Holy Priest BiS

[res]

Yes it scales with spirit, but it also becomes "worse" when you get T2 3set since the casting regen will become closer to the non-casting regen so that when the trinket procs you "gain" less mana.

For example let's say you have 100 mana per tick when not casting, and ALL of it is from spirit (about 308 Spirit).

If you have no talents and no T2 you "gain" 100 mana per tick when this tricket procs since your regen when casting is 0, making it a total of 600 mana.

With talents this makes it so that you regen 15 mana per tick when casting, so when the trinket procs you "gain" a total of 510 mana.

With talents + T2 3set bonus you would regen 30 mana per tick when casting, so when the trinket procs you "gain" a total of 420 mana.


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

When you gain more than 92 mana per tick from the proc (for example you have 8 mana per tick when casting and 100 when not casting), that is when it becomes better than mindtap talisman.

Since you would want to have 3set T2 and the 15% regen talent, your casting regen will always be at least 30% of your non-casting regen. The breaking point would be if you have 132 mana per tick when not casting and 39.6 per tick when casting from SPIRIT, that is when it becomes better than mindtap talisman.
This should occur when you have about 406 or 407 spirit if my calculations are correct.

So if you have over 406 spirit then Darkmoon Card: Blue Dragon should give you more mana than mindtap talisman, (THIS IS ASSUMING YOU CAST 2.5 SEC SPELLS WITHOUT INTERRUPTIONS FOR THE DURATION OF A FIGHT AND THE TRINKET WILL PROC ON EVERY 50TH SPELLCAST, ASSUMING THE FIGHT IS AT LEAST 125SECONDS LONG) however it is still very unreliable and I wouldn't even use it then.

---

The reason why this trinket used to be good it that it used to stack with both T2 3set and the talent so when it procced you got 130% mana regen - but that's not how it works on this server.

[res]

Looked it up, the Darkmoon card should stack with talent and T2 3set until patch 1.11 which would make it VERY POWERFUL.

Fixed a bug where characters were able to exceed 100% of their normal mana regeneration while casting.

also found this in the same patch:

Darkmoon Fair Card Blue Dragon: This item will no longer trigger from using any ability which does not cost mana.

So we have the nerfed version on nostalrius, when we should have the OP one!

[PriestInOurTime]

While it might be nice, I'm not sure on Nostalrius devs take on what Blizz considered bugs.

[Euronymous]

There's an error in the calculation of DMC:BD's mp5 average on the spreadsheet.
It states that 360 mana restored over 125 seconds amounts to 7.2 mp5, which was derived from doing 360/50.
The error lies in assuming that there are 50 intervals of 5 seconds within 125 seconds.
The correct value is double the amount ((360 mana)/(125 sec))* 5 = 14.4 mp5.

This item would never be considered so good if it was a measly 7.2.
Furthermore you can increase your uptime with clever use of rank 1 holy nova and prayer of healing (each target hit by those spells has its own chance to trigger the effect). This is concluded from testing on a very large sample of DME lashers.

This item is definitely a worthwhile pick.

Source :

MP5 (also known as MP/5) is a shortened term for the item bonus "Restores X mana per 5 sec." or "Mana per 5 seconds". Although the number given is based on 5 second intervals, the effect triggers every 2 seconds. For example, an item that grants 10 mana per 5 sec. restores 4 mana every tick.

http://wow.gamepedia.com/MP5

[res]

There's an error in the calculation of DMC:BD's mp5 average on the spreadsheet.
It states that 360 mana restored over 125 seconds amounts to 7.2 mp5, which was derived from doing 360/50.
The error lies in assuming that there are 50 intervals of 5 seconds within 125 seconds.
The correct value is double the amount ((360 mana)/(125 sec))* 5 = 14.4 mp5.

This item would never be considered so good if it was a measly 7.2.
Furthermore you can increase your uptime with clever use of rank 1 holy nova and prayer of healing (each target hit by those spells has its own chance to trigger the effect). This is concluded from testing on a very large sample of DME lashers.

I think you confuse ticks with 5 second intervals.

MP5 works in ticks, so if you have 2MP5 you would get 1 mana every 2.5 seconds.

There are 50 "2.5 ticks" in 125 seconds. (125/2.5)

You're trying to say there are 200 ticks( 100 5sec ticks?) in 125 seconds? I don't think I understand

Or maybe you're just trying to say that they are only 25 "5 second" ticks which would be correct ( if MP5 did tick every 5 seconds) but then you'd have to double the value of the ticks I stated and you'd end up with the same result.

(The priest in my example has 80MP5 when casting and 200MP5 when not casting.)

[Euronymous]

MP5 works in ticks, so if you have 2MP5 you would get 1 mana every 2.5 seconds.

There are 50 "2.5 ticks" in 125 seconds. (125/2.5)

This much we agree upon.

On the spreadsheet : "This amounts to 360 mana divided by 50 ticks (125/2.5) which is about 7.2 MP5."
In reality this calculation gives you 7.2 MPerTick. You need to double that to get how much the mp5 equivalent is.

[Jeniwyn]

Um, it does indeed seem like you have made a mess out of your DMF trinket calculation. Small detail before we start, with 100 regen ticks your inside 5SR regen would be 30 not 40. But since all other numbers have been with a diff of 60 rather than 70 I'll stick to that.

First of all, mana ticks every 2 seconds, not every 2.5 seconds. Thus, you get 450 and not 360 mana each proc, on average. 450 mana / 125 seconds = 3.6 mana per second = 18 MP5.

Secondly, even after that correction this is actually a flawed way to calculate it. The correct way takes into account that you might get a proc while a proc is already running. This leads to a few procs getting partially wasted. The correct way to calculate is.

uptime = 1 - (1-procrate)^(proclength/interval_between_chances) = 1 - (1-0.02)^(15/2.5) ≈ 11.42%
gain = uptime * manadiffpertick/2 * 5 = 0.1142 * 60/2 * 5 ≈ 17.1 MP5

The first 5 * 2.5 = 12.5 seconds of the fight this value is lower, on average 7.3 MP5. Not very relevant, think of it as negative 2 intellect.

So, clearly this is better than Shard of the Scale then, one might think. Alas, the assumptions going into these calculations are far too generous. First of all, it assumes 100% time inside the five-second rule. If we relax that assumption to 90% (still very high, probably too high for most fights) we need to reduce the value by 2/3*(1-0.9) ≈ 6.67%. (we are always in the FSR the first 5 seconds=1/3rd of the proc.) Just change the 0.9 to whatever value matches your fights.

Most important however is the assumption that a heal is going to land every 2.5 seconds. In very few fights outside of short MC fights, where this trinket is meh anyway, will this ever be true. Every 3.5 seconds is still probably too generous for most fights, but I think it is a decent number to start with.

Then, with the same spirit as in your example I'd estimate the value at no more than:

(1- 0.98^(15/3.5)) * 60/2 * 5 * (1-2/3*(1-0.9)) - 2 Int ≈ 11.6 MP5 - 2 Int.

Considering that my assumptions were still on the generous side and that I dislike the randomness of the proc, I would easily keep Mindtap over it at the spirit level discussed here. (About 290 spirit with buffs.)

The complete formula for any amount of spirit is:

(1-0.98^(15/avg_time_between_spells)) * (32.5 + 1.25*your_spirit) * (1-2/3*(1-fraction_time_in_5SR)) * (1-fraction_spiritregen_while_casting)

[res]

You are right, and all my other calculations on mp5 are based on 2 seconds ticks, I don't know how I could mess up that bad ( probably looked at the 15sec uptime of the trinket and though blizzard knew what they were doing).

About how often you cast a spell during a fight, It could be every 1.5 seconds for all I know, I based it on a MT-healing duty or something similiar where you would use 2.5 almost all the time and sometimes 1.5sec casts, which is when you would want to use this trinket (because of more casts per fight).

I will fix it when I get back from work, I did mess up pretty bad. I guess my excuse is I've had a lot on my mind lately.

Sorry for the crappy math, guys!

[res]

Okay, it should look better now, thanks the feedback guys, if you still think something is wrong or have any other opinions or questions, just keep posting it here in this thread.

I don't want the stickied item list to be MY LIST I want it to be OUR LIST, if anyone think they can make a better estimate of how to value stats etc. better than I have, go ahead and tell us about it here!
If enough people prefer another way of valuing stats I'm more than willing to give other people permission to edit the sheet if they seem to know what they're doing.

I'm not really that talented when it comes to maths, this just started off as me copying macalisters list and adding some items I thought were missing, but then he/she quit the server and my list was the only one that was up to date.

[Jeniwyn]

Okay, it should look better now, thanks the feedback guys, if you still think something is wrong or have any other opinions or questions, just keep posting it here in this thread.

I think this has been brought up before, but the assumption of 100% time inside the five second rule and spells landing at the interval of their cast time isn't even close to a realistic assumption outside of 1 min fights.

I have my own little private BiS list, it isn't even close to publishing worthy at the time however because it is unordered and have a ton of flexible variables going on, and it differs a lot from yours in many places. Mainly because of said assumptions. If you relax those assumptions you will get a much better list.