Jakani

There was some discussion some time ago about this, forgive me for not looking up (it's irrelevant anyway). I went ahead and checked again my self to be sure. Using default gear and buffs (which are mid-T6 level), Tri-Spec hemo and Combat Swords are basically equivalent (1851 combat to 1852 Tri-Spec).

And for what it's worth, changing to a 0/40/21 spec results in 1896, higher than both.

Surprised myself, actually. I expected Tri-Spec to beat out CP Hemo. Oh well.

Edit: Noticed something: the default mob type is set to Demon, meaning Murder isn't having an effect. Changing it to Any yields 1910 for combat swords and 1906 for Tri-Spec. This more closely aligns with our expectations, but doesn't change the fact that for many T6 bosses, 0/40/21 comes out as best Raid DPS spec.

 

drumbum

Well, Hemo+CP didn't beat Tri-spec until a recent version, so I'm wondering what changed to cause it. It appears to have happened before the new cycle sheet was added even. By comparison, Aldriana's sheet puts Hemo+CP about 50 DPS behind Tri-spec with the default gear set, and a few versions ago this spreadsheet seemed to agree with his, so at least it's worth looking into.

 

Left

His observation about Murder is an important one. Due to the nature of Murder, it adds more than 2% DPS. When Murder is disabled (eg, when the mob type is demon), I see a roughly 3% decrease in DPS in both the DPS Spreadsheet and the Gear Spreadsheet. (This 3% figure will vary slightly based on gearing and spec, but it's a good ballpark.)

Both Tri-Spec Hemo (of either variety) and CSwords include Murder. Hemo/Combat Potency does not. With the latest DPS spreadsheet, we're seeing that if the following two conditions are met, Hemo/Combat Potency might pull ahead in raid DPS contribution:

1) The mob is not murderable
2) The Hemo debuff estimate is attributed to the rogue's DPS.

I think further research is required.

EDIT: For the default gear in the latest Gear Spreadsheet (0_10_1), turning off murder (in order to simulate a non-murderable mob) yields the following results:

CSwords -> 1909 DPS
Hemo/CP -> 1800 DPS
Hemo/Tri-Spec -> 1797 DPS

Thus, the question becomes whether the Hemo debuff is worth 109 raid DPS. For a Hemo/CP build, CP yields energy at a rate of ~3 per second. This, coupled with your 10 base regen, is ~13 energy per second. 1/2 Cycle cost for 5 hemos plus a finisher is 35*5+25 (no Relentless Strikes) = 200. 200/13 ~= 15.4 seconds. Thus, every 15.4 seconds, 50 Hemo charges are generated and (presumably) used. The raid DPS contribution becomes (roughly):

Raid DPS Contribution = 50*42*(1+RaidCrit)*BossArmorReduction/15.4

Note that this ignores glancings, crit multipliers, and static damage multipliers. It's an approximation after all. For a raid crit of 30% and damage after mitigation at 75%...

50*42*1.3*0.75/15.4 ~= 132.95 > 109

In summary, yes, it might well be worth having a single Hemo rogue along for non-murderable end-game bosses. And if you do that, it's better to have a Hemo/CP rogue because their personal DPS on non-murderable mobs is very slightly higher.

 

Cragzop

I have some questions regarding the alpha spreadsheet.

Using the alpha sheet I get a combined dps total of 1591.Using 2.4.0.9 I get a combined dps of 1649.

I'm curious if anyone has any suggestions about what changed to cause a loss of 58 dps. If I look at the details the loss is almost entirely in the white dps area. It drops to 1001 from 1049. The other 10 dps loss is in the other dps category.

I'm also dismayed and a bit surprised to see that I cannot maintain a 5s/5r rotation (-0.21 sec slack time) as a trispec hemo rogue. Napkin math would imply a 10 energy/sec return rate and a need for 350 energy requirement for the 10 combo points. Assuming a 10% miss and 5% dodge increases that energy requirement to 407. This seems doable in the 46 seconds the cycle takes even without consideration of the 4 piece T5 bonus. What am I forgetting?

