• Vulajin

Version 0.1.3 is available via the link in the first post. Changes:

- Added red socket and 3 stamina bonus to Tunic of the Dark Hour. (For real this time!)
- Normalized instant attacks no longer incorrectly multiply AP contribution by 2.4 when a dagger is equipped.
- Added total health and projected Hemo debuff DPS to the header on data entry sheets.
- Added Power Word: Fortitude, Blood Pact, and Commanding Shout as buffs.

 

• Neto-

The effect that stacking Bloodlusts with Blade Flurry do is pretty impressive. I also love the flexibility of this spreadsheet, thanks a bunch Vulajin

 

jmredmon

Does this sheet work with OO?

 

• Vulajin

There should be no reason why it doesn't. There are no macros and it doesn't use any other special Excel features I'm aware of.

 

Karmon

As it stands, it does not work in OO. Though you do not use any Excle only specialities, it still suffers from differences in OO.
The most prominent incompatibility is the result of VLOOKUP/HLOOKUP when used in arithemtic if the result of the V/HLOOKUP is blank.
Excel treats that as zero, OO gives a type mismatch.

The only sheet that is affected is the 'Gear' sheet.
Replace all occurence of VLOOKUP() that is part of arithmetic (left or right side of + - * / operator) with N(VLOOKUP())
e.g Cell Gear!C12
becomes
This should work also in Excel so there is no need to maintain a OO_only version.

Also the named range 'cdsepc610' is missing. It is named cdsepc6010.

After correcting those it worked for me, except that in sheet DPSCalc1 and Cooldowns1 the circular refernces give an error ("does not converge") even after enableing circular references.

 

• Vulajin

Thanks for the OpenOffice fix, I'll be sure to implement that for 0.1.4. Could you please elaborate on the circular reference on DPS Calc 1 and Cooldowns 1? By design the sheet should have no circular references, so if one exists it is an error and I'll need to fix it.

 

Karmon

Sorry for the false alarm. I double checked with the original version you posted and it looks like I inadvertantly changed a cell that resulted in that said circular reference.

 

Rei86

So with the gear that my toon has for PVE gear I wanted to see how much DPS I would have if I went from Sword/Sword BOI/BOS to the Badge MH/BOS combo.

Well I thought that combo would do more DPS but it showed it as a 2DPS downgrade. I know thats not that "HUGE" of a difference but I'm one of thos lame min/max type of people:/

 

• Vulajin

Version 0.1.4 is available via the link in the first post. Changes:

- Fixed Gear sheet compatibility issues with OpenOffice.
- Reimplemented extra attack calculations in line with the new findings on same.

Link to extra attack findings

Reimplementing extra attacks based on these new findings literally increased my DPS estimate by 70. It also suggested that if I were to equip [Talon of Azshara] in place of the [Vanir's Right Fist of Brutality] I currently use, I would experience a gain in DPS (slight loss from the weapon swap itself, but gained back and then some thanks to poison talents). Please note that the beta disclaimer still applies, so I am not at all advocating everyone go out and switch from Vanir's to ToA even if you have one lying around.

 

Falian

I just played around with the sheet and found something strange, not sure if its a bug, but thought it might be good to post it here:

When looking at the DPS Table with the "Target Level" set as 71 (anything except Boss I assume) the chance to miss seems to be 1% too high.

Shouldn't the formula in Cell B13 (and C13) be
"=MAX(WENN(blevel="Boss";0,28;0,24+0,005*(blevel-clevel))-$C4;0)"
instead of
"=MAX(WENN(blevel="Boss";0,28;0,25+0,005*(blevel-clevel))-$C4;0)"

 

• Vulajin

Thanks for the report, that's definitely a bug and will be fixed in 0.1.5.

 

duvar

I have a few suggestions. This may sound stupid but I'm actually using this spreadsheet while levelling my rogue. After much work I did figure out how to add my own gear into the spreadsheet since this obviously only includes endgame gear. But what I would like is a) for the Agility to Crit factor to be automatically calculated based on the value of Player Level, and b) for Target Level to, instead of being hardcoded to 70, 71, 72, boss, instead be relative, e.g. +0, +1, +2, +3, ... This would allow the calculations to be more accurate for people who are not 70. I realize it's the exception rather than the norm, but I don't think it would be difficult to add, and the spreadsheet already supports non-level 70 players in various other ways anyway.

