Elitist Jerks
Register
Blogs
Forums


Go Back   Elitist Jerks » Class Mechanics » Death Knights

Closed Thread
 
LinkBack Thread Tools
Old 10/15/09, 7:28 AM   #886
dr_AllCOM3
Great Tiger
 
dr_AllCOM3's Avatar
 
Orc Death Knight
 
Blutkessel (EU)
I've updated 3.3 SS in the sim with what I think is correct, based on Ghostcrawler's posts. I haven't gone through the latest discussion, since I was on vacation. Go ahead and try it out.
Here's the code, it's pretty self explanatory: Pastey.net - paste bin.


Offline
Old 10/15/09, 7:41 AM   #887
akihex
Glass Joe
 
Minx
Blood Elf Death Knight
 
Non-US/EU Server (EU)
Originally Posted by NeuroMedivh View Post
From these results, I conclude that the 1.0438 multiplier is spot on as long as there's no partial resist, but whenever there is a resist, something gets lost. I'll play with the math more and see what I can figure out.
Back in one of the old unholy threads Zurm suggested that the displayed number for resists does not include all (or any) modifiers:

Unholy DPS Discussion
"Another side note, when you see a strike that says, say "Your SS hits X for 4000 (200 resisted)", you wouldn't have originally done 4200 damage, but actually quite a bit more. The (X resisted) is actually before modifiers/buffs/talents/crits, so in reality, if half a strike was resisted, you might actually see something like "Your SS hits for 2.5k (1.9k resisted)". Just some food for thought. I can't tell you how many time's I've seen 5500(900 resisted) when my SS crits that don't resist always break 8k."

Offline
Old 10/15/09, 7:55 AM   #888
Bensch78
Von Kaiser
 
Orc Death Knight
 
Zirkel des Cenarius (EU)
Originally Posted by dr_AllCOM3 View Post
I've updated 3.3 SS in the sim with what I think is correct, based on Ghostcrawler's posts. I haven't gone through the latest discussion, since I was on vacation. Go ahead and try it out.
Here's the code, it's pretty self explanatory: Pastey.net - paste bin.
Your sim.MainStat.StandardMagicalDamageMultiplier include the 10% from Rage of Rivendare, but as i explained and calculated before, only the following modifiers apply to the shadow part.

Your Shadow Part is 10% to much.

These apply:

Blood Presence (15%)
Desolation (5%)
Bone Shield (2%)
Epon Plaguebringer (13%)

These might apply:

Feroicious Inspiration (3%)

These do NOT apply

Outbreak and Rage of Rivendare, as they are no damage modifiers for Shadow Damage.


And the tests NeuroMedivh made prove my theory

Offline
Old 10/15/09, 7:56 AM   #889
Grey_Matter
Glass Joe
 
Blood Elf Death Knight
 
Aerie Peak (EU)
Using Neuro's numbers, and akihex's assumption, we can do some simple maths to see if the resist damage is reduced by the same %.

Find the theoretical amout lost from a resist, then divide by the actual resist we see in the CL.

(expected hit-actual hit)/resist
(1560-1404)/138=1.1304
(1479-1331)131=1.1297
(1519-1215)/269=1.1301


These all round nicely to 1.13. From this can we assume that the reason we see less resisted damage than we would expect is that Ebon Plague is not taken into account? This would be presumably as it factors in after a partial resist.

EDIT:layout

Offline
Old 10/15/09, 8:03 AM   #890
dr_AllCOM3
Great Tiger
 
dr_AllCOM3's Avatar
 
Orc Death Knight
 
Blutkessel (EU)
Originally Posted by Bensch78 View Post
Your sim.MainStat.StandardMagicalDamageMultiplier include the 10% from Rage of Rivendare, but as i explained and calculated before, only the following modifiers apply to the shadow part.
I didn't check the magic damage modifiers, yet. Does stuff like BP and BS really apply to the magic damage? I thought that they prevented double dipping, so that the magic damage only gains modifiers that aren't already included in physical part.


