating this as an extra damage boost and not trying to work it into the cycle calculation. Given how combat swords rogues end up using it, it makes sense.

It's 160 energy if you use it correctly, right? Use it within 1 second of the next energy tick and it'll give you 8 doubled energy ticks. That's exactly 4 Sinister strikes. Eviscerate costs 10 energy though, and you'll probably borrow a combo point from your "normal" cycle to get in 5e. Perhaps what you could do is have it add the 4 instants + 5e damage, then subtract whatever the "value" of a combo point and 10 energy is. Can one say that the damage of a combo point by itself is (Damage from Rupture + SnD per cycle)/(Average number of instants per cycle)? Does that properly discount Ruthlessness? Similarly, as 10 energy is 1/4 of a SS, so it would be 1/4 of avg SS dmg + 1/4 of the damage one combo point.

Now things are different with Combat Daggers. I played Daggers when I had no worthwhile swords at all to use, and generally went with 3s/5s/5r. 160 energy is 2 + 2/3 backstabs, which barely speeds up the cycle at all. All AR generally did was allow me to make one cycle at 5s/5r, which fits with the 3 extra combo points (counting the 60% complete finisher which has a 60% chance at getting you a combo point). Thus all AR does in terms of cycles is fit in a 5r in a shorter amount of time (plus the extra instants). Calculating exactly how much damage you gain (besides instants) from this is kinda tricky though. The best I can think of is calculating the damage of this shortened cycle and subtracting how much damage you would have done in the same time period with your normal cycle.

As for how much that disagrees with reality, well, there's two parts. First, if you have 3/3 Ruth you'll be gaining a small fraction of a combo point. You'll be getting .36CP from Ruth and 2.66... from instants. That's an extra .1266...CP. If you have 2/3 Ruth (not a great idea, but many prefer to get 4/5 Leth), you'll be short a slight fraction of a combo point. Secondly, there's the cost of the SnD you're not paying. Your 3s normally costs 10 energy, but here you're not using that finisher, so you're gaining 10 energy.

I'm probably missing something about using the extra CP to modify your (long) cycle, but I'm fairly certain that was exactly the modifcation I ran when I was daggers.

Shouldn't that have been 2.3.3 ?

Anyway, the talent tree switching works, but....
Saving DPS doesn't seem to work, it throws an error.

Run-time error '57121'
Can't exit design mode because Control 'CommandButton1' can not be created.

Here's the file : RogueDPS_2_3_2_10NA.xls

I copied the regular data fields from an older version, which normally works fine.
If there's any reason this won't work anymore however, just let me know.

And, there are 2 copy buttons behind the neck row in the gear_buffs sheet.
As well as 2 behind the ranged slot, helm, legs slot and the belt slot.
You might want to double check all the buttons on that sheet.

Buttons contain too much text in a number of cases.
On the talent sheet i see: Res, Save, Delete Talent, Res, Save.
But checking the buttons i can see there should be more text in them, so you may need to make the buttons more wide or reduce the text.

Gear sheet i see : delete comments, items, gem, ench, re, save, del, save dps.
Mostly the buttons on the left seem very cramped.

The upgrade buttons only show "up", the copy buttons "c".
And further down there's 3 more buttons that list : Re, Save, Del. With the rest of the text hidden.


Next, the talent combo boxes.
It doesn't crash, however the following is somewhat weird.
I selected Hemo swords. no problem so far.
Next i select Combat swords, and tell it not to switch. The popup now stays on Combat swords even though it didn't load combat swords. ( basically the combo box should switch back to hemo swords)
If i save the sheet and close, then load it again, it'll still show Combat swords in the combo box, even though the talents are quite clearly Hemo swords.
Not a massive bug, but pretty odd still.

Hi, I have been a "follower" of the rogue dps spreadsheet for a while now and wish to give many many thanks to DMM for the recent fantastic updates. I do however have a few questions.

My guild is currently tackling illidan, who hopefully will die this reset. Now Phase 1 of illidan, assuming you dont get a parasite gives pretty nice data sets to compare to the spreadsheet on. (I am using recount and suspending it between wipes so it is only recording the Phase 1 data). Now the dps spreadsheet shows my theoretical damage breakdown as;

Melee 37%
Haemo 27%
Finisher 20%
Others 16%

However my data sets show my melee to be between 65-70% of my total damage with Haemo making up about 20% and Finishers and others as the rest (10-15%).

I am a pretty experienced player and am able to follow cycles almost perfectly, 5s5r in my case but these numbers seem to show a huge difference in the predicted damage breakdown.

I am currently playing on the Eu realm Runetotem from Japan (don't ask.....) and have a more or less constant latency of about 400ms, could this be playing a role in the above discrepancy? My average dps is also about 200-250 lower than the theoretical maximum of the spreadsheet which is 1450ish.

One thing I have noticed with such latency is that my energy bar never ticks at 20, i regularly get 23 or up to 25 energy per regen tick. Again this being a latency related problem could it be affecting my cycles and do you have any suggestions on a way to combat it?

In my opinion, the best way to provide a uniform model for AR is to do the following:

1) Divide 150 energy by the instant attack energy cost (and floor) to get the number of additional instants
2) Determine the average number of combo points from those instants (usually 1 for 1, unless seal fate is in there)
3) Add in 0.6 combo points (or 0.2 or 0.4) for Ruthlessness, as appropriate
4) Taking Relentless Strikes into account, determine the damage and resulting energy cost of the resulting eviscerate
5) Add up the energy cost of the evis. plus the instants and see how much you exceed (or come in below) 150
6) Add the instant damage + evis damage per 5 minutes
7) Subtract the fractional cycle damage due to finishers of a normal cycle for the energy used beyond 150 per 5 minutes (or, if you are below 150 energy, add this fractional difference)

*Step 7 roughly represents loosening or tightening your cycle following an AR. Basically, it is how much energy you pulled out of your regular cycle to run the AR bit.

In practice, let's say you did this with sinister strikes in a "regular" combat build:

1) 150/40 = 3.75 -> 3 sinister strikes
2) 3 sinister strikes -> 3 combo points
3) +0.6 combo points
4) Evis at 3.6 combo points costs 35 - (3.6/5*25) = 17 energy; calculate the damage of a 3 and a 4 pt. eviscerate and weight accordingly
5) Total energy generated = 150; total energy used = 137; energy left = 13
6) Add up the damage and divide over 5 minutes
7) Add some additional damage for that 13 extra energy... for example, if you run 2s/5r, your energy cycle is (2*40 + 25 - 2/5*25 + 5*40 + 25 - 5/5*25) = 295. Find your cycle damage due to finishers, multiply by 13/295, and add this number divided over 5 minutes. (Basically, this represents a bit of extra energy dumped into your regular cycle.)

Or, let's do this with a Hemo AR build:

1) 150/35 = 4.28 -> 4 hemos
2) 4 hemos -> 4 combo points
3) +0 combo points (I'm thinking 0/31/30; so no ruthlessness)
4) Evis at 4 combo points costs 35 - 0 = 35 energy (no relentless); calculate the damage of a 4 pt. eviscerate
5) Total energy generated = 150; total energy used = 175; energy leeched = 25
6) Add up the damage and divide over 5 minutes
7) Subtract off some damage for that 25 leeched energy... for example, if you run 4s/5r, your energy cycle is (5*35 + 5*35) = 350. Find your cycle damage due to finishers, multiply by 25/350, and subtract this number divided over 5 minutes. (Basically, this represents a bit of energy stolen from your regular cycle.)

Same thing would apply for backstab...