Elitist Jerks
Register
Blogs
Forums


Go Back   Elitist Jerks » Class Mechanics » Death Knights

Closed Thread
 
LinkBack Thread Tools
Old 08/07/09, 12:49 PM   #1111
Consider
King Hippo
 
Consider's Avatar
 
Draenei Death Knight
 
Dragonblight
Originally Posted by slant View Post
Running the simulator with my character's stats (ulduar25 without hardmodes) rather than one of the specs provided in the sim, 3/13/55 beat 0/17/54 by roughly 50 DPS with all three sigils. The additional thread reduction and runspeed made it an easy decision.

This is a great example why it's important that everybody run the simulator themselves rather than blindly adopting the cookie-cutter spec, as that spec may not be optimal for your gear level.
It's also a great example of why you should post numbers/parses/simulations yourself, when you disagree with what everyone else has already shown. All of the other simulations have shown that 0/17/54 is superior to 3/13/55 (until you drop 2pt8 and the SotVH, anyways). Perhaps you did something wrong with your simulation? Perhaps your gear is from being specced blood, and thus is stacking armor pen to a high amount, which although not as inferior as it used to be, still isn't the most optimal route (and not what most Unholy players are coming from)? Perhaps this, perhaps that.

Proof/support. Gotta have it, and 0/17/54 does. Everything everyone has done (and posted) shows 3/13/55 being subpar, and that may not be the case, but thus far, it would seem so, and one can't claim otherwise without numbers.

United States Offline
Old 08/07/09, 1:02 PM   #1112
slant
Don Flamenco
 
Draenei Shaman
 
Drenden
Certainly. I hope this answers your questions. I would appreciate any assistance if I am in fact running the sim incorrectly.

I'm running DKsim 0.9.9, in blood presence and fallen crusader enchant, 200ms latency, 48 hour simulation times, with draenei aura and bloodlust deactivated. I tried all three sigils and ran each sim multiple times.

Templates:
3/13/55
0/17/54

Character (Ulduar25 non-hardmode geared, unholy 3.1 optimized with little ArP)
<character>
	<stat>
		<Strength>1604</Strength>
		<Agility>125</Agility>
		<Intel>43</Intel>
		<Armor>13851</Armor>
		<AttackPower>235</AttackPower>
		<HitRating>259</HitRating>
		<CritRating>744</CritRating>
		<HasteRating>237</HasteRating>
		<ArmorPenetrationRating>180</ArmorPenetrationRating>
		<ExpertiseRating>184</ExpertiseRating>
	</stat>
	<weapon>
		<count>1</count>
		<mainhand>
			<dps>232.5</dps>
			<speed>3.4</speed>
		</mainhand>
		<offhand>
			<dps>1</dps>
			<speed>1</speed>
		</offhand>
	</weapon>
	<EP><base>50</base></EP>
	<Set>
		<T72PDPS>0</T72PDPS>
		<T74PDPS>0</T74PDPS>
		<T82PDPS>1</T82PDPS>
		<T84PDPS>1</T84PDPS>
		<T92PDPS>0</T92PDPS>
		<T72PTNK>0</T72PTNK>
		<T74PTNK>0</T74PTNK>
		<T82PTNK>0</T82PTNK>
		<T84PTNK>0</T84PTNK>
	</Set>
	<ChaoticSkyflareDiamond>1</ChaoticSkyflareDiamond>
Priority:
<BloodPlague></BloodPlague>
<FrostFever></FrostFever>
<Obliterate></Obliterate>
<BloodStrike></BloodStrike>
<DeathCoil></DeathCoil>

United States Offline
Old 08/07/09, 1:20 PM   #1113
Woy
Glass Joe
 
Blood Elf Death Knight
 
Bleeding Hollow
Slant, have you tried running the sim again with the DeathCoilMaxRp priority from post 1006?
Unholy Gains Shadow of Death (Unholy DPS Discussion now with more 3.2!)

It might make the difference.

Offline
Old 08/07/09, 1:31 PM   #1114
Loothorde
Glass Joe
 
Orc Death Knight
 
Kargath
Originally Posted by slant View Post
<offhand>
<dps>1</dps>
<speed>1</speed>
</offhand>
Doesn't the 1 in dps factor in Dual wielding? I have that value set to 0. I would assume this would make all your simulations invalid.

EDIT/CORRECTION: It appears the new versions of the sim set it to 1. I was using an old xml file that I copied forward where the value was 0.

Last edited by Loothorde : 08/07/09 at 1:33 PM. Reason: I wasn't fully informed

Offline
Old 08/07/09, 1:39 PM   #1115
Nefiir
Von Kaiser
 
Undead Priest
 
Eonar
Slant, have you tried the sim at 5/5 Necrosis (taking 3 points out of Morbidity and 1 out of Corpse Explosion) using your 3/13/55 spec?

Last edited by Nefiir : 08/07/09 at 1:42 PM. Reason: clarified spec I was referring to

Offline
Old 08/07/09, 1:39 PM   #1116
slant
Don Flamenco
 
Draenei Shaman
 
Drenden
I ran each simulation three times using SotVH. The DeathCoilMaxRP priority change brings 0/17/54 roughly equivalent to 3/13/55 with my gearset and chosen specs. It doesn't impact 3/13/55, assumedly since RP generation is lower it's in less danger of being capped.

