• ebs2002

Found a bug that gave me the N/A error: Hand enchant lookup is:
Should be:

Note the range should be to 259, not 258 (259 contains the "none" option).

Note: I use 'none' as enchant/gems to calculate my SEPs before deciding on an enchant/gem (find the upgrade with the most SEP, select it, then re-calculate).

 

• landsoul

Fix uploaded.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

NobleHelium

I get a run-time error when I try to click the sort-by-SEP button for the item lookup tab. It says "Run-time error '438': Object doesn't support this property or method." Clicking debug shows that the line that causes the error is in Module6, subroutine SortLookup(), second line, which reads, "ActiveWorkbook.Worksheets("Lookup").Sort.SortFields.Clear". I am using the .xls version of the spreadsheet on Excel 2003.

You should probably add Wicked Monarch Topaz (16 AP + 8 Crit), because Inscribed Monarch Topaz (8 STR + 8 Crit) is not actually in the game right now. Also consider adding Stoneskin Totem, which increases armor by 1150, with 20% more talented. I imagine it doesn't stack with Devotion Aura, but I haven't tested it.

 

Eliminate2

Couple of questions.

1) Is there any way you can implement a possible switch to turn off all buffs, minus the warrior ones.

2) I see there is a 114ap option for bracers. I tried looking for it on wowhead. Didn't come up with what actually provides that bonus.

3) I ran into a bug with the 2.111, Whenever I try to swap out one of the +27str gem to +20 str gem. The SEP value go to #N/A. Even trying to switch it back to +27 str. It still never recovers from it. Breakin the spreadsheet. I'm on excel97.

EDIT: tried it out with various gems. The one that seem problematic was the +20 str. For example, I dl'd the sheet, did nothing to change anything. Try to swap the +27 str gem out of the chest to +20str gem. Hit teh SEP button. All value but STR goes to #N/A. Try switching the gem back to the 27str, then SEP button, still shows #N/A

Edit: it seems all the +20 stat gems does this as well.

 

Xerophyte

 

illusive_2008

Nothing huge but Signet of Bridenbrad is spelled incorrectly it is listed as Signer, just a heads up.

 

VirUzxZ

not sure if file on filefront hasn't been updated yet but tested FuryDPS2.111BETAExcel07 that i downloaded with the glove enchant set to any random enchant was ok, but the moment when i set it to none and then back to an enchant the spread sheet produced N/A's. just an fyi dunno if it's been fix'd

 

Harpo

I am still getting runtime errors using the 2001 version of excel. I have downloaded the default spreadsheet from filefront, adjusted the expertise to "0", as I am expertised capped. However, I get a run-time error 438. Are there any solutions to this? Am I doing something incorrectly? Also, is it possible for anyone to upload their "ideal" or "wishlist" setup on a .xml? I have been trying to do this manually, but without the sort, I am sure it will be inaccurate.

Otherwise an excellent spreadsheet though.

 

Do0mbringer

It seems in the latest version your missing the racial base crit. *I can't remember if its the same for all races, but I know for humans its around 3.18%.

 

Flamero

nvm misread

 

• landsoul

Okay, fix going up again. I still need some More racial stats. still missing a few.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

• ebs2002

So, I've noticed that the DPS estimated is higher than actual, so I dove into the intricacies of where the DPS values were coming from.

The first thing I noticed is that white DPS is significantly higher in the sheet than in WWS reports of tank'n'spank fights such as 0-drake Sarth, which is odd because white damage is hard to mess up on tank'n'spank fights. Heroics being off, BTs being off, WWs being off...all of that could be accounted for by user error. But white damage?

Upon inspection, it seems your chance to crit on a white attack is X% of your non-glanced attack possibilities, not X% period.

If you're attacking 100x, and you have 35% crit, 10% white miss rate and a boss with 25% chance to glance, then:
25 swings will be glancing
Of the remaining 75 swings, 10% will miss, and 35% will crit.

Right now, the spreadsheet does not factor in the glance possibility in this way.

