Resto PvE Compendium and General Discussion - Page 13

Jynnothyst · 02/25/10, 12:12 PM

I think I can clear up what its doing in case Erdluf's explanation doesn't make sense.

Let's look at two WG casts:
First cast hits 2 players and 2 pets
Second cast hits 2 players and 0 pets

The data inside the script will look something like this:

WG	Player	Pet	Both
1	0	0	0
1	1	0	1
1	2	0	2
1	2	1	3
1	2	2	4	***
2	0	0	0
2	1	0	1
2	2	0	2	***

I put *** next to the final version of each line. The table doesn't actually get a new line for each WG application, only for new casts, but its hard to show a single line being updated for each application. Basically, each time it finds a new WG being applied, it puts a tic into either the player or the pet column, and the both column. At the end, the data is either taken from the player column or the both column. So the summary for this encounter would be:

Player only:
2: 2
Both:
2: 1
4: 1

I wouldn't worry too much about the exact numbers for the lower counts, I view it simply as a guide on how to aim my WG better. If people are interested, the other way I considered showing the data was as a 6x6 table with pets on one axis and players on the other, that would show you all your casts in an x+y format, rather than a totals format. I can work on that sort of formatting tonight after work if that would be easier to understand.

Greenpower · 02/25/10, 12:54 PM

I would agree, if you carry spare glyphs to every raid, (which you do if you are serious about doing your best) that there can be instances where WG glyph>Swiftmend. If its a heavy healing kind of fight where WG is so good and we are stacked up and I"m doing 5x1 anyway, I usually replace the Nourish glyph instead. Hasted rejuvs are sooo much healing already, if your doing 5x1 in a 10 man with RR there shouldn't be really much of anything for the other healer to do, except heal the tank and the odd spot heal, so you shouldn't have to use Nourish. Just SM while moving and for tank spikes. Since its hard to keep 10 ppl rejuv'd when using RR in a 5x1, when I use SM its very annoying to lose that hot.

I'll grant you there could be times wg>sm glyphs, but keep in mind you can lose the Nourish instead.

Also I was a bit confused by the argument where you mentioned the 2 types of fights, heavy raid damage, and spike damage, and were arguing that in the spike type fights SM glyph is wasted because the hot isn't used anyway. Thats true in some cases.. But tell me your not using WG off cd in those fights either.. Wg is rather useless in a random spike dmg type of fight. A hot that lingers, does some healing, and buffs Nourish>a 6th wg that you probably shouldn't use more then swiftmend in that type of fight anyway.

My guild finally killed LK in 10 man last week, I had no trouble in the portal either time, but I almost always rock 2/2 ET. Its just situationally too good to miss.

I have started to have a problem however. I decided to go with 4pc 258 T9 and go for haste/crit in as many other slots as I can (thats gear slots, i use standard druid gemming practices, sp in most, reckless in some good sp bonus sockets). I definitely think its better throughput, and its much more controllable throughput than t10. I gave up living spirit and have kinda switched between NP for some crit and ET.

My problem is when you continue to write regen off as not needed it eventually catches up with you

with mostly haste/crit gear, LS not talented, revitalizing skyflare meta, and sp idol, I do find myself almost ooming. I definitely have to use Innervate at 60% and off cd, and pop an alchemist pot. Partly because of my healing style, I am healing nonstop each encounter and mostly gcd locked, which burns mana and doesn't allow for as much regen. I actually turned down an Althor's Abacus yesterday because I can't afford to lose 1 of my 2 solaces.

That said I'm not changing the way I gear, it just makes you manage your mana better and I actually kind of like having to do that again. In a 25 man I don't have an issue with 1-2 mana tides, hymns, full buffs, and an innervate from the feral/balance in emergency. I'll admit in 10 mans I end up using Idol of Awakening quite often.

oopsminded · 02/25/10, 1:49 PM

Originally Posted by Erdluf

He's saying that all 10 casts that hit exactly 4 players, also hit one or more pets. There was no cast that hit exactly four targets (when targets includes pets).

Well, that's what I thought, but tried, mentally, to see where all those casts that hit 1,2,3,4 players went in the "and pets" table and was hitting only dead ends (or very-very hard to believe ones, like casts hitting 1 palyer and 5 pets and such).

Another thing that I just noticed: Total Casts on Players: 55, Total Casts on Players and Pets: 55

*maybe he means Total Casts only on Players: 55, Total Casts on Players and Pets: 55

Erdluf · 02/25/10, 2:11 PM

Originally Posted by oopsminded

Well, that's what I thought, but tried, mentally, to see where all those casts that hit 1,2,3,4 players went in the "and pets" table and was hitting only dead ends (or very-very hard to believe ones, like casts hitting 1 palyer and 5 pets and such).

