Rogue DPS Spreadsheet - Page 20

Gogge · 12/22/06, 6:19 AM

Originally Posted by Vulajin

Originally Posted by Kink

Is there any method to search within a post? I have been tyring for a while now to find out if dagger/fist weapon specs require you to offhand a dagger/fist weapon respectively for your offhand to get the 5% bonus.

I parsed about 1000 white attacks with a sword in my main hand a dagger in my offhand (with dagger spec) and found that:

1) my tooltip did not reflect the dagger spec bonus
2) my observed crit rate closely matched the tooltip rate

I did not have the opportunity to do the same with a dagger main and sword offhand, but I have a feeling that dagger spec for both hands is based only on the main hand, which, while counterintuitive and silly, is what the Thottbot-junkies have been telling us for quite a long time.

I did some 2 hours of white attacks with the 1.7 speed Hakkari OH sword + a 1.8 dagger MH, with dagger spec, on a Blasted Lands mob and got an melee crit rate of around 27%. I swapped out the Hakkari OH for an +4 healing green 1.7 dagger from AH and did the same test for another hour before the dagger broke but I got roughly a 31% crit rate. I didn't feel like it was necessary to go another hour, that was enough testing to satisfy me at the time. Going by that I'd guess that the OH weapon has to be the same as the spec or it doesn't benefit from the added crit rate.

Kalman · 12/22/06, 2:37 PM

Now, here's something I was thinking about on beta.

Without imp SnD at all, for a Mutilate build, for my gear (~1800 AP raid buffed, ~34% crit, 10.56% hit), Fang/Pugio, assuming finishers performed under FW:

5 CP SnD: (65 + .5*60.6 + 1.5*1800)*.3*21*(.4*.7 + .34*2 + (1-.055-.4-.34)) = 2114
5 CP Evis: (904 + 1800*.15)*1.15*1.1*(.34*2 + (1-0.055-.34)) = 1908
5 CP Env: (720 + 1800*.15)*1.2*1.1*(.34*2 + 1-0.055-.34) = 1679

For the 60-63 Mutilate builds (i.e. before you can have Imp SnD again), you might actually choose not to SnD at all, instead favoring Evis/Envenom.

pf · 12/22/06, 3:20 PM

Originally Posted by Kalman

Now, here's something I was thinking about on beta.

Without imp SnD at all, for a Mutilate build, for my gear (~1800 AP raid buffed, ~34% crit, 10.56% hit), Fang/Pugio, assuming finishers performed under FW:

5 CP SnD: (65 + .5*60.6 + 1.5*1800)*.3*21*(.4*.7 + .34*2 + (1-.055-.4-.34)) = 2114
5 CP Evis: (904 + 1800*.15)*1.15*1.1*(.34*2 + (1-0.055-.34)) = 1908
5 CP Env: (720 + 1800*.15)*1.2*1.1*(.34*2 + 1-0.055-.34) = 1679

For the 60-63 Mutilate builds (i.e. before you can have Imp SnD again), you might actually choose not to SnD at all, instead favoring Evis/Envenom.

I dont see envenom ever becoming a viable raid dps alternative to eviscerate. Even if it does more upfront damage than evis, on a boss mob, I just do not see it being better overall dps considering you lose DP dps. It should be a very nice pvp move against high armor targets. Still debating on whether or not wound + crip or crip + deadly is better though (probably depends on how much healing the other team has).

Anyway, Mutilate without imp snd, snd will be pretty much be used only for stationary dps fights and fights like 4h where you can get an snd started while moving to the next mob. Also, glancing blows are now 76% instead of 70% moving your snd dmg value up to 2158.

pf · 12/22/06, 5:11 PM

http://rogue.bleedo.net/

I am probably not going to work on this much until after Jan 1. So enjoy the new changes. And merry Christmas.

Version 2.1.3 12-22-2006
Buffed mutilate cycle now uses buffed eviscerate values instead of unbuffed.
Buffed mutilate cycle now chooses between 4cp/3cp itself instead of falsely relying on the unbuffed cycle sheet.
Kiss of the spider is now properly modeled with Mutilate cycles.
changed ap factor of non-5pt evis's to match value given by thottbot.
Added eviscerate multiplier to nextstat-ap sheet for mutilate cycles.
Buffed finishers were following the wrong attack table. (slightly affects surprise attacks builds)

• Wodin · 12/23/06, 11:41 PM

Originally Posted by porkbone

Originally Posted by pf

Lethality provides a minor dps gain for combat swords. 15% more damage on sinister strike crits is not that much. If 5/5 combat potency works as described, it is much more powerful than lethality. In fact, it should give you ~10% more energy, ie 10% more SS damage. The only time you might want lethality over CP is for loatheb. There are reports that it might not proc off of glances, but noone has actually tested this that im aware of.

I spec'd for CP the day the patch came out, but I noticed during my first raid with it that it didn't seem to go off very much. Cycles were basically the same, rarely had excess energy, etc. To test it, I wrote a little mod that keeps track of energy ticks and melee hits.

Over a UBRS run with a total of 2180 MH/OH hits, this is what my energy ticked for:

1 tick of 14
4 ticks of 1
4 ticks of 45
2 ticks of 30
147 ticks of 15
1 tick of 60
1 tick of 50
3 ticks of 8
6 ticks of 17
34 ticks of 35
5 ticks of 18
4 ticks of 9
4 ticks of 19
6 ticks of 7
898 ticks of 20
5 ticks of 10
85 ticks of 21
1 tick of 2
137 ticks of 5
4 ticks of 11
1 tick of 46
3 ticks of 4
1 tick of 29
4 ticks of 12
89 ticks of 25
5 ticks of 36
5 ticks of 26
4 ticks of 13
5 ticks of 6
6 ticks of 55
34 ticks of 40
7 ticks of 41

The reason I recorded every tick amount was in case I somehow missed a stray CP proc by only counting ticks of exactly 15. As you can see, there were only around 150 ticks of 15. Assuming about 63% of the total hits were from the OH (2.7sp MH, 1.6sp OH), 150 procs comes out to not even an 11% proc rate. Even if the odd ticks brought the CP count up to 200, that's still not even 15%. Unless I'm missing something, CP looks broken.

If anyone wants to try the mod or make sure the code is sound, you can get it here. Just type /cpot for a tick/hit count, or /cpot reset if you'd like to reset the data.

I'm not sure how people missed this before . This seems really broken. Have you had any further data suggesting otherwise?

Graecus · 12/24/06, 3:23 AM

Next week's patch switching Imp SND and Imp BS.

koaschten · 12/24/06, 12:32 PM

5 bucks, there will be no patch between christmas and new year

Graecus · 12/24/06, 6:17 PM

Hehe, for my PVP's sake, I hope not. :P

• Vulajin · 12/24/06, 7:07 PM

Originally Posted by porkbone

Over a UBRS run with a total of 2180 MH/OH hits, this is what my energy ticked for:

1 tick of 14
4 ticks of 1
4 ticks of 45
2 ticks of 30
147 ticks of 15
1 tick of 60
1 tick of 50
3 ticks of 8
6 ticks of 17
34 ticks of 35
5 ticks of 18
4 ticks of 9
4 ticks of 19
6 ticks of 7
898 ticks of 20
5 ticks of 10
85 ticks of 21
1 tick of 2
137 ticks of 5
4 ticks of 11
1 tick of 46
3 ticks of 4
1 tick of 29
4 ticks of 12
89 ticks of 25
5 ticks of 36
5 ticks of 26
4 ticks of 13
5 ticks of 6
6 ticks of 55
34 ticks of 40
7 ticks of 41

The reason I recorded every tick amount was in case I somehow missed a stray CP proc by only counting ticks of exactly 15. As you can see, there were only around 150 ticks of 15. Assuming about 63% of the total hits were from the OH (2.7sp MH, 1.6sp OH), 150 procs comes out to not even an 11% proc rate. Even if the odd ticks brought the CP count up to 200, that's still not even 15%. Unless I'm missing something, CP looks broken.

My first question would be, what do odd tick amounts such as 5, 35, 46, etc. represent? My gut feeling is that some of your CP ticks got mixed in with something else, resulting in ticks of amounts that differed from the expected 15 enough that you wouldn't think to count them as CP ticks. Is there some other way we could track the procs, given that there's no combat log message for CP?

Given your swing total and the speeds of your weapons, you should have experienced about 205 CP procs. I'm not sure where your 35-energy ticks came from (possibly the 20 energy base tick plus a CP proc that occurred with "good" timing?), but if you add those to the count of CP ticks, that brings the count a good deal closer to 205.

I just don't know if that method of tracking the CP procs is going to be sufficient...it's impossible to tell what the various tick values actually represent (there are some pretty crazy ones in there). Selecting only those of exactly 15 or nearly 15 might not necessarily account for all CP procs. I'd like to see if we can come up with a more accurate way of determining whether CP is proccing at the proper rate before we start thinking it's bugged.

(edit) Actually, it should be more than 205 procs...205 is how many you'd expect if CP had a 15% proc rate, but it's actually 20%, and I'm just stupid and forgot. Should be more like 270.

Kalman · 12/24/06, 9:59 PM

35 = normal + CP on the same tick.
46 = tick of 21 + 25 Relentless proc on the same tick.
40 = CP + Relentless on the same tick
60 = CP + Relentless + normal tick

Not to mention, I think Pork has 4/9 BS, so that opens up even more possibilities for broken ticks (some of those ticks of 20, for example, were probably CP+BS ticks).

Sneakiest · 12/25/06, 8:31 AM

Originally Posted by Vulajin

My first question would be, what do odd tick amounts such as 5, 35, 46, etc. represent? My gut feeling is that some of your CP ticks got mixed in with something else, resulting in ticks of amounts that differed from the expected 15 enough that you wouldn't think to count them as CP ticks. Is there some other way we could track the procs, given that there's no combat log message for CP?

Given your swing total and the speeds of your weapons, you should have experienced about 205 CP procs. I'm not sure where your 35-energy ticks came from (possibly the 20 energy base tick plus a CP proc that occurred with "good" timing?), but if you add those to the count of CP ticks, that brings the count a good deal closer to 205.

I just don't know if that method of tracking the CP procs is going to be sufficient...it's impossible to tell what the various tick values actually represent (there are some pretty crazy ones in there). Selecting only those of exactly 15 or nearly 15 might not necessarily account for all CP procs. I'd like to see if we can come up with a more accurate way of determining whether CP is proccing at the proper rate before we start thinking it's bugged.

If you run SCT, you'll find that your gut feeling is right. Right now, on the live regular WoW servers I'm running an 11/40/0 build and SCT to measure ticks. The +15 energy from OH procs will merge in the log with other procs, if timed correctly, such as:
- Relentless
- 4/9 Bonescythe
- A regular tick
- A 2/9 Druid set bonus afflicted rejuvenation
- A combination of any of the above or any other energy sources (3/4 Stormshroud, Darkmantle, etc)

So the 20%ish odd rate seems correct and not broken. My guess would be the odd 46 ticks are a result from the usual non-perfect 20 energy ticks (~19-21) with relentless or any other possible combination.

Samurai · 12/29/06, 6:54 AM

Just a small note on raiding before TBC, unless they fix the Master of Subtlety bug, I think 11/0/40 will be the best dps spec (if you choose to use it of course, but I'm using it to offset the weapon skill nerf :P )

For those who haven't encountered this lovely little glitch, if you zone or log in/out while stealthed you keep the +10% damage bonus forever until you stealth / vanish again.

Kjaska · 12/29/06, 7:28 AM

Originally Posted by Samurai

Just a small note on raiding before TBC, unless they fix the Master of Subtlety bug, I think 11/0/40 will be the best dps spec (if you choose to use it of course, but I'm using it to offset the weapon skill nerf :P )

For those who haven't encountered this lovely little glitch, if you zone or log in/out while stealthed you keep the +10% damage bonus forever until you stealth / vanish again.

have anyone tested this yet or have any proof?

€: after kicking some deeprun rats, I can confirm this bug

tfman · 12/30/06, 11:24 AM

What is the overall consensus with TF (MH) dps currently? Using this spreadsheet it does not calc TF dps well where i'm unable to gauge the differences if i grab a GM MH and move TF to OH.
Now that GM weapons are easily obtainable i would like to be able to calc if moving TF to an OH would be a better option.

currently is there a better/higher DPS sword MH then TH besides Gressil. anyone able to have dps numbers for each weapon?

