Crepusculu

I wasn't necessarily suggesting that you should drop all your points in Improved Mind Blast to fit the scheme. The 0, 1, and 2 levels of mind blast I'm brushing off, but displaying for completeness.

Usually what happens in combat (with an odd number in haste) is that I'll run into situations where I have a fraction of a cooldown before MB cooldown is up. Since my DPS cycle is built around casting as many mindblasts as possible, the only real increase to my dps occurs around agreements in time, otherwise I'm waiting for the cooldown. If you take 5/5 mind blast, but the majority of the time you never actually use its last 0.5 or 1.0 seconds on the cooldown, then it seems wasteful.


The haste formula at 70 for getting a 5.5 second cast
5.5 = base / ( 1 + hasterating/1570)
We want to solve in terms of hasterating
hasterating = 1570 * (base/5.5 - 1)

For a 6 second base (4 GCD)
hasterating = 1570 * (6/5.5 - 1)
hasterating = 142.73

5 GCD
hasterating = 1570 * (7.5/5.5 - 1)
hasterating = 570.91

Imp MB back up to 4/5 for 5 GCD, 7.5 second base time.
hasterating = 1570 * (7.5/6 - 1)
hasterating = 392.5


Hense, I think 143, 393, and 571 are ideal targets, and to a lesser extent, 241

Addons
http://www.wowinterface.com/list.php?skinnerid=35267

 

rooj

I never thought that the primary benefit of haste was with MB since it is non-spammable and has a cooldown. I always viewed the primary benefit of haste was the increased output from MF.

 

Crepusculu

Unless my math is mistaken, MB and Death are at least double DPCT (damage per casttime), VT is triple, and Pain is SIX times DPCT over casting Mind Flay. You want to cast mind flay the LEAST as possible against your other spells. That's why I wasn't a fan of spellhaste before it modified global cooldowns.

Addons
http://www.wowinterface.com/list.php?skinnerid=35267

 

rooj

Increased haste allow you to cast MFs quicker (they channel faster) which means you can fit more MF ticks into other spell cooldown periods. MF is also spamable, which means you are not waiting on the cooldowns. This is why you see people with higher values of haste see MF as a higher % of their total dmg output.

Despite the fact that MF is our "filler spell" until cooldowns are up, it is THE spell which gets the highest benefit from haste. I don't see why calculating around MB cooldowns are helpful at all, when most of us run 5/5 IMP MB. Remember haste doesn't lessen the MB cooldown, it just conceivable starts the CD earlier.

 

tedv

I believe your primary claim is that it is better to delay the casting of Mind Blast by a half or full second if it lets you get another full GCD worth of cast time in. You claim that with 5/5 blast you usually waste 1 second of cooldown from casting other spells. Another way of looking at the claim is that with 3/5 blast, you occasionally have to wait an extra second to cast your next Mind Blast (which also defers the cooldown of all future mind blasts by a second).

I disagree that trimming down to 3 or even 4 points in Mind Blast would result in a negligible loss of DPS! The extra delays add up often at every haste value I've been at (up through 300), measured by the number of times the cycle works out perfectly with 5/5 improved blast (at least 75% of the time there's no waiting).

In my anecdotal experience, Mind Blast has cooled down but I can't cast it yet roughly once every 90 seconds, or 4 times in a typical fight. That almost always occurs when the Vampiric Touch and Mind Blast cooldowns happen to coincide. I suspect you aren't interrupting your Mind Flays at the 2 second mark to cast Mind Blast when the cooldown is up. Without breaking flays then yes, you won't notice the DPS loss from taking only 3/5 or 4/5 imp blast, but only because you weren't taking full advantage of the damage 5/5 blast provides in the first place.

 

Crepusculu

The only way you could possibly say mind flay receives the highest benefit from haste is simply that we are casting it the most. My definition of benefit is SCALING. All spells receive the same flat rate percentage from spell haste. The only reason haste geared players cast more mind flay is because everything else is on cooldown. People aren't casting more mind flays because it scales better, but rather that they are forced to use it.

Mind flay is call a "filler" for the reason of its poor scaling. Getting gear-stats for the SOLE purpose of casting more of your worst scaling spell is bad plan, because it will come at the expensive of improving your better scaling spells. Had I any choice about gear, I would remove all the crit, most of the hit, stick on 144 haste and stack everything else into pure shadow damage. Since I don't have that option, I will try to optimize haste around my better scaling spells. Hense, my small table.


@ Tedv
Thanks for your input.

I've tried clipping flays on a few Brut attempts, but wasn't able to determine how effective it was. I'll see if I can work out some numbers.

