or.... Boots of the Shifting Nightmare
41 stam
22 int
18 hit
59 shadow dmg
lol prob a dumb question but just a hard choice for me to figure out..... each set of boots have their own ups and downs... Thanks
The answer is to open the spreadsheet linked in the first page, add each item's values to your raid gear and see which one raises your dps the most. :P
Hello guys I am currently new to all this "theory craft" stuff and I have a couple questions. Currently I am a 70 Warlock on garithos and in a raiding guild doing ZA/Gruuls/Kara. I am geared decently you could say, I am putting out a good amount of damage about top 5 in gruul/Za and top 3 in Kara but I can seem to put of the damage per second i think i should be. For example in Gruuls last night I was number 4 in damage out damaging the warlocks and some good geared people but I was 8th in DPS with about 480 per second while the other warlock had 520 (was destruction with 1300 shadow dmg)
Hello guys I am currently new to all this "theory craft" stuff and I have a couple questions. Currently I am a 70 Warlock on garithos and in a raiding guild doing ZA/Gruuls/Kara. I am geared decently you could say, I am putting out a good amount of damage about top 5 in gruul/Za and top 3 in Kara but I can seem to put of the damage per second i think i should be. For example in Gruuls last night I was number 4 in damage out damaging the warlocks and some good geared people but I was 8th in DPS with about 480 per second while the other warlock had 520 (was destruction with 1300 shadow dmg)
When I'm Dpsing i use SL >> Corr >>> CoA >>> Immo >>>> Drian Life
Right now I'm running 1085 shadow dmg with fel armor 16.48% crit 10238 hp with imp and 8388 mana
can I get some suggestions on how to max. my dps but still able to pvp when I want to?
Thank You
Quite simply, the "DPS" statistic is a complete lie and totally irrelevent. No DPS meter, especially WWS, has ever come up with a good way of measuring sustained DPS.
They generally work by taking your damage, and dividing it by your "in-combat" time. The problem is, this "in-combat" time is a totally fudged hack. It's the worst on gruul where everyone stops doing anything during shatters--your dots continue to tick so the parse reads you as in combat, but no one else. The result is that, while your damage is flatly higher because of the dots, your dps is flatly lower. This is true of any movement fight, but at your content level Gruul is the one where you see it the most.
Moral of the story, only pay attention to damage, and never to any official DPS statistic. There are ways of cheesing it to increase your listed dps, but most of them end up with you doing less damage, not more.
Disregard anything that says "DPS" on it, Affliction specs (which at your gear level is what Warlocks should be mostly) suffer the most because of what was said above. Just look at total damage done in relation to the other people. That's your role.
"Average DPS" on WWS fixes the concerns with Affliction "Effective DPS". It takes total damage divided by total time alive for the fight which includes movement time.
On which fights - if any - do the 0/21/40 warlocks in your raid (have to) produce the Imp to buff raid hp? This does not count the one lock in the MT group that may have to bite the bullet because the affliction warlock(s) are unavailable for the night.
We had a discussion in my raid very recently on this topic, and I would like to hear outside opinions.
On which fights - if any - do the 0/21/40 warlocks in your raid (have to) produce the Imp to buff raid hp? This does not count the one lock in the MT group that may have to bite the bullet because the affliction warlock(s) are unavailable for the night.
We had a discussion in my raid very recently on this topic, and I would like to hear outside opinions.
Najentus is the only fight where this should be necessary, IMO, and only if you have spellfire mages running around.
Seconding Najentus, but can do without if your mages have decent gear. Possibly Bloodboil if you get stuck in a mage soaking group. It won't make or break the fight but it's worth losing 15% shadow damage over especially since you easily get threatcapped without it anyway.
There was an Illidan tanking gear question before.
I don't think there is a need for full sr. I went through ally my last months wws parses to check the exact numbers. I always go with 4 crafted parts(excl the leggings) + neck which leaves me at 321 sr buffed and lets me keep 4 t6. I never got hit for more than 8400 (luck?). In most cases the max hit for the entire fight doesn't exeed 7.5k. It also allowes me maintain a very decent dps through entire fight and aggro for the demon phase so everyone in reach can nuke their hearts out without any worries. Also, I have never died to shadowblasts when tanking(credits to the healers), have chugged a hs/healthpot on rare occasions to be on the safe side though.
Am I the only one that dares to wear less than full sr? What's the theoretical chance for me to get a completely nonresisted hit? I remember getting a 10k+ max hit a couple of months ago during learning the fight when i tried tanking without the neck(281 sr).
There was an Illidan tanking gear question before.
