Picked up my warrior 2 days ago, respeced arms and I've been trying to learn the rotation and mechanics of arms... so far I think I'm doing alright but I have a few questions.
On heroic (boss) dummies I run 3 - 3.4k, I test to1 million damage each attempt.
Here's two semi-random attempts I'll post details on
#1 (3050 DPS)
DW 18%
OP 18%
Melee 17.7%
Slam 12%
MS 11.8%
Execute 10.3%
Rend 8.2%
WW 3.9% (Bladestorm, I didn't stance dance to hit WW)
#2 (3401 DPS)
DW 18%
Melee 17.9%
OP 14.9%
Slam 12%
MS 10.7%
Execute 10.6%
Rend 10.2%
WW 5.6%
__________________________
Glyphs: Rending, MS, Execution
My rotation:
Rend (I don't clip rend, I let it fully expire for last tick of tfb then immediately OP/reapply rend)
OP
Execution
MS
Slam (towards end of swing timer)
I never have the spare rage gen for HS it seems, so I haven't even bothered trying to rotate HS in as a rage dump.
Anyway, to my questions (Sorry I'm long winded)
1) Judging by the percentages and rotation, what do you guys think my problem is?
2) When is the proper time to use slam... I've seen so many people argue and disagree on this, I don't know who to believe anymore. I typically try to use slam toward the end of my swing timer so my swing and slam hit at the same time.. Is this wrong?
3) Also a controversial topic it appears, should I regem for ArP or wait for more ArP gear before regemming? Or is ArP gemming even worth it at all? Armory is here if you want to look at my gear. (974 STR/3669 AP, 155 ArP[12.59%], 41.2% crit)
Blue post confirms that ArP is working as intended. >Source<
I have no idea where it will lead for max dps, but I would most definetly hold off until you get some Uld25 gear with all that armor pen and a weapon like Earthshaper (1005 damage and 1.76% crit). Axe spec with str is still good and there are many more chances to get great 2h poleaxes and axes than there are maces.
The only thing I could ask for now is a reason to use one two-handed sword as a warrior. Voldrethar is downright sexy.
Last edited by Tremulant887 : 04/18/09 at 3:10 PM.
- Rend Is Kept Up (Refreshed if dropped)
- Procs:
-- Execute
-- Overpower
- Bladestorm (Always with Trinket)
- Mortal Strike
- Slam
- Heroic Strike (As a rage dump only)
With the above rotation kept in mind, I have conducted two tests. Each consisted of 1 battle shout duration (5 minutes). The first one was with the same stats listed in the quoted post, and the 2nd test was done with an added [Elixir of Armor Piercing]. (Again, only for 1 Battle Shout Duration). Both tests were conducted on the Heroic Training Dummy. DPS was recorded with the Recount addon and I have recorded after all DOTs were removed. No other player buffs (or debuffs) affected these tests.
================
Test 1:
DPS: 2517
1. Melee (25.2%)
2. Deep Wounds (Dot) (17.5%)
3. Overpower (15.5%)
4. Mortal Strike (14.0%)
5. Execute (9.8%)*
6. Rend (DoT) (8.5%)
7. Slam (5.2%)
8. Whirlwind (4.3%)
================
Test 2: (With ArP Elixir)
DPS: 2718
1. Melee (23.7%)
2. Deep Wounds (DoT) (17.5%)
3. Overpower (16.2%)
4. Mortal Strike ((14.3%)
5. Execute (13.3%)*
6. Rend (DoT) (6.5%)
7. Slam (4.4%)
8. Whirlwind (4.0%)
* Although it was on the Heroic Training Dummy, I only used Execute when Sudden Death Proc'd.
================
================
One thing I forgot to mention is that 99% of my gems are strength gems. My Meta is Crit and my shoulders are gemmed with expertise / hit. Other than those two I have either [Bold Scarlet Ruby] or [Sovereign Twilight Opal] in the sockets. Going off these two tests and what others have stated I would not be surprised if my DPS went significantly higher if I were to modify my gems just a bit to throw in some more ArP.
One more question on rotation. When boss is sub-20% I have been hitting Overpower if it is up and execute otherwise, unless my rage is above 50ish in which case I will MS and then execute. Is this optimal at all?