Also I'm not sure if this is considered a bug, but I set the combat duration to 20s to hopeuflly adjust for the fact that the mobs I'm fighting have very low HP and usually die quickly. But it still tells me that Xs/Yr with X=1 and Y=5 is the optimal ability rotation. I don't believe this is true, because in a typical fight by the time I've done 1 S&D and have 5 Combo Points built up, the mob is really close to dead, and there's no way i'll get the full benefit of a 5 point rupture. 1 Eviscerate usually finishes it, or very close. So in this case Xs/Ye should be DPS.

 

• Aldriana

So, the trick is, spreadsheets of this sort are usually designed to model sustained combat, not the short fights one sees while leveling. When you're less than level 70 and not raiding, all manner of assumptions change. When leveling, you're usually attacking from the front, rather than the back; this changes the value of various stats. A typical mob while leveling tends to only survive for 15 or 20 seconds, tops, which *significantly* changes the balance of white and yellow damage. It takes time traveling from mob to mob, which effects the value of activated abilities and procs with cooldowns.

The point is that converting a spreadsheet from the usual assumptions for raiding to the assumptions for leveling requires a lot more than changing a few numbers. It's really a totally different ballgame. Thus, even if the changes you suggest were made... it still wouldn't really be that helpful for optimizing leveling. It's not what the sheet is designed for. So personally, I don't see a lot of point in making such changes.

 

• Vulajin

Just to make it clear, the "duration" parameter of the spreadsheet applies only to the modeling of cooldowns and their effects. It affects nothing else.

 

duvar

I agree it's definitely not what the sheet was designed for. But you did mention for example that while raiding you're usually fighting behind the mob and levelling you're in front of the mob. The sheet actually has a cell that allows you to choose the positioning, front or back. Anyway I'll definitely get over it if the Agility to Crit and Target Level fields aren't updated, just a suggestion. For what it's worth, even though I'm doing "levelling" fights and not raid fights, the dps numbers are accurate to within 2-3% when I compare the spreadsheet to my own numbers.

 

• Vulajin

Version 0.2.0 is available via the link in the first post. Changes:

- Mongoose corrected to 1.2 PPM.
- Corrected base miss chance for targets below Boss level.
- Reworked extra attack calculations to assume an extra attack can proc other types of extra attacks, but not the same type.
- Streamlined iterative calculation implementation for energy procs.
- Tweaked armor penetration calculations to allow EA if the user specifies an EA cycle with 2/2 Imp EA, even if EA isn't activated in Buffs.
- Added season 4 arena armor and off-set pieces.
- Added Fire-Toasted Bun as a consumable buff.
- Implemented all remaining unimplemented DPS talents.

A lot of changes that make this version substantially easier for me to use as a base for future efforts (for example, a Wrath version eventually), as well as a lot of accuracy improvements, so I went ahead and bumped it to 0.2.0. The beta disclaimer still applies though.

 

Karmon

I have trouble understanding an IMO strange behaviour of this sheet.
It has to do with cycle selection and energy pooling.

Opening the sheet it comes with a cycle of 5s/5r and 60/60 energypooling
When playing with the engerypooling in the rupture part e.g. setting to 0, you loose some 2 dps.
I can explain that as follows: you trade a ruture under AtoL with the a SS under AtoL, giving a slight DPS boost.

Now the puzzling part:
Setting the rupture part to a 4pt rupture the sheet gives a 6DPS increase and it does not matter how many energy you pool in the rupture part of the cycle.
I am at a loss what happens here.
Especialy why now it does not matter if you perform the rupture under AtoL or not.

Can it be explained by mechanics?
Is it a bug in the cycle/enegry model?

 

• Vulajin

Yes, it's a stupid stupid bug I just found that further cements the reasoning behind the beta disclaimer. However, it'll be fixed in 0.2.1. (For reference, for some reason I have a certain part of my cycle calculations figuring the energy required to generate the CP for all finishers based on the specified "Y" value instead of the actual CP value for that finisher.)

 

• Vulajin

Version 0.2.1 is available via the link in the first post. Changes:

- Fixed an error in cycle calculations related to Seal Fate cycle modeling.
- Specifying fewer than 2 points in Improved Expose Armor no longer prevents Expose Armor from working.
- Modified Deadly Poison calcs to accommodate Shiv and Envenom.
- Implemented Envenom as an alternative for use in Eviscerate-designated cycles.

Note that Envenom/DP interaction modeling is something I basically invented on the spur of the moment. I think it's probably reasonably accurate, but who knows. I'd definitely like people to make a point of double-checking it.

 

duvar

Sorry if this has been asked before, I saw a mention of trinket procs on a previous page, but not trinket use effects. It doesn't seem like trinket use effects are being taken into consideration in calculations. It would be nice if, for each trinket, it had a buff column, an amount column, and a cooldown column. Then perhaps another cell on the main page that allowed you to turn on/off calculation of trinket use effects, and any calculations done would assume 100% uptime whenever cooldown was available. Are there any inherent problems with this?

 

• Vulajin

Trinket use procs are being considered, unless there's a bug relating to them. They are treated the same as any other cooldown of the type specified on the Cooldowns sheet. In other words, if you specify fight duration = 0, their effects are averaged out as Effect * Duration / Cooldown. If you specify fight duration > 0, then a trinket with a 120-second cooldown is used in time with any other effects with a 120-second cooldown, and same for any other different cooldowns.

 

Gudgud

Hello,
It looks like there is a bug in the snd calculation uptime or i didn't get the why.
For example, when i put 5 snd and 3 rupture using backstab, the snd uptime shown is 100% (cell x27 in "Cycle").
For a full cyle, this uptime should be something like 85% (without ruthlessness). The problem (if there is one) looks like to be in the row 22 formula : the condition 'OU(cycley<X4-2;cycley>X4)' seems to make thing wrong.
Btw, thx u for this spreadsheet vulajin. Very interesting.

 

• Vulajin

Please don't use abbreviations like "thx" in your posts on this forum.

Let's do a quick sanity check here. A 5s/3r cycle requires about 6.6 Backstabs if you account for Ruthlessness, or about 396 energy, and about 10 energy for the Rupture and 0 for the Slice after Relentless Strikes. Total cycle energy is therefore about 406 on average. In most high-level gear setups I'm seeing about 13.5 energy generated per second on average via Combat Potency. 406 / 13.5 = 30.07 seconds, which is actually right within the realm of allowing 100% theoretical Slice uptime.

The spreadsheet corroborates this napkin math. There's no error going on; the formula you're pointing out has to do with my implementation of Seal Fate cycle modeling, and doesn't affect cycle calculations at all unless Seal Fate is activated.

 

Gudgud

Read again my first post vulajin please. I was talking about "without ruthlessness".

On speadsheet : 13,54 energy regen with my gear (T6 so no T4 bonus)
5snd/3r
-> 8 cp to build
Let assume relentless procs on the rupture too - best scenario among 60% of the scenarii -
8*60 energy = 480 energy to spend
-> 35.45 second to build cp > 30,45 snd time

The spreadsheet shows me 100% snd uptime that I cannot agree with.
Also tried it in game several times and i cannot repeat this cycle without breaking snd.

About the formula. Maybe not the right one i pointed but i find the "right" value when i put this condition true. But when i check the row 'Slice cp time', i cannot understand why it is the same value from columns 3 snd to columns 5 snd...

EDIT : This problem looks to be fixed in your newer version of spreadsheet (had it on beta 1-2-3 version). But it now looks good on the last sheet you realeased. Sorry for the missbug.

 

• Aldriana

Just as a public interest point: I took some time this evening to poke through the AToL modeling to try to resolve the previously noted discrepancies between this sheet and mine. The difference is as follows: my sheet computes cycle length based off your sustainable energy regen - that is, using a Haste Pot doesn't decrease the average length of your cycle. It may make that one cycle that it occurs during shorter, but it doesn't change anything in terms of the long term average. This is, of course, not precisely true, but it struck me as a reasonable approximation. This sheet, on the other hand, gives the full average-case cycle reduction due to these. This is also a reasonable approximation, but it's not immediately clear to me whether it's superior or not. My gut is telling me that the true answer probably lies somewhere in between, but I'd have a hard time proving that.

Regardless: what this means is that with buff setups with lots of activated haste, this sheet approximates the cycle length as being significantly shorter, which leads to a corresponding increase in Ashtongue uptime, at least in longer cycles.

That said: there also seems to be some manner of glitch in the AToL uptime calculations for shorter cycles; for instance, if I set the cycle with 1s3r with the default gear + buffs, it indicates that my overall uptime is 65.89%, but only 63.71% on CP Builders. It seems unlikely to me that energy queuing actually *lowers* the number you land within the buff, which makes me suspect some manner of glitch in the calculations.