I'm using Quartz with a thin line right underneath where my cast bar is. It's very subtle and non-intrusive; works great.
Did you set a spell on your Quartz options? IIRC you need to set a spell in the options dialogs for the add-on to poll the gcd from. I use Dispel Magic.

Having spent more time with the new Mystical Skyfire Diamond, I think it's actually a bit worse than the initial estimate of 24 spell damage. To really analyze the gem, you need to take into account the typical breakdown of each kind of spell cast (3 second, 1.5 second, and instant). To be really precise, you need to know for each kind, the chance of each kind of cast following it, but I won't go that in depth now. To wit:

A rough estimation says here is your typical breakdown with 0 spell haste:

In 60 seconds, a nearly flawless rotation will cast:

4 Vampiric Touch (6s)
8 Mind Blasts (12s)
4 Shadow Word Death (6s)
2 Shadow Word Pain (3s)
1 Vampiric Embrace (1.5s)
10 Mind Flay (30s)

Which leaves half a second for lag. A bit aggressive but good enough for the modeling. Now the breakdown per cast duration is around:

3.0 second cast (gives 1.0 seconds on proc): 50%
0.0 second cast (gives 2.5 seconds on proc): 30%
1.5 second cast (gives 4.0 seconds on proc): 20%

Making the assumption that each cast type has a 50/30/20 chance of following any type of cast (not necessarily true, but good enough for estimation), we assume there is a 50% chance a followup cast will take 3 seconds and a 50% chance it takes 1.5 seconds (either from cast time or GCD). We can break down the actual amount of hastable time as follows:

Proc after 3.0 cast: 50% chance 1 seconds
- 3.0: 50% chance (3.0s)
- 1.5: 50% chance (1.5s)
Net benefit: (3 * .5 + 1.5 * .5) * .5 = 1.125 seconds of haste time

Proc after 0.0 cast: 30% chance 2.5 seconds
- 3.0: 50% chance (3.0s)
- 1.5 then 3.0: 25% chance (4.5s)
- 1.5 then 1.5: 25% chance (3.0s)
Net benefit: (3 * .75 + 4.5 * .25) * .3 = 1.0125 seconds of haste time

Proc after 1.5 cast: 20% chance 4 seconds
- 3.0 then 3.0: 25% chance (6.0s)
- 3.0 then 1.5: 25% chance (4.5s)
- 1.5 then 3.0: 25% chance (4.5s)
- 1.5 then 1.5 then 3.0: 12.5% chance (6.0s)
- 1.5 then 1.5 then 1.5: 12.5% chance (4.5s)
Net benefit: (6 * .375 + 4.5 * .625) * .2 = 1.0125 seconds of haste time

Total benefit: 3.15 seconds hasted time per proc

Now it's worth noting that if you get enough haste rating, you can squeeze even more casts in the 4 second buff window, getting a larger benefit. You get a few benefits once you hit the 150 range, and more in the 350 I believe. The problem is that as you get more haste, a greater percent of your casts are mind flays, which make you much more likely to get the 1.0 second buff. Haste makes the best cases better, but the worse cases more frequent. I could work out the exact chances at each level of haste, but I doubt we'd see much change, especially given the already established conclusion that Ember Skyfire Diamond is the choice for endgame gear.

Anyway, at one proc per minute, MSD gives 3.15s / 60s of 320 haste. This is 16.8 haste rating, or only 2.8 spell damage higher than Ember Skyfire. I'll run some numbers at 150 haste (since that's how much I have) and see if the value goes up or down, but I'm not expecting much change.

Rather than the old "We only have two spells that can crit" argument, look at things this way:

We cast 5 spells, each of them makes up roughly 20% of our damage done. Sure this varies between about 15-25% and VT and SW: Death are almost always lower than MF, MB and SW: Pain, but it's a good working assumption considering the variance we get in damage done due to small sample sizes.

Also, critting isn't gaining a whole spell damage's worth of damage the only thing you cain from a crit (average crit damage) - (average non crit damage)

So, with a 10% crit rate:

20% of your damage is from Mind Blasts, of that, 0.5*0.1*0.2 = 0.01 or 1% of your damage comes from crits.

If you increase your crit rate to 20%, 0.5*0.2*0.2 = 2% of your damage will come from crits, a 1% gain from 10% more crit.

If you spent the same amount of your item budget on spell damage instead of crit, you would have a significantly higher increase in damage. This is why crit is "bad."

I know this doesn't look right, you'll say "but MB will now be a larger % of my total damage done" but the number of casts stays the same, you only gain that fraction of damage gained over the base from a crit.

But don't trust my word, look at real world data, here's a breakdown from a random WWS parse of someone I don't even know on Lurker where threat and raid damage shouldn't be a constraint: (Danô - WWS)

Damage done: 319,220

MB: 88 025 27% 19 hits averaging 2570, 10 crits averaging 3917 gain 1347 per crit, 34% crit rate.
SW: Pain: 65 645 20%
MF: 61 428 19%
VT: 56 118 17%
SW: Death: 39 990 12% 11 hits at 2060 5 crits at 3465, gain 1405 per crit, 30% crit rate
Devouring Plague: 8 014 2%

Bump this up to 45% crit rate:

MB: 16 hits for 41120, 13 crits for 50921, 92041 total, a whole 4K more damage
SW: Death: 9 hits for 18540, 7 crits for 24255, 42795 total, 3K more damage

Total damage: 326041, 102% damage with a 10% crit rate increase.

Again, the item value to increase your crit rate by anything meaningful would increase your DPS far more if it were spend on spelldamage instead.

With 10% spell haste (157 rating), Mind Flay casts in 2.72 seconds and all other spells take 1.36 seconds. Here's what a typical minute of casting looks like:

4 Vampiric Touch (5.44s)
8 Mind Blasts (10.88s)
4 Shadow Word Death (5.44s)
2 Shadow Word Pain (2.72s)
1 Vampiric Embrace (1.36s)
12 Mind Flay (32.64s)

That's 58.48 seconds total, with a cast time breakdown of:

2.72 second cast: 56%
0.00 second cast: 28%
1.36 second cast: 16%

With another 320 spell haste from the proc, the cast times drop to 2.30 for Mind Flay and 1.15 for all others. Here's the probability breakdown for each case. Sorry it's a lot of data and most of you won't read it, but it's there if you care.

Proc after 2.72 cast: 56% chance 1.28 seconds
- 2.3: 56% chance (2.72 of spell cast in 2.3 seconds = .42 seconds gained)
- 1.15: 44% chance (1.36 of spell cast in 1.15 seconds = .21 seconds gained)

Net time saved: (.42 * .56 + .21 * .44) * .56 = 0.183 seconds of additional cast time

Proc after 0.0 cast: 28% chance 2.64 seconds
- 2.3 then 2.3: 31.36% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)
- 2.3 then 1.15: 24.64% chance (4.08 seconds cast in 3.45 seconds = .53 seconds gained)
- 1.15 then 2.3: 24.64% chance (4.08 seconds cast in 3.45 seconds = .53 seconds gained)
- 1.15 then 1.15 then 2.3: 10.84% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)
- 1.15 then 1.15 then 1.15: 8.52% chance (4.08 seconds cast in 3.45 seconds = .53 seconds gained)

Net time saved: (.82 * .422 + .53 * .578) * .28 = 0.183 seconds of additional cast time

Proc after 1.5 cast: 16% chance 4 seconds
- 2.3 then 2.3: 31.36% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)
- 2.3 then 1.15 then 2.3: 13.80% chance (6.78 seconds cast in 5.75 seconds = 1.03 seconds gained)
- 2.3 then 1.15 then 1.15: 10.84% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)
- 1.15 then 2.3 then 2.3: 13.80% chance (6.78 seconds cast in 5.75 seconds = 1.03 seconds gained)
- 1.15 then 2.3 then 1.15: 10.84% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)
- 1.15 then 1.15 then 2.3: 10.84% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)
- 1.15 then 1.15 then 1.15 then 2.3: 4.77% chance (6.78 seconds cast in 5.75 seconds = 1.03 seconds gained)
- 1.15 then 1.15 then 1.15 then 1.15: 3.75% chance (5.42 seconds cast in 4.6 seconds = .82 seconds gained)

Net benefit: (1.03 * .322 + .82 * .678) * .2 = 0.178 seconds of haste time

That's a total of .544 seconds of cast time gained each proc, once per minute. So this lets you cast 60.544 seconds of spells in 60 seconds, or equivalent affect of 0.9% spell haste. That's the benefit you get from 14 spell haste rating.

In other words, as you gain more spell haste, Mystical Skyfire Diamond is even worse. So right this very moment, Ember Skyfire Diamond is probably better than Mystical if you've been farming black temple for a while.

I'm having difficulty trying to find the right metric to judge how good our shadow priests are at returning mana from wws logs. Is it the number of ticks of VT, or the average tick? I'm trying to find it so I can plug the mp5 I get from our shadow priests in Rawr.

In general? Their damage/DPS divided by 20. VT returns mana from all spells, not just the VT ticks. Dividing by 21 or 22 may be slightly more accurate due to DoT downtime.

For mana returns, (Mana gained from VT)/(damage done) 0.047-0.048 (4.7-4.8%) is good on static fights where you just stand there and wail on a boss. On fights where there are other things to do, 0.044-0.046 is about par. Under 0.04 is bad and needs improvement.

For single target fights, take the (DPS time)/3 for the optimal number of ticks for both SW:Pain and VT, they should be very close to 100% on static fights, lower as movemet increases but I don't know of any single target other than Supremus where less than 80% is reasonable. SW: Pain should have a higher uptime than VT on movement or interrupted DPS fights since it is instant cast.

(DPS time)/7.5 for 4/5 imp Mind Blast to determine the optimal number of Mind Blasts. If you're doing better than 60% you're doing OK. 5/5 would be DPS time/7, etc. It's casting time + cooldown for max number of casts.

(DPS time)/12 for SW: Death. Should be a higher percentage than MB on fights where you can use it, 70-80%+ is good.

That's for uptimes, it tells you if your shadow priests are actually casting constantly or not. Not casting 100% of the time is the biggest cause of low DPS.

Other things to look for are: Is he using his on use trinkets? What kind of potions and how often is he using them? Is his shadow fiend regularly getting close to 10 hits in? Does he have high DPS time (as a dot class it should usually be among the highest)? Is he dying to things he shouldn't? Is he consistently getting a large number of resists?

The absolute rock bottom minimum DPS is his average mind flay tick since the simplest thing is to stand and spam flay. If you can't do better DPS than that you're screwing up seriously.

Just discovered the MSD doesn't have 45s internal cooldown anymore, 36 seconds or less so that should make it a bit better at least.

http://elitistjerks.com/691498-post105.html

haven't seen anyone discuss this yet so a quick post about [Timbal's Focusing Crystal]. From the testing I did on dr boom last night it procs about 2-3 times a minute for 500 a hit. It scales with imp shadow bolt and with CoS which will put it up at around 600 a tick fully raid buffed. To the best of my knowledge it doesn't consume Imp SB. This means that it is adding about 25 dps or about 50 sp. Making the trinket about a 94 sp trinket which would put it as the top shadow priest trinket.

Aye, need to start farming for it - since it is certainly better than icon that I have at the moment. Probably hex head too. Blessings would be interesting.

Can it crit? I assume not.

yes it can crit, it seems to be based on base crit chance. If only it were on MB/SWD crit chance hehe.

The proc on [Timbal's Focusing Crystal] made up about 2% of my total damage on last night's Kalecgos tries.
(and there are actual shadowbolts flying)

Similar results, 2-2.5% damage dealt by the proc. It has an associated shadow bolt animation, and if you're standing still when it goes off your character will actually do a cast pose, but it doesn't interrupt any on-going cast or channel.

2x Tank trinket and that usless healing trinket so far. Hopefully Ill get Timbal's soon as Brut is right around the corner.

It's not one of your shadow spells, and therefore doesn't benefit from talents. I believe that on Kalecgos I had 7% resisted.

Still, 2% damage at roughly 1450 DPS is 29 DPS, or 58 spell damage. Including the innate damage, that puts the estimated spell damage equivalent at 102, just like the theorycraft estimates. I guess I'll have to farm one of these. Whether or not it's better than Hex Shrunken Head depends on fight duration, and it still seems worse than Skull of Guldan, but it's definitely competitive.

Your rating buster probably hasn't been updated for 2.4. I have almost as much int as a holy priest (630 for me, 700 for them). The primary difference there is that they use 5 int / 11 healing gems in yellow slots while I socket spinels. Draenic Wisdom is flat out better than Mageblood for any class with a meditation style talent.

Not even mentioning the fact it also gives a small boost in DPS, while mageblood does not.

My analysis resulted in ~29dps as well, after a few tries on Brutallus the other day.
The crit chance seems to be static at 5% at least it varied from 4.x-6.8% during the tries I was evaluating it.
From visual memory the internal cooldown+time-to-proc seems to be in the range of 15-30secs.

Did anyone noticed if it triggers VE/VT? I only thought about it now, but I'm guessing it does.

Short list of what people have commented on the Timbal's Focusing Crystal:
Affected by Shadowform, Misery, Shadow Vulnerability, ISB procs, CoS
Not affected by Darkness
Procs VE and VT health and mana returns
Does not consume ISB charges.
Crit rate of the proc might based on your char screen's crit rate (from your int and gear)

Average spriest will get 2-3 procs a minute, depending on how many MF ticks you get in or luck.

So by that logic it is better with haste than damage? Relating to the earlier discussion a haste set would get more tics from more casts of mindflay and therefore more procs & damage from the trinket.

Calistella - WWS is my WWS from Black Temple last night using Timbal's for the entire raid (replacing Blessings Deck). Echoing what was already said about it being 2% of total damage. It is definitely affected by the aura from Essence of Anger as well.