I am newly specced to combat and have some questions.
Spreadsheet Assumptions?
Entering my gear/talents (combat daggers spec, pre-kara/kara gear) with no buffs I get a cycle of 3s/5s/3r. On my DPS Details it list Backstab DPS with 37% (in cell Q3 on sheet Talents).
How am I supposed to work in the Backstab? Am I supposed to be using sinister strike to build the talent points?
Also as a combat spec rogue, do you usually start a fight being stealthed running to the target, or is it better to be unstealthed to get there faster and start the dps?
Also what is suggested for short fights, ie kara-trash?
I am newly specced to combat and have some questions.
Spreadsheet Assumptions?
Entering my gear/talents (combat daggers spec, pre-kara/kara gear) with no buffs I get a cycle of 3s/5s/3r. On my DPS Details it list Backstab DPS with 37% (in cell Q3 on sheet Talents).
How am I supposed to work in the Backstab? Am I supposed to be using sinister strike to build the talent points?
Also as a combat spec rogue, do you usually start a fight being stealthed running to the target, or is it better to be unstealthed to get there faster and start the dps?
Also what is suggested for short fights, ie kara-trash?
As a Combat-Dagger rogue, all you do to build combo points is Backstab. (Unless you literally can't get behind the target, in which case you'll Sinister Strike.) Yes, this means you build combo points more slowly than, say, a Sword Rogue, but the cycle takes that into consideration. Personally, I'm given a 3s/5s/5r cycle—you do two S&Ds in a row so that you can maintain S&D long enough to reach 5 combo points and Rupture. In practice, I rarely am able to get this cycle and maintain 100% S&D. More often I end up using what another spreadsheet has told me: 3s/3r.
WWS is a great tool for measuring player performance and diagnosing what's working or not on a fight. In my opinion all serious raids should be running it.
Thank you. I'll try to follow your instructions this evening and take my first WWS parse.
I have another question about that: does the Combatlog heavily compromise my FPS performance? I ask this as I play on a quite old P4 3.1 Ghz with 6800GTS.
I have another question about that: does the Combatlog heavily compromise my FPS performance?
Should not no. The combatlog is generated no matter what you do, the /combatlog only saves the data instead of only displaying it. Depending on the raid instance you do and the new combatlog you can depend on a file somewhere between 30 and 60mb after a raid on your harddisk.
Sorry for bothering you, but after registering on the WWS website, I've read some parses of other raids and now I have another question. Do I need a mod to let me separate every try at every boss? What does an add on like LoggerHead does other than starting and stopping the combat log? Thank you.
Sorry for bothering you, but after registering on the WWS website, I've read some parses of other raids and now I have another question. Do I need a mod to let me separate every try at every boss?
No. The java program will do that for you. Just open the program, add your log and click Host (or something like it), wait a bit and view the logs in your browser. Simple as can be especially after 2.4 patch.
What does an add on like LoggerHead does other than starting and stopping the combat log? Thank you.
It probably also resets your combatlog to a 200 yard range instead of the 40 yard (iirc) it normally has. You can also do this manually in the game files, very simple edit, or do it ingame via a macro or an addon. On most fights you wont need it tbh, but on some fights (say MH trash with infernals in the back) the extra distance is handy.
Hi
Using the latest version of the spreadsheet it tells me that using the build below with a 5s5r rotation my personal DPS will be 8DPS lower than combat swords - i was wondering as it seems pretty obvious that the raid buff from hemo will outweigh my personal dps loss of 8...so why doesn't everyone do this? Or am i missing something?
The spreadsheet is already taking into account an excepted raid DPS contribution from the Hemo debuff. Thus, what it is telling you is even with an estimate for the debuff enabled, you are still coming up 8 DPS short in total DPS contribution. If you want to see your personal DPS, disable the Hemo debuff contribution. (As an aside, if you had read the FAQ we wouldn't need to answer this question.)
