[Resto] Mists of Pandaria 5.3 - Page 13

Kjeldorian · 11/02/12, 7:55 PM

Note that it takes 2168 points per 1% crit, which is about 3.5 times weaker than 1 crit rating.

If you wanted to specifically raise your crit by 1% you would need 2168 Int to raise your crit by 1% versus using 600 crit rating to reach the 1% crit.

Note Stat Buff / HoTW / Leather Specialization are required for this evaluation, w/o HoTW you would need more Int per 1% crit.

Arentios · 11/02/12, 8:07 PM

Randomness is something of a valid argument against crit. Saying that it'll just result in overheal not so much; with Regrowth being 100% crit with the glyph you're only really worrying about HoTs. If your HoTs are all just overhealing on a crit then chances are the content you're doing doesn't demand anything near min-maxing.

Other points are the poor throughput conversion versus mastery in secondary stats and the fact that if you're using Regrowth one of your heals already has 100% crit. Point for point Mastery will give you more healing.

Payday · 11/03/12, 12:52 AM

Originally Posted by Hamlet

Old post, but guy who posted was talking to me in PM's and I think it's a good discussion for here.

--6652 is correct for the 9th tick of WG and Efflorescence. Table in the main thread doesn't specifically mention Effloresence, but the spreadsheet takes it into account--that part's worked the same since Cataclysm.
--In general, going from 3043 to 6652 haste gives you a tick added to WG/Efflo, and not much other benefit (a tick of RG if you're unglyphed, and a tick of Lifebloom). 12.5% healing added to WG/Efflo (9 ticks instead of 8) is about 5% total healing if those spells comprise ~40% of your healing (not sure how generous an estimate that is).
--3600 Mastery increases your Harmony bonus by around 7.5, as you said. At 20.0% base Mastery (low end, I think), this is around a 6.25% increase.

So they're actually pretty close. What it really depends on is
1) What % of your healing is done with WG/Efflo.
2) What stats you're actually choosing between. If all 3600 of the stat points you're using to get to the next breakpoint are coming right out of mastery/Int (Int on gems is very similar to mastery), I might not do it. But if you're giving up some crit/Spirit in there, the deal looks a lot better.

It's usually the case that your two best secondary stats don't compete with each other directly due to reforging. On Haste/Mastery items, you reforge the haste to Spirit, or not, based on whether you're going for the breakpoint--mastery is unaffected. Same for Crit/Mastery or Spirit/Mastery. On Crit/Haste or Spirit/haste items, you're going to reforge one or the other to mastery regardless (based on whether you're trying to move haste up or down, again).

So in practice, the bulk of reforged haste will pull out of crit/Spirit, and you only lose good throughput stats (Int/mastery) to the extent you have to regem to get the rest of the way there. So I expect that going for the haste breakpoint once it's practical is probably pretty reasonable, and will get more effective as your gear gets better (since you'll have more crit/Spirit whose best use is being turned into haste, and you'll eat less into mastery).

Generally, Rejuvenation is my top heal. WG is usually second or third depending on how effective my lifebloom is on the tank (tank damage vs raid damage fight). Swiftmend (ground hot) is usually below Tranquility.

The amount of burst raid damage in this tier means we can't just sit around waiting for WG to heal everything up. We have to use RJ to supplement as well as do it's main heal job. Swiftmend hot is really nice, but it can't always be used on a clump of people, especially on 10 man (my raid size).

I see where you are going with the reforge from spirit/crit instead of touching mastery and int, but I wonder about future gear choices. The pendulum effect of Rejuv, in particular, is my concern. Right now we can't afford to cast it enough to truly combat aoe damage, but when we can afford it with better gear, it can be a sledgehammer against it. Perhaps finding this breakpoint, if it exists, will help guide our choices. Spirit seems useful, but I don't notice a big effect on my mana pool with a difference of 2000-2500 (shifting it all to int). Keeping my gearing method the same it would take tiers of gear changes to see a bigger difference than that. Will this pendulum effect ever happen in MoP? I think it might take a mana cost change or set bonus (bigger than T14, perhaps an odd mechanic) for this to happen. Spirit increases alone most likely won't.

Currently and in the near future our only option to reach 6652 will take from mastery to a large degree. Even at a 6.25% bonus (additive) instead of my incorrect 7.5% (multiplicative), mastery still shines through as a clear winner over haste. Even when gear is capable of supporting 6652 without reforging from mastery, that haste would be better spent on further mastery, assuming haste can be kept very close to 3043. At no point until 13163 will reforging mastery to gain haste be a gain in healing. (again assuming haste kept at 3043) Can you agree with this statement? Can we stop furthering the '6652 haste is the next goal' mentality?

Old · 11/03/12, 10:03 AM

I can confirm on this. Second haste breakpoint is nowhere near the mastery lost. Until the day you will be way over the cap (3043) even after reforging all your haste out, there is no point considering going for 6652 haste.

Numiro · 11/03/12, 10:48 AM

Originally Posted by rmq

This is an argument held for ages. And I don't think it really has that much to do with crit's usefullnes. With one's preferred playstyle maybe. But if crit is listed as top priority stat on the first page most people won't be bothered by its randomness. This can happen if itemization cost of crit falls below itemization cost of mastery for example, or if value of mastery diminishes with higher level of gear.

Even if crit goes into 1.5 value over 1 point of mastery, I'd still not chose that over mastery, because it is unreliable. When you wipe from low healing it's not going to be because you didn't heal enough, it was because you didn't follow through with your healing plan. Every single fight I walk into I've got a set plan on how to deal with each event, if that plan doesn't heal as much as I expect it to (no crits) then my entire plan for the next time will be set of because I expect it to heal less (alot of crits), that's why I in general tend to never favour crit over say mastery.

Payday · 11/03/12, 12:31 PM

I don't see a problem with relying on crit with a druid. With how small our ticks are and how many ticks we have the difference between crit and not in a critical situation is not that large. If we were a class with direct heals I would agree not to rely on crit. A wild growth with 7 hits 1 crit or 5 hits 3 crits is not nearly as different as a light of dawn cast that is either entirely hit or entirely crit. In this way I think we can make use of crit as regular throughput.

Numiro · 11/03/12, 3:10 PM

Originally Posted by Payday

I don't see a problem with relying on crit with a druid. With how small our ticks are and how many ticks we have the difference between crit and not in a critical situation is not that large. If we were a class with direct heals I would agree not to rely on crit. A wild growth with 7 hits 1 crit or 5 hits 3 crits is not nearly as different as a light of dawn cast that is either entirely hit or entirely crit. In this way I think we can make use of crit as regular throughput.

The maths says yes, but the problem is that with a direct heal you tend to have more benefit out of an hot, when you hot someone up they're not going to be staying low, basiclly if you hot someone with a rejuvention, say that rejuvenation has 10 ticks (easier), 1 out of those have a chance to crit, you know that the 10 ticks will heal him up to full without crits. If it does crit, you've wasted that crit% because the guy would be full anyway, crit is to be favoured in a situation where everyone is constantly in need of more healing, but that is not the case in 99% of the raids so therefore I tend to lean towards the reliable and not the unreliable crit.

Healing is so much more having a plan and know when to spend mana and where compared to a DPS, random events like crits screw those plans up because you can get a streak where you have to spend more mana, that makes it alot harder to have those exact values of healing that is needed at the right time and for the right mana costs. Take Elegon for example, the last phase is very healing intense but yet you will have people topped of every couple of seconds, half the ticks we have during this time is going to be overheal, critting with a spell does not help us as much as having the extra mastery because it'll still need another heal but that extra heal won't be as big with mastery because you've already crit. The reason you can't rely on crits is so huge that it completly shuts it out for me as long as it's not miles ahead of mastery.

Lazerdollarz · 11/03/12, 4:39 PM

Your argument is illusory in the sense that mastery and crit affect HoTs in essentially the same way: a 1% increase in either is a 1% increase to all healing. Yes, sometimes crit HoT ticks will cause overheal, but if that's the case and you're that worried about overheal, you probably shouldn't have used a Rejuv in the first place. The same argument can be made for mastery overhealing. Any increase to healing will increase overheal in situations where rejuv is the appropriate spell to cast but the target is topped off prior to the HoT ticking fully, a common occurrence.

I severely doubt that in the space of one GCD anyone would be able to determine that all but one tick of a HoT is going to be effective healing if one of those ticks crits. 95% or more of all healing decisions are made with the knowledge that someone is at low, medium, or high health. The only things keeping crit less effective for resto healing are its greater itemization cost and the RG glyph guaranteeing a crit. Crit is a dump stat because it's less useful than mastery and int for throughput, not because it's useless.

I can assure you that nobody is going to be topped off for any considerable amount of time on hard more progression Elegon, among other heroic fights. Even for normal mode, if you're running an appropriate number of healers, your final phase overheal percentages are going to be incredibly low.

apostolis · 11/03/12, 8:36 PM

Crit is nice if you have incarnation and you keep casting lifebloom. Crit for rejuv is not worthing at least currently since how much crit can you have now ? 20% *inc raid buff*?

Go cast 50 rejuvs (without having any weapon proccing anything) with or without harmony up; i did that and did't even have a single rejuv tick to crit. So how is crit affecting rejuv besides the direct heal part of rejuv ? Does it worth in our current gear to stack crit for that? Dont forget we already have 100% crit in regrowth which is like the only direct spell we cast.

Probably you will say traq or swiftmend but are the ticks critting?

Kjeldorian · 11/03/12, 11:24 PM

Napkin Math that highlights the differences:

Test Case
Let's say I have do 10,000 Spellpower, 0% crit, and 0% mastery (meaning I only get the default 10% for being level 80+)

Assumptions:
That all our spells are effected by Spellpower and it averages out to 100%, meaning Spell + Spellpower Coefficient where the Spellpower Coefficient is 100% (ex RJ 36.9 and SM is ~130%). Not a perfect assumption but the result scales based on the spell coefficient for a more accurate analysis.

Assuming I have the spirit / crit meta [Revitalizing] 3% more spell crits, leather specialization (5% more int), and HoTW (6% more int)

Without changing anything my spell benefit from spellpower would be 10k x 110% (Mastery) = 11k
I gem for 1% crit or 600 crit rating, the benefit would be a 116.8 spellpower increase over the first case [10k x 110% (Mastery) x 99% (When I don't crit) + 10k x 110% (Mastery) x 206% (Crit factor w/ meta) x 1% Crit)
For the same 600 points of crit rating, I instead went for 600 points of Mastery (Gem equivalent), the change would be 250 spellpower increase the first case (10k x 112.5%)
Or instead I opted for 300 Int in lieu of 600 secondaries, my spellpower would go up by 367.3 over the first case (300 Int x 1.05 (Leather Specialization) x 1.06 (HoTW)), I'll ignore the Int benefit to crit to make this short, note that the number would be higher

TL

R / Summary
Mastery gets more bang per point than crit because of how the system has a requirement of more Crit per % than Mastery per %.

Post Script:
The difference gets smaller as you rely on more spells that do not rely heavily on spellpower but there is still a difference. Ex LB has a 5.7% coefficient for the HoT, in the test case above the differences become 6.65 SP for LB with Crit vs 14.25 SP for LB with Mastery.

Payday · 11/04/12, 12:43 AM

No one is saying crit is good and should be stacked, just that if you get a ilevel upgrade that happens to be crit instead of mastery it's still worth using. The difference in the intellect, spirit, and secondary value will be more than enough to offset. This has to be viewed in the grand scheme when balancing enchants and reforges, but in general picking up an ilevel upgrade with crit is still worthwhile.

Numiro · 11/04/12, 5:49 AM

Originally Posted by Lazerdollarz

Your argument is illusory in the sense that mastery and crit affect HoTs in essentially the same way: a 1% increase in either is a 1% increase to all healing. Yes, sometimes crit HoT ticks will cause overheal, but if that's the case and you're that worried about overheal, you probably shouldn't have used a Rejuv in the first place. The same argument can be made for mastery overhealing. Any increase to healing will increase overheal in situations where rejuv is the appropriate spell to cast but the target is topped off prior to the HoT ticking fully, a common occurrence.

I severely doubt that in the space of one GCD anyone would be able to determine that all but one tick of a HoT is going to be effective healing if one of those ticks crits. 95% or more of all healing decisions are made with the knowledge that someone is at low, medium, or high health. The only things keeping crit less effective for resto healing are its greater itemization cost and the RG glyph guaranteeing a crit. Crit is a dump stat because it's less useful than mastery and int for throughput, not because it's useless.

I can assure you that nobody is going to be topped off for any considerable amount of time on hard more progression Elegon, among other heroic fights. Even for normal mode, if you're running an appropriate number of healers, your final phase overheal percentages are going to be incredibly low.

I do know the average values of my spells and that's why I'm ignoring crit, because if I know that my spells heal for a certain amount in a raid enviroment then I can't really fall back on crit, I haven't been healing with the same healers throughout this expansion so far, but when I did in DS I had very very small amounts of overhealing because we knew and trusted eachother that once he did put out that hot that the target was either in a safe spot, or someone else needed help faster and that's in general why I've stopped calculating with crit when it comes to my on the fly healing per execution time / healing per mana calculations.

Elegon is a different fight, because the raid in general spends their time topped off, in between healing CD's they do drop, but that's 20% of the execution phase, not to mention that you'd waste more mana earlier on when you have a set amount to heal at every single explosion. For example I cast wild growth, then put up one rejuventation and I know that the other healer picks up the other ones as he sees that my hots are on those specific targets. There's so much about reliability and that you can't rely on crit that it loses so much favour in my eyes. Sure it's nice to have during specific phases, say elegon last 25-30%, but that is one out of 6 fights and merely 1/4th of the fight, so crit in general is something I don't value more then the fact that I can reforge it into mastery or haste.

Originally Posted by Payday

No one is saying crit is good and should be stacked, just that if you get a ilevel upgrade that happens to be crit instead of mastery it's still worth using. The difference in the intellect, spirit, and secondary value will be more than enough to offset. This has to be viewed in the grand scheme when balancing enchants and reforges, but in general picking up an ilevel upgrade with crit is still worthwhile.

The discussion started with me saying that I ignore the crit part of windsong, the buff uptime is insane for resto druids, almost to the point where it's worth more then the intellect buff, but because of that crit and haste being so minor the buff also loses in favour.

For example you're looking at 420 something int vs 1200 points split between crit/haste/mastery. If the buff is split even then you're looking at 420 int vs 400 mastery + 400 haste + 400 crit, and that is where I find that jade spirit is worth so much more.

Payday · 11/04/12, 12:07 PM

Originally Posted by Numiro

so crit in general is something I don't value more then the fact that I can reforge it into mastery or haste.

As is everyone else. We're just saying not to avoid ilevel upgrades with crit (that you will reforge) because crit is still decent. You make it sound like we're stacking crit.

LupeN · 11/04/12, 3:55 PM

• Hamlet · 11/05/12, 1:05 PM

Crit/Mastery:
Yes, I don't want to completely ignore the value of crit, but it's pretty clear that's just always going to be second fiddle to mastery.
--600 for 1% as opposed to 480 (slight distortions due to both being additive, and crit being boosted by meta, but not enough to make up that gap);
--Crit doesn't affect glyphed Regrowth, which is currently our most-used direct heal by far;
--Crit delivers the healing in a less efficient way. People debated above how much less efficient, but it's not that important since it's worse to begin with.

We only have two secondary stats for throughput (haste is technically one, but since its only use is breakpoints it doesn't enter into the comparison), and mastery is better, so crit is the stat that there's no reason to gear for. So it's the "dump stat" simply insofar as it's the one we'll always minimize, but it's not useless or anything.

--------------

Mastery vs. haste breakpoints:
Even my post above said not to go for the haste breakpoint if most of the tradeoff was in the form of int/mastery. But 6.25% benefit (3600 mastery) isn't that much higher than 5% (extra tick on 40% of heals). I think the main point in Payday's post above is that 40% is way too generous an estimate for the contribution of WG/Efflo, which is probably right. So if the value of the haste breakpoint really only affects 30% or less of total healing, it rapidly becomes less beneficial to take the tradeoff regardless of what stats you give up.