I don't think there is a need for full sr. I went through ally my last months wws parses to check the exact numbers. I always go with 4 crafted parts(excl the leggings) + neck which leaves me at 321 sr buffed and lets me keep 4 t6. I never got hit for more than 8400 (luck?). In most cases the max hit for the entire fight doesn't exeed 7.5k. It also allowes me maintain a very decent dps through entire fight and aggro for the demon phase so everyone in reach can nuke their hearts out without any worries. Also, I have never died to shadowblasts when tanking(credits to the healers), have chugged a hs/healthpot on rare occasions to be on the safe side though.
Am I the only one that dares to wear less than full sr? What's the theoretical chance for me to get a completely nonresisted hit? I remember getting a 10k+ max hit a couple of months ago during learning the fight when i tried tanking without the neck(281 sr).
There's no reason not to wear max SR. It's not really a DPS race fight so I wouldn't worry about that. I can't remember that max hit from the shadowblast, but it was probably around 10k unresisted.
Afaik in TBC 248 resist buffed removes any chance of ever taking a full hit, at least so it was theorized in some thread around here, prolly too well burried to even try to dig it up
Also, I don't use quite maxed SR but keep 4 pcs t6, 194 hit (Scryer Bloodgem ftw ))) ) but also getting paid to respec demo each week so really can't comment that much since I got it easy, as in I do whatever it takes to not die, my dps is irrelevant in a 25 min enrage fight.
I have a question, and my guild leader recommended me to these forums. Its been a long thread, and i must admit i didn't look at every page.. Nevertheless, i have a question regarding dmg, crit, & consumables. At the moment, i am Destruction (3/5 tier 6, waiting for the 4th piece ^_^).
W/ dmg gear, i have +1302 (+1382 shadow dmg when blessings deck is procced fully) shadow dmg, 28% chance crit (self buffed, no consumables)
W/ crit gear, i have +1234 shadow dmg, 32.06% chance to crit (self buffed, no consumables)
At my current situation, i have many choices when it comes to consumables. Ultimately, the questions comes down to A) whats the healthiest balance of crit/dmg for a destruction warlock? and B) which consumables result in the most benefit to dmg output? I have access to:
2) +23 crit & dmg elixir (Adpet elixir)
or +80 spell dmg (Flask of Pure Death)
I'm hoping somebody here has experience with the topic of dmg & crit, and consumables. I would've started my own thread, however i have to post 10 times in public forums to even do so.
What I've experienced recently is that a high crit rate (30% or so) can actually be detrimental on some fights unless you have the benefit of not only Salvation but also the Tranquil Air totem. I know I don't always get the benefit of the uber caster group (shadow priest and elemental shaman), and even if I did I would get a 3% crit bonus to go along with TA anyway. I know I plan on swapping out gear to actually lower my crit to about 25%-27% or so, which should also buff my spell damage by a fair amount. This should maximize not only my personal DPS but my ISB contribution as well (can't very well keep ISB up if you can't cast Shadow Bolts because of threat spikes).
Given that, I would suggest wearing your damage gear and using damage consumables. I know that is what I plan on doing.
Ultimately, the questions comes down to A) whats the healthiest balance of crit/dmg for a destruction warlock? and B) which consumables result in the most benefit to dmg output?
I recommend checking the spreadsheet for concrete answers regarding stat choices. However the options you're deciding between trade more spell damage for less crit rating. The original post also has a "general" guideline for ordering stat values and spell damage will almost always beat crit rating point for point.
My question: how accurately is haste rating modeled for affliction locks on a stand-and-cast fight using the Leulier spreadsheet? It's a long post, you've been warned.
The Leulier spreadsheet determines overall damage by calculating ratios of spells cast instead of developing a casting timeline (which some might call a casting rotation). According to the spreadsheet, spell haste simply reduces the cast time of shadowbolt, thereby increasing the number of shadowbolts that are cast relative to other spells that are cast.
However, I believe that faster shadowbolts may have a more pronounced effect for an affliction lock who uses shadowbolts as a filler between dot casts. It's widely accepted that a warlock should constantly be casting.. so if a dot will not expire before the casting time for that dot is up, then it's best to use a shadowbolt to fill until the dot has expired. It seems to me that faster shadowbolts should reduce the average time between when a dot ends and when a dot is refreshed.
Example: Assume that siphon life will expire in 1s. The warlock casts shadowbolt which takes 2.5s to cast. After 1s siphon life will expire but the warlock has 1.5s of shadowbolt cast left. When the shadowbolt is cast the warlock refreshes siphon life. If the warlock's shadowbolt takes 2.4s to cast, then the period of time when siphon life is down is only 1.4s instead of 1.5s.
I couldn't think of a way of integrating this change into the spreadsheet since I don't know how pronounced it is. I suspect it may be something to the effect of total haste increases your average dot uptime by the same percentage that it reduces spell cast time (this can't be true because otherwise you could get to 100% uptime with a reasonable amount of haste which just isn't possible).