It's doing:
ArmorReduc * (1 + Crit - Miss - Dodge - Glance)*DmgRange/WeapSpd

It should be:
ArmorReduc * ((1 + Crit - Miss - Dodge)*(1-Glance%) + (GlanceDmg))*DmgRange/WeapSpd, I believe. Or, if you prefer, you could create a modified value for crit/miss/dodge chances modified by 1-Glance% and reference those variables.

Does this sound right to you, or am I way off? It's hard for me to see the whole picture since I didn't design it


Also, it doesn't seem that your double-counting Heroic Strikes and the MH swings they replace. Since you have %Hsmh and %noHS, you should be able to take the whole MH portion of your White DPS and modify it by %noHS

 

Do0mbringer

White dps on the sheet will be higher. It doesn't calculate heroic strike dps. It calculates heroic strike addition. At least thats my understanding of it. As for glancing and crit, I've never heard that before, and in practice I do not experience that. Otherwise my effective white crit would be much lower. In your example 35% crit would result in 26.25% effective crit, which is something I haven't seen.

 

• ebs2002

You could be right, Do0mbringer. The reason why I'm making this assumption is based on a few observations.

WoW Web Stats links for reference:
Wow Web Stats
WWS Loading...

With all buffs, Daemondred has roughly 41% chance to crit (by plugging his gear into the provided sheet). His white attacks break down like this:
Landed = regular hits, so he had 179 white swings. 55 of those were crit, which amounts to 30% crit rate.
If you factor out the number of glances, he had 144 swings, 55 of them being crit amounts to 38.19% crit

Now looking at Craelon, he has a lower crit chance (closer to 37% or so). His white attacks break down like this:
159 swings, 48 crits = 30% crit rate. Pull out glances, 38.4% crit.

I think maybe we need to do more testing on crit rating, but these WWS are reporting white attacks critting much lower than they should be using a 1-roll system.

 

• landsoul

I have to calculate total rage income first to be able to get the amount of rage that can be used on heroic strikes. Only way to do that is to be able to calculate full white damage without heroic striking. After I get the rage income I can determine how much can be used on heroic strike. The cost of a heroic strike is the average rage gain from a MH swing, plus the rage cost of heroic strike. I then can find what % of main hand swings can be used on heroic strikes, and then add in the damage gain on those main hand swings only. If you look at the damage spread percentages to the right of the actual DPS numbers, you can see that they are more likely to be accurate with recount or a WWS.

Also I'm afraid glances don't work like that. The White DPS formula you see is actually an conveniently algebraicly simplified version of a larger problem. If you would like to see the full simplification, please send me a PM and I could help you understand it better.

Also, there has been new research that bosses have -4.8% to crit on all attacks, so that might be your explanation for the missing crit.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

• ebs2002

That would certainly partially explain the discrepancies that I'm seeing. I admit I was looking for a reason why the %ages were so far off. Maybe I just need a larger dataset to look at.

I thought the whole point of iterative calculations is that you don't have to do that much work.

10 Set # heroic strikes to 0.
20 Calculate white DPS, taking out heroic strike damage.
30 Calculate rage gained from white DPS.
40 Calculate percent of attacks you can heroic strike with.
50 GOTO 20

That's what I'm doing, and I'm finding them to be off. Maybe it's that my dataset is too small, but I'm consistently seeing white DPS % being much higher than actual WWSes reflect. I know that the sheet assumes a perfect rotation, but imperfections in your rotation would *increase* your white DPS, and all the WWS reports that I'm comparing this to are showing white DPS %age to be LOWER than what the sheet calculates.

The only explanations I can come up with is
a) The raider is jousting (ie dodging AEs, or doing something to stop auto-attacking while maintaining their specials rotation)
b) There is a problem with the way the spreadsheet calculates white damage

And because I'm using Sarth WWS reports with 0 drakes (which can for all intents and purposes be a tank'n'spank), I don't think it's (a).

 

• landsoul

Jousting could be a possibility. I would only look at Patchwerk fights honestly. On Sartharion you are dodging lava waves. I can experiment with finding a different way to calculate heroic strikes and see if it would change, but I highly believe I threw the idea out long long long ago due to some kind of roadblock, but it would have been over a year ago so it's not in my direct memory.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

illusive_2008

Im sure you already know, but I know it's a very common and wanted craftable helm until people get into Naxx. So figured I would post Spiked Titansteel helm is being updated to Spiked Titansteel Helm - Changes were made to this item in patch 3.0.8. World of Raids Gallery
Since I am sure alot of people are using that helm currently . Thanks for the great sheet and all time you put to keep it accurate and updated.

Also with change to TG hit, I have been seeing alot of different comments on Rage starving and needed hit. Obviously there is no magically number. But do you feel 200-250 is going to be enough for raiding. Any insight on what one should aim for would be great to give me idea. Mostly for heroics and 10 man Naxx offspec for myself anyways. I have around 240 hit, unless told otherwise I feel that is pretty solid number after patch. But I am always double checking for feedback from full time DPS wars to confirm my ideas :P

 

• ebs2002

The reason I didn't look at Patchwerk fights is because at high gear levels, he just dies too quickly, and popped CDs skew the results (2mins with a bloodlust/deathwish won't be as accurate as a 3.5min fight, or a 4.5min fight, due to the CD). Besides, I believe most guilds don't actually have their white DPS affected by Lava Waves, if you position him right all you have to do is rotate around him.

My guess is that the roadblock you encountered was that you were using OpenOffice, that forced you to manually refresh each iterative calc, and factoring in heroic strikes to white damage requires a lot more iterations than anything else you're doing. In Office, you have iterative calculations automatically, up to 32k iterations, so this is no longer an issue.

I was playing with the sheet to get a more accurate model, and from what I found you would only need to make a few changes:

- WhiteDPS should multiply the MH dmg by %noHS
- Hit/oth should multiply the MH portion by %noHS
I haven't double-checked my numbers on the rage calcs, but that should be all you need.

Then, let it iterate a bunch and presto, your white DPS will have the heroic strike swings removed (and your heroic strike
swings will have decreased slightly as your rage is decreased from less white swings, etc etc.

 

• ebs2002

There is no magic number for hit. Your rage generation is affected by your hit, crit, str and haste. The best way to find out is to run with 240hit and see if you're starved :-p

 

• landsoul

No matter how I try to simplify the heroic strike calculations, they always diverge as I suspected. Ebs2002, If you can find some way to make it converge by your means, please tell me how.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

nponoBegHuk

uhm... Heroic Strike calculations should be easy once you set it's rage cost to something like
Or maybe I understood wrong the point

EDIT: Okay I read more posts, but still can't see where the problem is. I did a DPS warrior spreadsheet some time ago for myself, would be glad to contribute to your awesome work but you'd had to PM me with precise description of the problem(s) because my forum reading skills aren't great

 

• landsoul

The thing is that's how its done now. People were wondering if it was possible to not do it this way once the heroic strike frequency is determined, and have the total rage gained from white DPS modified off that percentage, and recalculate the rage available for heroics and also remove the RageGeneratedByAverageMainHandSwing cost of the strike. I say it is not possible, because I have tried it different ways this way all diverging. Someone said it was possible, but haven't shown how yet.

Visit my Youtube Channel for an increasing selection of warrior videos, including Undermanning, PvP, and LK Raids:
http://www.youtube.com/user/LandsoulWoW

 

• ebs2002

Okay, so you're right; once I set all the variables at once it started to diverge.

The solution I came upon, then, is to have a WhiteDPS temp value that the rage value looks at, and keeping the RageGenMH field's functionality. The actual white DPS should be the current formula with the MH portion multiplied by %noHS, but the rage generation formula should be based off how much white DPS you'd do if you did no HSes.

Doing it this way reports the same %HS freq that the current sheet shows, but will report a more accurate whiteDPS portion (~800 instead of 2000). This somehow is causing the SEP calcs to cause an Excel crash, and I don't know enough about debugging macros or excel macros in general to find out how/why it's happening.

One other thing you may want to include in the rage calcs is the cost of slam and deathwish, as these would take precedence over Heroic Strikes

 

Rallik

Could you add mixology with a flask to the AP Flask choices? It's 244 AP.