Juggernaught

Noticed a few of the value drop down ranges in the Cunning Pet talent tree are wrong;

Boar's Speed
Spiked Collar
Avoidance
Lionhearted
Carrion Feeder
Great Resistance
Cornered
Wolverine Bite
Roar of Recovery
Bullheaded

Also Spiked Collar is listed as "Spiker Collar"

 

Shandara

Update:

The shot rotation currently assumes you wait with pressing your button until the spell/ability has finished casting. It doesn't take into use the spell-queue. Still working on that.

If I forgot any changes today's patch brought, let me know.

 

Ravenfire

Shandara,

2 errors and 1 technicallity with the newest sheet.

The errors: With Glyph of Serpent Sting selected the spam rate of Serpent Sting fails to change and neither does the damage Serpent Sting deals, although the duration of Serpent Sting does correctly increase. The extra 3 seconds basically means that Serpent Sting goes from [(total damage / 5) * 5 ticks] to [(total damage / 5) * 6 ticks], thus basically gaining 20% more damage per a full duration application. The extra 3 seconds also means the time between applications becomes 18 seconds from 15 seconds.

The technicallity is that on the Calculations tab you have Serpent sting as having a cooldown of 15 seconds. This is false as it CAN be spammed at the rate of the GCD, now obviously doing so yeilds 0 damage (since it'll never tick) and wastes tons of mana if done against a single target. Against multple mobs you can easily tab-sting-tab-sting-etc-etc.. and have 1 serpent sting per mob. I'd suggest you change the "Spammed DPS" calculation to be "SerpentStingDamage/SerpentStingDuration". This also needs to be reflected in the "Shot Rotation" tab as a non-MM hunter will "spam" Serpent Sting based on its Duration, not its cooldown.



Edit: In theory this Glyph SHOULD cause an increase in Chimera shot damage for a MM hunter by approximately 0.4*1 ticks worth of Serpent Sting. A meh glyph for MM's and OK glyph for Surv's, but I think both MM and Surv Hunters wouldn't mind it being changed to a straight +20% damage. MM's would get the same effect from Chimera as currently, but would also gain damage from the actual ticks they see before refreshing Serpent Sting via Chimera; While Surv hunters would also gain +20% damage but still maintain a slightly higher proc chance for LnL (chance to proc every 15 seconds vs. every 18 seconds).

 

Shandara

This is a bit odd, since I put in quite a bit of time to make sure the 3 seconds extra would work. The thing is, it's basically an extra 1 tick, so the DPS doesn't change, but as you say the 'use rate' of Serpent Sting should change.

And it should, at least in my settings the frequency of use on the Shot Rotation properly changes to the new duration, despite the 'cooldown' being 15 seconds. The same goes for the Mana Per Second column, it properly shows the decrease in mana use.

I agree that 'cooldown' is a bit of a strange word in combination with this sting. But it's the easiest way to limit the frequency of Serpent Sting without building in more exceptions in the logic behind the shot rotation.

On the calculations tab it is indeed in error, since it doesn't count the extra tick of damage.

I'll reorder where the damage increase is calculated so it makes more sense (it's done halfway somewhere now, which is why the discrepancy appeared).

Edit:
This should automatically fix the Chimera shot bug so it gains the extra damage from the glyph as well.

 

Ravenfire

Ahh I see.... the Shot Rotation tab has every thing correct except for an accurate "cooldown" and correct damage dealt. The Calculations tab, however, in addition to the "inaccurate" cooldown and total damage done also fails to reflect the new MPS via a change in the DPM (which the Shot Rotation tab DOES do).

 

Shandara

Exactly, I merely forgot about the presentation. Should be fixed next version.

 

Indora

The result of 3 hours without anything to do:

-Talents!C28 & K27 are protected
-default Buffprofiles aren't compatible
-Buffs!C21 & C49 are protected

calculations!
-E79 is weird. Shooting with a 2 second (<4 sec internal cd) frequency leads to enormous mana gain
In addition, it ignores average delay between end of internal cooldown & hit.
Here's my suggestion for this cell: =E78/(E77+1/E75/2)
(E75 should be Shots per Second then)

