Prot gearing confusion.

[Promdates]

Haste means more GC procs in the next patch aswell.

To put that to the extreme. If you are tanking 100 mobs (lk25 or whatever), and you have 0% and 50% haste, when do you think you will get the most GC procs?

Haste does not directly increase the proc chance of GC, but it increases the speed in which you can use that GC proc allowing for a second proc much faster.

Yes, while haste does allow you to use those procs faster, you're still going to have a chance to proc multiple GCs before you can use one if you're tanking 100 mobs (lk25 as you put it). This doesn't mean MORE PROCS. It means MORE APPLICATIONS OF PROCS, i.e. using it more. The two things are not the same. Haste will NOT proc GC, but will allow you to use it. Where as now haste can directly allow you to proc GC more BECAUSE you can hit your abilities more often. There is a difference.

[Deleted]

Yes, while haste does allow you to use those procs faster, you're still going to have a chance to proc multiple GCs before you can use one if you're tanking 100 mobs (lk25 as you put it). This doesn't mean MORE PROCS. It means MORE APPLICATIONS OF PROCS, i.e. using it more. The two things are not the same. Haste will NOT proc GC, but will allow you to use it. Where as now haste can directly allow you to proc GC more BECAUSE you can hit your abilities more often. There is a difference.

There is a difference in how it works, but the end outcome is the same.

Haste does not grant your more procs from GC, it does however grant you the ability to use more of those procs, which means you get more holy power from GC with more haste.

- - - Updated - - -

No, you will not. You start with 4 HoPo. J provides 1 HoPo. If CS misses, it will provide 0 HoPo for a total of 5 HoPo. 5 HoPo does not give us back to back SotR coverage.



No, it's not. If you're Expertise capped, when you're at 5 HoPo and a generator is available, you can press SotR, followed by the generator, and be sure to be at 3 HoPo. You will be ready to react with a SotR after the current SotR is over.

If you're not Expertise capped, when you're at 5 HoPo and a generator is available, pressing SotR leaves you with a chance to be stranded at 2 HoPo if the following CS misses. Not pressing SotR in anticipation of an upcoming boss special is wasting HoPo.

Sorry, I misunderstood you, thought CS was available. Okay, so you are saying 4 HoPo, J available and CS in 1 GCD? Big hit in 3-6 seconds.


You got 3 options that I see.

First one assumes that the CS hits J- > SotR+CS -> Sotr +X -> X -> CS -> J This ensures a coverage 2-4 seconds before and 2-4 seconds after the big boss hit (depeding on when it happens in that 3-6 second time frame).

The second one assumes CS gets parried

J -> SotR CS(parried) -> X -> X -> J+SotR -> CS - This will have the same effect as the above mentioned but will have a downtime of 0-0.5 seconds in the midde of the SotRs depending on your gear leve.

The third one is to delay CS.

J -> X -> SotR+CS -> X-> J + SotR -> CS - This guaratees SotR to be up 1-3 seconds before the big hit and 3-5 seconds after the big hit, depending on when it hits.

Delaying the CS is not optimal for holy power regeneration, but it is not something you should be afraid to do in case of preparation of a big boss hit. It also gives you a small damage increase so not the end of the world really.

The third option is probably what I would chose, since it grants me the highest troughput while also 100% guaranteeing a double SotR. It will probably give me a 0.1% lower SotR uptime for the fight, but that is just uptime that is "padding the meters". Rather take the dps increase over that 0.1% anyway. It is not uncommon for me to switch to J>CS priority in the middle of fights when I do not need more holy power than what I have to trade holy power for dps. That is something that is done consciously though to intentionally trade not-needed holy power for dps.

[Deleted]

So... You keep talking about how Haste would net you more HoPo (which is already untrue on it's own).

But you only get a higher potential HoPo, as you have to sacrifice HoPo generation due to the gear choices you made.

At least stating it that way, newbies should understand that your mindboggling way of going around doing things is in fact mindboggling, and that going for 15% expertise is just making things a lot easier, a lot safer, and overall generates you more HoPo.

The discussion was fun. Have a nice day, now.

[Wrathblood]

There is a difference in how it works, but the end outcome is the same.

Haste does not grant your more procs from GC, it does however grant you the ability to use more of those procs, which means you get more holy power from GC with more haste.