3/13/55 still wins for me due to threat reduction and movement speed. I'm not saying it's the spec for 3.2-- I'm just saying that everybody should run the sim using their individual character's stats and the spec they'll actually use in game (for example, I choose to invest a point in corpse explosion) and not rely upon the character templates that come with the program.

FYI, dualwielding is actually determined by the <count> variable under <weapon>. If it's set to 2, you're dualwielding.

0/17/54, DeathCoilMaxRP priority from #1006: 5850dps
0/17/54, Normal priority from #1112: 5815dps
3/13/55, DeathCoilMaxRP priority from #1006: 5850dps
3/13/55, Normal priority from #1112: 5850dps

United States Offline
Old 08/07/09, 1:43 PM   #1117
Consider
King Hippo
 
Consider's Avatar
 
Draenei Death Knight
 
Dragonblight
@ Slant | Using the stats you provided, I ran the simulation for each spec. I did use the whole DC max priority setting, as mentioned above. I think I also max Necrosis over Desolation in my specs, but if I do, I do so in both, so it shouldn't influence how the two appear relative to one another (and, besides, you had a point in CE! =p). I did it over 200 hours. The results are as follows, first 3/13/54, second 0/17/54.

Ability Total % Landed Hit% Crit% Miss% Average
Obliterate 88641553819 135414 50 49 0 6545
Plague Strike 76582570 1 34120 52 47 0 2244
Icy Touch 83679246 1 34018 67 31 0 2459
Blood Strike 194668035 4 67694 50 49 0 2875
Death Coil 649364365 14 108354 59 40 0 5992
UB 129882251 2 108354 100 0 0 1198
Frost Fever 235484834 5 221771 100 0 0 1061
Blood Plague 235765391 5 221533 100 0 0 1064
Necrosis 104618896 2 284418 100 0 0 367
Blood Caked Blade 130447955 2 85180 100 0 0 1531
Wandering Plague 186976441 4 175898 99 0 0 1062
Main Hand 871440457 19 284418 59 40 0 3063
Ghoul 522701353 11 664020 87 12 0 787
Gargoyle 200003128 4 68544 86 12 0 2917
DPS 6261        
Total Damage 4508030460 in 200 h        
Threat Per Second 3186        
Generated in 169s        
Template :03-13-55.xml(C:\Users\Zack\Documents\Downloads\DKSimulator0.9.9\DKSimulator0.9.9\Templates\03-13-55.xml)        
Priority :Unholy.xml(C:\Users\Zack\Documents\Downloads\DKSimulator0.9.9\DKSimulator0.9.9\Priority\Unholy.xml)        
Presence :Blood        
Sigil of :VengefulHeart        
RuneForge :FallenCrusader        
Pet Calculation :True        

And
AbilityTotal % LandedHit%Crit%Miss%Average
Obliterate198865737917330738594006012
Plague Strike179399359183071524702159
Icy Touch200786809183030673102418
Blood Strike4377780323165466584102645
Death Coil185946248916316865594005868
UB 371857208 3 316865100 0 0 1173
Frost Fever 577275407 5 555532 100 0 0 1039
Blood Plague 579074456 5 555055 100 0 0 1043
Necrosis 427631153 3 711044 100 0 0 601
Blood Caked Blade 318657624 2 213514 100 0 0 1492
Wandering Plague 458879332 4 440742 99 0 0 1041
Main Hand 2138330103 18 711044 58 41 0 3007
Ghoul 1306387536 11 1660050 86 13 0 786
Gargoyle 479329573 4 164342 86 12 0 2916
DPS 6291        
Total Damage 11323506460 in 500 h        
Threat Per Second 4276        
Generated in 469s        
Template :01754.xml(C:\Users\Zack\Documents\Downloads\DKSimulator0.9.9\DKSimulator0.9.9\Templates\01754.xml)        
Priority :Unholy.xml(C:\Users\Zack\Documents\Downloads\DKSimulator0.9.9\DKSimulator0.9.9\Priority\Unholy.xml)        
Presence :Blood        
Sigil of :VengefulHeart        
RuneForge :FallenCrusader        
Pet Calculation :True        

The gap may be much smaller at a gear level such as yours, but it is still there, and it still favors 0/17/54.

Edit: Failed at making a chart.

Last edited by Consider : 08/07/09 at 2:16 PM.

United States Offline
Old 08/07/09, 1:58 PM   #1118
slant
Don Flamenco
 
Draenei Shaman
 
Drenden
Consider, your simulations show a 30dps difference between the two specs. That's 0.4% of your total damage. You don't find threat reduction and runspeed a worthwhile tradeoff?

The readme.htm supplied with the simulator details how to correctly fill out your character template.

All simulations were run with the DeathCoilMaxRP priority. Morbidity is clearly a better investment than necrosis for 0/17/54, as it's geared for RP generation and deathcoil spam. It makes a much smaller difference for 3/13/55, but morbidity still wins. Morbidity also reduces the cooldown on death and decay, which is situationally quite useful.

0/17/54 (Necrosis): 5810dps
0/17/54 (Morbidity): 5850dps
3/13/55 (Necrosis): 5845dps
3/13/55 (Morbidity): 5850dps

United States Offline
Old 08/07/09, 2:06 PM   #1119
Consider
King Hippo
 
Consider's Avatar
 
Draenei Death Knight
 
Dragonblight
It's 30 dps at that (low) gear level. It grows substantially as your gear increases (until the point at which you break 2pt8 and drop SotVH, anyways). A note could simply be made in the OP saying that it's a minor trade off which some may find worth it. Besides, 0/17/54 *is* more dps than 3/13/54, however. Just not astoundingly so, until your gear improves.

I would never drop Morbidity for Necrosis. No, no, no!
0/17/54
3/13/55
Are the two specs used. The difference between Necrosis/Desolation has already been discussed, and the question of IUP/GF/CE are up to personal preference, and their benefit isn't shown in a simulation, thus I don't include them. Besides, they would affect both specs equally.

Last edited by Consider : 08/07/09 at 2:15 PM.

United States Offline
Old 08/07/09, 2:16 PM   #1120
slant
Don Flamenco
 
Draenei Shaman
 
Drenden
The morbidity/necrosis question was asked earlier in the thread. It wasn't in reply to you.

Again my standpoint is that people should run simulations themselves, with their precise character's stats, and determine which spec meets their needs rather then scanning the boards, grabbing a spec meant for a much higher gear level, and performing non-optimally.

Edit: Also, you wasted a point speccing into scourge strike in your 3/13/55 build, and didn't take any points in improved unholy presence in either build. Runspeed isn't simulated, but it is critically important in-game. This may account for the small difference in our simulator results.

Last edited by slant : 08/07/09 at 2:22 PM.

United States Offline
Old 08/07/09, 2:53 PM   #1121
 frmorrison
Protector
 
frmorrison's Avatar
 
Ashstrike
Human Paladin
 
No WoW Account
Run speed is a noticeable dps increase. The 15% run speed talent is more than a 1% dps increase (the new encounters require running from fire), so it is better than other Necrosis or similar talent, plus it allows you to get Icewalker or 32 AP.

United States Offline
Old 08/07/09, 3:26 PM   #1122
Vindice
Glass Joe
 
Vindice's Avatar
 
Orc Death Knight
 
<Hat>
Gorefiend
Originally Posted by Melizande View Post
I can only guess that (he) is using Glyph of Disease, which would just be such a waste of a Major Glyph.
On one hand, you'd get an extra obliterate, but I'm pretty sure you wouldn't generate as much runic power, and what glyph could you possibly trade out for it? I can't imagine the benefit of fitting that glyph in would ever outweigh the cost.
I think he may just be talking about throwing in a pestilence to help AoE, like when Auriaya guardian swarms, or ice blocks on Hodir.

Offline
Old 08/07/09, 3:33 PM   #1123
formina
King Hippo
 
formina's Avatar
 
Draenei Shaman
 
Alexstrasza
Originally Posted by frmorrison View Post
Run speed is a noticeable dps increase. The 15% run speed talent is more than a 1% dps increase (the new encounters require running from fire), so it is better than other Necrosis or similar talent, plus it allows you to get Icewalker or 32 AP.
That's hard to prove. I don't think there are enough encounters where the run speed is enough of an increase to DPS to offset the lost DPS on encounters where the talent points are trivialized. Northrend Beasts isn't entirely a great example of runspeed benefits when knockback distance can be shortened so easily in that arena. It can be a decent DPS increase, but it's too situational to be anything more than an option.

United States Offline
Old 08/07/09, 4:20 PM   #1124
Fargom
Piston Honda
 
Human Death Knight
 
Lightbringer
Originally Posted by formina View Post
That's hard to prove. I don't think there are enough encounters where the run speed is enough of an increase to DPS to offset the lost DPS on encounters where the talent points are trivialized. Northrend Beasts isn't entirely a great example of runspeed benefits when knockback distance can be shortened so easily in that arena. It can be a decent DPS increase, but it's too situational to be anything more than an option.
Run speed has always been a rather large dps gain on any type of a fight where you break contact with the boss for any reason. Even if you ignore the benefit of using the better boot enchant, the 15% speed is going to easily produce more dps than 1% per point. This has been a known mechanic for melee classes for a long time, at this point I would venture to say that anyone who strongly disagrees with it should be the one to prove otherwise.

Offline
Old 08/07/09, 5:15 PM   #1125
Miracleknight
Von Kaiser
 
Troll Warrior
 
Shattered Hand
Just a quick update: I will be revising the first post after I wade through the posts I haven't read in a while, sorry for the delay.

Also we debated quite heavily the benefits of the rune speed talent, and overall we found that you could take it or leave it, it came down to preference, I ended up Opting out of it and saw no significant changes in my DPS on any bosses.

Offline
Closed Thread

Go Back   Elitist Jerks » Class Mechanics » Death Knights

Thread Tools