Hunter Spreadsheets -- In development - Page 26

Carlaena · 08/27/07, 1:17 PM

Originally Posted by Cheeky

Factoring out Survival Instincts we get 25.4 crit rating and 57.5 RAP gain (ignoring Hourglass increase).

Umm still a bit confused .. Sorry!

Previously I stated that for BM spec, 1AP gives me 52.423% of the dps of 1agil and for SV spec, 1AP gives me 47.297% of the dps of 1agil.

I don't understand why we have such a small increase in simulated strength of agility when compared to AP. We can only see a 5ish percent change.

BM:
40agil = 40ap + 22crit

SV:
40agil = 57.5ap + 25.4crit
With survival, 1 agility gives us 43.75% more AP than BM and 15.45% more crit, but the sheet says there is only a 5% increase in agility dps contribution when compated to AP for a survival hunter, than that of BM hunter.

Why is the relative DPS contribution of Agility when compared to Attack Power for a Survival Hunter only 5% greater than a BM hunter. These numbers appear to show agility blowing AP, as a sole stat increase, off the face of the planet.

Surely the ratio AgilityDPS:AttackPowerDPS for survival must be more than 5% different than the same Bm related ratio.

Im confused.

Glaurong · 08/27/07, 2:11 PM

Are you adjusting the rotations as you change from BM to SV?

If you are comparing 1:1 rotations for both specs, of course you are going to get the results you did.

Also 1 point of agility costs roughly what 2 points of AP cost, so I don't see anything wrong with it contributing twice as much DPS per point.

Carlaena · 08/27/07, 2:31 PM

Originally Posted by Glaurong

Are you adjusting the rotations as you change from BM to SV?

If you are comparing 1:1 rotations for both specs, of course you are going to get the results you did.

Also 1 point of agility costs roughly what 2 points of AP cost, so I don't see anything wrong with it contributing twice as much DPS per point.

Survival has talents that increase every point of agility by 15%, and give you attack power equal to almost 1/4 of each agility point you have.

Beastmastery gains no buff for Agility, AP or Crit.

There is no way that Agil:AP DPS ratio for a survival hunter has only a 5% increase over a BM hunter.

Think about it. The BM tree doesnt buff AP nor Agil. The SV tree buffs AP by 4% and Agil by 15%. That alone should give Agil a 9% boost over AP. This is ignoring that massive dps contribution that EW gives us. Agility must be more than 5% better than AP for a survival hunter, than it is for a BM hunter. It must be.

I must be missing something because, in all probablity, its me that's wrong and not Cheeky's sheet.

Cheeky · 08/27/07, 3:06 PM

Originally Posted by Carlaena

Survival has talents that increase every point of agility by 15%, and give you attack power equal to almost 1/4 of each agility point you have.

Beastmastery gains no buff for Agility, AP or Crit.

There is no way that Agil:AP DPS ratio for a survival hunter has only a 5% increase over a BM hunter.

Think about it. The BM tree doesnt buff AP nor Agil. The SV tree buffs AP by 4% and Agil by 15%. That alone should give Agil a 9% boost over AP. This is ignoring that massive dps contribution that EW gives us. Agility must be more than 5% better than AP for a survival hunter, than it is for a BM hunter. It must be.

I must be missing something because, in all probablity, its me that's wrong and not Cheeky's sheet.

The math is all there, please look through the equation details. It is entirely possible that I have forgotten or missed something

That being said, BM Hunters benefit tremendously from Agility. In a non-clipping rotation, all DPS stats are 20% more valuable (roughly) through Serpent's Swiftness. Crit provides extra focus through Go For the Throat which BM pets can use for more specials to increase FI uptime. Focused Fire increases the effect of DPS-boosting stats by 2% more. FI is almost another 3% increase on its own.

I think you are looking at the problem from too simple a point of view. Hunter mechanics are complex, rich, and not always intuitive. And there are mathematical reasons why BM builds are doing superior DPS in raid settings. They do scale just as well as Survival or Marksman.

Locos · 08/27/07, 3:21 PM

I think the mathematical differences between Survival and BM have been clarified pretty well. From a contextual point of view it kind of makes sense to me why agility might not be hugely better for a Survival hunter over a Beast Master hunter (within the 5% you stated). From my experience the 1 agility being slightly better than 2 AP is pretty much dead on for BM.