I wrote a casting simulator to get an idea of the magnitude of the effect. It simulates casts over a given period of time. It works using a simple priority system.. if ua isn't up, cast it.. if ua is up but will end in less than 1.5+lag seconds, recast it.. if corruption isn't up, cast it.. if sl isn't up, cast it.. if everything else is up, then cast a shadowbolt.
I created a simulator because it's hard to model the effect of resists on a casting cycle.. without resists you would always cast corruption immediately after UA. Resists will break up this "cycle" and provide more accurate results.
The following conditions are set:
.1s of lag
2.5s base shadowbolt before haste
casting priority: ua > corr > sl > sb
no lifetap / dark pact included (this mucks things up a bit, I realize)
each fight simulation is 600s (10 mins)
1800 fights simulated for each % of haste from 0 to 10
Format of the results:
Haste %
UA uptime as a percent
Co uptime as a percent
SL uptime as a percent
shadowbolts cast in a 600s time period
Haste: 0.0
UA up: 90.6
Co up: 90.6
SL up: 94.6
SBs: 179.0
Haste: 1.0
UA up: 91.3
Co up: 91.3
SL up: 95.2
SBs: 181.0
Haste: 2.0
UA up: 92.1
Co up: 92.0
SL up: 95.9
SBs: 182.0
Haste: 3.0
UA up: 92.8
Co up: 92.8
SL up: 96.5
SBs: 184.0
Haste: 4.0
UA up: 93.6
Co up: 93.5
SL up: 93.5
SBs: 185.0
Haste: 5.0
UA up: 94.3
Co up: 94.3
SL up: 94.3
SBs: 187.0
Haste: 6.0
UA up: 90.7
Co up: 90.5
SL up: 94.4
SBs: 190.0
Haste: 7.0
UA up: 91.4
Co up: 91.2
SL up: 91.8
SBs: 192.0
Haste: 8.0
UA up: 92.1
Co up: 92.0
SL up: 92.5
SBs: 194.0
Haste: 9.0
UA up: 92.9
Co up: 92.8
SL up: 93.4
SBs: 197.0
Haste: 10.0
UA up: 93.6
Co up: 93.5
SL up: 93.6
SBs: 197.0
The number of shadowbolts increases regardless of the amount of additional haste. As a trend, uptime of the dots increases with additional haste until 5% and drops down to the 0% haste uptime levels around 6% haste (probably because of the constant .1s lag). 5% haste seems to be about the sweet spot for these conditions.. with a 4.4% increase in number of shadowbolts and a 4.1% increase in UA/Co uptime.
Doing some back-of-the-envelope math.. let's say a warlock does 50% damage from shadowbolts, 20% from UA, 20% from corruption, and 10% from SL. The spreadsheet would count 5% haste as 5% more shadowbolt damage (slightly less since the time gap between life taps will be slightly less)... so 5% haste would be a 2.5% increase in total damage. If we use the numbers generated from the simulation and count additional uptime as a linear increase in damage, then 5% haste would increase overall dps by 3.84%.
Like I've said, it's a crude estimate and the numbers are likely to change when factoring in ISB, nightfall, curses, and taps, but I do believe it's significant enough to try to somehow incorporate into the spreadsheet. My hunch is that spell haste is undervalued for affliction locks on the Leulier spreadsheet by 25-50%.
Currently DoT-gap is modeled as a single constant parameter rather than a calculated function, so if you really think it changes that much you could fiddle with the number itself as you get better haste. Your models provide a good baseline for what numbers to toss in. The DoT-gap probably could be modeled as a calculated paramter if you wanted to. A first approximation would be one-half of shadowbolt cast time, a more complicated model would probably include DoT resist rates. With the new casting mechanics the DoT-gap probably is a result of casting cycles whereas it used to be a result of human lag time, so its worth looking into.
In a realistic situation I suspect that it's appropriately undervalued because DoTs (aside from immo/UA) can be cast while mobile, which precludes shadowbolt clipping anyways. Of course, the spreadsheet is intended only to model brick-wall dps anyways so it's probably something that should be worked on.
I'm still a little fuzzy on 2.3 casting mechanics--does the human error factor into lag? Or does some sort of queue system deal away with that entirely if you're within some acceptable margin?
I'm still a little fuzzy on 2.3 casting mechanics--does the human error factor into lag? Or does some sort of queue system deal away with that entirely if you're within some acceptable margin?
I've wondered the same thing.
The other thing that I've wondered about is the effect of haste on unstable affliction.. I've never understood the haste mechanic well enough to wrap my head around this. I know that spell haste doesn't affect the GCD, so in perfect conditions reducing the cast time of UA doesn't help since you're limited by the GCD. However, if lag makes your UA effectively a 1.6s cast is there any value in reducing the UA cast time to 1.4s so that it's 1.5s with lag? I guess the question can better be phrased: are cast times and the GCD both affected by lag or just cast times?