Addons
http://www.wowinterface.com/list.php?skinnerid=35267

 

rooj

That is exactly what I mean. If increased haste allows us to cast more MFs with more effective ticks.. then guess what, it benefits more from haste than does MB and VT. (We agree that SWD and SWP don't benefit much from haste correct?).

The question then becomes does increasing haste create higher dmg output from MB vs MF? Once again, if you look at WWS and compare low haste higher dmg SPs vs lower dmg higher haste SPs, you will see that MF's % output dmg increases as haste increases compared to MB.

Sadly, I don't have the gear to show the difference or I would use Dr.Boom and show the % difference, but Tedv did make a post in the Help me melt faces that showed his numbers on brutallus as compared to numbers when he had less haste.

Now you make the very relevant point that MB' dmg over cast time is much higher than MF, which seems to indicate that haste should affect MB more than haste. While the dmg over CASTTIME is much higher, the damage over time is actually much lower because of MB's cooldown. IE Haste ONLY affects the casttime which is a small % of time between MB casts. At 5/5 imp MB the cooldown is still 5 seconds compared to a 1.5 second cast. Haste affects ALL of the time between casts on MF.

Let's take a quick look, (untalented)
MB at rank 11 is 1.5 second cast, with an 8 second cooldown for 708-748 dmg
MF at rank 7 is a 3 second channel no cooldown and 538 dmg.

Haste doesn't affect the 8 second cooldown, only the 1.5 second cast, while haste does affect the 3 second channel and there is no cooldown.

Without adding any haste ratings or +dmg gear, you have MB every 9.5 seconds = 708-748 dmg.
MF over the same period of time = 1614 dmg.

Your can reduce the casttime element of MB, but at 100% haste you will get .75 casts with still that 8 second cooldown.
while at 100%haste you will get a MF channel at 1.5 seconds.

MB with 100% haste = .75 cast time + 8 second cooldown= 8.75 seconds= 708-748 dmg in a 9 second window
MF with 100% haste = 1.5 second channels, which means 6MFs over 9seconds which = 3228 dmg.

Of course we should be doing the calculation at 5/5 Imp MB which reduces the CD by 2.5 seconds which means we have a cooldown of 5.5 seconds rather than 8.

With no haste this means

MB 0 haste = 1.5 sec cast with 5.5 second cooldown = 708-748 dmg in 7 seconds
MF 0 haste = 3 second channel with no cool down means 2 full channels and 1/3 of another = 1255 dmg per 7 seconds

MB with 100haste = .75 cast with 5.5 second cooldown = 708-748 dmg over 6.25 seconds
MF with 100haste = 1.5 second channels which mean 4 full channels and no ticks = 2152 over 6.25 seconds.


How exactly does MF not scale better with haste?

There may be a miscalculation with respect to the tick calculations on MF, but even if was off by a significant margin, I don't see how MF DOESN'T scale much better with haste than does MB.

The HUGE discrepancy between haste's affects on MB and MF is the primary reason that haste ratings start to affect dps at the same rate as +dmg at higher levels, even though the primary benefit of haste goes mostly to only 1 of our spells in the rotation, while +dmg benefits every single one of our spells in our rotation.

BTW, this does NOT mean that we should skip casting MB, it just means that MF scales much better with haste. Skipping MBs would mean we would lose out on the extra 748 dmg over the .75 seconds (which at 100% haste is only 1 tick of MF ), but it is because during the 5sec cooldown of MB we can now cast 3 channels of MF that haste gets so much benefit for us.

 

Dorvan

My understanding is that aggressive clipping is an important part of reaping the benefits of haste. Check out this thread starting at around page 28 or so for a lot of good math on haste, with various references to clipping popping up among the discussion

 

Anedris

Haste increases shadow priest damage. The damage increase comes in the form of mind flay ticks.

 

Missa

If my calculator in it's present state seems accurate, would anyone find a function that calculated out the minimal damage/haste necessary to reach a certain DPS helpful? I think it might help with those "Can I go to TK?" type questions.

 

Balkoth

Well, it really depends. Ideally, you'll have a spriest with full FSW and Spellstrike, as well as Belt of Blasting, badge gloves, Kara trash neck, one of the Kara cloaks, Bracers of Nimble Thought, two good rings, two good trinkets, badge wand, badge off-hand, and a good 1H weapon. But that same spriest could be walking into Kara with 1-2 less pieces of gear and can walk into BT/Hyjal with only a few more pieces of gear. It really depends on how much effort a person puts into their character and/or what a raid will allow.

Er, I'm assuming you need Quartz for it to work?

The brightest light casts the darkest shadow.

 

Malleus

Beaten to it.

 

Tymir

It should work with any casting bar mod. Just replace the mod's default .tga texture with the thirds texture and rename the filename to match ththe old texture's name.

 

◊ Kalman

It also represents an increase in MBs by way of MB cooldown beginning faster - a 0.25 second reduction in MB cast time is also an effective 0.25 second reduction in the cooldown in terms of how many MBs you cast, even though the actual CD remains the same.

 

Althor

Yes, but the relative increase in DPS from haste on MB is less than with Mind Flay because of that CD. The MB component of your total DPS (assuming you can use it every cooldown) is always going to be limited by that cooldown. The absolute fastest you can cycle MB is once every 6.5 seconds. The slowest (still assuming 5/5 IMB) is once every 7.0 seconds. So haste is giving you a maximum DPS increase of 7.7% of MB's DPS.
The slowest Mind Flay is 3 seconds. The fastest (in theory) is 1 second. So in theory haste can increase Mind Flay's DPS by 300%. But even assuming you cap Haste at the same amount as that needed to reach MB's cap then that's still 1 Mind Flay every 2 seconds which means haste gives you a DPS increase of 50% of MF's DPS. And while Mind Blast hits for more than Mind Flay, a 50% increase in MF DPS's is going to outweigh a 7.7% increase in MB's DPS on an absolute level.

So yes, Haste does give a benefit to Mind Blast (and to SW, VT and SW:P for that matter) but the compared to the benefit it gives Mind Flay these are minor improvements.

 

rooj

Not that it changes anything, but you can reduce MB to 6.25 seconds (see above post) 1.5 second cast with 100% haste would be .75 seconds with a 5.5 second cd which becomes 6.25 seconds. Of course that is the period of time between the start of one MB and the start of a second MB.

I think I read the relevant posts on clipping and the dps increase, but I am trying to figure out exactly why. With only a spell rotation of MFs, clipping doesn't actually do anything.

2 * (3 seconds channels of MF) = same dmg output as 3* (2 second channels of MF).

I THINK what clipping does allow is to keep all your other spells on cooldown, SWD, SWP, VT, and MB. Without clipping you will have more spells sitting without cooldowns which will lower dps. But on that theory, why not also clip after 1st tick if you see a CD up?
(And you shouldn't clip after 2nd tick if other spells are still on their CD, or dot still running).

Btw, for people with awkward pauses between cooldowns, i wonder if they have tried getting a 1 MF tick and then casting a new spell. For a non-hasted situation that would be 1 second of casting, with significant haste ratings, this could be measured in fractions of seconds.

I apologize if this conclusion has already been either confirmed or shot down.

 

Atemporal

The GCD is still capped at 1s, you can confirm this in game by using the UI hook found here and getting yourself bloodlusted with some haste gear on. This is why MB is currently chainable in 6.5s intervals at the absolute fastest, and why clipping MFs after 1 tick is not viable (you're still on GCD from the cast). Incidentally, this is also why Breath: Haste makes [The Skull of Gul'dan] less than optimal for KJ.

The purpose of clipping is to start casting your MB/SWD immediately when their cooldowns come back up, thereby increasing dps over simply finishing the last tick of MF. If you ever find yourself in a situation where you know you will have time for 4 ticks of MF and no more, it's best to clip twice after 2 ticks (the first clip to recast MF, the second to cast MB/SWD/whatever cd is coming up).

That said though, clipping requires a very steady latency to increase DPS. Personally I've found that it often hurts more than it helps due to unstable ping times, despite it being better on paper.

 

Balkoth

The benefit of clipping Mind Flay after the second tick is clipping it RIGHT after the second tick. Some examples with 0, using 1 Mind Flay (800 damage per tick), 1 Mind Blast (2500 damage for a 1.5 second cast), and 1 SW:D (1900 damage). General damage values pulled from Wow Web Stats.

2 seconds of MF, 1.5 seconds of Mind Blast, and 1.5 seconds of SW:D = 6000 over 5 seconds = 1200 DPS
1.5 seconds of MF, 1.5 seconds of Mind Blast, and 1.5 seconds of SW:D = 5200 over 4.5 seconds = 1156 DPS
2.2 seconds of MF, 1.5 seconds of Mind Blast, and 1.5 seconds of SW:D = 6000 over 5.2 seconds = 1153 DPS
3 seconds of MF, 1.5 seconds of Mind Blast, and 1.5 seconds of SW:D = 6800 over 6 seconds = 1133 DPS.
2.5 seconds of MF, 1.5 seconds of Mind Blast, and 1.5 seconds of SW:D = 6000 over 5.5 seconds = 1090 DPS
2.9 seconds of MF, 1.5 seconds of Mind Blast, and 1.5 seconds of SW:D = 6000 over 5.9 seconds = 1016 DPS

Looking at those values, we can draw some conclusions. First, clicking a MF at 2 seconds is ideal. Second, clipping a MF at 1.5 seconds is considerably worse than at 2 seconds, and a whopping 3 DPS higher than having a 0.2 second delay on the 2 second clip. In other words, unless you can't clip to within 0.2 seconds, aiming for the 2 second mark (or hasted equivalent) is best.

Third, clipping at 1.5 seconds or having a 0.2 second delay on the 2 second mark is still better than channeling all three seconds. Fourth, clipping the spell half a second after the 2 second mark or barely clipping the last tick is a drastic DPS loss.

By the way, keep in mind this is DPS over that window of time, not your average DPS for the fight.

Edit: Oh, and what do you folks think about gear listings in the first post? Wondering if 1, I should even bother listing some good items and just let people do the calculations through a value table, 2, list the best 2-3 items for each slot from each tier of "content," 3, list the top 10-15 items per slot, period, or 4, something that isn't included in the previous 3.

The brightest light casts the darkest shadow.

 

rooj

balkoth- Are those examples with 0 spell haste? I wonder if spell haste increases or decreases the value of clipping given its effect on both GCD and cooldowns. I ask because I assume with significant haste, the second MF tick happens sooner than 2s.

Atemp- I think you are right, because the GCD never would allow clipping right after the first tick, it makes no sense to only clip MF on the first tick, but does make sense to do so after the second tick.

 

Ntrails

As I understand it hasting those values would have one effect:- Making clipping mind flays MORE sensitive. If unhasted (and they are clearly unhasted values since the 2nd MF tick is at 2secs) being 0.2 second late is a DPS loss over clipping at GCD then hasted that value would reduce and it would be necessary to be within (at a gues) 0.1 seconds of the tick to actually improve DPS, which with latency to account for strikes me as tough.

However - that is a very very set circumstance and I am not sure how far it translates given the cooldown issues, we have taken a very extreme example without needing DoTs cast and with both MB and SWD available after 1.5 seconds. If for example the option was MB ready in 3 seconds, SW:P running out in 1.5, better to MF clip at 1.5 for SWP and then midblast OR get the second tick in then refresh SWP then MB....

 

kalanikta

Got another quick question, does anyone know if mass dispell is affected by normal +hit values ?

 

Acu

Hello, I am new here My shadow priest had 1263 shadow damage and 76 hit (with all the 5 points on Shadow Focus).

My dps suck though, couldn't get it pass 530 looking at wws.


I read this thread and also talking to other shadow priests, I can't understand how some of the shadow priests here can output over 1k dps!! Hell, 800 would be a dream come true to me!

This was my standart rotation:
VT
SW;P
MB
VE (after giving the tank some time to build agroo)
SW (only when my hp is above 50%)
MF (only if everything else is in CD or the dot weread off).


I have Dottimer (sp?) and "face melter" (where I have the above spell rotation set as my priority).

I also don't stand around "doing nothing" - I am always casting something if I don't have to move due to the fight.

I had oportunities to raid with other shadow priests and although they had less bonus shadow damage, they output a much higher dps.


So, after looking at the armory page of some priests from high end guilds, I noticed their gear is stacked with haste. Because of that I decided to swap some gems around and invest in some haste to see if it would make some difference.

Currently I am sitting at 1226 bonus shadow damage and 90 haste rating (or 5%). After many tries on Dr. Boom and looking at my wws, I can get to around 700 dps - what seems low compared to the reports I read in this thread and considering that in a raiding set I can't just stand dpsing all the time. Most of the fights require some moving, so I will lost some casting time.


After reading this thread I am going to resp to take one point from Shadow Power and max my Improved MB, besides that I don't see much else I can do. So, if any of you have suggestion/advices for me I would greatly appreciate it!!! Specially concerning how to maximize my dps output (of course without sacrificing mana regen to party members).

 

Balkoth

Give me a WWS link of you DPSing and I can see what I can discover. You're likely not casting enough, losing MF ticks, clipping DoTs, or all of the above.

The brightest light casts the darkest shadow.

 

lucilebluth

I think the issue might be your wws parse...are you running recount, or another in-game damage meter to compare with your wws numbers? Even if you were doing everything wrong, your dps shouldn't be that low with those stats.

However, the only way to give you meaningful feedback would require that you link your wws and an armory profile.

 

• DeeNogger

What the fuck is the mod 'face melter' ?

I write things The word of DeeNogger New Blog! 7/6/2009