First of all I'm not sure if this is the best thread for me to post this in because it does not directly have to do with the spreadsheet but because it concerns both the fury and the arms tree I thought it was best to post here.
After seeing the BiS gear list of both fury and arms I noticed that they don't vary much from each other. So it crossed my mind to dual spec for both a fury and arms raiding spec.
Now I have gone through the fury dps and arms dps threads and this spreadsheet thread and tried to figure out which fights favor one or the other spec.
Sofar I found that stationary fights like xt-002 favor arms dps. I myself raid mainly as fury but I find the idea fun and challenging to change during a raid to an arms spec if that gives better dps.
So my list would be: fury:
razorscale
ignis
iron counsil
hodir
thorim
freya
general vezax?
Basically gear-wise I dropped about 90 AP, 52 hit (unbuffed) to gain 45 ArP, about 0.9% crit, and 121 haste, which the spreadsheet was showing should be a 102 dps increase. I'll see what testing I can do tonight.
A loss of 52hit could change your heroic striking considerably; is it possible that you were HSing too frequently and rage-starving yourself?
I've found that any time my hit changes drastically, it takes me a day or so to get back into the rhythm of when to HS without thinking about it. Even if I'm HSing when I'm supposed to, my rotation can lag if I'm thinking about it too much.
Hello everybody,
I'm a new member, I'm trying to follow up with the posts about fury dps, I'm not doing great following up due to lots of things to do beside wow and reading forums. I have a question, as i couldn't understand the spreadsheet which i will try to find a post on how to read it, my question is that gear in the spreadsheet is it confirmed its the best dps gear so far? if not, what is the suggested best dps gear?
Finally, as we warriors depend on hit rating, my hit rating now is 209 which i know its low and i should increase it, but i was looking at ulduar gear, should i gem it as we used to do, strength? or use hit gems till i get to the hit cap again?
Hello everybody,
I'm a new member, I'm trying to follow up with the posts about fury dps, I'm not doing great following up due to lots of things to do beside wow and reading forums. I have a question, as i couldn't understand the spreadsheet which i will try to find a post on how to read it, my question is that gear in the spreadsheet is it confirmed its the best dps gear so far? if not, what is the suggested best dps gear?
In regard to your hit rating question, in almost every situation, you should do your best, either with gear or gems, to get to the yellow hit cap for your spec. It would behoove you to continue playing with the spreadsheet until you get the hang of it. It is a great tool and worth the time it takes to learn how to use it.
Finally, as we warriors depend on hit rating, my hit rating now is 209 which i know its low and i should increase it, but i was looking at ulduar gear, should i gem it as we used to do, strength? or use hit gems till i get to the hit cap again?
First of all, Warriors are no more dependant on Hit Rating than other mellee classes. I know some will contradict me, citing smooth rage generation as a reason, but above the soft cap, hit is not a warrior's main issue.
With 209 Hit Rating you are definitely not low on Hit Rating. As your Alliance-side, you should sit slightly below 10.5% to Hit with that. That's 2.5% above the soft cap and should be sufficient for reasonable HS Usage. At that amount of Hit Rating you should definitely NOT gem for Hit.
Even if you were only slightly above the soft cap, you should not gem for Hit. While it does make rage generation smoother and can prevent messing up your rotation due to miss streaks, it is worse than almost any other DpS-stat point for point (above the soft cap). In almost all cases, you can prevent messing up your rotation by using HS intelligently instead of mashing the button. If you want 100% HS usage in fights without incoming rage, Hit is probably the way to go at pre Ulduar gear level, but that will most likely not give you the maximum DpS.
Depending on your present gear, you should gem either for Strength oder for ArP, although with a few pieces of Ulduar Gear, ArP is very likely the way to go, especially if you're using Grim Toll or the Mjollnir Runestone.
But use the spread sheet to determine which stat is better at your level.
I've actually noticed that lately as well and it's been puzzling me. I swapped some gear out for last night's raid, that according to the spreadsheet should've changed my theoretical dps from 5235 to 5337, yet my dps on bosses dropped from the 4600 range to around 4000. It was really confusing, all raid buffs should've been the same (unless someone stealth-respecced), and other dps classes didn't seem to have problems. Unfortunately I was the only dps warrior (we have a feral druid & combat rogue), so I didn't have anyone else in for comparison. Tonight I'll pull my old gear out of the bank and try to compare more thoroughly. My latency seemed consistent, it's always in the 60-90 ms range. I had no rage problems with either gearset, I was able to keep up HS nearly 100%. It's possible I was hitting Berserker Rage more often than before to maintain rage, but I'm not sure.
