+ Reply to Thread
Page 4 of 11 FirstFirst ... 2 3 4 5 6 ... LastLast
Results 61 to 80 of 204

Thread: Cataclysm - The End of EHP?

  1. #61
    Join Date
    Jun 2009
    Posts
    2
    Well, tbh with you i really believe is icc where really ehp really confirmed to be absolute king...with the dodge debuff and such...and sorry but if you saying that lk or putri are not "high hitters" probably you never tank them bfore.sorry for any mispell (i'm italian) and yes, long time reader first time poster.

  2. #62
    Join Date
    Apr 2010
    Location
    Oklahoma
    Posts
    1,055

    Cataclysm - The End of EHP?

    Quote Originally Posted by tlitp View Post
    Helistar : when it comes to "EH is king lol", I'm afraid there's no room for an eppur si muove. You're deemed either a troll, either ignorant. Pointing at evident displays of secundum quid doesn't help at all. C'est la vie.
    I am assuming you are a fan of avoidance then. Nobody is saying you can't gear that way and be successful. I think that you are losing sight of the actual argument which you are accusing others of doing. The argument is that the most consistent and reliable way to survive is EH. The math in simulations has shown it to be true. The work involved in empirically proving it is daunting, however. The sheer number of logs that would need to be individually analyzed makes it unlikely someone will spend their free time doing it, especially if it is just to confirm what all the other methods have already determined.
    "he doens't need healing, he doesn't need healing, he doesn't nee-WHAOSHIT!wtf was that man!". Please stop leaning on TDR. -Teng

  3. #63
    Join Date
    May 2008
    Location
    Sweden
    Posts
    389
    Quote Originally Posted by Helistar View Post
    imagine gemming your red sockets with +agi/+sta instead of full +sta. Do you REALLY think it would make a significant difference?
    If you'd swap all gems for their native colors, i.e: Blue (Stam), Red (Dodge) and Yellow (Defense), yes, the EH of the given tank would go down a lot and I think on a lot of encounter it would make a huge difference in terms of tank-survival.

    Quote Originally Posted by Helistar View Post
    Raid wipes, not tank deaths.
    On most encounters the raid (i.e. anyone in it) will more often wipe the raid than the RNG-bad-luck-zomg-tank-death. That however has nothing to do with the fact that MT survivability is based on.. MT survivability. I don't care in that matter of the raid mostly costs us a wipe - it's a good thing but has little to do in the EH of the MT's discussion. What's on the table is if EH is better or worse for a tank. If 5% of the wipes is a wipe due to RNG based on a tank-class and you can shave that to 2% with higher EH - EH is king no matter what you try to say. You've lowered this case of wipage by more than 50%. I don't know of any boss where people have said "oh, hey, put on the avoidance set and let's hope for the best!". Bosses like LK HM and Halion HM have a lot of stacking damage - multiple sources of damage - some even unavoidable.

    In the current game where 1-2 hits can kill the MT, EH will be king most of the time. Adding 1% avoidance to a tank that will die with-in 1-2 hits won't really do much as healers will still spam-heal and not be using reactive-heals. Adding 1% avoidance to say 5-8 hits before the tank dies makes more sense - especially if we're sitting on a base of say 20% avoidance rather than 40%.

    Quote Originally Posted by Helistar View Post
    In the twilight realm there's a lot of movement for healers…
    Still our druid with is ≈10K more HP survives way better than I do as a Warrior with the same healer-setup.

    Other bosses like Sindragosa or Patricide have proven EH and/or CD-savvy classes are very efficient there. Then again, if Sindragosa wouldn't 2-shot people with her melee-swings (at the 0% buff with out full 277-gear that is) but her breaths where as high as they where now:

    1. You'd have a lot more use of reactive abilities and esp. reactive healing
    2. Healers would be in a situation where they can't spam-heal for 5 minutes straight - they have to conserve mana but still top the tank up at a given times
    3. You still have tank-swaps and thus want to conserve mana-usage here.

    With no 3 I could even see a tank being required to use a bandage (if these are actually powerful again).

