[GUIDE] 1.12.1 Fury warrior spreadsheet.

[Guybrush]

Hey. Just got my warrior to 60 so I've been looking at hit caps and the miss formula and it seems to be wrong on your sheet.

Net Hit MH/OH: E35+N35*0,04+B98
This is the pvp formula.

White Misses: 5+(EnemyLevel-PlayerLevel)+0,2*(EnemyLevel-PlayerLevel)-NetHit
Player level does not influence the miss rate, weapon skill does.

How it works on Mangos and I assume Nost as well:

=5% + IF((TargetDefense - WeaponSkill) > 10,(TargetDefense - WeaponSkill-10)*0.4% + 2%,(TargetDefense - WeaponSkill)*0.1%) - HitFromGear
(+ 19% for dual wield white hits)

Source:
https://github.com/mangoszero/server/bl ... .cpp#L2714

[vido]

Thankfully nostalrius doesn't use default mangos formulas.

The formulas in the spreadsheet do take into account weapon skill, notice that NetHit is dependant on weaponskill. The 0,2*(EnemyLevel-PlayerLevel) is there to correctly account for the difference in weapon skill and enemy defence and 5+(EnemyLevel-PlayerLevel) is true to the formulas used on nostalrius, which as far as I can tell are completely blizzlike.

Thanks for making me look through all of it again though, caught a couple mistakes. I'm going to add 24+(EnemyLevel-PlayerLevel) into the white hit formula. Also if PlayerLevel>EnemyLevel then EnemyLevel-PlayerLevel should be = 0, since you can't go below the 5% base miss chance before you apply weapon/defense skill differences.

Edit: It looks like the base 24% white miss chance isn't affected by level differences in vanilla (e.g. 27% base miss against a raid boss before you apply weapon/defense skill) so the formula will remain unchanged for now.

[Guybrush]

Thankfully nostalrius doesn't use default mangos formulas.

Did the nost devs say anything about the formulas they use or do you have any data to back this up? All I could find was that Nost was built on top of the mangos core.

The formulas in the spreadsheet do take into account weapon skill, notice that NetHit is dependant on weaponskill.

Its multiplying by 0.04%, which is the hit modifier vs players, not mobs.

The 0,2*(EnemyLevel-PlayerLevel) is there to correctly account for the difference in weapon skill and enemy defence and 5+(EnemyLevel-PlayerLevel) is true to the formulas used on nostalrius, which as far as I can tell are completely blizzlike.

Level difference does not account for skill / defence difference. A level 60 dwarf will not have the same weapon skill as a level 60 human.

These formulas are not blizzlike at all, if nost is in fact using them then its a pretty serious bug. This makes the hit cap 24.6% instead of the proper 28%.

[vido]

Yeah two level 60's don't necessarily have the same weapon skill however every level 63 mob has 315 defense which is all that matters, since the players weapon skill is taken into account in the NetHit calculation.

In TBC and later the white hit cap is 24% base miss + enemy and player level difference which leads to a 27% hit cap against a level 63 raid boss. 28% sounds like some arbitrary number, what do you base it on?

I've played on this server for a year and the formulas are correct afaik, the only thing I'm not sure of is the white hit cap since I've never tested it. I'll trust Deathwing's formulas in this case though since it's actually based on retail and not mangos.

[Guybrush]

Deathwing seems like a pretty legit source as well. All we need to find out is who the Nost devs trust more, Mangos or Deathwing.

I'll try and run some tests later on.

[Mooneyez]

I have a few questions.
All of them maybe due to opening the sheat with libre office instead of excel (don't currently have it). I can only look at the formulas and and the default numbers as it seems libre office reads something somewhere wrong which causes a "div/0" error to list in execute spam.

"White Hit Mod MH/OH" lists the same value for both mh and oh, is this an error due to libre? Because i find it wierd otherwise.

Most things look completly written from a TBC perspective, what parts are checked against nost data or vanilla data?

Glancing blows seems off for nost (still might be getting bugged formulas due to libre?) since most (from what i have gathered) seems to agree that nosts glancing formulas are closer to athans numbers then anything else.
Athan posted his statistics in a dicsussion on wowwiki, post date and link bellow.
--Athan 06:34, 20 July 2006 (EDT)
http://wowwiki.wikia.com/wiki/Talk:Weapon_skill

If i where to try to make it work in libre office where would you suggest i start looking. At what point would be the logical start of the data cascade?

Most things look completly written from a TBC perspective, what parts are checked against nost data or vanilla data? (just as a refrence if i where to do some rewriting in libre )

[vido]

Nothing was written from a TBC perspective, it wouldn't make much sense in the first place if you look at the weapon skill and white hit formulas. A few beta TBC talents were added at one point and a level 70 WF swing AP bonus if you set your character level to above 60 in the sheet. I kept those little bits in there when I was rewriting formulas, for no real good reason I guess. I suppose I'll have to get rid of it eventually since it's of no use.

I am planning to update the sheet with athan's formulas including some miscellaneous things like white hit cap vs lower level targets, some more slam related stuff, sword spec fix, weapons procs like BRE/felstriker etc.

White OH/MH hit mod is seperated to account for MH/OH weapon skill and the OH damage modifier is taken into account later on in the white DPS formula.

Regarding the div/0 error, there's a circular reference in my ragePS20 formula so I'll have to do something about that.

[Mooneyez]

Nothing was written from a TBC perspective, ---snip--- there's a circular reference in my ragePS20 formula so I'll have to do something about that.

Thanks for the quick response and explanation, makes more sense now.

[HammerDeath]

If I switch to a fat 2h and check slam, dps pre-execute greatly outperforms execute spam dps. Is this because execute spam dps is pure execute spam and nothing else? Even if thats the case, a differnce of almost 10% in favor of pre-execute dps is remarkable.

[vido]

The execute formulas were written with dual-wield fury in mind, where you simply spam execute more often than not. If you're using a slow 2h and especially if you're not spec'd into flurry, your execute spam dps will be terrible. The spreadsheet however doesn't take into account the usage of MS/BT/WW between swings which is what you'll have to rely on to do good damage with a 2h.

It's a bit tricky to get accurate execute dps formulas if you're using BT/MS/WW/Slam since it depends on how far you swing timer is progressed and your current rage, and this all changes depending world buffs/gear since at certain break points you'll be using BT/MS/WW over execute regardless of swingtimer.

Edit: Swapping to DW is probably the best option for execute phase, unless you're using a BRE. Though I'll have to add different options for execute rotations, for example continuing to use a slam in execute phase or MS+WW alone. I guess I'll end up adding 2 options for slam rotations to use outside of execute phase along with this, since I have a good idea on how to do it now.