Zyz · 12/31/06, 6:06 AM

The cycle information is in a different tab, if you want more info you should probably talk to Kalman.

Kalman · 12/31/06, 11:35 AM

CP/Cycle is basically a long term energy usage model; it's a calculation of energy delta per cycle (i.e. energy spent vs. energy restored). If energy delta is negative, you're using more energy than you regen during SnD uptime and can assume some SnD downtime; if energy delta is positive, you're regening more energy than you use, and that excess energy is assumed to convert directly into damage via extra instant attacks when already at 5 CP (essentially, the DPE value for your chosen instant attack is multiplied by the excess energy and that value is added directly to your damage per cycle time; cycle time is determined by either SnD time for positive energy delta or by required regen time for negative energy delta). Damage per cycle time is energy-derived DPS; add that to the white damage model and the miscellaneous damage models (poisons, enchants, etc.) and you have an overall damage model.

Basically, CP/Cycle started as a way to play with cycles to try and derive maximum efficiency cycles for various builds, and mutated into a generalized model of how energy converts into damage, both directly and via CP, given certain assumptions about how you choose to use your energy.

For modeling SF, the simplest method involves converting the CP/energy cost via crit and using that value instead of 1, but that ignores some refinements (like wasted CP from crits at 3/4 CP) and restrictions (sometimes you crit a lot in a row, but those excess CP can't be used efficiently via SnD, so they go into inefficient finishers; it isn't a big deal for something like Mutilate where you'll always have enough for SnD to go back up, but for a SF daggers build it can be problematic). If you want to derive a generalized model for waste based on cycle, *be my guest*. I doubt Pf would object much either.

If you want more info or to discuss further, hit me via email (kalman@radiationnow.net) and we can talk via email or I can give you my IM info (which I'm not so interested in posting publicly.)

pf · 01/09/07, 8:06 PM

http://rogue.bleedo.net/

Version 2.1.4 1-9-2007
Weapon Expertise changed to 5/10 skill.
Changed surprise attacks to follow WoW 2.0.3 changes.
Switched Imp SnD and Imp BS in tree location.
Fixed a div/0 error when using 0 ruthless and 0 seal fate with mutilate.
Mutilate and crusader should no longer cause problems with next-stat sheet.
Fixed Sinister Strike min/max formula. This had no effect on actual dps calculations.

Un · 01/09/07, 9:43 PM

Thanks for the quick update.

My guild is still raiding right up until the xpack (unless folks flake out here at the last minute). I figured I'd switch out of Mutilate with the SnD/Imp. BS change, and hit combat daggers again for this last week of raiding. Now I can determine if 15/31/5 or 8/41/2 is going to be a better choice. The changes to WEx and Surprise Attacks (while still completly underwhelming talents) might just be enough to justify losing out on Lethality+RS. Although 8/41/2 is going to have to be a *signifigant* dps upgrade over 15/31/5 for me to trust that it will be a true upgrade on any fight with time off target.

pf · 01/09/07, 11:53 PM

Originally Posted by Un

Thanks for the quick update.

My guild is still raiding right up until the xpack (unless folks flake out here at the last minute). I figured I'd switch out of Mutilate with the SnD/Imp. BS change, and hit combat daggers again for this last week of raiding. Now I can determine if 15/31/5 or 8/41/2 is going to be a better choice. The changes to WEx and Surprise Attacks (while still completly underwhelming talents) might just be enough to justify losing out on Lethality+RS. Although 8/41/2 is going to have to be a *signifigant* dps upgrade over 15/31/5 for me to trust that it will be a true upgrade on any fight with time off target.

From earlier discussions in this thread, 7/41/3 is better than 8/41/2, and 6/41/4 is best if you have 4piece bs.

Un · 01/10/07, 12:04 AM

Originally Posted by pf

Originally Posted by Un

Thanks for the quick update.

My guild is still raiding right up until the xpack (unless folks flake out here at the last minute). I figured I'd switch out of Mutilate with the SnD/Imp. BS change, and hit combat daggers again for this last week of raiding. Now I can determine if 15/31/5 or 8/41/2 is going to be a better choice. The changes to WEx and Surprise Attacks (while still completly underwhelming talents) might just be enough to justify losing out on Lethality+RS. Although 8/41/2 is going to have to be a *signifigant* dps upgrade over 15/31/5 for me to trust that it will be a true upgrade on any fight with time off target.