RobertM, I'm not able to duplicate your results. Here is what I did:
1) Opened my copy of the sheet
2) Imported your current gear from the armory
3) Set the spec to Combat-Daggers via the macro
4) Manually set all the buffs to "off", including removing Windfury from your mainhand weapon and replacing it with instant poison
5) Set Executioner, Mongoose, and Major Striking as your mainhand enchant
I got these results:
Unbuffed DPS (before adding enchants) -> 764.15
w/ Executioner -> +9.09 (+1.18%)
w/ Mongoose -> +14.97 (+1.92%)
w/ Major Striking -> +6.55 (+0.85%)
I'm not sure where you are getting such a low DPS base number from. I have "Melee Boss" selected (highest armor possible), so the number I'm getting should be the absolute lowest point. And I'm certainly not seeing the enchant behavior you are referencing.
A friend of mine was playing with the spreadsheet for the first time and noticed similar behavior. With all raid buffs (and boss debuffs) set to zero, removing the Mongoose enchant from his Guile dagger actually caused his dps to go up by a few dps. I tested it with my own build (41/20/0), zero'd out the buffs and debuffs and found the same result. Using the Guile and Retainers, my dps went up when i removed Mongoose (by .88 and .15 dps, respectively). When I did the same with my actual weapon, the Blade of Serration, my DPS did go up but by a very suspiscious and lowly 1.76 dps. The only default setting changed on the spreadsheet was the boss debuff, raid buff and consumables- all set to zero.
Ok... Its a bit odd in my eyes, but its still true.
I tried removing all the debuffs and buffs and it still wouldnt show similar results on my sword.
So I switched to a dagger, and surely enough it showed exec as being a loss of dps compared to no enchant.
But once I recalctualted AEP and Optimized, it started making sense again with exec having a positive effect on the dps like it should.
Its odd, or rather beyond my understanding of the sheet, but it still works.
Running the optimizer should not make *that* much difference. In particular, it should never cause a DPS loss. You say you are using Excel 2002? That might have something to do with it; I don't know. I'm going to try to replicate your exact process (fresh download and all) on Excel 2003 in a minute here.
The only thing I can think of is if you ran the optimizer with no weapon enchants on, it may set the actual uptime for all those enchants to "0". Then, using the expected vs. actual difference addition/subtraction, it might be taking off DPS because the default is higher than the "0" actual (left over from when you didn't have it on there). I'll try to check...
Ok, by starting from a fresh version of the sheet, I found the problem. (DMM, take note of this... it has significant accuracy impact for people without Excel or who aren't running the optimizer macro every time.)
Recently, DMM added the "Optimize" table to the spreadsheet, which is a feature that allows for DPS corrections when calculated proc uptimes do not match the default proc uptimes used by the sheet. In simple terms, this is what it does:
1] It estimates the uptime of a proc, such as Mongoose, and uses that to complete its cycle calculations
2] Based on the outcome of the cycle calculations, it revises its prediction of the uptime for the proc
3] It finds the difference between the estimated and calculated proc uptimes, then uses that number to predict an average stat difference. (For example, if Mongoose was estimated as up 50% of the time but calculated to be up only 30% of the time, then there is a 20% difference. 0.2*120 = 20 fewer average agility from Mongoose.)
4] It takes this stat difference and multiplies it by an expected DPS contribution of that stat to get a DPS difference
5] It adds (or subtracts) this DPS difference from your total DPS
All of this works fine, except step [4]. I'll get to that in a second. First, let me explain why this isn't a problem for people who use the macro. When you use the "Optimizer" macro, it makes successive approximations for your estimated proc uptime until the estimated and calculated uptimes of procs are virtually identical. Once you reach that point, steps [3], [4], and [5] above do nothing, because the difference is zero. Thus, you don't have any errors because the macro has corrected them all for you.
Ok, so what about if you don't use the macro between gear/spec/buff changes? Well, your expected proc uptimes will change. A lot. In particular, dropping Windfury and/or going from a MH sword to a MH dagger will have significant impacts on proc uptime (making them incorrect until the optimizer is rerun). In theory, steps [3], [4], and [5] are supposed to make the sheet respond well anyway by adjusting your DPS according to the difference in proc uptime. The problem, as I mentioned before, is with step [4].
In step [4], as I stated above, the sheet takes the expected difference in average values of a stat, then uses an expected DPS contribution from that stat as a multiplier to estimate the DPS gain or loss. Where does it get these expected DPS contributions from? Well, it seems to be pulling from a static table on the "Buffed DPS" tab, located at "Buffed DPS" B27:B34. (For agility, for example, it will compute 0.830 DPS per point of agility gained/lost, using a static value in cell B29.) The problem is, these expected DPS values aren't right, especially in low-buff situations. A well geared, buffed up rogue might get 0.830 DPS per point of agility, but a low-buffed rogue won't. And, these values are totally static. They don't update with the AEP/Optimize macro.