In order to fix Quartz so that it only shows your buffs make sure that you download Quartz r13. This is the latest version and I can confirm that it only shows your own debuffs again.
P.S. My main alt has and always will be an arms warrior hence why I am trolling here too
Sajakain, I am not convinced at all by your test. To be honest, I simply doubt that 45 ArP can increase your DPS by 200.
Looking at the numbers you provided, your execute damage % jumped from 9.8% to 13.3%, while your MS damage stayed pretty much the same.
Since I have no reason to believe that execute benefits more from armor penetration than mortal strike, I simply believe that you had more sudden death procs on the second test.
The decrease in slam damage also points in this direction, since more sudden death procs would lead to less free GCD to fill up with slam.
5 minutes is probably not enough time to see a real difference, too prone to RNG.
Sajakain, I am not convinced at all by your test. To be honest, I simply doubt that 45 ArP can increase your DPS by 200.
Looking at the numbers you provided, your execute damage % jumped from 9.8% to 13.3%, while your MS damage stayed pretty much the same.
Since I have no reason to believe that execute benefits more from armor penetration than mortal strike, I simply believe that you had more sudden death procs on the second test.
The decrease in slam damage also points in this direction, since more sudden death procs would lead to less free GCD to fill up with slam.
5 minutes is probably not enough time to see a real difference, too prone to RNG.
I would also agree with you. Something I have not posted: I did re-socket ArP. Guess what happened? DPS went down to 2300 - 2400. So while ArP may be nice, and could help, slightly at most, it's not the stat to go for. SD did proc more on the 2nd test, as the results show.
However, I would say a 5 minute fight would be as good as a 10 minute fight, going strictly off the Battle Shout CD. The only difference is time. You're still going to do the same rotation, just twice over. The only other difference that would come into play is the chance for more procs.
The question that remains, though, is how are some people putting out 3500 DPS when at best a handful of us can only put out, at best, 2500 - 2800? Is it just they're getting extremely lucky on procs? Or is it something else?
Rend
Bladestorm
SD
MS
OP
Slam when everything is on GCD
HS with overflow of rage
I did 10 separate 6 minute tests my lowest DPS was 3483 and my highest was 3772
My only buff was non imp BS
What were your procs like? What was the frequency of them? Your 54/17 doesn't differ from mine with the except of two talents. Your stats are just about the same as mine, save you have more AP and more Crit and between the two of us our gear balances out.... and to top it all off..our rotation is basically the same. But yet you're coming out with 1,000 more DPS?
Is this on the Heroic Training Dummy or the Grandmaster Training dummies? I can put out 3K+ on the Grandmasters.
Melee 19.3%
DW 15.6%
OP 15.5%
Slam 13.3%
MS 12.4%
SD 11.1%
Rend 7.9%
BS 4.7%
Total dps for this 10 minute fight was 3343.
I did three 10 minute attempts the others being 3302 and 3407dps, a little lower than when I did my six minute runs. Interesting RNG will always be RNG though I guess. My advice is that if you have over 30% ArP and have a side grade item or two with ArP on them + regemming it would be well worth it.
There is another arms warrior in my guild with that nice mace off 10man leviathan who does very similar dps on the dummies as well. ArP is the way to go.
EDIT: Sajakain Just from looking at the difference in my melee dmg and yours I am going to go out on a limb here and say that your problem may lie in not maximizing your GCD as good as possible. Also, I am using slam a lot more than you. Hope this helps.
Last edited by uglygreenguy : 04/19/09 at 12:01 AM.
I would also agree with you. Something I have not posted: I did re-socket ArP. Guess what happened? DPS went down to 2300 - 2400. So while ArP may be nice, and could help, slightly at most, it's not the stat to go for. SD did proc more on the 2nd test, as the results show.
You really can't base anything on a 5m parse; there's far too much RNG over that small of a sample size even just factoring crit chance and what crits before you even factor in things like SD procs.
The spreadsheet is using the correct armor formula based off the numbers GC posted, and it's really good... once you get enough of it, that is. It's showing at about 1.15 SEP on the spreadsheet when you're at ~700, which is pretty huge.