-H30 should be =Sum(H18;H29) instead of =Sum(H18:H29)
-Your Wild-Quiver calculation ignores increased auto shot frequence
-H34-H37 should be normalized (*2.8 instead of *RangedWeaponSpeed)
-H60 & H66 should use "AutoShotBaseDamage" instead of "AutoShotDamage", which includes Wild Quiver & Modifiers
-Chimera Shot's Serpent-Sting based Damage isn't increased by imp. SS
-K10 should be RangedWeaponDamage instead of -Dps
-K104 doesn't include Damage Adjustments
-N7: Marked for Death increase the damage by 5%, not by 10%
-N30 (EW calculations) include serpent sting, which cannot crit
-N136 =|= N153 (with x/3 L&L)
-N161 should be seperated to Effective Bonus for Arcane shot & -Chimera Shot

-Your Cobra Strikes model doesn't include double-proccs, here's my suggestion:

Uptime = 1 - chance not to procc ^ attacks in 3 pet GCDs
= 1 - (1 - critchance * 0.6) ^ ((1/steady-frequency + 1/arcane-frequency + 1/kill-shot-frequency) * 4.5)

Because of Kill Shot's high cooldown of 35 seconds, I think it's possible to ignore his possible higher critchance.
edit: it isn't possible to have Cobra Strikes & Sniper Training at the same time anyway <.<
In addition, N146 doesn't include possible less Specials per Second because of missing focus.



Well, that's all for now, haven't looked into pet Calculations yet.

Ah, I've forgotten the most important thing: please change these colors!

 

Shandara

Cheers,

• The H18;H29 thing luckily didn't change anything because it was all percentages, but that was quite a typo... heh.
• The judgement of wisdom thing is just a placeholder really. I really should redo it. Rather than model it, it should just give xx mana every 4 seconds, since the chance to proc is apparently 100%.
• Both Multi-shot and Steady Shot had errors in the normalization, thanks for noticing. I'll fix them for the next version.
• Buffs should be fixed/changed next patch, been working on them to try and make them save/load faster
• Are you sure about Marked for Death? The talent calculator says 2% damage/crit bonus for each talent point
• Cobra Strikes will now use the average time for a pet special as opposed to just the pet GCD and will only use the rate for pet specials that can actually crit.
• With the cobra uptime change, I actually looked closely at what I was doing and the uptime isn't used at all. Basically, once we know the uptime we know what part of the pet's special attacks is covered under Cobra Strikes and we can calculate the effective crit bonus. Basically, with an uptime of 33%, a third of your pet's specials should have a 100% crit rate, the other 2 the base crit rate. The combined crit rate should be 100%*Uptime + (100%-Uptime)*PetBaseCritChance. Subtract the pet base crit chance to find the bonus crit%.

As for the color scheme, I'm open to suggestions, I don't particularly like or dislike Cheeky's choices myself.

EDIT: The Judgement of Wisdom thing is confusing because it says mana per second rather than Mp5, it's fed back to the Buffs tab where all the Mp5 bonuses are gathered. The name of the calculation is in error, it should be 'Mana per 5 seconds'

EDIT#2:
On second thought, Judgement of wisdom should work like this:
1) You get a proc, internal cooldown starts
2) 4 secs later, internal cooldown ends
3) You need to fire a shot and it needs to hit, takes X time
4) you get another proc, total time elapsed 4+X

Mana gained is Y, mana gained per second is Y/(4+X)

 

Indora

• JoW: My suggested "model" is simply 4 seconds + average time from end of cooldown until you hit the target. That's more precise than just 4 seconds.
• Marked for Death: Uh, you're right, it was 5%, now it's 10%.
• The Cobra Strike Uptime is used in N149, but you calculated it again (N146/N145) instead of just using N147. ^^
• Colors: I really dislike lurid colors. Apart from that every color is fine.

(Lists look great!)

edit:
That's exactly what I meant.

 

Shandara

Update:
EDIT: And fixed JoW like discussed above, but forgot to write a change log entry about it.

EDIT2: Uploaded wrong version by mistake, should be correct now.

 

Indora

You forgot to hide the numerous calculations on the right side of the priority table. ^^

 

Shandara

I also noted I broke exporting/importing somehow. C'est la vie.

I'm not locking the sheet out of any sense of keeping it to myself, I just find it handy so I don't accidentally touch cells/sheets and delete/modify stuff without noticing.

On a different note, Arcane Shot and Multi-shot seem to be falling behind again as I fix/change things. But this is mostly due the unknown armor levels I guess in the case of Arcane Shot.

And I repeat my entreaty to whoever is in the beta to check their class base stats and base pet stats so I can add them to the spreadsheet.

 

Shandara

Update:
I've not really been extensive on adding gear, especially leather. Let me know if you want specific items added.

 

Indora

It really looks like Marksman will be the future raidtree (As long as you have enough mana...)

Btw, why do you still create a HasteLookUp table?
I think calculating Dps per Haste, Haste Softcap & Dps per Haste after Softcap is enough, you aren't using the Haste Table to calculate Itemvalues anyway.

Oh, and you forgot to unlock the Gathering Skill selection.

 

Shandara

For mana usage, the figures aren't really realistic, since I have no information on base mana pools at level 80, so I'm still using level 70 figures, thus making the abilities really cheap.

 

Cranch

Thanks for the work so far, including fixing the armory import!

One thing I noticed is that you don't turn off or reset autoshot when doing an aimed shot. For example, putting in aimed shot as the only shot in the shot rotation should drop autoshot dps by about a third, but it doesn't (tried it on v59.)

 

Shandara

Is this still the same on beta then?

If it is, it's easy to implement.

EDIT:
Never mind, this question has just been answered in the WotLK Hunter thread.

EDIT2:
Should be added to next release.

EDIT3:
Small bug-fix update and addendum:

WotLK Hunter DPS Spreadsheet

 

Ravenfire

Shandara,

Some new info on wild quiver for you; the proc deals nature damage and thus is not likely affected by armor. This info comes from the beta boards and at this time I don't know if it is supposed to deal Nature damage or physical damage (it is being reported as a possible bug due to the nature damage). I'll try and do some Doc. Boom tests during my lunch period to see if the WQ proc is directly related to the damage dealt by the Auto shot that procs it, or a static amount based on weapon damage. In the same vien I'll try and varify whether or not AS crits can proc WQ. I'll also try to check if WQ's damage is effected by debuffs that increase nature damage though that might be a stretch for my lunch break. =)

Edit: Bad grammer is bad.

Edit 2: WoWhead shows the spell "Wild Quiver Auto Shot" as dealing Nature damage and 60% of weapon damage. So I suspect that the bonus damage value will not be related to the AS that procs it crit or otherwise. It remains to be seen if WQ can crit on its own accord, though I highly doubt it.

Edit 3: Your current Wild Quiver "Damage per Shot added" calculations contain an error. Your calculation is: (# procs * Ave. AS damage * 0.6) / (60 / Total static haste) IE. your using H25 instead of H26 under that 60. This should be varifiable be the following equation: Ave. AS damage * 0.6 * 0.1
IMO its not worth correcting at this time due to the likely hood of Wild Quiver needing to be totally reworked to; not count AS crits as part of its damage, account debuffs (if applicable), and how WQ crits are calculated (if applicable).

 

Juggernaught

I noticed that Frenzy didn't seem to be efffecting my total dps.

On the "Pet" tab, cell C7 (Base DPS) is pointing at 'Pet Calculations'!F24. F24 is the base damgae of the pet's attack, not the dps of the pet's attack which is why the Frenzy talent seems to have no effect. It should be pointing to 'Pet Calculations'!F25 if I'm not mistaken.

EDIT
Ah, this also explains somewhat why Cobra Reflexes seems to take a big cunk of pet dps. The damage decrease is being counted, but not the speed increase.

 

Shandara

Cheers,

I've been moving stuff around and this cell displacement must've slipped, hehe. And because it pointed to a valid cell it didn't bug. This means a big dps boost for my popular BM spec again.

On a related note, does anyone know the exact amount Cobra Reflexes reduces damage by on beta?

You're correct, I'm not sure why Wild Quiver is like this in the sheet, but it's one of the first 'new' things I added back when I took over and it shows. Your suggestion will do for now until we know exactly how Wild Quiver works.

I'm guessing it is _supposed_ to be part of auto-shot damage and be purely physical dps, at least reading the tooltip suggests that.

WotLK Hunter DPS Spreadsheet

 

◊ Lactose

It seems to be identical to live. Posting something Cheeky had in some other thread/forum:

Look, Lactose, we'd rather you didn't eradicate the whole human race.
- Sam & Max

 

Shandara

Made a small update:
It's made me doubt the numbers a bit though. BM pets doing more than 1000 dps over non-BM pets (with full raid buffs, I don't really know yet whether having all those at the same time is feasible, I doubt it really).

It mostly comes from the AP scaling on melee attacks. Pets get a sick amount of buffs that improve AP and it shows in their melee attack damage.

Edit:
That said, we have pets in TBC doing 600-700 dps on certain encounters already, so it mustn't be that far-fetched.

WotLK Hunter DPS Spreadsheet

 

Ravenfire

Well my lunch period was.... productive but frustrating.

Test information
My level: 72
Dr. Boom: 68
RAP: 3007 (TSA & Hawk self buffs)
Crit %: 25.9
Spell Crit %: 10.9
damage range: 1007 - 1184
Talents: MM Limited build
No debuffs were put on Dr. Boom.

Things I learned:

1) /combatlog on beta doesn't seem to actually make a WoWcombatlog.txt file... Any help with this would be greatly appreciated.

2) Wild Quiver can and does proc off of any Auto Shot hits be they crits/blocked/normal.

3) The damage done by the Auto Shot that procs WQ doesn't appear to have any bearing on the damage dealt by the bonus shot; A verification of sorts for Wowhead's spell data: [Weapon % Dmg] Value: 60. And yes, it does do Nature damage.

4) Wild Quiver bonus shots do NOT use ammo (arrow stack doesn't decrease in size), and has no flight annimation despite having a firing annimation. No combat log means I cannot varify whether it is instant or has a flight time.

5) Wild Quiver CAN crit. Since I some how failed at making a Combatlog.txt file, I cannot get a realistic crit chance but my impression is that it is using Spell crit.

6) Not sure how Wild Quiver is calculating regular damage, but it appears critical strike damage is based on the physical crit damage modifier of 100%. Non crit range I saw was 512 - 584. Crit range was 1050 - 1268. I did not see a single instance of a WQ proc doing over 600 unless it was a crit, Based on the talents I ran this test with I am unsure as to how WQ crits are being calculated either


Future tests:
a) does Mortal Stirkes affect WQ crits
b) does Imp. Stings affect WQ (via nature damage)
c) does RWS affect WQ
d) does Marked for Death affect WQ
e) does Imp. Tracking affect WQ
f) does Focused Fire affect WQ


EDIT: Fixed a typo on one of my numbers.

Update: After playing around with Numbers for a bit I have a rough equation that matches the results I saw. While Weapon Damage is normally calculated as: [(Low end + high end)/2 + (ammo dps * base weapon speed) + (RAP / 14 * base weapon speed)], it appears that Wild Quiver ignores ammo quality when calculating weapon damage for purposes of it's damage as follows: [(low end + high end)/2 + (RAP / 14 * base weapon speed)]

Example: With Golden Bow of Quel'thalas, Timeless Arrows (which is what I used for my test)
w/ammo low end ---> 204 + (53 * 3) + (3007 / 14 * 3) = 1007 * 0.6 = ~604
w/ammo high end ---> 380 + (53 * 3) + (3007 / 14 * 3) = 1184 * 0.6 = ~710

w/o ammo low end ---> 204 + (3007 / 14 * 3) = 848.36 * 0.6 = ~509 * 2 = ~1018
w/o ammo high end ---> 380 + (3007 / 14 * 3) = 1024.36 * 0.6 = ~614 * 2 = ~1228

This covers the WQ range I saw for non crits (512 - 584) and also very nearly covers the crit range (1050 - 1268) assuming a pure 100% bonus damage on crits. The fact that it appears to not use ammo quality in its equation makes sence due to WQ not consuming any ammo when it actually procs.


UPDATE 2:

I went and conducted 3 more Dr. Boom tests last night, or I should say reconducted one and did two additional tests. Each test was 50 WQ procs worth of data, for a total of 150 controlled environment Wild Quiver procs. One third was conducted as a base line test, one third was conducted as a base line + maxxed Hunter's Mark test, and one third was conducted with a full compliment of "hopeful" talents and a maxxed Hunter's Mark. Talent builds for each test case will be included in this update.

Disclaimers: 50 procs per test doesn't seem like a big enough sample size to really verify all possible talents that have an impact on WQ, however 50 procs is 2-3 Dr. Boom's worth of auto shots and a significant amount of time (roughly 40 minutes per test). As for Crit chance purposes, all three tests can be combined due to a constant crit chance across all three tests of 25.9%, thus a total test pool of 150 procs.

Summarized Results

Test 1; Base Line
Talents: Base Line Test
Proc Break Down:
AS normals AS Crits WQ normals WQ Crits
36* 14 42** 8
Minimum
687 1444 513 1111
Maximum
802 1648 611 1264
Median
745 1555.5 575 1210.5
Average
743.9 1554.9 566.5 1197.1
Expected Averages (Hypotheticals)
743.9*** 1510.2 561.8 1140.5

* 2 AS were blocked for 49 damage each
** 3 WQ were blocked for 49 damage each
*** I used the [AS average seen]/[pre armor AS average] to find the effects of armor reduction to be 32%.