Another thing that I just noticed: Total Casts on Players: 55, Total Casts on Players and Pets: 55

*maybe he means Total Casts only on Players: 55, Total Casts on Players and Pets: 55

Consistent with my understanding of his Festergut tables
4 casts hit one player and no pets. (4 casts counted in the '1:' row on both charts)
1 cast hit one player and one pet. (Counted in the '1:' row for the first chart, '2:' row for the second chart)
2 casts hit two players and no pet.
1 cast hit three players and no pet.
1 cast hit three players and three pets.
10 casts hit four players and two pets.
2 casts hit five players and no pets.
10 casts hit five players and one pet.
24 casts hit six players and no pets.

That is 55 total casts. It matches both of his festergut charts.

art3d · 02/25/10, 9:26 PM

I've been looking for some advice regarding Valithria Dreamwalker encounter. I did it on 10-man, and using all of the help found here, I was able to heal through with a paladin. But we had to fly around together, so that we would have enough clouds each time we took a portal. In 25 man, it's obvious that a paladin and a shaman would provide grater hps, but our RL is worried that high stacks of Emerald Vigor could kill them without my HoTs. So we plan to send 4 healers to portals, with me healing all of us, while we fly around in two pairs. Could anyone suggest as to, what our role as restodruids is in this fight in 25-man?

Also, I would would like to share that using sFilter addon on Valithria, helped me heal her, as if I was a dps class, properly maintaining all HoTs, while running around searching for a portal. So, if anyone is using that, here is the code for a restodruid

:

  ["DRUID"] = {    
    -- Rejuv
    {spellId = 48441, size = 32, unitId = "target", isMine = 1, filter = "HELPFUL", setPoint = {"CENTER", UIParent, "CENTER", 243, 0}},
    -- Regrowth
    {spellId = 48443, size = 32, unitId = "target", isMine = 1, filter = "HELPFUL", setPoint = {"CENTER", UIParent, "CENTER", 280, 0}},
    -- Lifebloom
     {spellId = 48451, size = 64, unitId = "target", isMine = 1, filter = "HELPFUL", setPoint = {"LEFT", UIParent, "CENTER", 240, -80}},
    },

• Hamlet · 02/25/10, 9:36 PM

I'm actually not a fan of sending Paladins into the portals, since keeping them outside leverages Beacon better. Shamans are good to send in; we can do fine as well.

See if these help:
YouTube - Valithria Dreamwalker, 10-man Heroic
YouTube - Juggernaut vs. Valithria Dreamwalker, 25-man Heroic

Jynnothyst · 02/26/10, 4:26 PM

You guys are right, a tool that's not very easy to understand isn't very useful. I updated the printer so the results are shown in a table, like you can see below. I hope to have a new, complete set of results tonight from Marrowgar - Someone later on, depends how far we go. I'll run the glyph tonight as well.

As far as actually using the glyph goes, I think I'm going to keep it. I'll give the glyph of Rejuv (sub-50% one) a try, but I expect it to only do well on fights like Festergut or Blood-Queen, and I think my WG is just as useful in those. Having the WG glyph gives me that buffer body so that even if I hit a pet, I can still get 5 players, which is great for Festergut since we tend to run a group of 7 under him and then 3 spread out ranged. Similarly for Blood-Queen, we tend to run 2 tanks and 2+ melee dps, so I can usually get 4-5 people with a WG. If I didn't have the glyph, I'd likely cap at 3 or 4 players, since the other 1-2 would go onto the hunter and lock pets, which I don't really care about healing.

Here's the updated function:

# does all parsing in a single pass
def parseAll(file_in, player):
	# file_in - combatlog
	# player - resto druid
	fi = open(file_in, 'r')

	# data holders
	time = []
	players = []
	pets = []
	both = []

	# control parameters
	current_time = 0

	# main loop
	for line in fi:
		if line.find('Wild Growth') > -1:
			if line.find(player) > -1:
				if line.find('SPELL_AURA_APPLIED') > -1:
					# first we do a little parsing to get the timestamp
					temp = line.split()
					time_raw = temp[1].split('.')
					time_str = time_raw[0].split(':')
					time_num = int(time_str[0])*10000 + int(time_str[1])*100 + int(time_str[2])

					# now we compare the timestamp to the last event
					# new casts are more than a second after the last event
					if time_num > current_time+1:
						# for a new cast, we update all the lists
						time.append(time_num)
						players.append(0)
						pets.append(0)
						both.append(0)
						current_time = time_num

					# for all events we update what kind it is
					temp2 = temp[2].split(',')
					hex = temp2[4]
					type = hex[4]
					if type == '8':
						players.append(players.pop() + 1)
					elif type == '4':
						pets.append(pets.pop() + 1)
					both.append(both.pop() + 1)
	fi.close()

	# now we need to make the table
	# rows are players
	# columns are pets
	results = 	[
				[0, 0, 0, 0, 0, 0, 0],
				[0, 0, 0, 0, 0, 0, 0],
				[0, 0, 0, 0, 0, 0, 0],
				[0, 0, 0, 0, 0, 0, 0],
				[0, 0, 0, 0, 0, 0, 0],
				[0, 0, 0, 0, 0, 0, 0],
				[0, 0, 0, 0, 0, 0, 0]
				]

	# populate the table
	for i in range(len(time)):
		#print i
		#print players[i]
		#print pets[i]
		results[players[i]][pets[i]] = results[players[i]][pets[i]] + 1

	# print the table
	print ('\t\t\t\t\tPets')
	print ('\t\t0\t1\t2\t3\t4\t5\t6')
	print ('P\t0\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[0][0], results[0][1], results[0][2], results[0][3], results[0][4], results[0][5], results[0][6]))
	print ('l\t1\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[1][0], results[1][1], results[1][2], results[1][3], results[1][4], results[1][5], results[1][6]))
	print ('a\t2\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[2][0], results[2][1], results[2][2], results[2][3], results[2][4], results[2][5], results[2][6]))
	print ('y\t3\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[3][0], results[3][1], results[3][2], results[3][3], results[3][4], results[3][5], results[3][6]))
	print ('e\t4\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[4][0], results[4][1], results[4][2], results[4][3], results[4][4], results[4][5], results[4][6]))
	print ('r\t5\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[5][0], results[5][1], results[5][2], results[5][3], results[5][4], results[5][5], results[5][6]))
	print ('s\t6\t%d\t%d\t%d\t%d\t%d\t%d\t%d' % (results[6][0], results[6][1], results[6][2], results[6][3], results[6][4], results[6][5], results[6][6]))

And here's what the output looks like when I run it on my old Firefighter attempt, like I used before. Apparently I didn't have the the glyph back then:

					Pets
		0	1	2	3	4	5	6
P	0	0	0	0	0	0	0	0
l	1	15	5	1	0	0	0	0
a	2	26	8	0	0	0	0	0
y	3	15	12	17	0	0	0	0
e	4	18	44	0	0	0	0	0
r	5	37	0	0	0	0	0	0
s	6	0	0	0	0	0	0	0

Edit: Explained the data a bit better.

• Hamlet · 02/27/10, 12:25 AM

The tricky thing with the data is that, as I think some people have said, when you see a WG hit 6 targets including pets, you have no way of knowing who the "sixth" target is, a player or a pet. Absent any reason to believe otherwise, we can only assume the ratio of players hit to pets hit will remain the same as you go from 5 to 6 targets. So in any case, the Glyph represents a 20% increase. The only case in which the Glyph hasn't added that benefit, on average, is when exactly 5 targets are in range.

And none of this is even mentioning the fact that heals on pets aren't useless to begin with. I mean, someone has to heal them, and having them pick up AoE heals is probably the best result, since then they don't take anybody's attention. If a smart targeted heal like WG is hitting them, they needed the healing just as much as anyone else.

e: All that said, I really do like the program people have been using above and the general analytical approach. That data is still quite interesting to see, and a similar script might be useful for some other things. In fact, it might be neat to look at Trauma procs the same way.

Xasthur · 02/27/10, 6:04 PM

It seems that the final bloom of Lifebloom gets the critical heal bonus from Touched by the Light. Also, semi-curious if the bloom would double-dip in divinity for a 10% bonus if it behaves this way.

I apologize if this information is already general knowledge. I was always under the assumption that this already known until I brought it up with some guildies and was curious if Prayer of Mending has the same behavior and nobody seemed to have heard of it. Also - I couldn't find anything about it in the forum through the search function.

-tested on a Prot Paladin with 5/5 Divinity outside of my group

Fallenangel · 02/28/10, 2:00 AM

Originally Posted by Xasthur

It seems that the final bloom of Lifebloom gets the critical heal bonus from Touched by the Light. Also, semi-curious if the bloom would double-dip in divinity for a 10% bonus if it behaves this way.

I apologize if this information is already general knowledge. I was always under the assumption that this already known until I brought it up with some guildies and was curious if Prayer of Mending has the same behavior and nobody seemed to have heard of it. Also - I couldn't find anything about it in the forum through the search function.

-tested on a Prot Paladin with 5/5 Divinity outside of my group

I have seen very high crit values on PoM and the proc from glyph of PW:S on prot paladins. Didn't test it too thoroughly since it's an alt, but they were off the charts, like 2.8K from the glyph when I got 2.2K elsewhere.

Shelendil · 02/28/10, 1:03 PM

It must be on any spell that "belongs" to the person you cast it on when it comes to determining threat. I imagine earth shield falls in this category as well.

exschwizer · 03/01/10, 5:55 AM

Originally Posted by Shelendil

It must be on any spell that "belongs" to the person you cast it on when it comes to determining threat. I imagine earth shield falls in this category as well.

I can confirm that Touched by the Light works with Earth Shield. However, the threat of all these kind of spells is accounted to the caster of the spell and not the recipient, since patch 3.0 if I remember correctly.

Arythorn · 03/03/10, 2:05 AM

Just downed this tonight and one of the things I took to doing was running stacks of LB across all marked targets. LB actually accounted for >40% of both my and our other druid's healing. Toward the end of the fight, when we had 6 marks, I was actually looking for a little bit longer LB so I could refresh some RJs as well as keep the stacks up across all marks. Which made me think about potentially glyphing LB for this fight. I would've actually switched between attempts but it's one of those glyphs that I don't really carry around as I haven't had a use for it. But this seems like the gimmick fight where it just might fit depending on how you build your heal team and assignments.

My thought for the role I was playing would be a Swiftmend, LB, Reg Rejuv setup as I was largely providing mitigation across multiple targets (great use for a druid on this fight). Our other druid actually picked up the sixth mark as a specific assignment so perhaps Swiftmend, Rapid Rej., Nourish would be better if you get assigned a particular mark to keep up.

Anyway, toward the end it really became about me managing GCDs of keeping a 3 stack of LB across 6 targets plus RJ refreshes across all 6 (for extra healing plus Swiftmend utility) and that extra time on LB seems like it would be nice.

Anywho, I like the setup of druid underlay healing across all 6 targets as the damage gets pretty amped even with beacons , direct-assign healers spamming on them. RJ + 3-stack LB seems to be the most stable per-tick output across those targets for my money. Logs below for the interested:

World of Logs - Real Time Raid Analysis

Kermit · 03/04/10, 5:37 AM

Talent specs and hastecap

I know this issue has been discussed, but I don't think any conclusion was made.. I've just reached 850 haste (without sacrificing gems and enchants, other than using reckless gems for yellow socket bonuses) and I will soon swap some items that will bring me to 856+, so I'm wondering if it's time to respec. I could take 4-7 points out of balance and put into resto. What I'm wondering about:

* I still run 10-mans every week, where not all buffs are available (though main focus is in 25man). At that point CF is still useful due to lack of hastebuffs from time to time. Is this reason enough to stay with 3/3 CF?
* What talents would be most priority to spend the available talent points on? My guess would be 2/2 ET and the rest in Tranquil Spirit to enable me to do more nourish-spamming. How did you guys spec? Did you go with LS instead of TS?
* Would you recommend keeping NG, or are those points better spent elsewhere?
* With the resulting spec, would you recommend dropping swiftmend glyph for nourish glyph?

I'd be interested to hear your views about this.

thefool808 · 03/04/10, 9:40 AM

Originally Posted by Kermit

* I still run 10-mans every week, where not all buffs are available (though main focus is in 25man). At that point CF is still useful due to lack of hastebuffs from time to time. Is this reason enough to stay with 3/3 CF?
* What talents would be most priority to spend the available talent points on? My guess would be 2/2 ET and the rest in Tranquil Spirit to enable me to do more nourish-spamming. How did you guys spec? Did you go with LS instead of TS?
* Would you recommend keeping NG, or are those points better spent elsewhere?
* With the resulting spec, would you recommend dropping swiftmend glyph for nourish glyph?.

First off, why optimize your spec/glyph setup for 10 man when you are raiding 25 man content (and wearing 264+ gear)? Of course CF is of great value if you are under the haste cap, but it will be of no value for you in the harder (i.e. 25 man) encounters which you should be optimizing for.

However, I think a lot of these choices come down to preference, because those last few points aren't going to be adding much to the spells you most use in 25 man raiding (rejuvenation, wild growth). 2/2 ET seems like a very good place to put extra points, but those points can easily be taken from Living Spirit. I would stay away from Tranquil Spirit and just replace gems/trinkets/idol to maintain proper mana levels. I see very little value in Nature's Grace at the haste cap. Living Seed is also of little value, but it still is a good place to put extra points.

Recap: Empowered Touch, Living Seed, Living Spirit, Nature's Grace, Natural Perfection, and even Improved Barksin are all situational/preferential talents (preferable in my opinion in that order) that do little to increase the throughput of the spells that should be doing the large majority of our healing.