 

Trioxx

first i would like to thank you for all the knowledge i have gained about rogues from your forums.... My questions though are about the spreadsheet... I downloaded it today, but looking thru it, i notice a couple of different items not on sheet that i currently am using, and would like to know if there is anyway i could possibly get some help in figuring out how i should try to upgrade the quickest way

 

Aaberg

Scroll down on the "Gear_Buffs" pane, and enter the stats of your items in (Unmodelled xxxx); then choose it in the dropdown menu, where you would else find alternative gear.

 

Left

For your first question: you may want to check default buff sets, and you may want to check boss settings. In particular, see if it is set to a caster boss instead of melee, or something like that. Short of that, I'm not sure.

Second, 5 pt. SnD only lasts for 30.45 seconds (34.8 sec with 2pT4). Also, your needed energy calc is a little off. Ruthlessness essentially adds 0.6CP per finisher, thus you get 1.2 CP from that talent per cycle. You therefore need (10-1.2)*35 = 308 energy per cycle (actually, a little more after dodges). At 10 energy per second, that is 30.8 seconds (just slightly longer than 30.45). Thus, it seems like the slight down time it comes up with makes sense, from a ballpark standpoint. The ballpark puts you at 30.8 - 30.45 = 0.35 sec downtime; 4pT5 reduces that downtime slightly. It seems correct to me.

EDIT: Redid my calcs to account for Ruthlessness as well as the set bonus he was actually referencing.

 

todemax

It should be noted that the 4 set T4 bonus is bugged as hell, and have been for a long time. The calculations of it on the dps calc sheet makes references to cells on the unbuffed cycles sheet that makes no sense. This being said, I have no idea how to fix it atm. Also only unbuffed values are calculated.

EDIT: Clarification

 

Dontmindme

I'm guessing you mean the T5 4-piece bonus. I took a quick look and did a few forums searches. This can be much more accurately depicted but I'm missing key information. It appears the spreadsheet is currently using the assumption of a 1.2 PPM proc rate. I saw an old post saying the Gear sheet is using 2%. These are similar, but nowhere can I find good testing of the actual proc rate.

So if someone with 4-pc T5 could do a little testing...
Test #1: Does it proc from Instant attacks?
Using the Spirits in DM, turn oneself to a position where you are not auto-attacking and SS/Hemo until you see a proc or a few hundred attacks.

Test #2: Is it PPM or %-based?
Using very fast weapons (in one or both hands), say around 1.3 or 1.4 speed, make at least 1000 auto-attacks.
Using slow weapons, say 2.4 or greater, make at least 1000 auto-attacks.
That should be enough to tell if its PPM based or %-based.

Next step after that would be to make a lot of attacks under fixed conditions to really pin it down.

Of course if anyone can point me to good existing data, that might work too.

 

Cragzop

Thank you for the clarification on what I was miscalculating in the cycle times.

I'm still at a loss for the DPS drop though. I've looked and I don't see any changes in the buffs. Besides looking at the selectiosn themselves, I've compared the stat totals at the bottom of the sheet (below all the buffs) and they are essentially the same.

I've also verified the boss type amd armor values and they match (also if they were truly different would the damage attributed to Hemo still match?).

I think I'll try and dig into the white damage calculations. Something must be changed.

Update: It looks like the formula in cell c54 of the buffed dps sheet has been changed

from:
B54*bMHthit*(bMHasec+F3+uinstsec+D52)

to
B54*bMHthit*(bMHasec+F3+D52)

This is in the calculation area for the impact of WF. Uinstec is from the unbuffed cycles sheet (c17) and is labelled as the number of instants per second.

 

Left

I loaded your gear into 2.4.0.8 and the Alpha to compare, and was about to post when you posted your update...

Yes, I believe this was a bug fix. Instant attacks (yellow attacks) cannot proc Windfury anymore, and thus that reference is no longer valid. (I guess it was there from earlier versions of the sheet and the game when instants could still proc Windfury.) You'll see a drop in Windfury DPS because of this.

The other things that drop where MH/OH white DPS (very slightly) and Sword Spec procs. MH/OH white DPS drops a little because the new cycles model properly takes into account SnD uptime and downtime. Sword Spec procs drop a little because there are fewer attacks to proc them after the bug fix (since they proc from WF).

Thus, the result is you see your white DPS somewhat lower than before.

