Page 1 of 3
1
2
3
LastLast
  1. #1
    Keyboard Turner Ritic's Avatar
    Join Date
    Jan 2012
    Location
    New Jersey
    Posts
    8

    Guardian Legendary Meta

    Has anyone tried both the tanking and the DPS one to see how much DPS or damage reduction they do?

    Not really sure what one to pick, now I use the Agi/Crit one, and I have been thinking about going with the DPS one, unless there is a really big damage reduction on the stam one.

    Thanks for any help

    Ritic
    Area 52

  2. #2
    Warchief
    Join Date
    Jun 2009
    Location
    London
    Posts
    2,045
    The DPS one does about 15%-18% of my total damage on a boss fight. The tanking one procs often from what I've heard, but haven't personally tested it. Tanking one is amazing though too.

    As for which one you want, the answer is both. Also consider the content you're tanking. Doing reg modes ? Get dps one. Doing heroics ? Get tanking one.

  3. #3
    Quote Originally Posted by rated View Post
    Doing reg modes ? Get dps one. Doing heroics ? Get tanking one.
    Ah oversimplification, gotta love it.

    The DPS one has an absurdly high (bugged) procrate. You can expect ~40 PPM from it with decent haste. It looks like we're using feral's 1.721 modifier for some unknown reason, which helps explain why it's a good 15-16% of our DPS vs. ~8-10% of other tanks'.

    So unless you're having significant, raid-wiping issues surviving, or you're aoeing/otherwise have your personal DPS crippled (think Tortos/Primordius when not mutated on H), the 20-40k dps boost you can expect from capacitive is a good choice.

  4. #4
    With respect to the tanking meta, obviously your active time tanking the boss will affect your uptimes (as well as RPPM RNG). That being said, encounters where I've mostly been tanking non-stop (such as Megaera and Dark Animus), I have uptimes between 35% and 40%. Other encounters with tank swaps or lower active tanking time, uptime can drop to about 20-25%... while that may seem low, its proc rate is basically the same as if you were tanking non-stop if you take out all your dead time.

    Long story short, the tanking meta is a super Barkskin, having a longer duration and better uptime than our normal Barkskin.
    *Since I'm tired of reposting the link:* EH vs TTL and Hybrid Tax

  5. #5
    I used this graph in a brewmaster discussion but it works here as well:



    You die when the encounter damage > your survivability/EH, denoted by the black death zone. Increasing your survivability bumps your survival line towards the right, decreasing the black death zone. Equipping the tanking metagem doesn't bump the curve to the right, it sort of stretches it a bit because while the average mitigation is higher, then minimum remains the same.

    Basically, what the graph says is that while the tanking metagem proc increases the maximum survivability when all of your stuff is up, it doesn't actually reduce the zone during which you'll die by a noticeable amount. A cursory analysis tells me that out of every one thousand tank deaths, that metagem would prevent perhaps 7 or 8 of them - assuming 100% of the damage you take is physical. Since that's not the case, we're probably looking at 5-6 deaths per 1000 prevented with the gem.

    Empirically, to the best of my knowledge the ratio of successful tanks using the dps meta to tanks using the tanking meta on Ra-den/Heroic Lei Shen is something like 15:1. Go with the dps meta. Tanking one is stupid and retarded.

  6. #6
    Warchief
    Join Date
    Jun 2009
    Location
    London
    Posts
    2,045
    Quote Originally Posted by Braindwen View Post
    Ah oversimplification, gotta love it.
    What I said is correct. There's no justification for you being a dick about this.

  7. #7
    Keyboard Turner Ritic's Avatar
    Join Date
    Jan 2012
    Location
    New Jersey
    Posts
    8
    Thanks for all the advice, I picked up the crit meta today. I'm not having any survivability issues with the few heroics we have done so far, but I was thinking about maybe trying the tanking one for heroic Tortos if I'm going to be on the bats. But other then that the crit one looks like the way to go.

  8. #8
    What kinda tanking partner do you ahve are you the OT or MT, raiding 25m or 10m.

    Dps meta is really stong, so is tank meta. but if you tank with a pala like i do who is tanking most of the time, you should take the dps one, if you tank most of the time use the tank one. if you are raiding 25m take the tank meta.

    From what i heard tankmeta has about 25% uptime, dps meta procs 100% from autoattacks as a druid. plus procs from special so you will get atleast 5 stacks every 8 seconds. Since they are not so expensiv to change it might be good to have the other one in your inventory if you have dps probs or survival probs.

  9. #9
    Keyboard Turner Ritic's Avatar
    Join Date
    Jan 2012
    Location
    New Jersey
    Posts
    8
    I tank with a DK, and for the most part I'm the main tank in a 10 man.

    And that's also what I just did, I'm going to start with the DPS one but I did pick up one of each to have in my bags as well to change as needed.

  10. #10
    Quote Originally Posted by kaiadam View Post
    I used this graph in a brewmaster discussion but it works here as well:


    Density distribution graph, very interesting. However it took me a bit to understand the graph because it's not labelled and usually in physics we swap your x and y axis. It must be a sketch instead of calculation result I assume. I also assume x is boss' DPS or tank's max-allowed unmitigated DTPS, and y is probability distribution (which should be normalized). So the death probability is the overlapped area. If this is correct, it didn't support your point that tank meta is useless, and I don't think how you analyse the change of distribution is correct. Tank meta will first shift the whole distribution to the right by a little bit (stam increase), which obviously decrease overlapped area. Then also bump up the right hand of the DTPS peak and makes it asymmetric (reduction proc). Since distribution should be normalized, the overlapped area decrease too. If we want to look at the effect from this graph, ofc it will increase your surviving probability. And about how much it will boost you, it's a very subjective thing. I don't doubt that for top players like you mentioned, who has top gears and skills, in this graph, the boss DPS and tank DTPS peaks are far away separated (death chance is already low enough), then the tank meta boost is totally ignorable. But for a mid-class player, whose gears are not that good, raid is not that perfect either, these 2 peaks could be very close (death chance is high) and tank meta in this case could make a difference.

    Unfortunately my main isn't bear either, and I haven't done an actually simulation, so I can't tell how much it can actually affect a bear. But my idea is, if possible, get 2 helms and gem differently, and swap depends on fight, at least this is what I do to my DK. It's only 5 primal diamond after all. For a fully heroic geared tank, you probably don't need tank meta on H Lei Shen. But for, say, an ilvl 510 tanking H Tortos bats, then why not?

    PS: Also there is another fact that death chance is not the one truth. We are not soloing, healers are healing us. Perhaps I didn't die in 100 attempts but the pressure I threw on healers is not imaginary. I'm 100% sure I can one tank 10N Iron Qon on my dwarf dk with full haste dps gears (ilvl 496 ish i think, my off spec, haven't started gearing yet), get world #1 ranking on dps, with the cost to fk all our healers and probably kill a couple of dps. Sometimes a fail tank doesn't appear like he died a lot, instead he draws healing and mana from healers and let others die.
    Last edited by jasonleekungfu; 2013-04-30 at 11:29 PM.

  11. #11
    Quote Originally Posted by kaiadam View Post
    I used this graph in a brewmaster discussion but it works here as well:



    You die when the encounter damage > your survivability/EH, denoted by the black death zone. Increasing your survivability bumps your survival line towards the right, decreasing the black death zone. Equipping the tanking metagem doesn't bump the curve to the right, it sort of stretches it a bit because while the average mitigation is higher, then minimum remains the same.

    Basically, what the graph says is that while the tanking metagem proc increases the maximum survivability when all of your stuff is up, it doesn't actually reduce the zone during which you'll die by a noticeable amount. A cursory analysis tells me that out of every one thousand tank deaths, that metagem would prevent perhaps 7 or 8 of them - assuming 100% of the damage you take is physical. Since that's not the case, we're probably looking at 5-6 deaths per 1000 prevented with the gem.

    Empirically, to the best of my knowledge the ratio of successful tanks using the dps meta to tanks using the tanking meta on Ra-den/Heroic Lei Shen is something like 15:1. Go with the dps meta. Tanking one is stupid and retarded.
    Well, I went to your brewmaster post to see if I could get some info concerning the graph, but the graph is pretty useless unless you provide specifics (even simple things, such as what each axis is measuring for starters). The graph above could be replaced by a smiley face and it would say literally the same thing: next to nothing of value. Please provide proper labeling and data used or just don't use the graph at all. jasonleekungfu beat me to posting about this, addressing my concerns.