    Imo the above scenario would be more fun then the random hit-hit-hit-death or hit-breath-hit or what ever killed our tanks when learning the fight in less than 1s (parry-haste could actually be a pain here). This RNG, in my opinion, bullshit isn't fun and we can counter it today by one main mean - EH and stacking as much of it as we possibly can.

    Again, show me an avoidance tank going for Sindragosa with the 0% buff. Good luck surviving that.

    To add to the above - if the per-hit-damage is lowered compared to our HP-pool the RNG of hit-hit-hit will do one thing mostly and that is requiring our healers to be reactive on healing, perhaps even redirect healing to the MT and possibly the MT being reactive on CD-usage. Today it goes so fast sometimes you have no chance of reacting (or server-client-lag kills you). This is not "what the hell happened there" once in a life-time event - it's far to common of an issue in the current game. It might be removing this and putting strain on things like mana-conservations makes the game easier or more boring, I can't say but it's too much 100-50-100-50-100-0 today.

  4. #64
    Join Date
    Feb 2009
    Posts
    47
    Quote Originally Posted by sifuedition View Post
    I am assuming you are a fan of avoidance then.
    He didn't write anything of the sort.
    Don't assume that "EH optimizations are overrated" is synonymous of "avoidance rules". It isn't.

    The math in simulations has shown it to be true. The work involved in empirically proving it is daunting, however
    And this doesn't strike you as a problem? The validity of the results of a simulation are based on the fact that the simulation correctly models whatever you're trying to simulate.

    I have nothing against EH in itself, actually I use TTL calculations to choose my stuff, but from what I see the few % I can gain with uber-optimization are irrelevant, I can't remember a single wipe where looking at the death log, 3% more hp would have saved me. It's always a 30K hit sending me 10k+ negative.

    What I don't like is that people treat it the same way as GearScore. It's a number, the bigger the better, it's based on information which may (or may not) reflect reality, we don't even care about the assumpions (which are quite big) used to create that model, but we'll swear by it? Well, not me.

    I give you an example of why I find that at times things get exaggerated a bit:

    Quote Originally Posted by Roarc
    If you'd swap all gems for their native colors, i.e: Blue (Stam), Red (Dodge) and Yellow (Defense), yes, the EH of the given tank would go down a lot and I think on a lot of encounter it would make a huge difference in terms of tank-survival.
    down a lot, lot of encounters, huge difference. Ok, in my stuff, if I switch all gems the change in stamina is a grand total of 174 (I might have missed a socket bonus, BTW this is counting everything, even gems I would never consider switching because the socket bonus is complete crap). My total stamina is 3119 (unbuffed). That's a 5.5% difference... Negligible? Not really. Significant? Neither. The 10K difference between you and the druid is more around 20% than 5%.

    BTW the example you make: if 5% of the raid wipes can be reduced to 2% by EH optimization, that's a 3% reduction of raid wipes (100%->97%). Optimizing the remaining 95% of wipes would be a lot more effective.
    And I'd like to see the logs for this, because as I wrote above, cases where 2-3K HP more would have saved me I've seen none... so 3% less wipes? I don't buy it. Yes I know I'm boring, I always want real data

    Imo the above scenario would be more fun then the random hit-hit-hit-death or hit-breath-hit or what ever killed our tanks when learning the fight in less than 1s (parry-haste could actually be a pain here). This RNG, in my opinion, bullshit isn't fun and we can counter it today by one main mean - EH and stacking as much of it as we possibly can.
    For the rest of your post I completely agree with you that the current approach of tank-from-100%-to-death-in-2s is completely idiotic, and in this case avoidance is equivalent to playing the lottery (even more so with the wonderful idea of the dodge debuff, which screws up the DR curve, making avoidance not only unreliable, but also numerically less effective than pure stamina stacking in TTL calculations, the only thing it affects is DTPS). But in this scenario (tank dead in 2 hits) it's not like 2% HP more will be more reliable than 2% avoidance in saving your ass.
    The lack of reactive CD usage is something I particuarly dislike, because it becomes "press the button because the timer on BigWigs told you so" instead of "use your judgment".