 

Stormveil

I dwnlded the spreadsheet but I have huge copy buttons that keep me from inputing certain items. I can't remove them at all. Is this a bug? If these are macros then none of them work for me. Any suggestion?

 

hannigaholic

If you're using OpenOffice then download the appropriate version. There isn't an OO version of the CurrentAlpha or CurrentBeta versions though, so you'll have to navigate to View -> Toolbars -> Form Controls. Turn Design Mode on and manually delete each of the macro boxes that are getting in the way.

 

Dontmindme

Okay, so from the sound of it, it appears the "Alpha" version seems rather stable. I'm hoping to find some time this weekend and wanted to make sure I have my "to do" list up to date:

1) Add Season 4 gear.
2) Go over T5 4-piece bonus (think I'm going to use 1 PPM - only proccing off white attacks unless someone can give me good reason or tests showing something different).
3) Apply any other small bug fixes not already implemented.

Maybe...
4) Work through and implement the new armor penetration model.
5) Rehab the Hemo Debuff Estimate including new customizable options. (raid crit rate, non-buff raid armor pen)
6) Look at the Mac button issue. I've already taken a quick look and am not seeing anything. I don't see doubled buttons. So if any Mac users have found a work around...You could try changing the size of the buttons and tweaking the column size. Let me know if any of that helps.

Even without Envenom and Ghostly Strike, I'm probably going to make the next an iterative version as neither seems to be the optimum option.

 

typ0ninja

I did some ogre spirit smashing today to work on Deathmantle 4pc.
I did the turn to the side test for specials and it did indeed proc off of my hemos.
Typoninja - WWS

I then auto attacked them for a while with both fast and slow weapons.

I realize it's less than 1000 hits but did not catch this till I had already uploaded the
reports, hopefully this still helps.


Fast Weapons (Trackers blade MH, Merciless Gladiators Quickblade OH)
Gordok Spirit - WWS

Slow Weapons (Rod of the Sun King MH, Blade of Infamy OH)
Gordok Spirit - WWS

edit: yes I have seen it proc off sap as well

 

schnaxine

I don't have a screenshot, but I can confirm that the Deathmantle 4pc can even proc off sap.

 

Dontmindme

Okay, so by looking at this I see 9 procs off 671 attacks using fast weapons and 20 procs off 541 attacks using slow weapons. This seems to indicate that it is more likely a PPM effect than %-based. So I have something to go on.

The next step to really pin down the proc effect would be to make a lot of auto-attacks (probably want at least 10000 but the more the better) with fixed same speed weapons and a set amount of haste. This means no Mongoose and no trinkets like DST and no Slice and Dice. Also, the actual combat logs would be preferred for the next phase as we can look for double and triple procs by looking at how long before the buff drops.

So if anyone wants to help with this:
1) Link the combat log
2) List the weapon speed
3) List your haste rating

 

Dontmindme

Posted a beta candidate for 2.4.2.1... (both Excel and Mac versions for testing)

A quick note about the new Deathmantle 4-piece modeling. Due to circular references, one does not "see" the bonus until the AEP macro is used (or the Optimize column is copied over as per Open Office). I chose 0.5 PPM as it appears to be a PPM effect and 0.5 seems to put the proc rate in the expected ballpark. 1.0 PPM seemed too much.


2.4.2.1 dmm
Added
Season 4 PvP gear
Some ZA and Sunwell boss armor estimates.
SnD and Rupture uptimes to stat block
Changed
Boss Armor Estimates fonts changed back to 9 pt
Twin Blades now based on proc mechanics (1 PPM - 45 second cooldown)
With the upcoming availability of gems from badges, default fill-in gem changed to Glinting Pyrestone
Brand new cycle model (Thanks Left and Todemax!)
All previously 15% procs now 16.67% (or 1/6th).
Blackened Naaru Sliver estimated as 45 sec cooldown - 10% proc rate
Deathmantle 4-piece modeled as a 0.5 PPM effect, no cooldown
Fixed
Bug with gems from Armory Import
Sunder Armor/Expose Armor formulas
Saved Builds issue overwriting first name in list
Stat block % changes to be compatible with Open Office
Incorrect references in Extra Attacks block
Corrected AR pro-rating model.
Energy expansion SnD formula (missing racials)

 

Zurgat

Just imported 2407 into 2421 beta, and it seems to work mostly ok. (disabled options and talents when importing)

The issue mentioned in my earlier post also seems to be gone :


The only oddity that i can currently see, is that the -armor field is displaying " 370037 " which can't be correct.
Also, this prevents proper calculation of any item with the -armor attribute.

-= Random Ravings - RSS Feed =- Rogue and Hunter stuff here. As well as guides to get you trough your spare time.

 

Dontmindme

The issue with importing from that version of the spreadsheet seems to be the "Expose Armor" debuff. For some reason, because Expose Armor didn't exist in 2.4.0.7, it is copying the text from the column above it into that cell.

All I did for the buttons was make them slightly smaller. It's possible they were crossing over the column lines.

 

Wickedblade

Question about Expertise rating....
I read back some time ago that Weapon Expertise also gave a 1% bonus chance to hit boss level mobs. This was supposedly built in to not make it to op for other situations and make it more of a raiding talent. I never heard of this being changed and was wondering if it still has the same effect. I also am curious if the effect (+1% Hit against boss level mobs) was actually coming from the 10 Expertise or if it was an added effect of Weapon Expertise itself in the combat tree? If it infact does come from 10 Expertise then that would explain why the Shard of Contempt is rated so highly in the spreadsheet.
If this isn't the case then is Expertise overvalued greatly? I ask this because to my knowledge Expertise reduces your opponents chance to dodge, parry & block your attack.....In raiding environments proper positioning means you are behind your target which means they can't parry or block and I'm pretty sure they can't dodge either so in these cases unless Expertise has other built in effects it would be useless if you're positioned correctly behind the mob.
And last but not least does Expertise have a cap similar to Hit rating and do they work together with each other which I've kinda already worked into my paragraphs above if in fact it does give 1% extra hit chance per 10 expertise.

 

Giantlol

Mobs can dodge your attacks from behind but cannot parry or block, expertise is NOT overvalued greatly(at least not for warriors, and I'm sure its about as important for rogues)
Pretty sure mobs have a 5%+ chance to dodge from behind as well >_>

 

hannigaholic

This is hardly a priority, especially given that respeccing for raids is so cheap, but anytime you have a talent build that includes no finishers (IE the sheet recommends a 5s cycle and nothing more) there are a lot of #DIV/0! errors all over the place. It seems that it's the 'Other DPS' figure that is causing the problem. This is in the 2.4.2.1 Beta NA version of the sheet.

In case you're wondering, the reason I've ended up with such a build (0/3 Improved Snd, 0/3 Ruthlessness and 0/1 Relentless Strikes) is because I'd like to know which Shadowstep Variant build gives the most dps for doing the odd Heroic or Karazhan while PvP specced on a raiding break.

 

drumbum

Well, before the introduction of expertise, the Weapon Expertise talent gave 10 weapon skill instead. Against a raid boss, this effectively reduced your chance to miss by 3.5% (barring the presence of any additional weapon skill from gear or racials -- but the specifics aren't really important anymore), which is the same effect of hit rating. This is why the hit rating cap was lower before expertise replaced weapon skill. This is perhaps why you think the talent awards some hit rating.

However, since the change, the talent no longer affects your miss rate at all. The only beneficial effect while attacking from behind is that it reduces your dodge chance (and yes, mobs do dodge from behind). However, this effect is essentially the same benefit of hit rating with one exception -- expertise helps special attacks while hit rating beyond the 9% special hit cap does not. Since it's virtually impossible to not be over the special hit cap (only 4% from gear combined with 5/5 Precision will reach it), expertise rating is better point for point than hit rating, at least until capped.

Finally, yes there is a dodge cap similar to the hit rating cap. It depends on racials and talents though, so check the first post of Roguecraft 101 for that information.

 

Sneakerpimp

I too am getting errors with the "Other DPS" figure. In addition to the #DIV/0! errors, also gives negative values with certain gear / buff configurations (such as removing the weapon buffs with the default configuration).

Edit: Just checked and am getting negative values in 2.4.0.9, too.