So, basically what ends up happening, is that unless you run the macro, the spreadsheet ends up misestimating the DPS contribution of the proc uptime difference. In RobertM's case, with his gear and no buffs, 1 point of agility for him is worth 0.31 DPS (as determined by running the AEP macro), not 0.830 DPS. Thus, when it calculates the difference for Mongoose, it takes off over twice the DPS that it should. Same thing for Executioner (it is using 0.125 DPS per point of armor penn. when it should be using 0.04).
The workaround is: use the AEP/Optimize macro. However, there needs to be some way to make the Optimize table look at the right DPS estimates per stat point. It would be an improvement (but not a total fix) to make it look at the AEP table on the first page. (It wouldn't be a total fix, as the user still has to run the macro at least once to get themselves into the right ballpark for their gear and spec. And if they are running the macro, it fixes the problem anyway.) A better fix would be to get a rough derivation of stat value without running a macro (which is, incidentally, also what needs to happen to fix OpenOffice AEP). So, we'll need to think about that.
Moral of the story: if you are using a gear, spec, or buff set significant different from sheet defaults, you really need to use the macro or else your estimates won't be right, either for AEP or for procs.
EDIT: Another workaround is this...
1] Fill in your gear, spec, and buffs
2] In the bottom of the "Gear_Buffs" section, find an empty line (eg, row 136)
3] Enter 1 agility in the "Agi" column
4] Note and record the DPS change
5] Type the DPS change into cell B29 (DPS value for agility) of the "Buffed DPS" sheet
6] Delete your "1 agility" buff
7] Repeat steps 3-6 for all other stats
It's a pain, yes, but it is effectively a way to calculate your own OpenOffice AEP, or to calculate AEP in general by hand without a macro
The best solution would be to remove the optimizer entirely by updating the sheet with formulas that are compatable with excel's and OO's ability to run iterative calculations automatically.
This was implimented in the warlock spreadsheet almost a year ago when trying to solve for optimal time between lifetaps to maintain infinite mana or to run out after 10 mins of fighting.
For rogues this is obviously a much more complex equation because there are a much larger number of layers in the iterative calculation. When the optimizer was first released my initial reaction was to copy the formulas and turn iterative calculations on but this does not work in all cases on the optimize sheet most noteably on weapon proc enchants.
I would run it before running any of the search functions (item, gem, or enchant) and you should always run it any time you need an accurate benchmark given the gear, talents, and buff you currently have active.
After some trouble, I got the spreadsheet working in OpenOffice on my Mac. However, it gave me a very unexpected result. With my current gear (turning Windfury off, allowing Cold Blood on Mut, and leaving everything else at default), Mutilate comes out to significantly more dps than CDags (1119 vs 1069). I was under the impression that the builds were generally on par. In fact, if I assume everything's going to have Mongoose on it, Mutilate with [Emerald Ripper]/[Guile of Khoraazi] (my current weapons) comes out to almost the same (1132) as fist spec with [Vanir's Right Fist of Brutality]/[Guile of Khoraazi] (1134). This seems suspect, because Vanir's has so much more DPS than the Emerald Ripper.
Can these findings be correct? Is it a spreadsheet error? Am I doing something wrong?
That's actually a reasonable result. Mutilate works out well at lower gear levels because it has quite a number of static elements that contribute more at low gear levels than higher ones: high static damage contribution to the combo point generating attack, high damage from poisons, and a large number of finishers. Completely unbuffed, I sometimes see Mutilate come out ahead of Combat Swords in the spreadsheet (depending on specific gearing). It's just that at higher gear levels, all those static components don't scale up, so Mutilate starts to fall behind a bit.
EDIT: Also, in your specific case, a 1.6 speed offhand will hurt fist spec some, as you won't get as many combat potency procs. (Also, it not being a fist or a sword for fist/sword spec will hurt you a little.) Same for Combat Daggers - reduced combat potency procs.
In Open Office one can also do the "Copy & Paste Special" as suggested on the Optimize page. This is what the macro is doing anyway and should achieve the same accuracy. The static values became necessary because the non-Macro AEP broke with modeling changes (although prior to that it was getting far off anyway). The plan was to eventually rebuild that section and make those values dynamic again. Unfortunately, I just haven't found the time to do much with the spreadsheet lately besides small bug fixes and minor additions.
Hi everyone! I'm hoping to get some feedback about an inconsistency between the DPS and Gear spreadsheets.
I currently wield the badge fists, but recently acquired the Blade of Infamy. I have enough honor to buy the Merciless Gladiator's Quickblade tomorrow. Plugging the Blade of Infamy/Merciless Gladiator's Quickblade into both spreadsheets (and changing weapon specialization accordingly) I get the following results:
DPS spreadsheet: +5 DPS
Gear spreadsheet: -12 DPS
I'm using version 2.4.2.2 of the DPS sheet and version 0.10.3 of the Gear sheet.
If the discrepency was only a point or two I'd go back to swords and think nothing more of it. But the difference between spreadsheets in this case seems rather large. The idea of going back to swords being a downgrade of 12 DPS might be enough to make me shelve the Blade of Infamy until I can get the OH from Mother Sharaz (my guild just killed RoS and we don't have everyone outfitted in full SR yet for lack of Hearts, so killing Mother might be awhile).
Any ideas why the spreadsheets show two different results going from badge fists to swords?
Any chance of making a change to the Upgrade Options?
Today if I click upgrade I get a list of every item considered better by the spreadsheet in the form of Unbuffed DPS number. For example if my current chest piece is 52.61 Unbuffed DPS (which it doesn't show anywhere that I can find unless I select a lower version chest piece and upgrade from it then looking at my current chest piece in the list). The list will show any item with a Unbuffed DPS number greater than 52.61 but it doesn't not put it into perspective the current item. Is it a small upgrade or a large upgrade. Should I focus on my chest piece versus my pants, etc?
What I think would be nice is either another column - Difference (the difference between my current item and the one in the list). Or add my current item to the list to get the DPS number. This would show me if a chest is 10pts versus legs being 15 pts.
It looks like all the math is there to do it... If it is already there and I am missing it somehow please flame away... and then point out how...
Hi everyone! I'm hoping to get some feedback about an inconsistency between the DPS and Gear spreadsheets.
I currently wield the badge fists, but recently acquired the Blade of Infamy. I have enough honor to buy the Merciless Gladiator's Quickblade tomorrow. Plugging the Blade of Infamy/Merciless Gladiator's Quickblade into both spreadsheets (and changing weapon specialization accordingly) I get the following results:
DPS spreadsheet: +5 DPS
Gear spreadsheet: -12 DPS
I'm using version 2.4.2.2 of the DPS sheet and version 0.10.3 of the Gear sheet.
If the discrepency was only a point or two I'd go back to swords and think nothing more of it. But the difference between spreadsheets in this case seems rather large. The idea of going back to swords being a downgrade of 12 DPS might be enough to make me shelve the Blade of Infamy until I can get the OH from Mother Sharaz (my guild just killed RoS and we don't have everyone outfitted in full SR yet for lack of Hearts, so killing Mother might be awhile).
Any ideas why the spreadsheets show two different results going from badge fists to swords?
I believe this was covered a page or two back. Basically it comes down to how the sheets model windfury. I believe this sheet asserts that sword spec can proc off a windfury attack while the other does not. Therefore a mh sword is worth more in this sheet than in the other. I'm not sure if either has been proven correct.
Last edited by Professor Hurt : 06/23/08 at 5:10 PM.
There is strong evidence that WF can proc Sword Spec and that Sword Spec can proc WF.
Although there is still the question of whether one can get chain procs like WF -> SwS -> WF. I'm beginning to believe it cannot, so the answer probably lies somewhere in-between both estimates, but I'm guessing closer to that as presented in this spreadsheet as 3rd generation string procs are going to be a fairly low probability and have a much smaller affect on overall DPS than a 2nd generation effect like not including WF -> SwS.
I should include the caveat that I do not know how the Gear spreadsheet calculates this, but it was reported in Adriana's other project that WF was showing about a 1% increase over that presented in the Gear spreadsheet. Given that this sheet is also showing higher WF numbers, I'm inclined to believe the newer project.
The short answer seems to be that the Fists are roughly equivalent as the power of Sword Specialization makes up for slightly weaker DPS on the weapons.
Today if I click upgrade I get a list of every item considered better by the spreadsheet in the form of Unbuffed DPS number. For example if my current chest piece is 52.61 Unbuffed DPS (which it doesn't show anywhere that I can find unless I select a lower version chest piece and upgrade from it then looking at my current chest piece in the list). The list will show any item with a Unbuffed DPS number greater than 52.61 but it doesn't not put it into perspective the current item. Is it a small upgrade or a large upgrade. Should I focus on my chest piece versus my pants, etc?
I'm not sure if I'm missing something, or if you are.
First, the "upgrade" sheet shows a column heading of "unbuffed DPS" still (along with a blank "buffed DPS" column), but this is just an artifact of the old sheets. What this column really shows is the "DPS difference" a gear upgrade will provide, relative to whatever DPS you had when you last clicked the "save DPS" button.
As an aside, the current versions have eliminated the display of unbuffed DPS entirely on the Gear_Buffs sheet. In order to see unbuffed DPS changes now (if you're interested in your solo/grinding performance I guess), you go to the Talents page and change "disable buffs" to Yes, I believe.
Anyway, back to the upgrade macro. All this macro does is take a snapshot of your current DPS with your current gear, then cycle through all other gear options for that slot. If a new piece of gear gives a DPS increase greater than the "minimum DPS" amount you specified in the "Update Search Options", that piece of gear will be copied over, along with the DPS difference (cell C8) and the "source" of the item.
You can make that DPS difference amount be relative to any arbitrary set of gear (typically your current set) by first selecting that set of gear, then hitting the "Save DPS" button, which will reset the DPS different cells to 0.
So, for the base case, wear all your current gear, then hit the "save dps" button. then hit the upgrade button(s). All the gear which is an upgrade will be copied over, along with the DPS increase each provides. You can then sort the Gear_Upgrades tab by "unbuffed DPS" (really DPS difference) to get the biggest upgrades up at the top, if you'd like. This gives a convenient visual representation of which items give you the biggest bang for you buck.
I don't know if it was built this way on purpose, but the current setup lets you do a lot of trickier things too. This is sorta hard to explain, but I'll try to give an example.
Right now I am a fist/sword build, since those are the best weapons I have available to me. But I am interested in picking up offspec weapons (say, daggers) as they are available (and not needed by anyone in my raids for raid use). How do I do that?
Well, I start as above. Set up my current gear and spec, then "Save DPS" to zero it out. Then switch the Weapon MH and OH options to daggers in the update search options. Then I switch to the current daggers I have (from Karazhan), and change my spec to CDags or Mutilate. My DPS difference shows at a large negative value, because my daggers are way worse than my raid weapons (and the specs are a little inferior to boot). But that's OK, leave it showing those negative values.
Now go ahead and hit the update button again. It will take a snapshot of my current DPS, then cycle through all daggers looking for ones that are an upgrade. Most of them end up being worse than my fist/sword DPS, but as long as they are upgrades to what I had selected when I clicked the upgrade button, they get copied over. So now I have a bunch of daggers on the Gear_Upgrades sheet, all with negative numbers in the "unbuffed DPS" column. Perfect!
Now I sort that page by unbuffed DPS, and all those daggers end up at the bottom of my list. So I now know which daggers are upgrades to what I currently have (so I know to pick them up if they drop and no one else wants them), but I can also see that none of them are actual raid upgrades to my fist/sword combo, so I never waste a roll/DKP/whatever on them thinking they are a raid upgrade.
There is strong evidence that WF can proc Sword Spec and that Sword Spec can proc WF.
Although there is still the question of whether one can get chain procs like WF -> SwS -> WF.
- Each extra attack proc is an independent event that can occur simultaneously with other extra attack procs on a single attack.
- Windfury Attack procs can proc additional Windfury Attack procs.
- Windfury Attack procs and Sword Specialization procs can proc one another.
- Sword Specialization procs cannot proc additional Sword Specialization procs.
- Consecutive extra attack chains don't have any arbitrary cap on the number of consecutive attacks that can occur.
06/19/08, 2:01 AM
Similar Threads