    In the interest of time, since I can't refute a graph with little substance to it as presented, I believe what's trying to be shown is that the tanking meta is not a reliable as an EH boost. Since it is a RPPM-based meta, once cannot predict when you'll intially receive the 20% physical damage reduction. However, from a Guardian perspective, the meta is likely more useful even in the realm of EH compared to other tanks. Specifically speaking, when a Guardian is concerned about physical burst damage (Snapping Bite, Talon Rake, triple ability overlap on Horridon, etc), unlike other tanks we cannot use our active mitigation to guarantee any result, which is why we tend to rely on major non-AM cooldowns to prevent burst. In this respect, the tanking meta likely provides more benefit to Guardians as an EH meta compared to other tanking classes, even if the proc is not guaranteed.

    That being said, the tank meta is likely designed to be a TTL proc (aka, a time-to-live proc), reducing overall damage physical damage taken throughout an encounter. Unfortunately, as has been the case for many expansion, TTL is a little less tangible to EH (and even less tangible to most people with respect to DPS/damage output), which is why it's generally brushed off. In our Guardian forums, TTL is akin to RPS (rage per second) with how we're designed, since we increase our rage generation to allow more SD/FR usage, leading to us living longer and taking less damage over time. While I could spend all day talking about this concept and how the tanking meta contributes to this and allows more decision-making as Guardians, I'd like to keep this specific post as short as possible.

    Now, I'm not saying the DPS meta isn't viable or desired for tanking, I just think the benefits of the DPS meta is getting blown out of proportion in comparison to the tanking one. This week I'll make an effort to use the DPS meta get a more comprehensive numbers output, but to let the kitty out of the bag now, the total raid damage gain from me using the DPS meta thus far has averaged around 0.4%. Now, we've had 1% wipes before on heroic progression, but I guarantee that those are in the vast minority. The biggest point I can make is that the benefits of the legendary metas are both useful, and that their individual use should be tailored to the individual and their specific raid.
    Last edited by exochaft; 2013-04-30 at 11:34 PM.
    *Since I'm tired of reposting the link:* EH vs TTL and Hybrid Tax

  12. #12
    Quote Originally Posted by jasonleekungfu View Post
    Density distribution graph, very interesting. However it took me a bit to understand the graph because it's not labelled and usually in physics we swap your x and y axis. It must be a sketch instead of calculation result I assume. I also assume x is boss' DPS or tank's max-allowed unmitigated DTPS, and y is probability distribution (which should be normalized). So the death probability is the overlapped area. If this is correct, it didn't support your point that tank meta is useless, and I don't think how you analyse the change of distribution is correct. Tank meta will first shift the whole distribution to the right by a little bit (stam increase), which obviously decrease overlapped area. Then also bump up the right hand of the DTPS peak and makes it asymmetric (reduction proc). Since distribution should be normalized, the overlapped area decrease too. If we want to look at the effect from this graph, ofc it will increase your surviving probability. And about how much it will boost you, it's a very subjective thing. I don't doubt that for top players like you mentioned, who has top gears and skills, in this graph, the boss DPS and tank DTPS peaks are far away separated (death chance is already low enough), then the tank meta boost is totally ignorable. But for a mid-class player, whose gears are not that good, raid is not that perfect either, these 2 peaks could be very close (death chance is high) and tank meta in this case could make a difference.

    Unfortunately my main isn't bear either, and I haven't done an actually simulation, so I can't tell how much it can actually affect a bear. But my idea is, if possible, get 2 helms and gem differently, and swap depends on fight, at least this is what I do to my DK. It's only 5 primal diamond after all. For a fully heroic geared tank, you probably don't need tank meta on H Lei Shen. But for, say, an ilvl 510 tanking H Tortos bats, then why not?
    Yeah, my professors hated my lab write-ups too.

    I took a few liberties to emphasize the overlap, such as moving the survival line lower than it normally is. Normally that amount of overlap (I think it's roughly 3%?) means that a tank dies something like every 60-90 seconds on average which obviously doesn't happen. I also enlarged the encounter damage curve to make it intercept the survival curves at a higher point, AND doubled the effect of the meta-gem to point out that there actually was a difference - if I stuck to the original numbers I had there was like a 2 pixel difference instead of the current 5 pixel one. Also, yes, that 324 stam probably moves the second curve maybe 0.1% to the right ... but it was low enough that I didn't really worry about it.

    But all technical details aside, what I wanted to illustrate with that graph is that the metagem increases your average survivability, but not evenly. The metagem is something like a 7% physical damage reduction but it's sporadic and random, so while a 7% physical damage reduction (what is that, 3k mastery for bears? 5k?) might shift the entire curve to the right, all the metagem is doing is stretching it towards the right. So when you're already unkillable it makes you more unkillable, when you're vulnerable, it doesn't do much to make you less so. Its benefit towards tank survivability is somewhere in-between 3% dodge and X mastery, for whoever actually maths out what 7% physical DR is. That seems very poor to me, not nearly enough to give up the 30-40k dps that the dps version would give us.

    There's no way that I could do an even semi-realistic formulation of tank EH vs encounter damage so I gave up on it immediately and only drew the graph for broad illustrative purposes. I should probably at least label my axes though. Oh well, -10 points I guess.

    My argument for brewmasters was that we could simply reforge 3k of crit/haste into mastery, then use the dps meta, and end up with higher survivability AND higher dps compared keeping everything the same and using the tank meta, but I guess that applies less to bears as crit and to a lesser degree, haste, contribute to guardian survival, somewhat.
    Last edited by kaiadam; 2013-04-30 at 11:40 PM.

  13. #13
    Quote Originally Posted by kaiadam View Post
    I used this graph in a brewmaster discussion but it works here as well:



    A cursory analysis tells me that out of every one thousand tank deaths, that metagem would prevent perhaps 7 or 8 of them - assuming 100% of the damage you take is physical. Since that's not the case, we're probably looking at 5-6 deaths per 1000 prevented with the gem.
    Is that based solely on damage taken / damage mitigated / damage avoided ? As in, does it include healer mana conservation and what impact that plays in tank survivability ?

    Just asking questions, as the main tank in my guild is a bear, and I want to be able to have a more-informed discussion about his choice. I imagine he'll go with the dps one, as the extra rage means more choice/control rather than reacting to a proc, but ... yeah, again, just trying to gather info.

  14. #14
    Yes/yes/no, completely ignores healer mana. In general, having tanked and healed on and off for WoW all the way from vanilla, I've never been in a situation where healers were actually going OOM because of a tank's damage intake, or even close to it - and I've been in both camps, so I more or less ignore healer mana conservation.

  15. #15
    Quote Originally Posted by rated View Post
    What I said is correct. There's no justification for you being a dick about this.
    Except, it's a gross oversimplification and downright incorrect.

    There are heroic fights where DPS checks are far more meaningful than tank survival. There are heroic fights where tank death doesn't occur due to physical damage pretty much ever (see: Jin'rohk, any death there is almost certainly due to distracted healers not topping off the tank after lightning storm, or static wound issues, both cases involve primarily magic damage). Even encounters like horridon, where tank survival is certainly called into question, you shouldn't just immediately rush to the tank meta. What if, like was my situation, the other tank taunts horridon off you as rampage starts, and you only tank briefly at the end when he runs out of CDs? Capacitive is worth a solid ~20M damage on Horridon himself, that's ~20M less health you need to tank through, and allows you to have reliable CDs up for a larger portion of your (already very short) tanking period.

    And then, consider normal mode guilds with horrid healers, where the tanks are largely left to fend for themselves. Tank meta looks surprisingly attractive.

    Normals=Capacitive, Heroics=Indomitable is not "correct". It might be for some people, but not everyone.

  16. #16
    The "tank" meta is worthelss for EH as Daught mentions, since it's completely unreliable in relation to when times you will potentially be killed occur. It's great for TDR, but that only helps with healer mana really.

    Ahanss is also right when he says it's completely dependent on your group. If you have trouble staying alive (for whatever reason) start with the tank meta. If you don't start with the DPS meta.

    The "correct answer" is to get both, and have a helm for each.

  17. #17
    Quote Originally Posted by kaiadam View Post
    Yes/yes/no, completely ignores healer mana. In general, having tanked and healed on and off for WoW all the way from vanilla, I've never been in a situation where healers were actually going OOM because of a tank's damage intake, or even close to it - and I've been in both camps, so I more or less ignore healer mana conservation.
    This!
    Average damage doesn't really matter a lot. The amount of mana a healer uses to heal a tank who on average takes 20% more damage, is lower than the amount they'd use to heal a tank who 10% of the time spikes down and is in need of immedeate healing.
    It is also the second tank which will stop healing from going to other raid members, and therefor let them die.
    As a tank your main priority should be to minimize RNG! As kaidam is saying, the tanking meta will not stop you from spiking.

  18. #18
    The big problem like alot say is RNG. you could argue yes but druids already have rng with their dodge, and adding another proc, will lower the change from spiking.

    So, proc doenst help you with onhit procs from bosses like dodge does.

    to make things easier, lets say you have 100k life. assuming a big hit comes that will do 80k dmg to you and you have 100% life. you will take 80k dmg and survive, if the meta procs you will take about 64k dmg you will still survive and need to get healed back up again. And like other people say from a healer point of view he will heal you and they dont care about that 16k dmg more or less.

    now nr 2, still 100k hp. now there will be a special that hits for 120k, 5 seconds before it hits, you use SI/shieldwall/etc. so you will take 60k dmg, and survive again. If the meta proccs 3 seconds before the attack, you will take less dmg yes but you would have survived the attack anyway.

    On attacks where you would actually need the meta to proc you cant hope for that, or on big hits if you use your cooldown proberly you dont need it either.
    As far as i know meta only proccs when you get hit or absorb physical dmg. Now as a druid lets say you taunt get a hit meta proccs and you have your SD uptime is 100% during the meta proc phase, as a druid you will not stop using SD or i wouldnt tbh, as pala or warrior maybe you could delay your SotR /shieldblock, during that time. But the proc would be wasted for druids.

    5-6 out of 1000 deaths, that is very low. Still it may help, if you get really lucky and the meta proccs and you survive, and then get the kill on a boss. On the other hand the dps meta does ALOT of dmg as a druid and i mean alot it does about 20-30k dps depends on the boss, which is a huge amount in 10m. Now you can argue, if you had the dps meta and with your extra dmg would have killed the boss a bit faster, maybe that the last big hit that would have killed you was still on cooldown and you kill the boss cause you didnt die. in 25m the changes are very small that this would ever happen due to your 30k dps, but so are the changes the tanking meta will actually save you were you woud otherwise be dead.

    The only part were the tanking meta is really good is when a big attack will kill you the tankLMG proccs, and you might only have to use barkskin plus meta to survive, and save SI or other CDs for later
    Last edited by Viromand; 2013-05-01 at 10:24 AM.

  19. #19
    Keyboard Turner Ritic's Avatar
    Join Date
    Jan 2012
    Location
    New Jersey
    Posts
    8
    Last night I tested both on Heroic Tortos. For the First 3 bosses (Heroic Jin'rokh, Heroic Horridon, Normal Council) I used the DPS meta and it worked great, about 18-19% of my damage.

    Last night we started working on Heroic Tortos and I'm the tank on the bats, we have our DK on the boss. We had some problems with keeping the shield up on me on some pulls. I did use both meta's on this encounter to test them, but I don't think I seen a difference using the tanking one over the DPS one when it came to my shield dropping, but seen a big difference in my DPS.

    We are 2 healing, and keeping 2 DPS on bats full time (monk & shammy)... any tips on tanking the bats? I think I'm going to go back to the DPS meta for this, If the bats die faster will it be easier for the healer and me to keep my shield up, or even know I did not see a big difference would the tank meta keep the shield up better then the bats going down a little faster?

    Thanks again

  20. #20
    Sorry it's taken me a while to get back with some actual numbers for the DPS meta gem while tanking, life gets hectic sometimes.