Offline
Old 10/15/09, 8:09 AM   #891
NeuroMedivh
Von Kaiser
 
NeuroMedivh's Avatar
 
Human Death Knight
 
Greymane
Originally Posted by Grey_Matter View Post
Using Neuro's numbers, and akihex's assumption, we can do some simple maths to see if the resist damage is reduced by the same %.

Find the theoretical amout lost from a resist, then divide by the actual resist we see in the CL.

(expected hit-actual hit)/resist
(1560-1404)/138=1.1304
(1479-1331)131=1.1297
(1519-1215)/269=1.1301


These all round nicely to 1.13. From this can we assume that the reason we see less resisted damage than we would expect is that Ebon Plague is not taken into account? This would be presumably as it factors in after a partial resist.

EDIT:layout
I believe you are correct. Going back over, I find that the formula ((PhysDmg x .75 x 1.15 x 1.05 x 1.02) - Resist) then x1.13 for EP gives the actual amount done.

Though to be honest, it wasn't the resist discrepancy I was concerned about; it's the fact that the shadow portion is using Spell Hit and the x1.5 spell crit formula.

Of course, the 10-20% resist rate on the shadow portion is also worrysome.

Offline
Old 10/15/09, 8:10 AM   #892
Bensch78
Von Kaiser
 
Orc Death Knight
 
Zirkel des Cenarius (EU)
Originally Posted by dr_AllCOM3 View Post
I didn't check the magic damage modifiers, yet. Does stuff like BP and BS really apply to the magic damage? I thought that they prevented double dipping, so that the magic damage only gains modifiers that aren't already included in physical part.
My Calculations and Neuromedivh´s numbers prove that.

World of Warcraft - English (NA) Forums -> Scourge Strike On The PTR ( Information)

Look at point 4 of GC´s post.


------------------------


Originally Posted by NeuroMedivh View Post

Though to be honest, it wasn't the resist discrepancy I was concerned about; it's the fact that the shadow portion is using Spell Hit and the x1.5 spell crit formula.
That´s a bug. Look at point 1 of GC´s post.

Offline
Old 10/15/09, 8:58 AM   #893
Sulika
Glass Joe
 
Human Death Knight
 
Bloodfeather (EU)
Originally Posted by NeuroMedivh View Post
From these results, I conclude that the 1.0438 multiplier is spot on as long as there's no partial resist, but whenever there is a resist, something gets lost. I'll play with the math more and see what I can figure out.
Looking at your numbers it seems that the partial resists are being calculated before the 13% damage increase from Ebon Plaguebringer is being applied.

SS Physical SS Shadow (resist) x1.0438(Final - Shadow) / Resist
14951404 (138)1560(1560 - 1404) / 138 = 1.1304
14171331 (131)1479(1479 - 1331) / 131 = 1.1298
14551215 (269)1519(1519 - 1215) / 269 = 1.1301
35073294 (324)3661(3661 - 3294) / 324 = 1.1327
14791235 (273)1544(1544 - 1235) / 273 = 1.1318
39623722 (366)4136(4136 - 3722) / 366 = 1.1311
38073179 (703)3974(3974 - 3179) / 703 = 1.1308
34703260 (321)3622(3622 - 3260) / 321 = 1.1277
38513618 (356)4020(4020 - 3618) / 356 = 1.1292
33723168 (311)3520(3520 - 3168) / 311 = 1.1318

As you can see in every case you just need to scale the amount resisted up by 13% to keep in line with the non-partially resisted data. So the partial resist is just being calculated before the ebon plaguebringer scaling is applied. It is worth noting that this does not end up costing you any DPS since the partial resist would just remove more damage if it was applied after, it just makes the numbers reported look odd.

Edit: Seems others have reached the same conclusions.

Offline
Old 10/15/09, 10:57 AM   #894
skurdt
Glass Joe
 
Blood Elf Death Knight
 
Throk'Feroth (EU)
Is it possible to conclude some stats weights from what we know or is it still too early to tell ?

Offline
Old 10/15/09, 11:09 AM   #895
Consider
King Hippo
 
Consider's Avatar
 
Draenei Death Knight
 
Dragonblight
Although I can undertand Outbreak not double-dipping, RoR seems somewhat odd. Regardless, actually getting on the PTR, I receive similar numbers - an approximate 1.0438 modifier.

Which is still a clear improvement over the current SS and, with crit working properly on the shadow half, probably still enough for Reaping. If you assume a 200% crit modifier (which isn't affected by Viscious Strikes), and a 35% crit rate, that brings the shadow half up to 1.40913, which places your average SS at slightly above 9600 (if you use 4k as a base number for the physical portion). If, however, you assume the crit rate of the shadow half is affected by Subversion/Viscious Strikes, putting it at 50% crit, you then get a modifier of 1.5657, which puts your average SS at slightly over 10,000.

If the crit modifier on the shadow half is supposed to be affected by Chaotic/Relentless and/or Viscious Strikes, that would further inflate the numbers.

Either scenario would be enough for Reaping to be superior to the current Dark Conviction. I say current because DC's value will change as well - it will go down a lot with the 4p t9 nerf/drop, and go up some because of the SS change. Most likely down more than up, which would give Reaping an even wider lead.

Anyways, calculating the value of Black Ice is simple enough, Jackinthegreen.

Damage without Black Ice: 1 + 1.0438 = 2.0438
Damage with Black Ice: 1 + (1.0438 x 1.1) = 2.14818

Essentially a 5.107153% increase in total SS damage.

2H Weapon Specialization, on the other hand, provides nearly the same bonus (4% exactly, as the second portion doesn't double dip), except for the fact that you lose much less speccing into Blood than you do speccing into Frost (partially because there's nothing in Frost aside from Black Ice and Annihilation, partially because Black Ice is an expensive talent, and partially because you have to go into Blood for Subversion anyways).

When you consider the UB nerf, fewer DCs being casted (due to the GoIT change, and the fact that even without GoIT, a 20 second rotation produces less RP), and the nerf/eventually loss of 4p t9, on top of it's weaker affect on SS, Black Ice is going to be even less competitive than it is now.

Edit: It's still too early to calculate EP values, as the sim isn't quite right yet. If I were to guess (emphasis on guess), haste will be around 1.6, ArP will be around 1.4-1.8, and crit (without 4p t9) will be around ~1.8. Merely guesses, however, based on current/past values.

Edit 2: The sim is still showing Reaping as a dps loss, regardless of the combination I try. SS simply isn't doing quite as much as damage as I thought (various reasons as to why). Epidemic is definitely a go, however. Black Ice is not worth it, as expected. Without 4p t9 or with it nerfed, BA > DC, point for point.

Last edited by Consider : 10/15/09 at 11:53 AM.

United States Offline
Old 10/15/09, 11:28 AM   #896
Bloody_sorcerer
back in my day, we tanked uphill in the snow
 
Bloody_sorcerer's Avatar
 
Blood Elf Death Knight
 
Malfurion
Originally Posted by Bensch78 View Post
These apply:
Blood Presence (15%)
Desolation (5%)
Bone Shield (2%)
Epon Plaguebringer (13%)
These might apply:
Feroicious Inspiration (3%)
These do NOT apply
Outbreak and Rage of Rivendare, as they are no damage modifiers for Shadow Damage.
To generalize this, all buffs, debuffs, and talents that have the Effect: Apply aura: mod % dmg taken ("All" or "some number") will modify scourge strike's damage. If, in parenthesis, it says "All", it is a buff that applies to all magic damage, implying shadow damage (curse of elements does this). Otherwise, it will be a bitmask, which will follow the standard WoW spell bitmasks (API COMBAT LOG EVENT - WoWWiki - Your guide to the World of Warcraft see Parameter Values section), so all bitmasks containing 32 will buff scourge strike, which includes blood presence, desolation, bone shield, ferocious inspiration. Likewise, Outbreak and Rage of Rivendare are notably missing this, and as a result do not have any affect on scourge strike shadow damage.