I have a question about how much ArP i should be going for exactly. I have Grim Toll and i remember reading and seeing some math that said i should just use Grim Toll and socket/gear ArP till im at the cap with when Grim Toll procs. So with Sunder + FF + Battle Stance i need ~916 ArP rating i think with Grim Toll. Take off 612 of that and im at 304 ArP rating which i pretty much hit with straight gear just about.
Thing is when i go to the spreadsheet and enter my setup and then convert the gems from STR to ArP i am still getting an increase in DPS. So with Grim Toll approximately when does socketing ArP become inferior to STR? the cap?
LVL 83
For a level 80 target, C=15232.5. For a level 83, C=16635
The cap for Armor Penetration then is: (armor + C)/3
the cap therefore is (13083+16635)/3 = 9906
you can essentially reduce the boss armor to 3177 with 100% ArP through equip unfortunately Blizzard didn't mention how exactly spec and stances are calculated into this. It could be that it is additive and just added into this calculation in which case you would just have a 100% ArP cap.
EDIT: Well I did some testing with a hunter pet and I have to say Blizzard is telling us a big pile of crap. As of now there is no real ArP cap.
Here are the results:
The pet had 18830 armor at lvl 80 putting the cap at 11354 armor reduction or 60.29% with Blizzards formula.
This would put the pet capped at 7476 armor. I used buff food and pots to reach 52.05% ArP through equipment and Thunderclap as damage since it's very constant
Alone Grim Toll puts me above 100% ArP and ever ArP effect still pushes damage above any possible cap. I am pretty sure that bonuses like stance or spec or sunder armor are totally seperate from ArP from your equip. So you can essentially reach the ArP cap Blizzard mentioned by reaching 100% ArP and reduce the rest through Sunder Armor. I definetly have to do more testing on this.
Basically what I'm looking for is what is the lowest attack power/crit you can have after gemming arp before it will get better than gemming strength. for example as axe spec with str gems, I have 3379 ap, 36.94 crit, and my hit is a little high but thats cuz of gear, nothing gemmed.
I calculated how much ap and crit i would lose after going mace/arp gems, and i would be sitting around 29.5 crit with 3091 ap, yet have 287 armor pen rating, not including mace spec or battle stance. I'm wondering if this would be too low to have a significant dps increase.
Right now I'm sitting at 3246 AP, 39.43% crit, 230 hit (/gg Draenei racial), and 515 ArP rating without mace spec. With ArP food I'm at 545 rating which is like 45% or so. I wasn't quite sure what to expect when I switched all my gems but so far it seems to be doing it's job. I'm still quite steadily in top 5 damage atm.
Oh and another thing for those who never got the ring from 10 man Sarth 2D (like me, fuck you Sarth) the crafted JC ring with a 27 gem gets you a whopping 72 rating.
Since we are in battle stance, 10% is automatically taken of and we get 94.05% to cap (without the Grim Toll proc)
With Grim Toll proc: We would only need 44.36% arp in battle stance (no mace)
44.36 * 12.32 = 546.52 Arp rating
So, as Mace spec that would put the cap at:
546.52 - (15% * 12.32 = 184.8) = 361.72 from items, with Mace Spec, right?
If this is the case, what are the thoughts on not putting 5/5 into Mace spec, but only putting 3/5 in? 546.52 - (9% * 12.32 = 110.88) = 435.64 -- Raises the amount of ArP you can gain from gear (since it's almost literally everywhere), and putting the extra 2 points into something like Weapon Mastery (if expertise is needed). I know it's off the standard train of thought, but it'd allow for more unique itemization, depending on what type of gear is available in ulduar (Since we obviously don't have all of the gear discovered yet).
546.52 - (15% * 12.32 = 184.8) = 361.72 from items, with Mace Spec, right?
If this is the case, what are the thoughts on not putting 5/5 into Mace spec, but only putting 3/5 in? 546.52 - (9% * 12.32 = 110.88) = 435.64 -- Raises the amount of ArP you can gain from gear (since it's almost literally everywhere), and putting the extra 2 points into something like Weapon Mastery (if expertise is needed). I know it's off the standard train of thought, but it'd allow for more unique itemization, depending on what type of gear is available in ulduar (Since we obviously don't have all of the gear discovered yet).