Except that its not. At present, all that Haste will increase both your generation of GC procs *as well as* your usage capacity by some amount (lets say 4%). In ~2 weeks, that Haste will only increase your capacity. Remember, our ability to use a GC proc is only going to be increasing by a tiny amount at that point (perhaps 0.05 seconds per GCD). Given 30% avoidance and tanking two targets with 1.5s swing timers, we're only going to be averaging a GC proc every ~9 seconds. The odds of us getting 2 procs within the same GCD isn't really that low, it'll probably happen almost once/minute.

The problem is that shrinking the length of GCDs by such a small amount doesn't really help very much. Cutting the chance of getting procs during the GCD by 4% means you're going to be able to take advantage of 2 very closely spaced procs, roughly once every half-hour during which you're tanking 2 mobs, during which time you'll get ~190 GC procs. So, as it stands now, tanking 2 mobs, increasing your Haste by 4% will increase your number of GCD procs by ~2%, and will increase the number of GC procs you can take advantage of by 0.6%. And I agree that it would get more valuable as you're tanking mobs the math for which gets complicated rather quickly.

So, yes, Haste still increases your ability to take advantage of GC procs, and I'll admit by more than I expected. However, you're still significantly overstating the value of this (except, I'd agree, in really extreme cases. If you're literally tanking 10+ mobs for an extended period of time, this could be a thing), and worse you're playing fast and loose with terminology in a way that could be confusing for someone who didn't know the mechanics.

[Nillo]

Outside of Ra-Den there really is no place where you'd actually need reliability for back to back ShoR for a prolonged period of time.
Pretty much everything else can be handled by just keeping 3-5 HoPo for each boss special.
Even Horridon is pretty consistent with his Triple Puncture and there are only 2 exceptions that could throw you off balance:
Charge/Cleave before Triple Puncture - no probs since this way Triple Puncture is delayed anyway.
Direcall - that's pretty much the only time you'll want to have 5 Hopo and a J ready, but that's regardless of being exp capped or not.

[Promdates]

However, you're still significantly overstating the value of this (except, I'd agree, in really extreme cases. If you're literally tanking 10+ mobs for an extended period of time, this could be a thing), and worse you're playing fast and loose with terminology in a way that could be confusing for someone who didn't know the mechanics.

This isn't uncommon. He did actually mention if you were tanking 100 mobs (i.e. soloing LK25H).

[Wrathblood]

This isn't uncommon. He did actually mention if you were tanking 100 mobs (i.e. soloing LK25H).

Sure, as I said, if you're really tanking 10+ mobs for an extended period of time, this could be a thing, and soloing LK25H would be a case of it. But this isn't something that comes up often during normal raiding and he's still playing fast and loose with his terminology which is half the issue.

[Deleted]

there really is no place where you'd actually need reliability for back to back ShoR for a prolonged period of time.
Pretty much everything else can be handled by just keeping 3-5 HoPo for each boss special.

Though the thing is, if you're Expertise capped, you can maximize SotR output, and still have a SotR available whenever you need. Using SotR at 5 HoPo with a generator ready means you'll always be back up to 3 HoPo at least, ready to place a SotR.

If you cannot rely completely on CS connecting, and you need 3 HoPo, you have to keep HoPo banked and available, which lowers your SotR uptime.

[Deleted]

So... You keep talking about how Haste would net you more HoPo (which is already untrue on it's own).

https://www.dropbox.com/s/ydz9fvfrvv...exp_h_cap.html
https://www.dropbox.com/s/3din2pnuy7...aste_prio.html

See? I got more HoPo from droping exp (down to 9.49%) for more haste!
Problem?

In any reasonable situation it is a HoPo gain. A minimal one, but it is there.
But please continue and ignore facts.

[Deleted]

So... You keep talking about how Haste would net you more HoPo (which is already untrue on it's own).

Fairly easy to mathematically prove that you are wrong in more cases than you are correct with that statement. Only at extreme haste levels is that true. Which has been said by others than me in this thread.

But you only get a higher potential HoPo, as you have to sacrifice HoPo generation due to the gear choices you made.

The situation was a once in a million situations of all stars alligning that you painted up. It is highly unlikely to ever happen in any real boss fight, in any case, I covered the dangerous part of the fight with a double SotR + I gained more dps than I would have done if I did not. That to me is a win/win. If that means that I lost about 0.5-0.75 HoPo over the course of the fight, that would only have been used to cover some irrelevant 10 man melee swing (as in 10 man, the bosses melee hits are irrelevant on most fights, not all but most). I still gained more HoPo over the course of that fight, which kinda balances out that extreme scenario, but as I said, either way, I gained more damage done and still covered the boss mlee swings. That seems like a better performance in my eyes.

Still, it would be like me saying "What if your SS procs 4.1 seconds before a big boss melee hit and your SS has a tick speed of 4.11 seconds. That extra haste would have made it 4.09 secods causing it to proc before the big boss melee hit causing you to take far less damage from that big nuke!"

You realise how silly that scenario sounds? It is a once on a million, just as yours is. I feel so many people always claiming "oh missing a CS causes deaths!". That have never happened to me since 5.0
What I cant help to ask myself is, how often did that extra tick speed on SS save my life? That is something that I do not know but as long as that number is above 0, it has served me better than the expertise would. Even if it saved my life 0 times, I still gained more dps out of the haste.
0 deaths with more dps > 0 deaths with less dps. So atleast for me, being under expertise cap has never caused me any issues. I know most of the people in here that also played below the expertise cap says the same things. You have 5 holy power for a reason. Use it.

At least stating it that way, newbies should understand that your mindboggling way of going around doing things is in fact mindboggling, and that going for 15% expertise is just making things a lot easier, a lot safer, and overall generates you more HoPo.

That going for expertise hard cap generates more Holy Power is just a straight out lie. I really dont see anything mindboggling about this however. The situation that I just metioned was my own way of min-maxing. That is something that I do not expect the general paladin population to do. The only thing minboggling here is your need to fabricate lies to support your claim. Which agains show how weak your standpoint is. Fact remains that haste>expertise with realistic haste and expertise levels yields the highest throughput. The only time that expertise>haste in those scenarios (for 10 mans atleast) is if you are terribly bad at managing your holy power. In which case, yes, I would suggset expertise cap to solve your own problems with managing your resource system.

Though the thing is, if you're Expertise capped, you can maximize SotR output, and still have a SotR available whenever you need. Using SotR at 5 HoPo with a generator ready means you'll always be back up to 3 HoPo at least, ready to place a SotR.

If you cannot rely completely on CS connecting, and you need 3 HoPo, you have to keep HoPo banked and available, which lowers your SotR uptime.

Just no... It does not lower your SotR uptime at all. You have a cap of 5 HoPo, not 3.

The discussion was fun. Have a nice day, now.

Meh, it was kinda old, thought this discussion was settled back in 5.0/5.1. Apparently still some people have your conservative believes, I honestly did not think that. Have a nice day!

Except that its not. At present, all that Haste will increase both your generation of GC procs *as well as* your usage capacity by some amount (lets say 4%). In ~2 weeks, that Haste will only increase your capacity. Remember, our ability to use a GC proc is only going to be increasing by a tiny amount at that point (perhaps 0.05 seconds per GCD). Given 30% avoidance and tanking two targets with 1.5s swing timers, we're only going to be averaging a GC proc every ~9 seconds. The odds of us getting 2 procs within the same GCD isn't really that low, it'll probably happen almost once/minute.

The problem is that shrinking the length of GCDs by such a small amount doesn't really help very much. Cutting the chance of getting procs during the GCD by 4% means you're going to be able to take advantage of 2 very closely spaced procs, roughly once every half-hour during which you're tanking 2 mobs, during which time you'll get ~190 GC procs. So, as it stands now, tanking 2 mobs, increasing your Haste by 4% will increase your number of GCD procs by ~2%, and will increase the number of GC procs you can take advantage of by 0.6%. And I agree that it would get more valuable as you're tanking mobs the math for which gets complicated rather quickly.

So, yes, Haste still increases your ability to take advantage of GC procs, and I'll admit by more than I expected. However, you're still significantly overstating the value of this (except, I'd agree, in really extreme cases. If you're literally tanking 10+ mobs for an extended period of time, this could be a thing), and worse you're playing fast and loose with terminology in a way that could be confusing for someone who didn't know the mechanics.

I am sorry but I think you misunderstood the intentions of my post. I thought I was clear when I made such an exaggerated statement as tanking 100 mobs that I wanted to demonstrate the existance of the effect, not how effective the stats was. By taking an extreme example such as that it is easier to show "Look, it happens". All I was saying with that statement was that the saying that "Haste does nothing for GC in 5.4" was straight out wrong. I was not trying to argue how good haste is for GC procs or trying to base how good it is one that you are tanking 100 mobs in an actual boss fight. Come 5.4 with the new GC, there is going to be huge differences between tanking 1 mob, 2 mobs or 5 mobs for your GC procs. At 1-2 mobs haste does not probably grant you that many, if any, additional HoPo from GC. Though if you are tanking 3+ mobs, it is probably far more valuable than it is now. Once you start getting above 5 mobs it is going to be incredibly strong.
In the end, it is very hard to quantify this effect in an overall as the variable of how often you are getting hit is too big.

The thing is that in most discussions around paladin tanking we are talking about incredibly small numbers. Just bringing out some of my old 5.0 calculations (note this is no longer accurate, just an example), it was calculated that 100 haste rating vs 100 expertise rating at a certain haste/expertise level would increase your holy power regenation by 0.0113 holy power per minute. That means that 1000 would have been a difference of 0.113 holy power per minute. When you are talking about such extremely small numbers it can be quite hard to grasp the difference. That is why I did my overly exaggerated example to easily demonstrate the phenomena.

Still, when you see these small numbers, 0.113 holy power per minute, you begin the grasp that even if haste would allow you to gain one additional GC proc used in 10 minutes, while that seems small, would be equal to 0.1 holy power per minute, which in these comparison are extremely huge. That is the problem here, we are talking about such small numbers, it can be hard to actually get the hold that 1 extra GC every 2 minutes or whatever is actually quite a big deal.

- - - Updated - - -

https://www.dropbox.com/s/ydz9fvfrvv...exp_h_cap.html
https://www.dropbox.com/s/3din2pnuy7...aste_prio.html

See? I got more HoPo from droping exp (down to 9.49%) for more haste!
Problem?

In any reasonable situation it is a HoPo gain. A minimal one, but it is there.
But please continue and ignore facts.

I do not think we are going to turn this one, he seems way to conservative, and reason+logic do not often apply to conservative people.
Talking about not hard capping expertise is a bit like telling the church back in the dark ages that the earth is round. You are more likely to get burned for witchcraft than to be heard for your reasoning.

[Deleted]

Fairly easy to mathematically prove that you are wrong in more cases than you are correct with that statement.

Show me that math. I've done the numbers, you've been handwaiving. If you actually do the numbers, you'll notice who's right, and who's wrong.

You realise how silly that scenario sounds? It is a once on a million, just as yours is.

Actually, in CS > J > X > CS > X > J > CS > X > X > repeat, that 2nd J has a 1 in 3 chance of bringing you to 5 HoPo, with the 3rd CS right after, meaning that this situation appears every 27 GCDs, a good 20 times per fight. The chances of a Triple Puncture or Snapping Bite or anything similar coming up right after is not a 1 in a million chance, it is actually very likely to come up more than once during the course of a fight.

The situation that I just metioned was my own way of min-maxing. That is something that I do not expect the general paladin population to do.

So you recommend people to gear a certain way that will often get them in a situation, where you do not expect them to be able to press the right buttons, and thus die.

That going for expertise hard cap generates more Holy Power is just a straight out lie.

Except that I included the math to prove it. I also mentioned the difference is tiny enough, that it doesn't matter either way if you gear Expertise or Haste for HoPo/minute.

What does matter is that Expertise generates reliable Holy Power. With Expertise hardcap you can reliably keep yourself at 3+ HoPo without ever wasting any HoPo. Without Expertise hardcap, the situation can and will arise where you have to jump through loopholes to assure yourself having 3+ HoPo available for the moment it's required.

(You also claimed it's a DPS increase, without proving it. In general, fillers do more DPS than HoPo generators, but they also have a much longer cooldown, leaving us occasionally starved with buttons to press. By using more fillers, you might starve yourself on more buttons to press and actually lose DPS.)

Not only does your gear style (not capping expertise) lead to a more complex gamestyle (You can't rely on always having 3+ HoPo to react, so you actually have to shuffle generators around, something the people that are looking around for help surely aren't gonna be comfortable about from moment 1), there is yet zero proof that the gamestyle pays off.

So far, you've proven the following advantages of Haste over Expertise capping:
- More SS ticks. (One additional SS tick every 1m40s to be exact. That's sure going to be a lifesaver.)
- Less wasted GC procs. (Though the amount of additional HoPo gained from this is minimal, smaller than the fluctuation of HoPo missed from not being Expertise capped. In addition, when we get a GC proc, we're less at danger of dying, so getting a 2nd GC proc on top of the first one is not going to do much to our survivability.)

These two advantages are both so minor that they do not weigh up against the unreliability of HoPo income. Keep in mind that while Haste and Expertise provide the same HoPo income, Expertise cap will spread this better around automatically, assuring that each string of 3 hits is at least partially covered. Without Expertise cap, if you SotR at 5 HoPo and a generator, you would have strings of 3 hits that are not at all covered with SotR, making you more vulnerable to burst damage. You could SotR earlier, as you said, at 4 HoPo for example but that only increases your chances of being HoPo starved at the wrong moments. And of course you'll lose out on SoI procs.

There is tons of math detailing that Expertise is way better than Haste for survivability. You want to waylay that? Bring the math to prove it wrong.

[Theck]

Dropping 1876 expertise rating for more haste (till I reached 50%) increased my DPS, my HPS from SS, and my SotR uptime, it lowered my DtPS. It lowered my SoI heals too, but in simcraft, it never over-heals, which makes it rather unrealistic. It increased my so called "TMI", but TMI doesn't has much to do with IRL bosses.

I realize I'm digging back a few pages here, but I wanted to point out that TMI is probably more relevant to IRL bosses than anything else being discussed in this thread. If you're really concerned with surviving unexpected spike damage (i.e. a sudden string of back-to-back-to-back unavoided melees), then it stands to reason that a metric measuring the severity and frequency of those events would be fairly relevant.

You can argue back and forth all day (and it seems you all have) about what's better for predictable spikes. If you're banking HP and playing perfectly, you can certainly get by with less than 15% expertise to gain some haste. I don't think it's advisable for any number of reasons, but you could certainly do it.

That said, most of the discussion here has been about what to tell beginner tanks. How many of those tanks do you think bank HP perfectly? I'd venture to say that none of them do. In fact, I'll go one step further and suggest that nobody in this thread does (myself included), even the ones who seem to enjoy puffing their chests up and saying "well, L2Play." If none of us can do it that well, how do you expect a beginner to do it?

Players like Slootbag, Treck, Absalom, and other world-first players gear for 15% expertise too for that exact reason: they do not play perfectly. There is so much going on in heroic encounters that it is often just not realistic to perfectly prepare for every predictable spike. Serious raiders are in the business of increasing reliability, which is why they tend to prefer 15% expertise - it gives them more breathing room when it comes to making minor rotational or hp-banking mistakes.

So, sure, you'll probably see a slight increase in HP generation by shifting expertise into haste. But it comes at several costs. Less reliability and more complicated planning and banking are obvious downsides. A less obvious downside is that your HPG will in fact be burstier, because those parried CS's create longer periods during which you have no HPG. This is why the TMI measure goes up - you're creating longer periods of no SotR coverage, which naturally means a higher chance to take multiple hits in a row without SotR mitigation. Not even perfect banking can mitigate that downside.

In my mind, the more compelling reason to drop expertise for haste is increased DPS, because there the difference is a little more pronounced. And as has been brought up several times in other threads, tank DPS matters a lot more on 10-man than on 25-man. I certainly think you can make an argument that it could be worth dropping some expertise for more haste to increase DPS on certain encounters. But I also think that's well within the realm of "advanced tanking techniques," as it requires that your rotational play be very precise. It's not something I'd ever suggest to a beginner, even at 10-man damage intake levels. Especially because that beginner tank is likely to be undergeared for the content.

[Promdates]

The difference in the DPS output that Wanko was posting from SimC is only 0.005371170981178% difference, or 1711.5 dps. You're making a case for less than 2000 dps. I can't even begin to wonder why you're making such a huge stink about that, but it bottles my mind.

[Deleted]

The difference in the DPS output that Wanko was posting from SimC is only 0.005371170981178% difference, or 1711.5 dps. You're making a case for less than 2000 dps. I can't even begin to wonder why you're making such a huge stink about that, but it bottles my mind.

Hey! 0.0055371170981178% difference is still higher than 0.0052%!

Both are obviously extremely small, and possibly ignorable, but then we wouldn't had anything to discuss here.

[Theck]

Regarding whether haste or expertise grants more time-averaged holy power: while I think the calculation is sort of irrelevant, if people are going to continue to focus on it, we may as well make sure it's done right. So let's hope I did this correctly.

Holy power generation can be modeled as follows:

HPG(h,e)=(1+h/42500)*((1-[0.15-e/34000])/4.5 + (1 - [0.075-e/34000])/6.75)

Where h is your haste rating, e is your expertise rating, and terms in brackets are assumed to be lower-bounded by zero. If you plot that, you get something like this:



It's hard to tell much from that plot though, as the x and y scales are different. What we really want to know is whether a small amount of haste rating dh gives more HPG than a small amount of expertise rating de. To do that, we differentiate the above equation and solve for |dh/de|. This is tricky because of the max function implied by brackets, so we get a piecewise result:

if e<7.5%
|dh/de| = 42500/34000*(1+h/42500)*(1/4.5+1/6.75)/((1-[0.15-e/34000])/4.5+(1-[0.075-e/34000])/6.75)
if e>7.5%
|dh/de| = 42500/34000*(1+h/42500)*(1/4.5)/((1-[0.15-e/34000])/4.5+(1-[0.075-e/34000])/6.75)
again, using the same bounded-by-zero convention for brackets. If you plot that, you get this:



Note that for dh/de<1, haste gives more HPG than expertise. That condition is not satisfied at all below soft cap, so expertise is always better for HPG below soft cap. Above soft-cap, it's trickier to see, but at low haste levels haste>expertise, and at high haste levels expertise>haste. The breakpoint seems to be a little over 31%.

Yes, I've ignored Grand Crusader income here, but that effect is incredibly small whether you're looking at 5.3 or 5.4 mechanics. Wrathblood is basically correct about the small reduction in GCD time being negligible for the purposes of Grand Crusader clashes. For a regular boss with a 1.5+ second swing timer, it's almost completely irrelevant. It starts to be worth thinking about once you have a handful of adds attacking you, but even then it's going to be a fairly small effect because it's 30% of ~30%. And at that point, you're proccing it often enough that every GCD is a holy power generator anyway, so it's mostly a question of DPS.

[Deleted]

We could still discus whenever the lost GC procs from the missed CS are making a difference or not. (At least till 5.4.) Afaik they currently won't proc from missed CS.

PS: Those plots are assuming that we use Judgment on CD, right?
PSS: I have sometimes the feeling that it is rather pointless to spend so much time arguing about something, that has a effect <1%. But then again, you guys forced me to do math (or at least try it), so I hate you guys. GN.

[Deleted]

We could still discus whenever the lost GC procs from the missed CS are making a difference or not. (At least till 5.4.) Afaik they currently won't proc from missed CS.

They do which is why you put a 1.12x multiplier on the value of CS hopo generation when you calculate the value of expertise.

[Deleted]

Note that for dh/de<1, haste gives more HPG than expertise. That condition is not satisfied at all below soft cap, so expertise is always better for HPG below soft cap. Above soft-cap, it's trickier to see, but at low haste levels haste>expertise, and at high haste levels expertise>haste. The breakpoint seems to be a little over 31%.

I was sooooo confused there for a moment, when you had other text there. :P Was trying values with your first formulae, and then tried with your second formulae, and had no idea what was going on. Tried to ask you about it, but when I quoted your post, there was totally different text in it, and it all made sense again!

[Deleted]

But this doesn't explains at all why on SimCraft you get lower HPG with exp hard-cap, than with lower expertise, but higher haste.

gear_expertise_rating=5100
gear_haste_rating=20250

vs.

gear_expertise_rating=4100
gear_haste_rating=21250

The second one has consistently higher HPG, and better uptime.

[trystero]

The difference in the DPS output that Wanko was posting from SimC is only 0.005371170981178% difference, or 1711.5 dps. You're making a case for less than 2000 dps. I can't even begin to wonder why you're making such a huge stink about that, but it bottles my mind.

Minutiae is what internet forums do best...