  5. #65
    Join Date
    Sep 2009
    Posts
    721
    Amusingly enough, we minimzed tank deaths significantly on Halion by...regemming for expertise over stamina. Because big hits weren't killing us - parry hasted hits were. In that specific case, it was far better to lower the chance of a parry hasted hit (since there would still be parries from time to time) than it was to gain an absolute amount of EH that wouldn't save anyone.

  6. #66
    Join Date
    May 2008
    Location
    Sweden
    Posts
    389
    Quote Originally Posted by felhoof View Post
    Amusingly enough, we minimzed tank deaths significantly on Halion by...regemming for expertise over stamina. Because big hits weren't killing us - parry hasted hits were. In that specific case, it was far better to lower the chance of a parry hasted hit (since there would still be parries from time to time) than it was to gain an absolute amount of EH that wouldn't save anyone.
    We had similar findings learning Sindragosa given it was a greater difference simply swapping tank-classes (i.e. a DK over a Warrior). This is quite fight-specific but it very much favours a slower weapon class and a class with a lot of CD's. We never swapped stam for expertise in gems but we did go with expertise-food and focused on expertise-heavy armor. Being at 30-ish expertise was a far deal better than being around 15. After a few buff-rounds (10 or so %) this was a non-issue however (more EH ).

  7. #67
    Join Date
    Sep 2009
    Posts
    721
    After a few buff-rounds (10 or so %) this was a non-issue however (more EH ).
    Actually, I'd argue strongly that on Sindragosa it wasn't extra EH on the tank - it was extra healing throughput. Sindragosa is a hugely healing intensive fight, and having 10-15% more healing on things like shields and hots on it is a huge deal. By comparison, you know for certain that EH at at least the 5% buff level wasn't that big a deal for Sindragosa, given that you could have had that but chose to go Expertise. Then again, the tradeoff on armor is rarely EH vs Expertise, it's usually dodge vs expertise.

    And honestly, through WotLK the biggest finding and the most common result of fights was that the tank with the most effective CDs was the victor. EH rarely was the primary decider on the best tank class. Now, within that tank class EH was likely better than anything else (though again - Halion was an exception, and one could argue Anub and a couple others) - but comparing two tanks of different classes solely by EH usually wasn't the help people often think it would be.

  8. #68
    Join Date
    Jun 2009
    Location
    cluj napoca,romania
    Posts
    3
    There are so many types of tanks.The ones with huge HP,the ones with huge damage input (pala tank anyone!?),but I prefer my type of tanks..acceptable hp so you don't get 1 shotted,and avoidance all the way.That and the mitigation (a bit dodgy on a shieldless therefore without block dk).So I prefer to get lesser damage instead of having a huge HP this even meant to go south if your healer decides to take a nap

  9. #69
    Join Date
    Nov 2009
    Location
    WI, USA
    Posts
    2,614
    Ok... so reforging is out now. And like I mentioned several pages ago (back on page 1), reforging does not result in a loss in stats, the item level of the item reforged remains constant (meaning the stats remain constant). The percentage only pertains to how much of a given stat you may reassign. So if you take 21 (of 53) haste rating and convert it to hit rating, you get 21 hit rating.

    What this means is that people will probably continue to gem and enchant solely for effective health. And reforge stats like hit, expertise, or avoidance stats into other avoidance stats. So a tank that is over hit capped will reforge hit to avoidance (because they can't reforge to effective health they would choose to reforge to avoidance). So in Cataclysm I think that people will gem and enchant for effective health because that is still the higher priority and the only means of getting it. However reforging will be used to maximize avoidance (ratings). It can also involve reforging one avoidance stat to another to minimize the effect of deminishing returns. If you are high on parry you can reforge parry rating to dodge rating to get higher benefit as an example.

    Reforging will largely be a balancing act between threat and avoidance, usually transfering between expertise and hit ratings to and from dodge and parry. This balancing act is as your present situation requires. If there is a parry hastable boss like Sindragosa for instance, expertise increases in priority naturally.
    Last edited by Quinafoi; 08-05-2010 at 08:10 AM.
    "In anything, if you want to go from just a beginner to a pro, you need a montage." /w TankSpot WTB Montage for Raiders.