Originally Posted by Esoth View Post
Torture/terrorism/war/taxes? Whatever. But if this motherfucker doesn't realize why the heroic strike change is complete shit...
Originally Posted by Malan View Post
STOP SHITTING UP THE THREAD MALAN

Offline
Old 10/15/09, 12:30 PM   #897
Consider
King Hippo
 
Consider's Avatar
 
Draenei Death Knight
 
Dragonblight
14/0/57 (without Reaping) EP values. I used the same stats Bensch had used a couple pages back, except I doubled the ArP value (100 -> 200), as its value would obviously go up, so most people would likely have higher - and its value is so dependent on how much you already have.

EP:50 AttackPower 1
EP:50 Strength 3.08
EP:50 Agility 1.28
EP:50 CritRating 1.9
EP:50 HasteRating 2.29
EP:50 ArmorPenetrationRating 1.71
EP:50 ExpertiseRating 2.22
EP:50 BeforeMeleeHitCap<8% 3.4
EP:50 SpellHitRating 1.34
EP:10 WeaponDPS 6.35
EP:0.1 WeaponSpeed 412.7

Crit was 2.1 with 4p t9. Agi was 1.43.

All seems fine, except haste. There can't be any way that's right. It's even higher than the current value - and without the current GoIT and with Epidemic, we should have plenty of free GCDs. It doesn't make any sense, unless I'm missing something. The other numbers all seem logical enough, however.

Doing seperate sims, increasing haste (aside from it's normal function of increased white/ghoul/garg damage) appears to increase the number of abilities you do (SS, BS, DC, etc). That shouldn't be the case with Epidemic. PS IT SS BS BS DC DC // SS SS BS BS DC, with 5 RP leftover. That's 12 globals, or 18 seconds. Even if you used HoW on CD, which the sim is likely doing, that's still only 13 - 19.5 seconds. Either way, I don't see how the sim is getting haste to increase yellow damage, which it somehow is - and by doing so, is giving a high haste value.

If you do three sims (one with X stats, one with X stats plus 100 haste, and one with x stats plus 200 AP), and then manually calculate how much dps the "X stats plus 100 haste" gains over the base "X stats" run from auto attack, ghoul, and gargoyle alone, and then divide that difference by how much the 200 AP gave, you get a much more reasonable EP value of 1.505.

Last edited by Consider : 10/15/09 at 12:52 PM.

United States Offline
Old 10/15/09, 1:46 PM   #898
Afabar
Don Flamenco
 
Draenei Death Knight
 
Chants Eternels (EU)
Originally Posted by Consider View Post
Doing seperate sims, increasing haste (aside from it's normal function of increased white/ghoul/garg damage) appears to increase the number of abilities you do (SS, BS, DC, etc). That shouldn't be the case with Epidemic. PS IT SS BS BS DC DC // SS SS BS BS DC, with 5 RP leftover. That's 12 globals, or 18 seconds. Even if you used HoW on CD, which the sim is likely doing, that's still only 13 - 19.5 seconds. Either way, I don't see how the sim is getting haste to increase yellow damage, which it somehow is - and by doing so, is giving a high haste value.
It's hard to say exactly why you have this issue. A Few possibility:
- There is a problem with the sim due to the recent changes to support epidemic less priority mode.
- High latency.
- Your character haste is giving you a lot of disease clipping.
- the Random Number Generator don't like you.

I reproduce it so I will try to take a look at it by analyzing the combat log. However, you are right on what you have done manually, this is how the sim does too but with 50 haste / 100 AP.

Offline
Old 10/15/09, 2:45 PM   #899
Velisella
Glass Joe
 
Blood Elf Death Knight
 
Arthas
Scourge Strike on PTRs

* 1) Scourge Strike should crit for 200% damage like other DK spell effects.
* 2) The Shadow part of the damage should never miss if the physical part hits.
* 3) The Shadow part of the damage should never be decreased additionally to the physical damage. It should not "double dip" from resilience et al. Example: My physical portion that hits for 2000 is reduced to 1000. The Shadow part should then hit for 1000 (modulo bonuses) and not 500 or whatever.
* 4) The Shadow part of the damage will still be increased additionally by talents and anything else that buff Shadow damage. It will "double dip" in the DKs favor. Example: My physical portion hits for 2000 and is reduced to 1000. The Shadow portion hits for 1500 because I have +50% Shadow damage.


Scourge Strike Damage
Only the physical damage can be reduced. The Shadow damage can only be buffed, since it is essentially already reduced by the physical mitigation, resilience or other defensive effects.

The intent was just to split the damage into part physical and part Shadow. We didn't want the split itself to be a penalty, if that makes sense.


This was posted by GC on the forums today.

Offline
Old 10/15/09, 3:38 PM   #900
Drelnax
Glass Joe
 
Orc Death Knight
 
Dunemaul
Needs Sim Testing

I've got a build suggestion that could use some Sim testing, if someone doesn't mind. (I'm stuck at work and can't do it myself)

Suggested build: http://talent.mmo-champion.com/?deat...Z,Q6F3eO,10571
Yes, the glyph's selected in this build have been chosen intentionally.

I waited to hear from GC on the "intended" functionality of the SS changes, to create this build. Based on what GC has outlined; I'm now bringing this to our Unholy DPS community for review, scrutiny, suggestions, and Sim testing.

This build w/glyphs will extend BP to 24sec and FF to 21sec. Allowing for a longer rotation and more RP to use. this "SHOULD" also give us a sizable dps increase as well; if I correctly understood GC's explanation of what the shadow damage portion of SS will benefit/increase from, as well as correctly built the following rotation suggestion.

Rotation Suggestion: First Half- PS,IT,SS,BS,BS,HoW,Garg
Second Half- SS,SS,SS, DC DUMP

This rotation suggestion seems to leave an extra 3RP at the end, so long as my math is correct. Please let me know if it's not; so that I can go through and fix it.
This rotation should be close to 24sec in length, possibly ~1 or 2 seconds shorter. The way I've done this allows you to repeat the rotation over and over, for the duration of a boss encounter. If you do, you will need to substitute the use of Garg in the first half with a DC DUMP. Based on a ~4 to 5min encounter duration you should be able to use the Garg twice. More if the encounter is even longer.
You will have to figure out the timing of when to revert back to Garg use roation, based on each boss encounter that allows for more than one use of Garg. Typically though, you should use it every time it's up while your damage increasing procs/effects are active on you.

This will allow us to make the most of our PS/IT disease applications without clipping them and reapplying them with ~1 or 2 seconds where the diseases aren't up, allowing a latency of ~160ms.

I haven't had a chance to test any of this out yet as my work and home life keep me rather busy as of late. I will get some testing of my own done, as soon as time allows and will also post results of said testing.
Happy testing to all you decide to help out with this. My general goal of this, is to have provided a "fresh/different" perspective on possible performance/dps increases that may have previously overlooked.
Best case scenario for me would be that this suggestion does give us the sizable dps increase I think it will. Worst case scenario...LodeRunner or someone else at EJ deems this post/suggestion worthless and I receive the "BanHammer"

Last edited by Drelnax : 10/15/09 at 3:40 PM. Reason: To fix broken hypertext link

Offline
Closed Thread

Go Back   Elitist Jerks » Class Mechanics » Death Knights

Thread Tools