If you have Sunder + FF + Mace Spec + Battle Stance you get 49% ArP
That leaves 51% to get which is roughly 628.116 from gear/buff food contributions which will be pretty difficult to do while maintaining the hit cap etc. just from what gear that's been discovered so far.
I am pretty certain you should stick to 5/5 mace. At least for the time being.
EDIT: Although Grim Toll is now one of the best trinkets in the game if you could get within 20% of ArP cap and can fill your trinket slots with Greatness and one of the many nice trinkets from 10/25Ulduar I think you would see better returns.
If you have Sunder + FF + Mace Spec + Battle Stance you get 49% ArP
That leaves 51% to get which is roughly 628.116 from gear/buff food contributions which will be pretty difficult to do while maintaining the hit cap etc. just from what gear that's been discovered so far.
I am pretty certain you should stick to 5/5 mace. At least for the time being.
EDIT: Although Grim Toll is now one of the best trinkets in the game if you could get within 20% of ArP cap and can fill your trinket slots with Greatness and one of the many nice trinkets from 10/25Ulduar I think you would see better returns.
Going with boss armor at 10643, we get sunder and ff applied before armor pen cap is calculated. Armor is then reduced to: 10643 * .8 * .95 = 8088.68
(8088.68+15232.5)/3 = 7773.56 (104.05% to cap)
In battle stance and fully mace spec, 25% is subtracted from the cap which leaves 79.05%. When grim toll procs, it gives 612 armor pen rating or 612 / 12.32 = 49.68%.
79.05 - 49.68 = 29.37% or 29.37 * 12.32 = 361.83 armor pen rating
So if the boss armor is 10643 and the debuffs to armor such as sunder and ff are applied before the armor pen cap is calculated, then in Battle Stance with Mace Spec 5/5 and Grim toll, the armor pen rating needed to cap is 361.83 or 29.37% armor pen.
Now what about Shattering Throw? 20% armor removed for 10 seconds. Since it is a debuff on the boss, I would assume that would reduce the boss's armor an additonal 20% when used. Then armor pen cap is re-calculated? If so, seems like you would time that for additional damage during Heroism.
If it works that way then the new cap for that 10 seconds would be:
10643 * .8 * .95 * .8 = 6470.94
(6470.94 + 15232.5) / 3 = 7234.48
6470.94 = (7234.48 X) / 100 WHERE X is the Percent needed to Cap
647094 = 7234.48 X
X = 89.45%
Then 35% from Battle and Mace 5/5: 89.45 - 35 = 54.45% armor pen or 670.82 armor pen rating
So if you can time that after your first or second grim toll proc, you could maximize your armor pen for the boss fight if not cap without grim toll.
If by some lucky chance you have all that and grim toll up, then you would need:
Basically what I'm looking for is what is the lowest attack power/crit you can have after gemming arp before it will get better than gemming strength. for example as axe spec with str gems, I have 3379 ap, 36.94 crit, and my hit is a little high but thats cuz of gear, nothing gemmed.
I calculated how much ap and crit i would lose after going mace/arp gems, and i would be sitting around 29.5 crit with 3091 ap, yet have 287 armor pen rating, not including mace spec or battle stance. I'm wondering if this would be too low to have a significant dps increase.
Dont worry dude, My reply with a very similar question got deleted, nps.
I investigated a little bit a few days ago, to get familiar with ArP and its new mechnaics.
For that I skilled fury (with Unending Fury), to be able to use BT for testing. (It's damage varies only by +-1)
I equiped two 1h Swords and either gear with no ArP on it or gear with 9,25%.
I made 8 tests on the Heroic Training Dummy.
1. completly no ArP
2. ArP over gear
3. only SA (5 stacks)
4. only Battlestance
5. ArP over gear with SA (5 stacks)
6. ArP over gear, SA (5 stacks) and Battlestance
7. SA (5 stacks) and Battlestance
8. ArP over gear and Battlestance
#
AP
Damage befor mitigation
Damage done
Mitigation
1
3259
1792.45
1055
41.14%
2
3471
1909.05
1159
39.29%
3
3259
1792.45
1149
35.84%
4
2859
1572.45
957
39.08%
5
3471
1909.05
1263
33.84%
6
3057
1681.35
1152
31.48%
7
2859
1572.45
1043
33.61%
8
3057
1681.35
1057
37.07%
Then I played with an few possible models in calculating those mitigations with 10643 armor.
In the end following brought me exact numbers:
with
C = 400 + 85 * targetlevel + 4.5 * 85 * (targetlevel - 59) (out of the blue post for the half armor value)
A = armor
x = sum of arpen buffs(mace spec/battle stance/gear arpen)
y = product of arpen debuffs(sunder/ff) (1 - 0.95 * 0.8)
My problem is, if I want to get correct numbers with 10643 armor, I have to use for C = 15232.5, but this would mean that I attack an level 80 mob, not an level 83 boss.
Do I want to use for C = 16635 then the boss-armor would have to be something arround the area of 11620 to 11640
Any suggestions or perhaps affirmations for the armor?
I investigated a little bit a few days ago, to get familiar with ArP and its new mechnaics.
For that I skilled fury (with Unending Fury), to be able to use BT for testing. (It's damage varies only by +-1)
I equiped two 1h Swords and either gear with no ArP on it or gear with 9,25%.
I made 8 tests on the Heroic Training Dummy.
1. completly no ArP
2. ArP over gear
3. only SA (5 stacks)
4. only Battlestance
5. ArP over gear with SA (5 stacks)
6. ArP over gear, SA (5 stacks) and Battlestance
7. SA (5 stacks) and Battlestance
8. ArP over gear and Battlestance
#
AP
Damage befor mitigation
Damage done
Mitigation
1
3259
1792.45
1055
41.14%
2
3471
1909.05
1159
39.29%
3
3259
1792.45
1149
35.84%
4
2859
1572.45
957
39.08%
5
3471
1909.05
1263
33.84%
6
3057
1681.35
1152
31.48%
7
2859
1572.45
1043
33.61%
8
3057
1681.35
1057
37.07%
Then I played with an few possible models in calculating those mitigations with 10643 armor.
In the end following brought me exact numbers:
with
C = 400 + 85 * targetlevel + 4.5 * 85 * (targetlevel - 59) (out of the blue post for the half armor value)
A = armor
x = sum of arpen buffs(mace spec/battle stance/gear arpen)
y = product of arpen debuffs(sunder/ff) (1 - 0.95 * 0.8)
My problem is, if I want to get correct numbers with 10643 armor, I have to use for C = 15232.5, but this would mean that I attack an level 80 mob, not an level 83 boss.
Do I want to use for C = 16635 then the boss-armor would have to be something arround the area of 11620 to 11640
Any suggestions or perhaps affirmations for the armor?
I am assuming BT means bloodthirst otherwise what I write would be a fallacy. Your test 1,2,3 and 5 are in battle stance or in berserker. If in berserker wouldn't you get more str than battle? Therefore your bloodthirsts would be inaccurate sinces it scales with attack power.
Yes, sorry, BT means Bloodthirst.
And as you can see, I listed AP.
Not only the berserker stance influenced my AP in the tests, but also the different gear did.
And the damage befor mitigation varies throught that, too. (AP*0,5*1,1 with Unending Fury)
Don't worry, I didn't use always only the same damage from test #1.
Besides, I didn't wear anything which could proc. The only proc in the test was Rampage but that's of course uninteresting.
Furthermore, it was 05:00 o'clock in the morning so nobody but me stood at the dummy and could have destroyed my test results by debuffs.
Edit:
But i found an mistake:
"y = product of arpen debuffs(sunder/ff) (1 - 0.95 * 0.8)" is incorrect
y = product of arpen debuffs(sunder/ff) (0.95 * 0.8) is correct
My problem is, if I want to get correct numbers with 10643 armor, I have to use for C = 15232.5, but this would mean that I attack an level 80 mob, not an level 83 boss.
Do I want to use for C = 16635 then the boss-armor would have to be something arround the area of 11620 to 11640
Any suggestions or perhaps affirmations for the armor?
Armor mitigation is based on attacker level, so it makes perfectly sense that the value used in calculation is YOUR level and not the boss level.
04/18/09, 2:23 PM
Similar Threads