    Just a couple of hi-lights from this week's heroic 25man raid thus far:
    Jin-Rokh: 0.87% raid damage increase
    Horridon: 0.74% raid damage increase
    Council: forgot to start my logs again, so I didn't record it.
    Tortos: sat myself since I didn't need anything off of him, sorry
    Megaera: 0.95% raid damage increase
    Ji-kun: 1.02% raid damage increase

    How your raid executes the fights will likely contribute to how much damage the proc will do. For example, I tank adds full-time on Horridon, so someone on the boss more often would likely see an inflated number. Ji-kun was using my personal logs, and since we two-tank it my logs may not reflect a proper damage output from all the raiders. Megaera has us not killing the green head, so the Vengeance scaling for the proc is ridiculous while tanking the green head, but at the same time completely useless towards killing the boss.

    Now, it was mentioned that a good use for the DPS meta is that it cuts the fight length down. Absolutely it will... but how much? For example on Jin'rokh, the damage increase from the meta shortened the fight by less than 2 GCD's using the encounters overall average DPS (again, just a generalization since timing is everything, and for all examples GCD =1.5 seconds). Horridon, using just the average encounter DPS, I saved the raid roughly 5 GCDS, which could be very important... however, it really depends when the procs occurred and what was being tanked (since I tank adds full-time, there's more variance in GCDs from that alone). I already mentioned Megaera and the largest portion of the proc damage going into a head that we never kill. For Ji-kun, I saved the raid 4 GCDs, but we all know that logs on that fight are very subjective since people are always coming and going (including myself), so the total contribution was likely much less.

    Now, I hesitate to talk about this next part since this is even more subjective to numerous factors... but compared to previous kills where I used the tanking meta over the DPS one, I've felt a noticeable drop in survivability when not using the tanking meta. If we take this week's Megaera kill, it was 15 seconds faster than last week. Great, awesome... but I was definitely a squishier tank. Overall, using the DPS meta, I took roughly 2k more DTPS (damage taken per second) in the fight. Alright, we can chalk that up to RNG, but let's look a little closer at melee attacks taken:

    Using the tanking meta -> 16.4mil damage from melee attacks overall, 97 hits averaging 169.7k, 89 dodged (47.6% dodge rate), 6.6mil damage absorbed
    Using the DPS meta -> 17.9mil damage from melee attacks overall, 89 hits averaging 201k, 82 dodged (47.9% dodge rate), 6.5mil damage absorbed

    Taking 9% more melee damage and having auto-attacks hit roughly 18% harder on average on a 15 second shorter fight executed in the same manner is definitely noticeable.

    While I could delve into other fights with hefty physical damage (such as Horridon and Iron Qon), getting all the specific numbers for is very time consuming, and even just a cursory investigation into those numbers shows similar if not more drastic differences in damage taken just by switching meta gems. However, the point remains the same: both the DPS and tanking meta have their uses. If you and your healers can deal with taking extra damage, potentially save your raid a GCD or two with the DPS meta (perhaps boost your ego with high WoL ranks, too). If survivability is an issue, use the tanking meta to help mitigate physical damage, which is typically the largest single source of damage you'll take on a fight as Guardian.


    Quote Originally Posted by Ritic View Post
    We are 2 healing, and keeping 2 DPS on bats full time (monk & shammy)... any tips on tanking the bats? I think I'm going to go back to the DPS meta for this, If the bats die faster will it be easier for the healer and me to keep my shield up, or even know I did not see a big difference would the tank meta keep the shield up better then the bats going down a little faster?

    Thanks again
    If you're tanking bats, the DPS meta will likely not help you kill the bats much faster since it's a single-target spell. While the Vengeance boost will make the proc hit harder, you're only hitting one of the bats with it. If you hit Tortos with the proc every time, aka you pick up the bats and park yourself directly attacking the boss, then it might be beneficial. Still, the tanking meta might be more useful, because it's going to proc a LOT with that many things attacking you, and your main problem (the bats) will not likely go down much faster using the DPS meta.

    I suppose the real question is what problems are you experiencing on the encounter that's preventing the kill. If it's your own survivability, err on the side of caution with the tanking meta. If it's a problem with your healer, again, the tanking meta is the likely answer. If the bats are dying too slowly, maybe the DPS meta... but it's still not going to kill them much faster if at all. If the bat situation is fine and the problem is everything else, pick whichever meta you want.
    *Since I'm tired of reposting the link:* EH vs TTL and Hybrid Tax

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •