Petrified Twilight Scale

Get help with your character's gear

Moderators: Fridmarr, Worldie, Aergis, Sabindeus, majiben, lythac, Digren

Re: Petrified Twilight Scale

Postby yappo » Thu May 20, 2010 6:04 am

On a minor sidenote. Is that the full datamine, ie did MMO publish the full PTR loot-table? If so, then plate-tanks got pretty shafted with a grand total of two drops.
yappo
 
Posts: 1107
Joined: Wed Sep 17, 2008 4:15 pm

Re: Petrified Twilight Scale

Postby Zobel » Thu May 20, 2010 8:05 am

yappo wrote:On a minor sidenote. Is that the full datamine, ie did MMO publish the full PTR loot-table? If so, then plate-tanks got pretty shafted with a grand total of two drops.

Of course it's the full datamine. There are zero tank drops on 10 man normal. Clearly.
User avatar
Zobel
 
Posts: 722
Joined: Wed Dec 19, 2007 6:31 pm

Re: Petrified Twilight Scale

Postby Shoju » Thu May 20, 2010 8:44 am

Wouldn't surprise me if there aren't. Seriously Blizzard can DIAF over their stupid 10 man loot distribution this xpac.
User avatar
Shoju
 
Posts: 6391
Joined: Mon May 19, 2008 7:15 am

Re: Petrified Twilight Scale

Postby yappo » Thu May 20, 2010 9:33 am

Zobel wrote:
yappo wrote:On a minor sidenote. Is that the full datamine, ie did MMO publish the full PTR loot-table? If so, then plate-tanks got pretty shafted with a grand total of two drops.

Of course it's the full datamine. There are zero tank drops on 10 man normal. Clearly.


You're right of course. About as likely as Inscription getting zero upgrades through three x-pacs. Oh, wait...
yappo
 
Posts: 1107
Joined: Wed Sep 17, 2008 4:15 pm

Re: Petrified Twilight Scale

Postby pfunkmort » Thu May 20, 2010 7:26 pm

yogg saron 25 man.
Image
[2.Trade][Alessan]: Is WoW becoming an online kart racer in 4.0.3?
[2.Trade][Tehron]: Yes
User avatar
pfunkmort
 
Posts: 99
Joined: Mon Apr 13, 2009 3:11 am

Re: Petrified Twilight Scale

Postby lythac » Fri May 21, 2010 2:41 am

Yogg25 is different, not every boss in an instance should drop an item that is used for every possible spec. But in an instance there should be at least 1 piece of loot useful for every spec.
Ryshad / Lythac of <Heretic> Nagrand-EU
User avatar
lythac
Moderator
 
Posts: 2694
Joined: Wed Sep 24, 2008 8:10 am

Re: Petrified Twilight Scale

Postby bldavis » Fri May 21, 2010 9:32 am

yappo wrote: You're right of course. About as likely as Inscription getting zero upgrades through three x-pacs. Oh, wait...


lol
Image

Brekkie:Tanks are like shitty DPS. And healers are like REALLY distracted DPS
Amirya:Why yes, your penis is longer than his because you hit 30k dps in the first 10 seconds. But guess what? That raid boss has a dick bigger than your ego.
Flex:I don't make mistakes. I execute carefully planned strategic group wipes.
Levie:(in /g) It's weird, I have a collar and I dont know where I got it from, Worgen are kinky!
Levie:Drunk Lev goes and does what he pleases just to annoy sober Lev.
Sagara:You see, you need to *spread* the bun before you insert the hot dog.
User avatar
bldavis
 
Posts: 7411
Joined: Mon Sep 07, 2009 12:04 pm
Location: Searching for myself. If i get back before I return, please have me stop and wait for myself.

Re: Petrified Twilight Scale

Postby tlitp » Sat May 22, 2010 9:51 am

Meloree wrote:
bldavis wrote:it looks nice for physical fights with avoidable damage, but like you said, whats the DR gonna be like?

With my gear, it's about 8% dodge, FWIW.

Define CHEs/NCP
Let a Contiguous Hit Event be a string of at least two non-avoided, back-to-back NPC (auto)attacks. Let NCP(N) be the probability of non-CHEs (under the assumption that CHE/non-CHE form a complete probability field) within a size N string of NPC (auto)attacks.


Computational example of NCP
Using a value of total avoidance equal to a, compute NCP(3) :
hit-hit-miss, p=a*(1-a)^2
hit-hit-hit, p=(1-a)^3
miss-hit-hit, p=a*(1-a)^2
∑p=(1+a)*(1-a)^2. Do note that using the conventional Bernoulli method, one runs into the problem of overlapping probabilities (read [1],[2] on TS).


Computing NCP(N) using Markov chains
Let (i=<N,j=<2) be the states of a MC, with index i describing the number of attacks performed by the NPC, and j describing the number of contiguous non-avoided attacks. The permitted transitions will be :
Code: Select all
hit                    (i,j)  ->(i+1,j+1) for i<N and j<2, p=1-a
miss                   (i,j)  ->(i+1,j=0) for i<N and j<2, p=a
CHE shift              (i,j=2)->(i+1,j=2) for i<N        , p=1
attack counter reset (i=N,j)  ->(i=N,j)                  , p=1

Example of transition matrix for N=4
Code: Select all
  0,0 1,0 1,1 2,0 2,1 2,2 3,0 3,1 3,2 4,0 4,1 4,2

[  0   a 1-a   0   0   0   0   0   0   0   0   0   0,0
   0   0   0   a 1-a   0   0   0   0   0   0   0   1,0
   0   0   0   a   0 1-a   0   0   0   0   0   0   1,1
   0   0   0   0   0   0   a 1-a   0   0   0   0   2,0
   0   0   0   0   0   0   a   0 1-a   0   0   0   2,1
   0   0   0   0   0   0   0   0   1   0   0   0   2,2
   0   0   0   0   0   0   0   0   0   a 1-a   0   3,0
   0   0   0   0   0   0   0   0   0   a   0 1-a   3,1
   0   0   0   0   0   0   0   0   0   0   0   1   3,2
   0   0   0   0   0   0   0   0   0   1   0   0   4,0
   0   0   0   0   0   0   0   0   0   0   1   0   4,1
   0   0   0   0   0   0   0   0   0   0   0   1]  4,2

From the TM one can compute the stationary distribution, and then the NCP(N).


Sample results for non-heroic SN
Meloree's gear
Gear_Defense=852
Gear_Dodge=684
Gear_Parry=456
Gear_AGI=32

boss_base_swing_speed=variable
boss_parryhaste_flag=0

Code: Select all
CotT active     NCP1[%] NCP2[%] aNCP[%] rNCP[%]  aA[%]
swing_timer=1.0   12.3   22.8    10.4    84.6    8.4
swing_timer=1.5   29.3   42.1    12.7    43.3    8.4
swing_timer=2.0   38.7   51.4    12.6    32.5    8.4

CotT inactive   NCP1[%] NCP2[%] aNCP[%] rNCP[%]  aA[%]
swing_timer=1.0   42.6   59.7    17.1    40.2    8.4
swing_timer=1.5   60.7   74.1    13.3    21.9    8.4
swing_timer=2.0   68.3   79.5    11.2    16.5    8.4

NCP1 - NCP(floor(10/boss_swing)) before SN
NCP2 - NCP(floor(10/boss_swing)) after SN
aNCP - absolute increase in NCP
rNCP - relative increase in NCP
aA   - absolute increase in avoidance
All values are rounded down.

Over the studied parameter field, the effect of SN is superlinear. Also keep in mind that using CHEs of size 2 is somewhat extreme, as most bosses do not regularly (i.e. outside of their "cooldowns") two-shot tanks with autoattacks. Higher values of the size of CHEs increase significantly the corresponding NCP values, ceteris paribus.
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

Re: Petrified Twilight Scale

Postby bldavis » Sat May 22, 2010 5:40 pm

tlitp wrote: /snip


ok i admit, i dont fully understand it all completely, but

is all that saying that it would be about 8% dodge with meloree;s gear, so he was about right?

/headache

(tbh im having trouble in trig so ....)
Image

Brekkie:Tanks are like shitty DPS. And healers are like REALLY distracted DPS
Amirya:Why yes, your penis is longer than his because you hit 30k dps in the first 10 seconds. But guess what? That raid boss has a dick bigger than your ego.
Flex:I don't make mistakes. I execute carefully planned strategic group wipes.
Levie:(in /g) It's weird, I have a collar and I dont know where I got it from, Worgen are kinky!
Levie:Drunk Lev goes and does what he pleases just to annoy sober Lev.
Sagara:You see, you need to *spread* the bun before you insert the hot dog.
User avatar
bldavis
 
Posts: 7411
Joined: Mon Sep 07, 2009 12:04 pm
Location: Searching for myself. If i get back before I return, please have me stop and wait for myself.

Re: Petrified Twilight Scale

Postby tlitp » Sat May 22, 2010 9:37 pm

bldavis wrote:is all that saying that it would be about 8% dodge with meloree;s gear, so he was about right?

That ~8% avoidance gain is correct, I was merely pointing out a clear case of non-WYSIWYG. SN's (and by extension other dynamic avoidance effects) true effectiveness is greater than what one can actually observe in terms of avoidance gain.

It's mostly the evolution of a concept. People tried to gauge avoidance's effectiveness in real scenarios - an attempt that had two explicit issues and one implicit one :
  1. simulation, thus inherently non-deterministic
  2. the values of N/k
  3. a fundamentally skewed perspective of the problem; what we're interested in is not the probability of exactly k contiguous hits in N attempts, but the probability of of at least k CHs given a size N string of attempts
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

Re: Petrified Twilight Scale

Postby Meloree » Sun May 23, 2010 4:52 pm

Honestly, to me, it largely comes down to the uncontrollable nature of the trinket-proc. If I could put it up for the Soul-Reaper haste effect reliably, that's enough avoidance that I'd be seriously interested, especially in combination with 4pc. On the other hand, Arthas hits hard enough to randomly trigger the proc lots of not-so-useful times, reducing the trinket's value to me. On the other hand, at a fight like heroic Festergut, it's going to be very useful because no matter when the proc happens, it's helpful at reducing the likelihood of 18-hit strings.

Worth noting, I used the phrase "about 8%" primarily because I could be wearing just about any gearset in armory. The gearset Tiltp used is ICC25H threat-set, which places it rather lower on the DR curves that almost anything else. Note things like hit leggings and rings, rather than armor equivalents, and SoV libram, over a dodge-rating libram. Last Word is also pretty low on the avoidance stats.
Meloree
Maintankadonor
 
Posts: 1420
Joined: Wed Mar 12, 2008 10:15 am

Re: Petrified Twilight Scale

Postby Invisusira » Sun May 23, 2010 4:58 pm

bldavis wrote:also wats the uptime gonna be like on that?

I had set a Power Aura to alert me when my Corpse Tongue Coin - which also has a sub-35% ability - proc'd, and it actually actives FAR more frequently than you'd imagine.
User avatar
Invisusira
Moderator
 
Posts: 9021
Joined: Sat Oct 06, 2007 6:23 pm
Location: alt-tabbed

Re: Petrified Twilight Scale

Postby Awyndel » Sun May 23, 2010 9:22 pm

Well if we wear this it will be for the armor. So the proc is not really relevant, just a nice bonus.
User avatar
Awyndel
 
Posts: 672
Joined: Sat Feb 14, 2009 8:49 am
Location: The Netherlands

Re: Petrified Twilight Scale

Postby svedka » Sun May 23, 2010 9:28 pm

Would it interact negatively with Ardent Defender or WotN?

edit:: was it changed back to dodge?
svedka
 
Posts: 20
Joined: Mon Jan 19, 2009 1:33 pm

Re: Petrified Twilight Scale

Postby tlitp » Mon May 24, 2010 1:34 am

Deliverance makes SN even more interesting. Assuming activation immediately after the proc of SN :
using an avoidance-heavy template
Gear_Defense=1000
Gear_Dodge=1000
Gear_Parry=500
Gear_AGI=32

boss_base_swing_speed=variable
boss_parryhaste_flag=0

Code: Select all
CotT active     NCP1[%] NCP2[%] aNCP[%] rNCP[%]  aA[%]
swing_timer=1.0   39.3   54.0    14.7    37.4     7.4
swing_timer=1.5   57.9   69.8    11.8    20.5     7.4
swing_timer=2.0   65.8   76.0    10.1    15.4     7.4

CotT inactive   NCP1[%] NCP2[%] aNCP[%] rNCP[%]  aA[%]
swing_timer=1.0   79.3   91.3    12.0    15.1     7.4
swing_timer=1.5   87.4   94.9     7.4     8.5     7.4
swing_timer=2.0   90.3   96.1     5.8     6.4     7.4

NCP1 - NCP(floor(10/boss_swing)) before SN
NCP2 - NCP(floor(10/boss_swing)) after SN
aNCP - absolute increase in NCP
rNCP - relative increase in NCP
aA   - absolute increase in avoidance
All values are rounded down.

4pt10 makes SN push the probability of not getting at least two CHs, within the duration of the effect, to 50% (worst case scenario in ICC), respectively well over 90% (outside ICC).

Now, the real question is : are these numbers significant or not ? As they're not equal to 100%, most "tanks at heart" (pun intended) will still say "I'm wearing it for the armor. EH iz king LOL". Heck, even an almost sure event is not good enough for them. :P
User avatar
tlitp
 
Posts: 556
Joined: Mon Jul 27, 2009 3:25 pm

PreviousNext

Return to Gear Discussions and Advice

Who is online

Users browsing this forum: No registered users and 1 guest

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest