So the actual math. I don't want to use a theoretical spirit value because ilevel is different every patch. We might have more or less spirit on our gear, we might have more gem slots available for spirit. The trinkets can add more or less regen now that Rapture doesn't benefit from temporary spirit effects.

Instead, we should make an equation based on the proposed initial conditions and solve for how much spirit is needed.

For napkin purposes, you could say I need enough mana for 7 minutes. So

420 seconds * (burned mana per second - average mana regen gained per second) = 300,000 mana.

Well, ignoring power infusion, pw:s spam 8 times and solace once would be 8788.1356 mana per second, so plugging that into the above formula we can solve to find how much mp5 we need, but this isn't accounting all of the variables I listed in the previous post, so this is merely an estimation.

420*(8788.1356-x)=300000 or x = 8073.849886 or 40,369 mp5.

For some more napkin math, shadowfiend does 3% per swing, 8 swings, every 3 minutes, or 666.66% mp5. And solace is going to be cast every 10.62 seconds, so we know that would be 1412.43 mp5. If you subtract these two mp5 values from the above mp5, we only need to come up with 38,290 mp5 from both spirit mp5 and rapture. Here's a formula:

spirit regen + rapture regen = 38,290 mp5

(300000*0.02+(1.1287*0.5*x)) + (((1.5*x)/18)*5) = 38,290

6000+0.56435x+0.4166666x=38290

0.56435x+0.4166666x=32290

0.981016x=32290

x=32914.855619

or you would need 32,290 spirit to be able to sustain pw:s spam for 7 minutes.

But that's just the napkin math. I don't want to evaluate something that isn't going to ever be viable this xpac, so I'm not going to do the proper math for how much spirit you would need. What we could do is take a realistic amount of spirit and assume rapture will proc off cooldown (in the case of an aura fight), and see how long we would actually last on a napkin.

13,000 spirit was proposed in bis t14 gear. But there are some traps when looking at gear. Secondary stats are weighted simply by decisions from the item designer. Also, the scaling between ilevels isn't linear anymore, so you don't know how your gear is going to be next tier. But tier 14 is the current tier, so that's something we actually know, I just wanted to raise concerns about predicting the future of ilevels and scaling. So:

300000 / { [8788.1356] - [(300000*0.02+(1.1287*0.5*13000)) + (((1.5*13000)/12)*5) + 666.6666 + 1412.43 ] }/5 = time until oom

Solve that and you can sustain bubble spam for 51.76 seconds.

These disagree with the OP because shadowfiend with a 90k return would be the return value of a fiend under heroism. My fiend is much lower at 72k. Also, I just did a pw:s with pw:solace for the ease of napkin. also, our pw:s numbers differ (18,300*0.75*0.85=11666.25). Also, Kales is proposing that the priest stand around idle for 23% of the encounter, which is a strategy arguably. Sustainability also differs when spirit shell enters the equation.

But what's going on in the original post is the rotation is taking a break from bubble spam to cast spirit shell once per minute. If the rotation is changing that drastically, I can't just eyeball it with napkin math. I could do a theoretical breakdown of each second per encounter, but the consideration that must be thought of here is that pw:s only lasts on targets for 15 seconds. If you cast 151 shields for 178 seconds during an encounter, that's a maximum of 2265 seconds of total shield uptime across the raid. Across 10 people, that's 226.5 average uptime per person over a 420 second encounter. Even if all of these shields were effective healing, that's still a 53.9% effective use of shields for hps.

But even if that were the case on an aura style fight, the shield spam style is supposedly for stopping burst damage, right? Well, between bursts, the window of time you're looking at before the burst is very closely tuned to 15 seconds, or else your shield will start going into overhealing. So at that window, you could do 12 casts, for 12*(19428+(1.871*spellpower))*(1+mastery). Alternatively, you could use that time to build evangelism stacks to do 4 prayers of healing, or 1 well placed cascade and 3 prayers of healing, or 1 well placed halo and 2 prayers of healing, all of which would be half of the mana cost, and wouldn't stop all other healing for 15 seconds (so averaged to be more overall hps and dps).

I guess what I'm getting at is that this proposal is an interesting idea, but it isn't practical. Don't get me wrong, I'm a fan of bubble spam (see my elegon video), but this change isn't really going to make pw:s spam viable, and I would only lean on it if extreme burst aoe was about to wipe the raid and my spirit shell, barrier, inner focus, level 90 spell were all on cooldown AND my other healers were oom

Instead, we should make an equation based on the proposed initial conditions and solve for how much spirit is needed.

For napkin purposes, you could say I need enough mana for 7 minutes. So

420 seconds * (burned mana per second - average mana regen gained per second) = 300,000 mana.

Well, ignoring power infusion, pw:s spam 8 times and solace once would be 8788.1356 mana per second, so plugging that into the above formula we can solve to find how much mp5 we need, but this isn't accounting all of the variables I listed in the previous post, so this is merely an estimation.

420*(8788.1356-x)=300000 or x = 8073.849886 or 40,369 mp5.

For some more napkin math, shadowfiend does 3% per swing, 8 swings, every 3 minutes, or 666.66% mp5. And solace is going to be cast every 10.62 seconds, so we know that would be 1412.43 mp5. If you subtract these two mp5 values from the above mp5, we only need to come up with 38,290 mp5 from both spirit mp5 and rapture. Here's a formula:

spirit regen + rapture regen = 38,290 mp5

(300000*0.02+(1.1287*0.5*x)) + (((1.5*x)/18)*5) = 38,290

6000+0.56435x+0.4166666x=38290

0.56435x+0.4166666x=32290

0.981016x=32290

x=32914.855619

or you would need 32,290 spirit to be able to sustain pw:s spam for 7 minutes.

But that's just the napkin math. I don't want to evaluate something that isn't going to ever be viable this xpac, so I'm not going to do the proper math for how much spirit you would need. What we could do is take a realistic amount of spirit and assume rapture will proc off cooldown (in the case of an aura fight), and see how long we would actually last on a napkin.

13,000 spirit was proposed in bis t14 gear. But there are some traps when looking at gear. Secondary stats are weighted simply by decisions from the item designer. Also, the scaling between ilevels isn't linear anymore, so you don't know how your gear is going to be next tier. But tier 14 is the current tier, so that's something we actually know, I just wanted to raise concerns about predicting the future of ilevels and scaling. So:

300000 / { [8788.1356] - [(300000*0.02+(1.1287*0.5*13000)) + (((1.5*13000)/12)*5) + 666.6666 + 1412.43 ] }/5 = time until oom

Solve that and you can sustain bubble spam for 51.76 seconds.

These disagree with the OP because shadowfiend with a 90k return would be the return value of a fiend under heroism. My fiend is much lower at 72k. Also, I just did a pw:s with pw:solace for the ease of napkin. also, our pw:s numbers differ (18,300*0.75*0.85=11666.25). Also, Kales is proposing that the priest stand around idle for 23% of the encounter, which is a strategy arguably. Sustainability also differs when spirit shell enters the equation.

But what's going on in the original post is the rotation is taking a break from bubble spam to cast spirit shell once per minute. If the rotation is changing that drastically, I can't just eyeball it with napkin math. I could do a theoretical breakdown of each second per encounter, but the consideration that must be thought of here is that pw:s only lasts on targets for 15 seconds. If you cast 151 shields for 178 seconds during an encounter, that's a maximum of 2265 seconds of total shield uptime across the raid. Across 10 people, that's 226.5 average uptime per person over a 420 second encounter. Even if all of these shields were effective healing, that's still a 53.9% effective use of shields for hps.

But even if that were the case on an aura style fight, the shield spam style is supposedly for stopping burst damage, right? Well, between bursts, the window of time you're looking at before the burst is very closely tuned to 15 seconds, or else your shield will start going into overhealing. So at that window, you could do 12 casts, for 12*(19428+(1.871*spellpower))*(1+mastery). Alternatively, you could use that time to build evangelism stacks to do 4 prayers of healing, or 1 well placed cascade and 3 prayers of healing, or 1 well placed halo and 2 prayers of healing, all of which would be half of the mana cost, and wouldn't stop all other healing for 15 seconds (so averaged to be more overall hps and dps).

I guess what I'm getting at is that this proposal is an interesting idea, but it isn't practical. Don't get me wrong, I'm a fan of bubble spam (see my elegon video), but this change isn't really going to make pw:s spam viable, and I would only lean on it if extreme burst aoe was about to wipe the raid and my spirit shell, barrier, inner focus, level 90 spell were all on cooldown AND my other healers were oom

*AND*I had half of my mana bar to dump.