Note: Previously I had stated that the range of normals was covered and the crit range was vertually covered. I know now that the crit range IS covered. I forgot that I had a +3% crit damage Meta in my helm and once I took it into account the hypothetical crit range lines up perfectly with initial crit range I saw. I also was conducting this test barely into lvl 72 and as such my bow skill wasn't maxed out for my level. Blocked damage amounts were added back into the actual hit values to improve the accuracey of the data.

Test 2, Base Line with a Maxxed out Hunter's Mark
Talents: Base Line & Hunter's Mark
Proc Break Down:
AS normals AS Crits WQ normals WQ Crits
36* 14 32 18
Minimum
835 1631 647 1298
Maximum
976 1990 768 1570
Median
921.5 1842.5 702.5 1474
Average
921.6 1839.4 709.0 1472.6
Expected Averages (Hypotheticals)
918.6 1864.7 716.1 1453.7

* 3 AS blocked for 49 each

Conclusion of test: Hunter's Mark DOES effect the damage of Wild Quiver.

Test 3, Full Talent effects
Talents: Full Talent Set
An explanation is needed here. Due to careful planning of talent point distribution, I figured it would be possible to test all the talents at once so long as no talent gave the same Hypothetical bonus values to WQ. Thus; 1% from Imp. Tracking, 2% from Focused Fire, 5% from RWS, 10% from Marked for Death, 30% from Imp. Stings, and 30% crit bonus from Mortal Shots.

Proc Break Down:
AS normals AS Crits WQ normals WQ Crits
36 14 38 12
Minimum
975 2378 744 1812
Maximum
1229 2950 871 2093
Median
1026 2531.5 809.5 1973.5
Average
1041.2 2546.2 807.7 1961.3

Right away I can see that Imp. Sting has zero impact on WQ and that WQ gains full benefits from Mortal Shot. I believe that Imp. Tracking doesn't work at the moment in Beta so I'll calculate the expected averages based on that also. Modifiers in use (non crit): 1.02 * 1.05 * 1.1
Modifiers in use (crit): 1.02 * 1.05 * 1.1 * 2.339

Expected Averages (Hypotheticals)
1082.2 2531.2 843.6 1973.2

With the average numbers significantly less then 10% off from the expected values, Marked for Death is confirmed as having an impact on WQ proc values. I do feel reasonably sure that RWS does impact WQ and while I strongly suspect that FF also impacts WQ; with the sample size I conducted this test for, the margin of error for the results forces me call the evidence inconclusive.

Critical Chance of WQ
Since all 3 tests were conducted with the same ranged critical strike chance of 25.9%, I can combine all 3 sets of data for a larger and thus more accurate calculation.
Crit Chance Data:
150 procs
8 + 18 + 12 = 38 Crits
38 / 150 = 0.25333 or a 25.33 Crit rate

This is significantly close enough to my ranged crit chance for me to call it the same, thus; WQ uses Ranged Critical Strike Chance as its own.

 

Juggernaught

Sniper Training seems a bit broken.

It should increase the damage of Steady, Aimed and Explosive shots by 2%/4%/6% and give 5% crit to Kill shot on targets below 30% health.

From what I can see Snipe training in the spreadsheet is currently adding 5%/10%/15% to Kill Shot and Explosive Shot.

------------------EDIT
Also seems tha a number of pet talents have the wrong CD and focus values. Demo Screech (Carrion Bird) and Pin (Crab) are two that I spotted - I'm using Hunter - WotlkWiki - Wrath of the Lich King Information as reference.

Also Furious Howl doesn't seem to do anything - should be 5% extra damage on next attack for all party members.

 

Shandara

Cheers,

The Sniper Training should be fixed next release.

Furious Howl should be working, but I average the effect over the attacks done during the cooldown so the effect is barely noticeable. I have, however, no other way of representing the 'on next attack' effect, because there's no way to predict or model which attacks receives it.

The way I do it is as follows:

You/your pet do X attacks per second, the cooldown is Y; so X*Y = Z attacks made over the cooldown time.
Each of those Z attacks gains an equal share of the 5% damage bonus.

That bonus is added to the hunter/pet's damage adjustments on the calculations tab.

WotLK Hunter DPS Spreadsheet