  10. #70
    @Quinafoi

    This is a scary trend in my opinion. Sure, reforging hit once you're over the hit cap sounds like a good idea. However, I would probably reforge to expertise and then start working on avoidance. Once you get into reforging one avoidance stat for another is where I see the general community dropping off the deep end even more than it is now.

    I would not be surprised if Reforging saw some form of major redesign pre-release because, in tank terms especially, it makes stats even more mathy than they are now in WotLK. How deep into the DR of each stat do you get before it becomes move valuable to reforge towards another stat? What if there is an instance with a Dodge debuff again. Do you reforge ALL your gear to give parry instead? In my mind I see a vision of tanks literally writing Aggathon-style dissertations to figure out their reforging priorities and hearthing to town in between certain fights to reforge to a more fight faborable stat. Now this vision might be a bit extreme but I would rather overkill it and be prepared than be naive about the whole thing and get caught off guard once Cataclysm raiding begins for my guild.

    None of this is bad except that the general trend of Cataclysm is to simpligy the math involved in this game, not further complicate it.

    What I would rather see is forced gem color choices and a larger variety of gems within each color. Sure, it eliminates any real customization for gear however none exists currently anyway...every class has that one certain gem that they stack into every socket pretty much. DPS Warriors? ArPen. Unholy DKs? Str. RDPS? SP for the most part. If you were to force gem color and then provide more choices as to gems, you would cut the EHP of all tanks across the board but if you tune fights to the new level of EHP, you can actually begin to make gemming a much more involved and entertaining enterprise.

  11. #71
    Join Date
    Nov 2009
    Location
    WI, USA
    Posts
    2,614
    I didn't say it would be easy. I just said thats likely how it will end up, at this time it is still just an educated guess. You will gem and enchant for effective health because you can and that is still highest priority overall. With reforging however you can't a lower value stat turn that into effective health, so reforging will be a balancing act between threat and avoidance. Expertise in the case of parry-hastable mobs falls under both threat and avoidance, however against a non-parry hastable mob only falls under threat. The choice to reforge for expertise is probably the most complicated part because some encounters may make that better than pure avoidance would be because of parry hasting. Hit has a measurable target to shoot for so that won't be too complicated. Dodge and parry, although having complex calculations, there is a way with the use of spreadsheets or tools like rawr to calculate which is currently more benefitial for those trying to maximize the effect, however there will probably be some generic guidelines to shoot for to simplify it for 99% of the time (like try to get 2% higher dodge than parry before getting more parry).
    "In anything, if you want to go from just a beginner to a pro, you need a montage." /w TankSpot WTB Montage for Raiders.

  12. #72
    Join Date
    Jul 2010
    Posts
    6
    Reading the title of the topic, I assumed that it was talking about Cataclysm but I might be wrong...Anyone gots real conclusions from the beta that could give us some informations ? Main questions are the actual Time to live for random tanks in dungeons or in the places were mobs are supposed to hit quite strongly, how many time a healer can spam its greater heal without gettint out of mana, what is the average duration of a fight.

  13. #73
    Join Date
    Sep 2009
    Posts
    721
    None of that information is remotely set, Menethal. Tank health is poor, tank TTL is bad, healers are regularly feeling weak (and weaker as they level up), and things are completely unbalanced right now. There aren't any heroics, much less raids, so it's hard to say. There's simply nothing reasonable to judge things on.

  14. #74
    Join Date
    Feb 2009
    Posts
    47
    Yes, I agree with felhoof, it's way too early to judge. Something which was evident from my two instance runs is that Zerg-lol-AoE mode is dead. Also, boss fights are more complex, closer to the easier raid bosses than the dungeon bosses (and note, this was in normal mode). But with half of the skills and talents not implemented/not trainable/not working, it's really hard to get a complete picture.

  15. #75
    I think the situation always depend on the need of cutting edge progression. Even in WotLK, there are some fights that favor avoidance/resistance more than armor/stamina. When people are first tanking DBS, they would choose avoidance gear over EHP gear. And for Sindragosa, some forst resistance gear is nice compared to the normal armor sets. Tanks are always supposed to adjust their gear according to different encounters.

    However, nowadays EHP gear is generally the best tanking gear, as healers don't suffer from mana problems, and end game bosses hits really hard, like Heroic Professor P3 or Heroic LK. This concept could be changed if healers would suffer from lack of mana in CTM, but the essential thing is whether the tank EHP is big enough to survive the boss. If your EHP is so low that you couldn't survive even 2 successive hits, then your healers would be forced to spam healing on you, regardless of your hp level, and that would be a far more waste of mana than the benefits from your increase in avoidance. The possibility of conserving mana arises from enough EHP from the tank, and I highly doubt the bosses will hit so light that EHP buffer won't be a problem, so my guess is that EHP will mostly likely still be the top tanking status.

  16. #76
    Join Date
    Dec 2008
    Posts
    4
    Any thoughts on the possibility of dual-wielding mobs/bosses in Cataclysm, and how that would have an effect on our choices of EHP/Avoidance?

  17. #77
    Join Date
    Feb 2009
    Posts
    643
    Quote Originally Posted by Anticipation View Post
    Any thoughts on the possibility of dual-wielding mobs/bosses in Cataclysm, and how that would have an effect on our choices of EHP/Avoidance?
    I'd assume it would increase the power of avoidance seeing as it would essentially cut a single hit's damage in half reducing the damage of a 10 hit streak to a 5 hit streak from a non dual wielding mob reducing the chance of a total streak that does X amount of damage in half. So overall it would be less risky and dangerous banking on avoidance.

  18. #78
    Join Date
    Feb 2010
    Location
    Portland, OR
    Posts
    2,043
    Quote Originally Posted by krc View Post
    I'd assume it would increase the power of avoidance seeing as it would essentially cut a single hit's damage in half reducing the damage of a 10 hit streak to a 5 hit streak from a non dual wielding mob reducing the chance of a total streak that does X amount of damage in half. So overall it would be less risky and dangerous banking on avoidance.
    Hmm?

    A) Don't assume dual-wield means each hit hits for less. That's probably a variable they would play with to increase/decrease fight difficulty.

    B) With DW, even if you avoid one attack, there's still a second one coming in, which could land while it wouldn't keep the 'streak' alive, it would result in continuous damage. You'd need double avoids to result in the same 'break' for healers to catch up than you would with single wield.

    C) Just thinking about it, imo this makes armor even more valuable for the same reason armor is more valuable in ICC. You're going to take more hits. They may be smaller, but they are likely to be more consistent regardless of 'true' streaks because there will be fewer true 'breaks'. If you have 50% avoidance against a single wielding target, you have a 50% per swing to avoid any damage that swing timer. Against dual-weilders you only have a 25% chance (.5 * .5) to avoid any damage that swing timer. Thus the focus would be making the damage you take smaller, making you easier to heal up.

    With damage intake more steady, even if smaller, survivability may not be so much on avoiding, but minimizing incoming damage.
    An introduction into WarTanking (no longer updated as I've retired from WoW - the concepts will still be mostly accurate but the numbers no longer will be.) - http://www.tankspot.com/showthread.p...101-The-Primer

  19. #79
    Join Date
    Sep 2009
    Posts
    721
    And the best way to avoid incoming damage per itemization point is....avoidance!

    Yes, dual wield (or any boss that swings a lot of times) increases the value of avoidance. Basically the rule of thumb is that the closer you can get to avoidance becoming statistically correct, the better it is. This is achieved by either having a very high amount of avoidance so that large streaks of non-avoided attacks are statistically impossible, or it's achieved by taking so many hits that it is statistically unlikely you'll get huge streaks. You see the latter when facing many, many mobs; at that point avoidance becomes the best bet (that and static reduction of damage).

    So yes, assuming that a bosses' DPS remains the same and it's a choice between taking one slow hit and two fast ones, avoidance is favored in the second case.

  20. #80
    Join Date
    Dec 2009
    Posts
    22
    Quote Originally Posted by felhoof
    (...) having a very high amount of avoidance so that large streaks of non-avoided attacks are statistically impossible (...)
    At avoidance 0.999, a string of contiguous non-avoided attacks is improbable. At avoidance 1, is impossible.
    Last edited by tlitp; 08-20-2010 at 11:51 PM.

+ Reply to Thread

Bookmarks

Posting Permissions

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