As far as Survival goes I am not sure. Realistically, the +4% to AP in the survival tree is a hedge against +15% agility. Some things I can think of are that Survival has much higher crit rates than BM, so each point of crit is likely going to be more valued to a BM hunter (diminishing returns for survival). Plus, higher crit rates will affect kill command rates, which I am assuming hit for much more average damage from a BM hunter. I am not sure what effect +20% attack speed has, but it seems like that would favor agility and crits as well. Although I think that may depend on how many shots you are getting off. Just because I fire an auto shot faster than a survival hunter doesn't mean I am getting more total shots off in a given time-frame (it does seem likely though).

Anyway, those are just some ideas. I have never specced survival to check the accuracy of the spreadsheet, but I do know that my actual results as a BM mirror the sheet pretty well.

Norek · 08/27/07, 3:30 PM

Sorry for the like 2 day response time >.> been very swamped lately

Originally Posted by Cheeky

Since the Gem locations are unprotected, you can modify the Validity for the cells yourself. Select all the gem cells simultaneously and change the Validation to be =Gems (default will be =INDIRECT(xxx) - different for each cell). You should be fine doing that.

Ok so I feel kinda dumb here since I couldnt figure out where exactly to make this change, but only because in the validity options window for a cell, the only section you could type in was like INDIRECT(Gear.$AC$10) - and for some reason this didnt click (working on stuff late at night bad)

So for anyone else who might be brain malfunctioning like myself, just erase that and =Gems in the box.

Thank you =]

Fuulish · 08/27/07, 3:52 PM

Originally Posted by Cheeky

Yep. I'm hoping to do JoW. Any others you are interested in?

I'm not sure if this is feasible, but would modeling weapon buffs (like Mana Oil or Sharpening stones) or non-stat enchants (like Spellsurge from a party mate) be possible? I'm interested in looking for sweet spots of best dps + longevity.

Thanks much

Bikiniwax · 08/27/07, 3:57 PM

Originally Posted by Fuulish

I'm not sure if this is feasible, but would modeling weapon buffs (like Mana Oil or Sharpening stones) or non-stat enchants (like Spellsurge from a party mate) be possible? I'm interested in looking for sweet spots of best dps + longevity.

Thanks much

If I remember correctly, sharpening stones will no longer work for ranged Crit next patch.

Cheeky · 08/27/07, 3:58 PM

Originally Posted by Fuulish

I'm not sure if this is feasible, but would modeling weapon buffs (like Mana Oil or Sharpening stones) or non-stat enchants (like Spellsurge from a party mate) be possible? I'm interested in looking for sweet spots of best dps + longevity.

Thanks much

You do realize the weapon buffs are already modeled, don't you? They are hidden in the "Buffs" tab.

I list the weapon names, and there are drop downs for the possible buffs, including stones and oils.

Spellsurge seems like a really, really bad idea for a Hunter. Do our shots or Kill Command even count as spells for the purposes of triggering it? (I know shot casts are possible through silence, but KC isn't.)

(Sharpening Stones will have to be removed with the changes in 2.2)

Norek · 08/27/07, 4:10 PM

Quick note for Open Office users - the ammo cell has the same problems gems do, changing the Validity to =Ammo fixes it

Norek · 08/27/07, 4:24 PM

OK, playing with the buffs on the sheet, I turn on stuff like Greater Blessing of Kings, or might, or fort, and non of them will effect my stats, but turning on say, FoRA does give me the AP increase. Hope I didnt find another Open Office issue >.> lol

Raba_Storm · 08/28/07, 8:19 AM

Originally Posted by Cheeky

Can you verify which version you are seeing this on (check Change Log tab). There was a problem with an un-sorted ring list several releases ago, but I think it should be fine now. I'm not seeing this problem on version 27.5 (my beta-test version.)

I'm using version 27 straight out of first page.
Violet signet vs. correct data

Fendryl · 08/28/07, 11:42 AM

A few things of note for v27:

The Load button on the gear doesn't seem to work fully, it won't load back previously saved trinkets.
Haste Rating in this version is 10.51:1, not PTR's 15.7:1, correct?
How are scorpids handled on the pet tab? It'd appear the current form isn't represented as the pet dps is fairly 'normal' levels, but interestingly enough scorpids still appear to dps cats, ravagers, and wind serpents.

And if it hasn't been said multiple times already, thank you for such a fantastic tool =)

Cheeky · 08/28/07, 12:04 PM

Originally Posted by Fendryl

A few things of note for v27:

The Load button on the gear doesn't seem to work fully, it won't load back previously saved trinkets.
Haste Rating in this version is 10.51:1, not PTR's 15.7:1, correct?
How are scorpids handled on the pet tab? It'd appear the current form isn't represented as the pet dps is fairly 'normal' levels, but interestingly enough scorpids still appear to dps cats, ravagers, and wind serpents.