Basically gear-wise I dropped about 90 AP, 52 hit (unbuffed) to gain 45 ArP, about 0.9% crit, and 121 haste, which the spreadsheet was showing should be a 102 dps increase. I'll see what testing I can do tonight.
Relying on "empiric" observations is probably not the best way to go when monitoring your DpS-output. Especially if it's only for one night of fights. Log your fights and compare the numbers, and by that I don't mean the raw DpS-numbers but the reasons for differences in DpS. Did you miss some specials due to white hit miss streaks and rage starvation, HS usage, length of fights (this is a major factor for fury warriors), unlucky Crit percentage, missing buffs/debuffs, etc.
If the loss of 52 Hit does not take you below the soft cap, I see no way that trading 90 AP and 52 hit for 45 ArP 0.9% Crit and 121 Haste can lead to a loss of DpS, especially not 600 DpS. Something else must be seriously wrong.
anything new about excel solver to put in the best suited gems for max dps??
If anybody could tell us the right fields and secondary terms solver would be the ultimate tool for gems in our spreadsheet...
AND: Landsoul, you should impove the dropdown menu from the item selection with telling which items we alredy heave.
e.g. the drop down looks then for for helemts we have like.... all items we currently have in our inventory are marked with (x)! so its easier to try different items.
Guise of the Midgard Serpent (x)
Obsidian Greathelm
Warhelm of the Champion (x)
Garona's Guise (x)
Conqueror's Siegebreaker Helmet
Helm of the Furnace Master(x)
Valorous Siegebreaker Helmet
Well, a nice solution would be a simple checkbox right to every item in the lookup tables... that automaticly adds "(x)" after the name if you activate the checkbox...
I have not enough excel skills to realize it but maybe you can...
These are two inferior enhancements; any other enchant is better, and there's an orange 8 strength + 8 hit gem cut.
But, if you want to see it, you can use the phoney stats section to add the stats you're missing.
I've actually noticed that lately as well and it's been puzzling me. I swapped some gear out for last night's raid, that according to the spreadsheet should've changed my theoretical dps from 5235 to 5337, yet my dps on bosses dropped from the 4600 range to around 4000. It was really confusing, all raid buffs should've been the same (unless someone stealth-respecced), and other dps classes didn't seem to have problems. Unfortunately I was the only dps warrior (we have a feral druid & combat rogue), so I didn't have anyone else in for comparison. Tonight I'll pull my old gear out of the bank and try to compare more thoroughly. My latency seemed consistent, it's always in the 60-90 ms range. I had no rage problems with either gearset, I was able to keep up HS nearly 100%. It's possible I was hitting Berserker Rage more often than before to maintain rage, but I'm not sure.
Basically gear-wise I dropped about 90 AP, 52 hit (unbuffed) to gain 45 ArP, about 0.9% crit, and 121 haste, which the spreadsheet was showing should be a 102 dps increase. I'll see what testing I can do tonight.
Who is usually applying your major armor debuff? (Though you did specify that you were the only 'dps' warrior, so perhaps you do have a prot' who was there in both scenarios.) Just a thought, when I saw 'only' and 'warrior'. Losing Sunder would be a pretty dramatic DPS loss, if that was in fact the case.
I just wrote an VBA Macro which translates the lookup list into any other language, you only need a new sheet with the english word and the equivalent translation.
The macro works but every time the original names are changed (even without macro - changing by hand the name) the SEP calculation stops with error #NV ??
Whats going on? why is the name not changeable sometimes?
If we can get this working, a new sheet in your spreadsheet can translate all items into any desired language.
Pls Help. Why does sep calc stop when changing names in Lookup table?
EDIT:
I tested it. some items can be renamed successfully.
But e.g. renaming "Might of the Leviathan" into any other name makes the #NV error. there are other items i cannot specify now but why the hell is this??
I found a problem with gem selection (possibly only '97/03 excel) a few builds ago and thought maybe it would have been corrected with the latest version, however, it persists.
When placing a +20 or some of the +10/10 gems into the leg's first gem slot, various N/A errors occur across the sheet.
This happens with Valorous Dreadnaught Legplates, Plated Leggings of Ruination, Legguards of Cunning Deception, and Conqueror's Siegebreaker Legplates.
I am sure these are not the only pieces that can induce this error; I wanted to rule out the error being caused by the item itself.
The following gems will cause the error.
+20: arp, exp, str, crit, haste, and hit
+10/10: agi/crit, agi/hit, exp/hit, str/crit, str/haste, and str/hit
I would like to add/reiterate that this only occurs with the first gem slot and not subsequent slots and is for the leg section only.
I've been using your spreadsheet for quite a while, and it's always been a really good measure for me - at least until recently. As you well know, the gear in non-hard mode Ulduar has the general theme of hit rating and armor penetration, and while equipping the pieces shows a nice increase on your spreadsheet, I'm finding my DPS is falling off drastically.
I've tested many gear setups with high hit and low hit - and the high hit setups were almost never worth the DPS shown in the spreadsheet. My speculation on that is, that the spreadsheet accounts for fight durations of 10 to 20 minutes. In 10 to 20 Minutes hit rating may be worth what the spreadsheet told you, because you miss a lot less on whitehits. But in shorter fights, there's not such a big possibility to miss lot of whitehits. That's leading to a possibility of DPS loss due to more or less useless hitrating (over the style cap).
I've tested many gear setups with high hit and low hit - and the high hit setups were almost never worth the DPS shown in the spreadsheet. My speculation on that is, that the spreadsheet accounts for fight durations of 10 to 20 minutes. In 10 to 20 Minutes hit rating may be worth what the spreadsheet told you, because you miss a lot less on whitehits. But in shorter fights, there's not such a big possibility to miss lot of whitehits. That's leading to a possibility of DPS loss due to more or less useless hitrating (over the style cap).
Absolute Bullshit you write here. Shorter Fights = More RNG dependant -> less Hit = higher difference between max and min DPS.
Lets say you could Run with 0 Hitrating in a 1 Minute Fight. you have around 60 Attacks (as MS). You can have a lucky try and 0 misses. Or you Miss 20 Times. Dps difference = 50% or 33% (only depends on which dps number you take to compare). If you run a 10 Minute Fight its more likely to miss around 48 Times out of 600 Attacks with 0 Hitrating.
That just wasn't the point. I'm aware of what you've written. However, it's one possibility why he lost DPS, because the additional hit rating he gained didn't helped him much.
I've tested many gear setups with high hit and low hit - and the high hit setups were almost never worth the DPS shown in the spreadsheet. My speculation on that is, that the spreadsheet accounts for fight durations of 10 to 20 minutes. In 10 to 20 Minutes hit rating may be worth what the spreadsheet told you, because you miss a lot less on whitehits. But in shorter fights, there's not such a big possibility to miss lot of whitehits. That's leading to a possibility of DPS loss due to more or less useless hitrating (over the style cap).
RNG is RNG. You are going to have the same "chance" to miss no matter what the duration is.
RNG is RNG. You are going to have the same "chance" to miss no matter what the duration is.
chances don't change based on how long the fight lasts... ^^^ correct, having more or less hit rating doesn't matter for longer or shorter fights? im not sure where anyone gets the idea that it does, you will still have the same chances.
The big thing that ive been wondering is, most of the dps warriors that are posting know that ArP > STR and yet im seeing hardly any ArP socketed or implemented even on those discussing how much better ArP is than STR. Is there a reason behind this? Just afraid the calculations are untrue? or dont have enough ArP gear to make gemming effective? im sitting at 40% or so unbuffed without battle stance and im still in mainly naxx/OS/maly gear (yea a month of farming ulduar and one warrior tier piece has dropped).
The big thing that ive been wondering is, most of the dps warriors that are posting know that ArP > STR and yet im seeing hardly any ArP socketed or implemented even on those discussing how much better ArP is than STR. Is there a reason behind this?
I think this is often because those warriors play both Arms and Fury, and want to keep str gems in any item that they use for Fury, and this is usually quite a few bits.
I think this is often because those warriors play both Arms and Fury, and want to keep str gems in any item that they use for Fury, and this is usually quite a few bits.
I believe it's also because a tremendous amount of ArP seems to be required to make it more worthwhile -- so it may be so with BiS gear, but perhaps not with whatever set is currently being run.
Had an idea last night and wanted to use your Spreadsheet to find out. Unfortunatly, you don't have "Executioner" as an option for weapon enchants.
So looking at your setup, and trying the best I could to understand it, I think I managed to simulate having an Executioner enchant.
This is where I'm wondering if I've made an assumption, or have a misconception about ArP, so do rip it apart if you find fault.
Using the same formula for Mongoose to figure out Uptime (both are 1.2 PPM), I calculated
Uptime*ArP rating/Amt per 1%
0.59638*120/12.316238841
Which works out to be 5.81% average extra armor pen.
(Weapon Speed 3.5, 0 haste, specced for Blood Frenzy)
I made the character completely nude, gave him [Ironsoul], but reduced the stats to 0. (using the phoney stats table)
So it came out to just being a naked Mace specced Human Warrior with a no stats Ironsoul. (I tried with other weapons, and still noticed a dps increase with executioner over massacre)
I then added "+IF(C19="None",X111*(120/AL61/100),0)" (This allows for easy comparison between Executioner and the other enchants already available) at the end of the formula in cell C62 (Average ArP) where
X111 is the Uptime formula
AL61 is the ArP value per 1%
Conclusion:
Executioner is better than Massacre for an Arms Warrior. This could potentially benefit fury warriors, though I think this proof enough to get Executioner into the Spreadsheet so further testing can be done with the Fury Sheet.
Can someone double check that I haven't screwed up? (Beserker will always be better than executioner, barring only theoretical situations, but perhaps a return to PvP weapons? At least for Arms based on this.)
When I tested it, berserking was always better than executioner. I left out executioner in current builds because of the many different penetration procs available and the notion that there was a cap to armor penetration, which isn't anymore. If I get time to rework the armor penetration system to where I can include all armor penetration procs (not just 2 at a time) then I will do it.
For now, just assume berserking is better for arms and fury in both pvp and pve situations.
You may have the same chance to miss. Yet how often you miss is dependent on RNG and the amount of samples or duration. Didn't we all learn this in 3rd grade. A classroom of kids each tossed a coin in the air. 1 kid got heads 78 times. Another got tails 80 times. All sorts of different ranges here. When adding up the 30 kids with 100 throws each for 3,000 throws total our number came out to 43% tails. This shows exactly how much difference there is in RNG. The less samples the greater the difference can be. It also shows a quite large sample is needed for it to work out right. A 3,000 sample wasn't high enough in our test. We repeated this test 4 times for a total of 12,000 samples and got 48% tails.
A computers ability to generate random numbers is based on what program and platform they're using. For example windows using php rand actually generates a repeating pattern. Where as Linux using a bitmap generator does not. Not exactly a fair comparison as one is a True RNG and the other a Pseudo RNG. Are they using 16 zeros and ones or a 0-100 system. Blizzard says they use 0-1 But i take it to mean either .001-1 or .0000000000000001-1 or something in between. Corrective algorithms can be used to help offset the incorrectness of the RNG to be more consistent. Doubt Blizzard does this. Without a corrective algorithm even a True RNG after an extremely large sample size it can still be off as much as 0.725%. These are only a few examples of the many possible factors involved with RNG.
So a short duration fight you may get unlucky with RNG and miss more than usual. That is a fact. What makes it worse is when you miss back to back. It's the 2 offhand misses in a row that screws up fury rotation. I even get 3 in a row now and then. You can decrease this chance of happening by increasing your hit rating. This is the main benefit of going with a higher hit rating to reduce this from happening. You use talents for rage generation. You use hit rating for rage generation. You combine the 2 in a way to make rage generation consistent. If it isn't consistent you fix one or the other or both.
Too much hit can also be a waste. To much rage generation due to talents can also be a waste. You then need to find a sweet spot. Enough consistent rage generation to not screw up your rotation. You then swap gear or talents for a dps gain while still being in this sweet spot. Crit and ArP also effect this as do other things. Rage and the control of it is our biggest determining factor for dps.
Last edited by catch22atplay : 05/10/09 at 11:43 PM.
05/07/09, 1:46 PM
due to lots of things to do beside wow and reading forums. I have a question, as i couldn't understand the spreadsheet which i will try to find a post on how to read it, my question is that gear in the spreadsheet is it confirmed its the best dps gear so far? if not, what is the suggested best dps gear? 
Similar Threads