From earlier discussions in this thread, 7/41/3 is better than 8/41/2, and 6/41/4 is best if you have 4piece bs.

7/41/3 was better for me before the Imp. SnD <-> Imp. BS switch.

Unfortunately I only have 2pcs of T3.

I'm showing:

7/41/3: 650.74
8/41/2: 655.97
15/31/5: 645.92

I think I'll go with boring 15/31/5. I'm sure 8/41/2 would get me better dps on 100% time on target fights, but I'm guessing it's not a high enough margin to also gain better dps where there's any kind of energy regen off target.

EDIT: Am I doing something wrong on the spreadsheet? I used 7/41/3 for a while in 2.0 before switching over to Mutilate, and it did seem to be the dominant CD spec on paper and in practice... but is that really still the case? It seems like losing 1 rank from Imp. BS would be a huge loss, whereas before it was just 1 pt from Imp. SnD.

pf · 01/10/07, 1:26 AM

doh, forgot about snd switch.

pf · 01/19/07, 11:26 PM

New lvl 70 version of the spreadsheet. Lots of new features and changes. Anyway, I am not sure how much longer I am going to keep updating this sheet. I am not planning on purchasing TBC in the near future and as such, my account will expire in ~3 months. I do plan on updating it fixing any small bugs left in the sheet that are found within the next week or so. If anyone is willing and able to pick this project up, I will gladly help them with any questions they have. Simply PM me on this forum and I will get in touch with you. I updated this all to lvl 70 stuff because I had come too far with working on the lvl 70 version when deciding to quit the game and move on. Anyway, enjoy the spreadsheet.

Version 2.2 - 1/19/2007
Added lots of TBC gear/weapons, and removed most previous gear/weapons.
Added some improved buff ranks since with 25 man raids, it will be more difficult to always have the best.
Added new consumables, removed a few old ones.
Added option to choose whether or not to take into account boss level defense value. (need to make +.1% crit dynamic)
Added ring enchants.
Added some more colors and rearranged things on the gear/buffs sheet to hopefully make it easier to enter things.
Added Mongoose Enchant. Seal Fate does not receive benefit from this enchant currently and its slightly undervalued for those builds. Also, mongoose causes next-stat data to be slightly off.
Added a haste rating stat and converted old hastes which fixed a few minor math bugs relating to non-snd/blade flurry hastes.
Added sockets and socket bonus functionality.
Updated gear/weapons to follow ratings system instead of straight % increases.
Updated stats to the new rating system
Updated ranks for the following: Mutilate, Sinister Strike, Backstab, Hemorrhage, Eviscerate, Rupture.
Updated buffs to lvl 70 ranks.
Updated Poisons and Windfury to lvl 70 ranks.
Changed how dual wield spec affects mutilate slightly to match in-game testing.
Combat Potency now somewhat modelled in next-stat +hit/skill/haste, not 100% correct, but uses a base DPE conversion.

Whorpe · 01/19/07, 11:48 PM

Not having thing like Gurubashi Dwarf Destroyer and any Bloodfang makes it anti-climactic to see my dps increase while getting new TBC gear :D

Un · 01/20/07, 5:51 PM

Thanks Pf.

Although sad to hear you're moving on. These spreadsheet contributions have been an amazing help to sorry sods like myself who's math skills have atrophied to a grade school level, and previously used terribly crude and inaccurate hunch's for comparing gear.

Best of luck to you whatever you choose to go on to.

• Maestroquark · 01/22/07, 1:11 PM

Originally Posted by Whorpe

Not having thing like Gurubashi Dwarf Destroyer and any Bloodfang makes it anti-climactic to see my dps increase while getting new TBC gear :D

It's pretty simple to add them back in. I'd recommend the bottom, though, as adding some stuff back in alphabetical order on Bracers/Gloves sheet was bugging out on me.

Quick-fix bug: selecting "None" on slots that can be socketed is bugged, and causes all values to become undefined. Putting in "0 0 0" or such for the "None" sockets fixes it.