And if it hasn't been said multiple times already, thank you for such a fantastic tool =)

I've run into a problem where the definition of a range is exceeded by the number of items, enchants, and gear cells. So I left trinkets out, since they are the easiest thing to reselect. I'm going to see if I can work around this though, it's annoying to me too.

Haste is still based on live stats, like everything else is. The Beta version has been updated. I'm going to release when PTR goes live (which doesn't look like this week.)

I do a complex set of calculations on Scorpid Poison. It's all based on the average RAP of the Hunter, and I also average out the effects of Beastial Wrath on it. Even just taken at non Trinketed and BWed values Scorpid Poison is an incredible DPS ability, far out of line with all other pet specials. I was going to provide the Poison totals as well for max trinket usage and BW, but it seems moot now.

You are welcome.

Xenoborg · 08/28/07, 12:07 PM

Try downranking the level of the scorpid poison to 4 instead of 5 (4 does less damage, but has a longer duration, making it less likely the stack will fall off)

from my gear and pet
Serpent with reflexes and breath 6: 342 dps
Scorpid with reflexes, rank 5 and claw 9: 355 dps
Scorpid with reflexes, rank 4 and claw 9: 397 dps

edit: this is just a personal observation, but it doesn't seem like 5 stacks are up even close to 94.37% suggested by the calculations page

edit2: actually, maybe it is up that much on a stationary target where the pet just attacks. Here's a WWS on my pet on tidewalker. His average tick is for 278, which if it ticks every 2 (is this right) thats 139 dps, which would put his average stack size across the fight to be in the 4.1 area

peteriong · 08/29/07, 1:27 PM

Recently I read the discussion on TKAsomething forum about the proc rate test of Don Santo. The autoshot only test by Cheeky over 5000 shots shows 7.6% proc rate. Another auto+steady shot test by donsantos with 20000 shots shows 3.95% proc rate. There is a huge difference between two test results. The major uncertainty coming from no UI is able to record proc while the santos buff is up. So I decided to start a test with the proc rate monitored by myself.

Here is my test setup. Each time I have only 150 ironbite shell on my character. Then I use 1:1 auto steady rotation shooting on lv 70 Void Conduit in Netherstorm. I am using a shot rotation macro so I can concentrate on monitoring my buff icons. After I run out of ironbite shell I will feign death and record how many times the buff is up (including refresh). I also record other information recorded by RECAP. This includes how many times the buff is up (not including refresh), buff average duration, combat time. Then I reset Recap and head to vendor for buying another 150 ironbite shell. I finished 36 set of 150 ironbite shell tests, total 5400 shots. My talnet build is 41/20/0. With 15% quiver my autoshot speed is 1.96. During the test I keep viper aspect. This test is performed on Taiwan WOW Onyxia server with patch version 2.1.3. My ping is around 100~150ms.

Here are my test results:

(1) Proc rate

Santos' Blessing manual record proc times every 150 shots:5,5,11,8,7,7,5,3,12,7,4,8,7,4,11,8,6,8,8,8,2,11,8,5,4,9,7,10,4,5,6,5,11,2,14,8

Santos' Blessing manual record proc 253 times in 5400 shots

Proc rate = 4.69%

(2) Estimated Up time

Santos' Blessing RECAP record Duration Counts every 150 shots:
5,4,6,6,5,5,5,2,6,4,4,3,4,3,5,5,5,5,5,5,2,8,6,5,1,5,6,4,2,4,4,3,7,2,6,4

Santos' Blessing RECAP record Duration Counts 161 in 5400 shots

Estimated 'Buff up' time every 150 shots = Duration Counts(By RECAP) * Estimated Average Duration(By RECAP)

Estimted total 'Buff up' time = 2090.8 Sec

Total combat time = 5376 Sec

Estimated percentage of 'Buff up' = 38.89%

(3) Theoritical Up Time

Assuming 4.69% Proc rate

Total time for shooting 5400 shots = 5376 Sec

Shots per second = 1.004

Shots in 10 seconds = 10.04

Theoritical percentage of 'Buff up' = 38.26%

(4) Auto shot only test

I also run two 1600 Shots auto shot only test. No manual record for times of proc.

Santos' Blessing RECAP record Duration Counts every 1600 shots:
59,70

Santos' Blessing RECAP record Duration Counts 129 in 3200 shots

Estimted total 'Buff up' time = 1507.1 Sec

Total combat time = 6283 Sec

Estimated percentage of 'Buff up' = 23.99%

Shots per second = 0.509

Shots in 10 seconds = 5.09

Theoritical percentage of 'Buff up' = 21.69%

(5) Discussion
(a) Assuming 4.69% proc rate, the theoritical percentage of 'Buff up' is quite close to percentage of 'Buff up' estimated from RECAP for both tests.

(b) Manual recorded auto+steady shot 4.69% proc rate seems reasonable compare to 3.95% from donsantos' test. But RECAP recorded auto shot 4.03% proc rate is far below 7.6% from Cheeky's test.

(c) For the auto+steady shot test, 2082 autoshot hit, 620 autoshot crit, 2083 steadyshot hit, 615 steadyshot crit. There are 2 more autoshot than steadyshot due to lag spike during test. None of the 5400 shots miss.

(d) My original test target is Dr. Boom. This spot is close to Area 52 so I take less time to travel back for buying 150 bullets. But I notice he occasionally turns towards me, causing some shots being partially blocked. Without knowledge whether block or miss affect santos proc or not, I decided to test on Void Conduit which won't block.

(e) Assuming 4.69% proc rate and knownledge of shots per second, I am curious is there anyway to estimate 'how many proc while buff is up' and 'how many proc while no buff is up' theoritically? From the results of 5400 shots test, There are 92 procs while buff is up and 161 procs while no buff is up. 36.36% proc is not able to be recorded by RECAP.

khel · 08/30/07, 6:12 AM

Originally Posted by Cheeky

Yep. I'm hoping to do JoW. Any others you are interested in?

We use JoW, JoC (talented), and JoL in most fights. Having a retribution paladin makes it pretty feasible.

Nice work peter.

Glaurong · 08/30/07, 7:55 PM

Hey Cheeky, I noticed a few things are incorrect in the pet calc tab.

Glancing blows are still wrong, glancing rate should be ~25% and damage should be reduced to 76%.

I also don't see you doing any resist calculations for LB. 17% of all LBs should be completely resisted and all mobs have an innate 24 resist to all schools that cannot be overcome in any way. It amounts to ~5% partial resists.

LB can also crit, it seems to pull it from the 10% crit talent. Spells crit for 150% damage.

See Loading... for an example.

Cheeky · 08/31/07, 12:27 AM

Originally Posted by Glaurong

Hey Cheeky, I noticed a few things are incorrect in the pet calc tab.

Glancing blows are still wrong, glancing rate should be ~25% and damage should be reduced to 76%.

I also don't see you doing any resist calculations for LB. 17% of all LBs should be completely resisted and all mobs have an innate 24 resist to all schools that cannot be overcome in any way. It amounts to ~5% partial resists.

LB can also crit, it seems to pull it from the 10% crit talent. Spells crit for 150% damage.

See Loading... for an example.

Glancing blows have been corrected, I just haven't published (will do this weekend at the latest). I've also been accumulating a ton of statistics on LB, since it currently interests me. I should have that more accurate now as well. I appreciate any logs you could provide. More data is always better.

Cheeky · 08/31/07, 1:22 AM

Excel: http://www.havoc-guild.us/Cheeky/Hunter_DPS_Calc.xls
Open Office: http://www.havoc-guild.us/Cheeky/Hunter_DPS_Calc_OO.xls (Validation fixes for gems and ammo)

Change Log:
- New gear ([Mantle of Perenolde], [Shard-bound Bracers], [Madness of the Betrayer])
- Gear corrections ([Boneweave Girdle], [Belt of Deep Shadow], [Clefthoof Hide Quiver], [Worg Hide Quiver], [Blackened Leather Spaulders], [Chestguard of the Conniver], [Bladed Shoulderpads of the Merciless])
- Updated Glancing Blow mechanics for Pets. Only L73 mobs can be glanced, and it is at a 25% chance for 76% damage. (6% effective reduction of melee DPS)
- Fixed a bug where Kill Command was not counting pet Hit/Crit rates (Thanks Aesa!)
- Added checks for Metagem requirements. Currently this only applies to the 2nd socket of the Helm gear item.
- Added a simple model for Madness of the Betrayer, based on a 1.5 ppm for the effect. This model is not as accurate as the 4 piece Beast Lord set, since I average the result, which is slightly inaccurate since armor is not a linear effect.
- Now properly handle Resilience socket bonuses on head items.
- Added a model (developed by Sayeth of Gorgonnash at TKA Something - A World of Warcraft Hunter's Resource) This model determines the averaged gain of the trinket based on the benefit per stack, and your time spent at that level of stacking.
- Added benefit of averaged out Bestial Wrath to Kill Command Damage if specified in the Settings and Results tab.
- Reformatted sheet to better display information users are interested in.
- Changed hit mechanics to work off of 0.1% chance to hit penalty for the first 10 points of difference in weapon skill vs. defense, and 0.6% for more than that. This results in a base miss chance on level 73 targets of 9.0%. These number have been confirmed by extensive testing by members of Elitist Jerks, most notably Olgas for Hunters.
- Added in Crit % and Resist (both full and partial) for Lightning Breath ability. This is still a work in progress.

As always comments, criticism, ideas and insults are always welcome.

Sapa · 08/31/07, 4:23 AM

Originally Posted by Cheeky

Glancing blows have been corrected, I just haven't published (will do this weekend at the latest). I've also been accumulating a ton of statistics on LB, since it currently interests me. I should have that more accurate now as well. I appreciate any logs you could provide. More data is always better.

All done with 0/24/37 or 0/20/41 spec.
Loading...
CillitBang - WWS
CillitBang - WWS
CillitBang - WWS
CillitBang - WWS

Hope it helps. If you need more or in clean .txt format let me know.

Xoran · 08/31/07, 6:50 AM

Hello Cheeky,

I just noticed that for the secondary pet ability the intervall on which it is used never goes down to the cooldown even if there's enough focus left. This is for Bite first and Claw/Gore second. Otherwise with Bite second it is never used even if there's enough focus left.
Sure enough I overlooked something, but can you please explain this? =)

And as a request I would like to see the ap proc of the exalted Hyjal ring modelled, which btw has lesser stats in your sheet as currently mentioned on wowhead (24agi 28sta 58ap vs 29agi 43sta 60ap).

Cheeky · 08/31/07, 10:03 AM

Originally Posted by Xoran

Hello Cheeky,

I just noticed that for the secondary pet ability the intervall on which it is used never goes down to the cooldown even if there's enough focus left. This is for Bite first and Claw/Gore second. Otherwise with Bite second it is never used even if there's enough focus left.
Sure enough I overlooked something, but can you please explain this? =)

The pet has a global cooldown, so if you are using Bite every 10 seconds, you cannot Claw/Gore every 1.5s as well.

Originally Posted by Xoran

And as a request I would like to see the ap proc of the exalted Hyjal ring modelled, which btw has lesser stats in your sheet as currently mentioned on wowhead (24agi 28sta 58ap vs 29agi 43sta 60ap).

Could you link the item (or use itemstats) in question? Being 2/6 in SSC I'm not in position to know the item well myself.

Inudemon · 08/31/07, 10:20 AM

Originally Posted by Cheeky

Could you link the item (or use itemstats) in question? Being 2/6 in SSC I'm not in position to know the item well myself.

Band of the Eternal Champion - Items - World of Warcraft

However, no one in game is currently Exalted with Hyjal so there's no way to test the proc rate, if it has an internal cooldown, etc. I believe a couple of Forte members will be Exalted next week though, but I'm not sure if any of them are melee classes.

I have a question about the spreadsheet though. I recently downloaded version 28, when I go to the Shot Rotation tab and try to change it to a Max Special rotation, the sheet changes it to an Auto -> Steady rotation. I've checked around for an option I could have disabled and haven't been able to find anything. I have macros enabled and all of that good stuff of course, it's always worked fine for me up until now, is it a bug in this version or just user error?

Cheeky · 08/31/07, 11:09 AM

Originally Posted by Inudemon

I have a question about the spreadsheet though. I recently downloaded version 28, when I go to the Shot Rotation tab and try to change it to a Max Special rotation, the sheet changes it to an Auto -> Steady rotation. I've checked around for an option I could have disabled and haven't been able to find anything. I have macros enabled and all of that good stuff of course, it's always worked fine for me up until now, is it a bug in this version or just user error?

Right now the logic for it will only fit in an Arcane or Multi if you can fire it before your next Auto Shot is due. If you don't have enough dead time in the rotation it won't add them. I'll see if I can make the code more intelligent there, maybe allow for a degree of forced clipping to be specified.