Which still puts it at 4.5 or higher crit rating for one point of damage. Remember that since Mind Blast only crits for 1.5 times normal damage, critting isn't as big of a deal as it is for just about every other class, save for returning slightly more mana.

It does turn out to be theory if it happens to proc at an inopportune time when you can't actually get the cast off (which can often happen now the buff duration is down to only four seconds). I'm not saying there is a better meta out there, just saying theory craft is, indeed, sometimes theory craft and in practice things don't always work out.

I think this has already been mentioned in this thread, but you'd have to miss a proc something like more than once every 6 times for the crit meta to be better. Considering that we only cast three spells that will consume the proc, the law of averages tells us that even if you don't actively attempt to use the proc on MF, even if you completely ignored it and used your regular cast rotation, it's still going to be the best gem available.

The buff is currently six seconds, and I generally just reorder my casts if so. The only situation where things even get a little messed up is when it procs at the start of a Mind Flay (3 seconds left when cast finishes) and you need to cast both Shadow Word Death and Shadow Word Pain. In this case I cast one of them, then the hasted flay, then the other. It's still a net increase in damage. Plus the proc averages once every 50 seconds, so you can still waste 1 in 6 procs and get one hasted flay per minute.

Missing 1 in 6 is enough to get a Focus proc one per minute. As far as when [Chaotic Skyfire Diamond] is a good choice, I ran some quick napkin math and the answer is "It's never better than 12 spell damage". Here's the math:

Generously assume 50% of your damage comes from spells that can crit, and generously assume a 40% crit rate. Without the diamond, this means 60% of the spells do 60 shares of damage and 40% deal 60 shares of damage, so the crit portion is 20 shares of 120 total. With 3% extra crit damage, this increases to .53 * 40 = 21.2 shares of damage, or an increase of 1.2 (from 120 to 121.2). That's a 1% increase in damage from sources that crit, which is a 0.5% increase in overall DPS. At 1200 DPS under these other crazy assumptions, that's still only 6 extra DPS, which is at best comparable to the 12 spell damage metagem. (Worse actually, because it doesn't come with run speed.)

So it's never good.

Except it doesn't work quite like that tedv. Most classes who are reliant on crit have talents that add to the crit damage modifier, and CSD works with that. So CSD ends up being more than a 3% bonus to crit damage in that case. If I can I'll hunt down the exact modifiers, but it I believe ranges from 3 to 6% depending on build.

Yes, but for shadow priests, it's junk and his analysis is accurate. We're not saying CSD isn't a good caster gem - we're saying it isn't a good gem for spriests.

MSD might approach the benefit of 48 spell damage if I could always consume the proc with a Mind Flay without disrupting my cast cycle, but that's just not how things work out for me in actual game play. Aside from the fact that I am a human being and will occasionally hit something other than Mind Flay by mistake while the proc is up, there are plenty of boss mechanics that make it less likely to be able to cast a Flay within the first few seconds following a proc.

Off the top of my head:

Winterchill - I get Tombed or get D&D on top of me right after proc
Anetheron - An Infernal add dies right after proc and I can't reach boss in time for Flay
Kaz'rogal - Very friendly fight, MSD is very strong here
Azgalor - Silence/RoF on me immediately after proc
Archimonde - Running away from Doomfire, Air Burst/Fear right after Focus proc
Naj'entus - I have stop casting and move to click a spine out of a neighbor
Supremus - Very friendly in phase one, but I'm frequently out of Flay range altogether during kite phase
Shade of Akama - Pretty friendly other than running down any loose adds in phase one
Teron - Very friendly fight, MSD is very strong here

Etcetera, etcetera. I'm sure you can all think of more.

On fights that require lots of mobility, like Archimonde and Supremus phase two, my DoTs can be the bulk of my damage and the +12/runspeed meta is arguably a better choice (not to mention that the extra run speed helps with survivability, assuming for some reason you don't use +9sta/runspeed on your boots). It's a similar situation for the 2-piece Spellstrike bonus and for the stacking +dam effect from the Darkmoon trinket.

And yes, it's still better to cast a proc'ed Mind Flay than it is to refresh a DoT, but that's still a little bit of missing uptime on that DoT that I would otherwise have. I occasionally trade off some minor dps losses in return for the substantial gains from the proc.

I certainly intend to keep using the MSD, but I don't entertain any fantasies that it even starts to approach something like 48 spell damage in any real-world scenario. I strikes me as more reasonable to value it at one-third to two-thirds of whatever I theorycraft the ideal value to be, shading upwards for static tank and spank scenarios, and downwards for high-mobility fights that don't always afford the luxury of staying in a full burn cycle.

The problem with this analysis is that for most examples, when you can't mind flay, you can't cast other damage spells either, so you're also getting a reduced benefit from potentially having a spell damage metagem instead. Those examples basically come down to, "In fight X you only get to spend 60% of your time doing DPS, so only 60% of focus procs help". That's true, but increased spell damage also only gives 60% of the theoretical benefit in that situation as well.

The only situations that matter are fights where you are out of range for mind flay but in range for mind blast and DoTs, and I can't think of any fights like that. You list infernals on Anetheron, but our shadow priests are instructed to stay on Anetheron full time (since the shadow priest ramp up time is so long). And for Supremus, I normally have no problem staying in Mind Flay range, although I do move around a lot to stay at max range.

(Defining 100% expected procs as the number of procs you would be using effectively on a 100% DPS time fight)
60% DPS time is far from 60% focus procs being useable, as you could just dps for a full 60% block then not dps for 40%, losing at most 1 focus proc, or you could dps-stop-dps-stop losing at most 2 focus procs, etc. Basically you should never have something as low as 60% focus procs on a 60% DPS time fight, the effect DPS "breaks" have on focus procs is in fact much much lower. To get 60% of your focus procs you would need to have 6s breaks after 40% of your spells, which has no direct connection to your % DPS time.

The problem is not spending X percentage out of Mind Flay range during a fight; the problem crops up in any situation where I can proc Focus and be unable to consume it with a Mind Flay before it expires. My point is that this can happen not only due to player error but can also be more likely due to certain kinds of boss mechanics.

Let's take a hypothetical boss mechanic where he puts a 10 second debuff on me any time I damage him; he is immune to all of my abilities while that debuff is up. How useful is MSD in this extremely contrived situation? It's worthless. I'll never be able to make effective use of the improved cast time. By contrast, I would still receive the full benefit of a +12 spell damage meta, just over a greatly reduced number of casts.

The key difference is that the benefit of +spell damage is always realized immediately at cast time. Proc-based abilities never are. Their benefit is only realized in future casts. This is a risky proposition any time I cannot guarantee that I will be able to cast another spell a few seconds after the one I'm casting right now. There are plenty of such boss encounters already in the game, and they are not friendly for any proc-based gear, including the MSD.

My secondary point is that using the proc only with a Mind Flay will occasionally require me to forego an optimal DoT refresh cycle and will cost me a small but non-negligible dps penalty due to slightly lower DoT uptime. This makes the net benefit less than 48 spell damage, even in an otherwise ideal scenario.

My final point is that I am a human being and not a robot. I will occasionally consume Focus procs with Vamp Touch or Mind Blast becase that was the next spell in my rotation and I didn't notice that the proc was up in time to change to a Flay. This will happen more often when I am tired, when I am learning a new boss, or when I am in a situation where I am distracted because I have to move a lot to avoid boss abilities that will otherwise kill me.

All of these factors add up to reduce the ideal gain you would receive from an MSD, which would be something close to the extra half a mind flay every minute that you cited earlier in the thread. That is not an incorrect statement; it's just a statement based on theory and not on practice.

To get a statement based on practice, people would have to raid with two separate gear sets that were otherwise identical with the exception of the socketed metagem and report the results of their combat log parses over many, many hours of raiding.

You are correct that the gain is not really 1/2 a MF every proc, not even close. However the loss of procs isn't nearly as bad as you describe. Overall I think the "stand and cast" dps value is what was taken into consideration from those using the simulation to evaluate the items, so the only loss over that simulation is if your next cast cannot happen becuase of fight mechanics. The % of the procs you're going to lose to fight mechanics is actually rather small. For example if a fight gives you 20s DPS time then 20s no-DPS time, you'll have 50% DPS time, however you will only lose 1/(spells cast in 20s) of the procs, which is much much smaller than 1/2. Fight mechanics lower the benefit from MSD - yes - but only a little bit. And the DoT rotation problem is probably taken into account in the simulation, although I'd go and double check this. All in all no matter what factors you'd take into account for losing MSD effects, I don't see how some spell dmg meta can outdo it (although I can see why you would favor head pieces with no meta slots )

Is there a way to mathematically model the approximate dps difference between a shadow priest with 4/5 Shadow Weaving and 5/5? Assuming that only 1 shadow priest is attacking that target, assuming its the same priest with the same gear and skill and assuming that the mob is up for only ~15-20s? The reason I ask is I'm wondering if 5/5 weaving has a significant gain on trash mobs, specifically the Hyjal trash waves.

I remember there being an equation on how to check the percent chance that the shadow weaving debuff will fall off a target with each percentage of shadow weaving and your +hit value. I cant think of how to mold that equation (which I would have to dig up) into the damage values of each spec. If only I had payed more attention in Stats class...

It'll be something along the lines of (1 - ((K)*H))^N, where K is your Weaving chance (0,0.2,0.4,0.6,0.8,1), H is your hit chance (0-0.99), and N is the number of spells you cast in a given 15 second time period.

So, for 4/5 weaving, hitcapped, using an N of

.25 VE
.5 SWP
1 VT
1.2 SWD
2 MB
2.5 MF

~7.5 in a perfectish rotation, call it 6 since casts don't always line up neatly.

Equates to a *very* low chance of it falling off (~0.008%). The bigger problem is that it can lead to higher stacking times.

(At 3/5, it's about a 0.5% chance of it falling off at any given cast.)

For Hyjal trash, stackup effects and the fact that it's hard to stack weaving on multiple targets simultaneously is probably more significant than 4/5 vs. 5/5.

Sorry... this is slightly off in a different direction here.... but since I cannot find an answer anywhere... I figured I would post and see if you guys can come up with anything.

I heard a rumor... yes... a rumor... there is a +spell damage cap. Is this true? If so, what is the cap? What is the formula to calculate the damage cap, and is there anything that can boost you beyond that cap effectively? Not pots, flasks or food or anything like that... Like a buff that allows you to go off the chart.. .like Imp Divine Spirit that increases your damage/ healing by 10% of your spirit.

I heard 1200 damage is the effective cap on damage, beyond that you get nominal returns on that +spell damage. I heard the only possible way to go beyond that are with the foods/ pots/ flasks, and Imp Divine Spirit.

Just looking for a little help from the expert theorycrafters here at Elitist Jerks!

Thanks
Tack

First, please read the posting rules, especially the parts about using complete sentences and not signing your posts.

Second, there is no spell damage cap. Your friends are either morons, misinformed, or pulling your leg.

As you've alluded to, not only does it affect you, but also the warlocks in your raid as well. I still run with 5/5 and on rare occasions we run with 3 Shadow Priests (usually 2). This would seem overkill, but...running with less than 5/5 SW lowers your DPS as well as your fellow raid member DPS, especially in situations where you are fighting multiple targets (mainly aoe). This is most noticeable in Hyjal for me. As I've mentioned in this thread, I attempt to stack misery and SW on all targets in an aoe situation. Not only is it huge DPS for me, but also improves AOE dps. Really, the numbers are probably small enough to be marginal (it's only 2% one out of every five casts) and hugely relative to your play style. Honestly though, it's a huge annoyance to me to cast a spell on a target and not see SW get stacked.

I have been running with 4/5 shadow weaving for ages. With at least two shadow priests in the raid (and we almost always run with 3, so 2 is a given) I think the percent chance of SW falling off is so low that it wont fall off but once every... year? 10 years? My question was more how much dps would be gained from a target being "weaved up" faster by a 5/5 shadow priest instead of a 4/5. There is a high chance that in a given situation it will be the same time (ie the chance that the 4/5 priest hits SW for each of his first 5 casts). There is a larger chance, I would assume, that the 5/5 would weave it faster, most likely 1-2 casts faster, and I am wondering if there is a way to put a concrete number on his dps gain because of that.

1-(.8^5) = 67.232% that at least one out of your 5 casts would fail to apply weaving. That's pretty big.

Having said that, I run with 4/5 as typically the only shadow priest in the raid now. Not having a weaving apply on the ramp up is pretty insignificant. What would be significant would be having it fall off, but the odds are extremely low. How low? Lets assume in a 15 second interval you are casting at least 6 spells. ( 1 VT, 2 MB, 1 SWD, 2 MF) The odds of at least one of those applying weaving would be: 1-(.2^6)= 99.9936%.

One can argue the trash perspective, or for something like Kael weapon phase, but if you're only putting 1 SWP up then you're losing 2% damage at most when it fails to apply. The bigger thing to get up anyways in that situation is Misery.

Is not Bracing Earthstorm Diamond somewhat underrated for a meta gem? If I did the math right when I first got my T6 helm, it's the best overall passive meta for a helm.

9 passive +dmg and 2% reduction in threat (never a bad thing).

I know with theorycraft it doesn't come out on top, but in practicality it's very nice. I'm usually in favor of things that do not have procs or uses, as it makes it easier to focus on other stuff that is more important.

Speaking from personal experience, threat is next to no issue in T6 content. All of hyjal's bosses have "low" armor (I believe 6200) which really boosts the tanks threat. Further, there is usually a mechanic that prevents you from standing and nuking for long enough to catch up to the tanks threat. Rage and Kaz'rogal are the only fights that can go uninterrupted (if you get good luck with DD and the marks of Kaz) but I have never had to slow down on either fight. BT fights are similar, Terron is a straight nuke fest, but again because of the low armor and truly unlimited rage for the tank, there seems little chance of a threat issue. Recently I did a Najentus without a paladin in the raid (read: no BoS) and with VE up full time I still didnt have to slow down. Considering that Najentus has regular armor, and the VE agro is tremendous from the consistent raid damage, I think its safe to say the minus 2% threat meta is unnecessary.

Further, the 2% less threat is only useful if you would have otherwise needed to slow down or stop dps (in which case 2% less threat is in a way 2% more damage). However, if you would have been able to keep dpsing without the 2% threat modifier, than the meta was a complete waste. I think the latter (a complete waste) will be the case for the majority of boss encounters, if not all of them, in T6.

That's pretty much the case for me too. The only boss where I have threat issues in T6 content is Gurtogg Bloodboil, especially if I don't get Tranquil Air + Salvation.

You don't need to explain BT to me as if I've never seen it...

It's the best passive Meta Gem for a shadow priest. Why do people only want the half cast one? I've been running with Bracing for several months now, and have had no regrets. Just wondering why more priests don't do the same.