Does your model account for Damage Per Cast Time loss at the end of the fight? i.e. you should not recast Corruption with 8 sec left in the fight and whatnot. On a 600 sec fight that can be a percentage or two uptime discrpancy. Not that I think uptime is that important anyway.
As an aside, regarding the OP, I think saying to "maximize your Damage Per Cast Time" is materially different than the tip under affliction that says something like "get all dots up and then shadowbolt". :P
My question: how accurately is haste rating modeled for affliction locks on a stand-and-cast fight using the Leulier spreadsheet? It's a long post, you've been warned.
The Leulier spreadsheet determines overall damage by calculating ratios of spells cast instead of developing a casting timeline (which some might call a casting rotation). According to the spreadsheet, spell haste simply reduces the cast time of shadowbolt, thereby increasing the number of shadowbolts that are cast relative to other spells that are cast.
...
Like I've said, it's a crude estimate and the numbers are likely to change when factoring in ISB, nightfall, curses, and taps, but I do believe it's significant enough to try to somehow incorporate into the spreadsheet. My hunch is that spell haste is undervalued for affliction locks on the Leulier spreadsheet by 25-50%.
Thoughts?
My thoughts?
For me as a destro lock who spams Shadowbolt, haste barely beats +damage, and I won't stack too much of it. After the first minute I basically have a LT + 4xSB rotation, and haste doesn't speed up that lifetap.
As affliction, nearly half the time you're casting spells that don't even benefit from haste. Most affliction warlocks also use Immolate, which does not benefit either.
Haste is nice because it makes your spells land faster. It's convenient when fearing, banishing, and in 5 mans. But for consistent dps on bosses, haste sucks compared to other stats for affliction. And most haste pieces have just that: haste and +damage. No hit or crit.
Although I have to admit, hit is a tad overrated as wel, on the end game gear. I'm currently trying to find ways of ditching my excessive +hit, since I'm far over the cap with an elemental shaman.
For me as a destro lock who spams Shadowbolt, haste barely beats +damage, and I won't stack too much of it. After the first minute I basically have a LT + 4xSB rotation, and haste doesn't speed up that lifetap.
Did you not read my post? It's about the effect of haste on filler spells between dot casts particularly applying to affliction locks.. clearly this doesn't apply to destruction locks because most destro locks don't use dots. I should have specified that I was looking for thoughts on the math of the model, not just general opinions about playstyle.
As affliction, nearly half the time you're casting spells that don't even benefit from haste.
The point of my post is that spells that aren't traditionally affected by haste (UA, Co, and SL) ARE affected by haste.. not because their cast times are reduced but because the average amount of time between dot end to dot refresh is shorter because of the shorter cast of the filler. The model is a simplified demonstration of this.
But for consistent dps on bosses, haste sucks compared to other stats for affliction. And most haste pieces have just that: haste and +damage. No hit or crit.
This is such a useless statement.. saying one stat "sucks" means almost nothing. I will agree that spell damage and spell hit are more powerful stats than spell haste for an affliction lock.. I don't think that anyone is disputing that. The question is: by how much? The game gives us a fixed number of items with a fixed set of stats. By understanding the value of 1 point of X stat to overall damage done we can get an idea of how to choose one piece of gear over another. Obviously I can't just take every piece of gear that has spell hit and spell damage on it because I'll be well over the hit cap. So I have to find the best compromise of other stats. The quintessential example at my gear level (and progression on my server) is the Loop of Cursed Bones vs. Vindicator's Pendant of Conquest.
Does your model account for Damage Per Cast Time loss at the end of the fight? i.e. you should not recast Corruption with 8 sec left in the fight and whatnot. On a 600 sec fight that can be a percentage or two uptime discrpancy. Not that I think uptime is that important anyway.
No, it doesn't account for exceptions in the last 30 seconds of the fight.. that's because the model is intended to find a realistic amount of dot gap given lag and resists. Since the same rules apply for 0% haste up through 10% haste, it shouldn't matter how the last 30 seconds are handled. As for the effectiveness of haste, one could argue that haste is particularly valuable because of the chance of getting just 1 more shadowbolt in before the kill once you hit the "stop casting dots, start spamming sb" phase.
Without accounting for the Felguard, Cursed Bones is better overall for any spec vs Ritssyn's.
The version 1.17 Warlock spreadsheet accounts for the extra 11 AP the Felguard gains when it gives stat weights.
In addition, spell haste makes 2 Tier 5 stronger (since you cast more SB in the same amount of time).
It is up to you which is better, but it is a close call either way.
12/17/07, 8:16 AM
))) ) but also getting paid to respec demo each week so really can't comment that much since I got it easy, as in I do whatever it takes to not die, my dps is irrelevant in a 25 min enrage fight.
Similar Threads
