It would pay to spend the first 2 mins of the fight sitting on your hands rather than the last 2 mins. Your party will be full mana at the start, possibly not so at the end. While as shadow we do damn good damage for a support class, in reality thats all we are. Number 1 job is to be returning mana so it'll be best to be returning it when your party needs it as opposed to keeping them topped up when they probably don't need it as much.
In any fight with an aggro ceiling, the optimal strategy is to front-load your dps. The reason is simply: you can't predict how long it will take you to do your total damage quota accurately enough to end at the same time as the rest of the raid does. If everyone nukes hard from the start they'll end up doing the same total damage and the boss will fall over earlier. This is the same as the optimal strat discussions on Conquest's boards back when Broodlord was relevant.
Furthermore the longer you draw out the fight, the closer the healers will come to the threat cap - and they can't shed aggro or hide in a corner when they hit the cap. AoE healing classes build threat fast - normally not noticable but you can easily see this with Chain Heal shamans on VR if you lose some dps early.
Honestly? If I was using Lootzor as an aspiring shadow priest, I'd probably just value +damage and leave everything else blank. Okay, there's some value to having some +hit when you're starting out and crit has some value too, but it's better to be in the mentality of +damage > all. The 6% +hit you need for bosses is probably going to come naturally on the gear you get and crit really is extremely poor early on, it only gets a little better as your sustainabiility and damage increases.
Why? If 6 crit is equal to 1 dmg you could very easily put 1 point per crit rating and 6 points per spell dmg and get more accurate results on lootzor for example... If you have more data no reason not to use it to make better gear choices, as there may be items where it'll actually make the difference.
Alright, this is my first post here on Elitists, and while I am a dedicated raider, im not really up to speed on all this theorycrafting, so I figured I would ask the experts, or atleast the guys who knows more about it then me, aka You.
With spellhaste being normalised at 15.7 for both casters and melee after patch 2.3, I have been looking into changing things up a bit. While I have been reading about the MB rotation discussions (To have 1/5 imp or 4/5 or 5/5) I have still not been able to find a post adequatly explaining the synergy between spellhaste, MF and MB rotations.
aka, at what Spellhaste rating will you be able to squeese in that extra MF between MB cooldowns.
Ex. With (?) spellhaste, I can lower my imp MB to 3/5 and squeese in an additional MF in the rotation between DoT applications.
As I said, im a bit new to theorycrafting, but I think the above example illustrates my problem.
For now, I have 4/5 imp MB, and 0 spellhaste. (Have some gear, but not using it yet) and I usually run my dots, then MB, MF, MF, MB in my rotation. Pretty standard for most bossfights anyway.
In any fight with an aggro ceiling, the optimal strategy is to front-load your dps. The reason is simply: you can't predict how long it will take you to do your total damage quota accurately enough to end at the same time as the rest of the raid does. If everyone nukes hard from the start they'll end up doing the same total damage and the boss will fall over earlier. This is the same as the optimal strat discussions on Conquest's boards back when Broodlord was relevant.
Furthermore the longer you draw out the fight, the closer the healers will come to the threat cap - and they can't shed aggro or hide in a corner when they hit the cap. AoE healing classes build threat fast - normally not noticable but you can easily see this with Chain Heal shamans on VR if you lose some dps early.
Very valid points, thanks. Gives me something to think about
In any fight with an aggro ceiling, the optimal strategy is to front-load your dps. The reason is simply: you can't predict how long it will take you to do your total damage quota accurately enough to end at the same time as the rest of the raid does. If everyone nukes hard from the start they'll end up doing the same total damage and the boss will fall over earlier. This is the same as the optimal strat discussions on Conquest's boards back when Broodlord was relevant.
Furthermore the longer you draw out the fight, the closer the healers will come to the threat cap - and they can't shed aggro or hide in a corner when they hit the cap. AoE healing classes build threat fast - normally not noticable but you can easily see this with Chain Heal shamans on VR if you lose some dps early.
Combined with the fact that many classes have special abilities for mobs at 20%.
So you can go fullout again at 20%, with fulll mana and CD's.
In any fight with an aggro ceiling, the optimal strategy is to front-load your dps. The reason is simply: you can't predict how long it will take you to do your total damage quota accurately enough to end at the same time as the rest of the raid does. If everyone nukes hard from the start they'll end up doing the same total damage and the boss will fall over earlier. This is the same as the optimal strat discussions on Conquest's boards back when Broodlord was relevant.
Furthermore the longer you draw out the fight, the closer the healers will come to the threat cap - and they can't shed aggro or hide in a corner when they hit the cap. AoE healing classes build threat fast - normally not noticable but you can easily see this with Chain Heal shamans on VR if you lose some dps early.
Actually this isn't necessarily true. When my guild first started voidreaver I was the only shadow priest and I was aware that I needed to slow down my damage to ensure that I survived until the end (or very near the end of the fight).
You are correct that I will do the same damage whether I start fast or slow (and indeed if I start slow I risk not being able to catch up later). However, my contribution is more than just my damage - if I die early then shadow weaving and misery will drop off the mob and cut the damage that our mages/warlocks can do.
The same applies for any class applying debuffs to the mob such as scorch or curse of XXX (although less of an issue as curses last 5 minutes). It will also apply to shamans who drop totems.
With spellhaste being normalised at 15.7 for both casters and melee after patch 2.3, I have been looking into changing things up a bit. While I have been reading about the MB rotation discussions (To have 1/5 imp or 4/5 or 5/5) I have still not been able to find a post adequatly explaining the synergy between spellhaste, MF and MB rotations.
aka, at what Spellhaste rating will you be able to squeese in that extra MF between MB cooldowns.
Ex. With (?) spellhaste, I can lower my imp MB to 3/5 and squeese in an additional MF in the rotation between DoT applications.
You should not be removing points from improved mind blast to fit additional MFs in. Here are some numbers on 390 spell haste rating; this is the maximum amount of spell haste currently available in game on cloth, including healing pieces.
(100/(100+(390/15.8)))*3.0=2.406s MFs.
Even at 850 average MF ticks (which you wouldn't get with this much haste gear):
(850*3)/2.406=1059 dps from MF.
Assuming 2400 average MBs, which is on the low end:
2400/1.5=1600 dps from MB.
This math is intentionally skewed in favor of MF, and it is still not even close to MBs dps. Bottom line: Take 5/5 IMB and use it on cooldown. Anything less is suboptimal dps.
About spellhaste and MB/MF/MF/MB subcycles:
With zero latency and perfect chaincasts, you would need 198 spellhaste to fit 2x MFs directly in between a 5/5 IMB cycle. In reality, you end up needing slightly higher than that to compensate for lag.
However, the short version of it is that spell haste is not an all or nothing deal- incremental gains will result in visible dps increases. As you gain spell haste you will start to notice your % usage breakdowns on WWS skewing more towards MF; this is the result of being able to fit in additional ticks in between small amounts of movement or other odd breaks in your casting sequence that would have been wasted otherwise. It helps to visualize it if you think of MF not as a channeled spell with 3 ticks, but as 3 separate spells each with one tick. As each tick gets shorter, you can start to fit them in during times on fights where you would otherwise have had to wait to cast.
The old thread haste discussions: pages 26 / 27.
Dedmonwaukeen's simulated haste analysis is here. Take note that dedmon's sim uses raw +shadow values far lower than you will actually have raid buffed in T6/haste gear, which results in his realized haste dps increases being lower than in-game. This is however a good illustration of how haste is not strictly a step function.
Actually this isn't necessarily true. When my guild first started voidreaver I was the only shadow priest and I was aware that I needed to slow down my damage to ensure that I survived until the end (or very near the end of the fight).
You are correct that I will do the same damage whether I start fast or slow (and indeed if I start slow I risk not being able to catch up later). However, my contribution is more than just my damage - if I die early then shadow weaving and misery will drop off the mob and cut the damage that our mages/warlocks can do.
The same applies for any class applying debuffs to the mob such as scorch or curse of XXX (although less of an issue as curses last 5 minutes). It will also apply to shamans who drop totems.
You need to not die. You don't need to DPS. You can and should maintain weaving/misery while doing minimal threat via Rank 1 SWP without VT.
Melador> Incidentally, these last few pages are why people hate lawyers.
Viator> I really don't want to go all Kalman here.
Bury> Just imagine what the world would be like if you used your powers for good.
I believe the estimates are fairly accurate. They are within 10% of the simulation that was linked, even if the spell damage numbers aren't correct. (For example, the .5 * Mind Flay DPS * Haste% says the first simulation will get a 45 DPS increase from 200 haste rating. In reality it went up by 50 DPS.)
Given that your unhasted mind flays will average at most 900 DPS even with the best of gear, 1% haste will be at most 4.5 DPS. That means you gain .288 DPS per haste rating. Since 1 spell damage is around .45 DPS, the absolute best case is 1 spell haste rating = .64 spell damage. In practice, .6 is more realistic. I'll note that this contradicts my previous claims on spell haste, but I won't bore you with the details of how my old math was wrong.
At any rate, spell haste is not useless and it's far better than spell crit, but you should never try to stack it over spell damage. The only spell haste item in the game which is best in slot is the trash drop caster ring from Black Temple (and that's just best for your second ring). Still, some items get close. The bracers from Rage Winterchill are only 2.5 spell damage better than Nimble Thought, and only if you get a Crimson Spinel for them.
My chief concern is about it's actual effect. All right, the ring gives me a 2.94 second Mind Flay. So what?
If I finish a double Mind Flay at 5.88 seconds I cannot use that 0.12 seconds for anything. I could come up with a dozen other situations which have the same problem. Even a triple Mind Flay would only save 0.18 seconds. What can you actually do with that?
The only answer I've been able to come up with is that it helps compensate for lag slightly. You get a tenth of a second to mess up on and not have to start that VT later. Has anyone come up with a practical answer? Yes, getting a 2.75 flay means you can get and consistently use 5/5 Improved MB...but what about a 2.94? 2.8? 2.87? Etc.
We can use that .12 seconds to start the cast of our next spell. Unless your latency is perfect and your reaction time is 0 (see:impossible) we will never have a "perfect" rotation where all of our spells line up exactly, at least for not a sustained amount of time. Even if we did, there's still the 1% resists that would throw it off eventually. If you factor in any movement time, which is required on almost every fight in the game, the "perfect rotation" doesn't exist. Therefore, a quicker Mind Flay cast will eventually lead to increased DPS somewhere. Whether it is finishing that MF .12 seconds early so you can move, or recast a dot after you've moved then flayed, etc. You are right though, there isn't much room for increased DPS there by stacking a minimal amount of haste.
In situations where you are low on mana or otherwise threat capped such that you can't cast SWD or MB having haste allows you to maintain a slightly higher dps mana efficient cast cycle that can sustain itself to that next pot timer or that next full burn phase. The more mind flays I can fit in between VT/SWP refreshes when low on mana the better, because at (I believe 1400 +dmg) those mind flays are essentially free.
You guys are assuming the tanks reach their max threat during the fight, which honestly doesn't seem to really happen with decent tanks in a 10 minute fight. Yeah as time passes they will generate aggro slower and slower, however they will theoretically never reach the cap, and practically they will not even get to the point where they're "close to the cap".
If you want to verify this theoretically you could assume each tank's TPS, cooldown on knockback and threat reduction, and figure out the threat your current tank would be holding at any given point in time. Even back on broodlord tanks seemed to not reach that point with a raid that doesn't slack (low threat generation + fight takes too long because some people simply don't deal damage).
So in my case since I'm using the [Handguards of Absolution] and can get the set bonus and satisfy the meta socket bonus of each of those it's best to use the [Bracing Earthstorm Diamond], that is unless the combo of:
Is worth more then it. Or to sum it up , is [Mystical Skyfire Diamond] > 6 SP (given the T6 Helms socket bonus) even in it's current form (45 second cooldown) and given that it may proc on worthless abilities like VT and MB.
You're neglecting the value of crit rating. For exmample when you're considering the imbued unstable diamond, while 15 crit rating is quite significant when the loss you're looking at is 4 spell dmg loss, taking 6 crit = 1 dmg your spell dmg loss becomes 1.5 isntead.
You're neglecting the value of crit rating. For exmample when you're considering the imbued unstable diamond, while 15 crit rating is quite significant when the loss you're looking at is 4 spell dmg loss, taking 6 crit = 1 dmg your spell dmg loss becomes 1.5 isntead.
That would get a lot more complicated because of socket bonuses and those meta gem requirements, going on a strait SP conversion of 6 crit to 1 SP the only gem that would/could be worth it are:
I'm still fairly convince that the [Mystical Skyfire Diamond] will end up the best even with it's periodic screw ups of being used on any spell other then MF.
My post was more for a "don't just ignore crit" rather than actually trying to affect your comparison. Heck I wouldn't even ignore spirit's spell dmg benefit but that's just me
[Mystical Skyfire Diamond] effectively gives you half a mind flay every minute. That's not just theory; it really works out that way. That's +1300 damage per minute, or +21.6 DPS. That's the same value as +48 spell damage. Nothing else comes close, even when you lose 6 spell damage socketing a shadowsong amethyst.
In reality you only lose 1 spell damage though, because the T6 helm socket bonus is +5 spell damage if you socket a blue gem. So it's a net +47 spell damage.
[Mystical Skyfire Diamond] effectively gives you half a mind flay every minute. That's not just theory; it really works out that way. That's +1300 damage per minute, or +21.6 DPS. That's the same value as +48 spell damage. Nothing else comes close, even when you lose 6 spell damage socketing a shadowsong amethyst.
In reality you only lose 1 spell damage though, because the T6 helm socket bonus is +5 spell damage if you socket a blue gem. So it's a net +47 spell damage.
It's not exactly that since it can work on MB and VT but it's about a 66% chance of working on MF. Still the best though.
It's not exactly that since it can work on MB and VT but it's about a 66% chance of working on MF. Still the best though.
The trick is to modify your rotation when it procs so that you cast mind flay before casting mind blast. For example, if you cast death and focus comes up when mind blast is on cooldown, you cast mind flay instead and then blast right after. The only times you miss focus procs are when your rotation has you cast two 1.5 second casts in a row and you get focus on the first. In my experience this happens less than 15% of the time. And because focus tends to proc around once every 50 seconds (45 second internal cooldown and 15% proc rate), this works out to around one useful proc per minute.
Im too stupid to figure out how much +shadow & mp5/spirit yields 100% mana returned. You guys are the theorycrafters i will never be.
I have +1000 shadow dmg. And 65mana/5s while casting. I know its attainable with raid buffs bc ive done it before, but for this question i am a 1-man show. Within a month or two i will be at 1200-1300 shadow bonus unbuffed, possibly sacrificing some mp5. if the 300 mana return meta is a big deal to you guys, sure factor it in. And i full burn, mind blast and sw;d cd's are rarely waiting on me. After 5-6 minutes, fights like prince, i only cycle VT and Pain while i enjoy my wand spec, and those 3 combined are unlimited, there is no way to screw that cycle up and run out of mana (actually u suck at life if u do that and cant regen more than you use)
Pls understand full raid buffed is a completely different thing for me. I have 1 more spriest with me and one of our officers is constantly throwing noobies in and out of our group bc they forgot to roll a spriest and be awesome. Every VR run we finish with 100% mana along with any of the lucky healers that gets to stay in the group for more than 25 seconds.
Simply tell me if i need 100mp5 while casting paired with 1k shadow. Or do i need 1400 shadow and my 65mp5. Im really bad at math, i just watch blue numbers scroll and listen to people kiss my a**. u guys are awesome
12/12/07, 8:11 PM
Similar Threads
