Jayhawks > Ducks
At least in basketball, we suck at football, but I guess we'll see how the new coach and QB work out.
I doubt the idea of gearing for Effective Health is going anywhere. In the Burning Crusade, healer mana was an issue and we could take multiple hits and yet this was the expansion that EH gained the majority of it's recognition and following. (Which isn't to say that EH was or wasn't the smartest way to gear in TBC)
WotLK has been pretty heavily slanted towards EH and I feel they want to get away from that in Cataclysm. Certainly they don't want us avoiding every trinket that has Block, Dodge or Parry rating on it.
It's a fine line though and as Aggathon indicates earlier in this thread, I don't want to have to do complex math in order to determine whether or not an app is gemming correctly. I believe this aligns with Blizzard's goals for Cataclysm too though. They want that perfect scenario of multiple stats being advantageous, but that finding the "right" answer isn't incredibly hard. Whether they can accomplish this or not remains to be seen.
What I'd like to see is socket bonuses improved in Cataclysm. I like that we have the option to ignore socket colours, but I feel that this approach should be discouraged as much as possible. Some ideas of how they could change that:
- Give items a proc if they match the sockets
- Double the stats of all gems socketed if they match the sockets
- Make an item change visually when it's socketed properly (sure it won't stop us min/max folk, but it's still got a certain cool feel to it)
- Make socket bonuses more unconventional like Increased Run or Mount Speed, Reduced Fall or Spell Damage
- Make split socket colours that allow you to say gem blue or red and if you gem blue it gives you extra Stamina and if you gem red it gives you extra Dodge
- In the case of set pieces, make socket bonuses affect skills like increasing the duration of Thunder Clap by 5 seconds or reducing the cooldown of Charge by 1 second
I feel like if they approached gemming in a more creative way, it'd steer us away from gemming in a single obvious way. I'd even like to see some items work like meta-gems where they only work if you're using the right number of blue, red and yellow gems.
An introduction into WarTanking (no longer updated as I've retired from WoW - the concepts will still be mostly accurate but the numbers no longer will be.)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
It's been discussed that they very well may remove bonus armor and the like from stat allocation, but what about all that armor we already have? By reducing avoidance and increasing Stamina across the board, armor becomes even more effective.
One example that immediately comes to mind: with avoidance lower, a trinket like Unidentifiable Organ will proc more and mitigate damage through armor at the same time.
Armor will more reliably reduce damage, and therefore maintain healer mana in a much more even way. Sure, we'll replace most pieces because of massive stat inflation, but a couple of these ICC armor pieces are going to hold up for a little while at least, especially if they don't put armor on items anymore.
Also, block is getting a buff in raids now as well, as it'll be mitigating a lot more damage in the new model and be a percentage mitigation on block. You'll want block rating pretty bad methinks.
Last edited by Byronius; 07-22-2010 at 12:27 PM.
All tanks very well "stack" Stamina.
Failure is not an option. It comes bundled with the software.
The fact of the matter is that, as long as they make bonuses something that are meaningless and not worth achieving, people won't want to achieve them. We get that they don't want them to be considered "necessary", but there are ways of getting an optimal balance, such as what Vene suggests. These concepts can be applied beyond socket bonuses too.
I'd love to see those suggested gemming ideas come to pass. The only problem might be that eventually people might start stacking odd gear for a specific gem bonus set, which could be used to break some encounters. I could see it being one of those "balancing nightmares".
Stacking stam in every socket has never felt good to me. The way gemming is set up it might as well not even exist. Every class has a specific way they gem. They basically could have just baked another 90 stam into Pillars of Might and taken out the sockets. You can't use gems to excel since you're balanced around proper gemming, so really gems just opened up a whole new way for people to fail.
Also, if threat is going to be something we will have to watch again stacking parry will be good for Hold the Line. I guess we'll just have to see how much health we need to survive ~3 hits and go from there.
Aim for some sort of softcap of EH that lets you survive some bad rng or some unavoidable magic attack + 1 melee hit and go for more avoidance or threat stats after that. Think thats the way cata is going to go and its definitly more interesting then blindly stacking EH.
I doubt we'll see a lot of change - stamina will always be a good stat for tanks, as will armor.
The only way to really fix it would be to give other tanking stats more of an effect when gemmed for, but I doubt we'll see that happen.
Originally Posted by Kahmal
as long as tanks can die, effective health will be the thing that trumps all. Whether health pools will be big enough to buffer long enough for EH to lose its spot as the thing to go for is something we will only know when the raid gear and boss numbers are known, which won't be for a while given that they are still iterating talents and in the case of paladins the whole class design.
In my opinion, Effective health will lose importance relative to avoidance IF AND ONLY IF bosses cannot kill a tank in the cast time of a greater heal. Will this come to fruition? only time and a score of beta builds will tell
With Cataclysm going the way "we want less numbers", you can expect all the "idealized model" optimization to go the way of a pure academic exercise.....
This is going to sound a bit far fetched but I've been playing Final Fantasy XIII a lot during some wow-downtime and its actually opened up some interesting aspects of future combat system of WoW.*
To be honest this is better in FFXIII than WOW. You have a greater padagrim shift, not always the "tank dps and healer" fight. Some fights are more effective going heads on with dps, the switch everyone that can to healing and then back to dps. You have party-wide damage bursts. Some fights have timers and thus requires you to do this or simply going tank, dps and healer your healing output is to low to recover the bursts of damage.*Heals are normally smaller and faster and given you have the move "one-two hits kill you" bosses, a lot of encounters are not and it does open interesting options.*
What FFXIII has whats somewhat different to WOW is you character can do everything in the end (i.e be a buffer, debuffer, dps, healer or tank). Some will excel more at one or the other and you will also gear for one or the other but the underlaying structure is as such a tank can in combat "stance dance" and do dps. This brings an interesting dimension to the game. Yes, you are running a party of three people but in all fairness this could scale to 5, 10 or 25 man parties as in a game like WOW. How would that work? Well.. A class like the Warrior has stances built in and IMO I see no real issue buffing each more profound.*
- One defensive (tank) stance. Its only here you effectively can tank. This can be controlled with damage taken and threat, possibly available abilities (taunt for example).*
- One offensive / buffing stance. Fury would fit in well.*
- One offensive / debuffing stance. Arms would fit well.*
All the above would work and the diversity between a tank-spec and a dps-spec will still be very far apart just though talent points spent and gear.*
Not sure, might be me having a crazy vision here but the above would for me be very refreshing.*
What was said in regards to gemming is also something Blizzard should look into. To keep the system open perhaps build gem-bonuses to talents. An easy solution would be simply having gem-bonuses be mastery points but that's a fairly lame way of doing it and won't scale at end-game. Improving talents based on gemming is interesting but would ultimately make things tricky to balance (i.e event X is ease-mode if the tank has 5s longer Thunderclap though gemming). What they defiantly should do is make gem-bonuses much greater than today.*
The poster that said EH is dead in ICC clearly hasn't done LK25M HM or HALION 25M HM. EH is still king here, the damage output is insane even considering the x% buff we've got in the former. Druids rule as LK MTs, simple as that, and even when MT healing and CD usage is done correctly, RNG can wipe the raid. This IMO isn't something that should be in the game. It stupid, random and not based on skill.*
So, in the end, less 1-2 shot moments IMO is a good thing. If this means we will or won't focus on stamina that doesn't matter to me THAT much. Making avoidance mean more is a good thing but we can't end up to complex of a system to min-max.*
This is contradicting.RNG can wipe the raid.
BTW I asked for hard data, not for memories of the raid. Do you have statistics on tank deaths which indicate that a +x% difference in EH makes the difference? Humans are horribly bad at evaluating risks at low percentages, overestimating them a lot (the typical example is the "OMG you have one wrong gem, do you know how many times we wiped at 1% and it would make the difference?" aimed at DPS. In reality wipes at 1% due to lack of DPS are rare, but people remember them a lot more vividly than then other 50 wipes where DPS died because of wrong execution).
The vast majority of our wipes on HLK were due to defile.
Most of our tank deaths could be attributed to an enraged, frenzied shambler not being tranqed/stunned properly and/or not using tank cooldowns. More EH would not have saved against a 150k hit.
On LK himself, most of our tank deaths happened when we did not use a CD on soul reaper and a defile caused healers to move out of range. In this situation it's theoretically possible that some of them could have been survived with more EH, but it's very doubtful; tank deaths tended to be on the order of 20-30k over. Note that these didn't cause wipes all that often.
That being said, we went from several tank deaths happening every night at 20% (with every druid standing by on res duty and soulstones on the LK tank) to only a few tank deaths happening (which were mostly recoverable) to a rare tank death a night as we went from 15-20-25-30%. It's hard to say how much of that is experience with the fight, the buff itself, healing increasing, etc. Gear wise our tank's gear hasn't changed significantly since 15%, so that's not it.
I could try to summarize our tank-deaths on our LK and Halion tries using a DK, Warrior, Paladin and Druid but it would take a tremendous amount of time. From our experience things where just far easier in term of keeping a Druid up compared to the other classes on certain parts of these encounters. You can blame healers not knowing the event or what ever but with a very similar setup, our Druid just survived better.
What you bring up Felhoof is controllable RNG - the Shamblers are relatively controllable. What they fail with however is the level of un-controllable RNG from:
- Tranq. resist and the fact only ONE class can sort this out
- RNG line-ups of abilities and the fact you have few options to deal with this (enraged / frenzied shambler that does a melee + shockwave combo is nice).
All in however the Shamblers are relatively good mobs to deal with and mostly all classes can survive them similarly. I would say EH is the thing you go for surviving thing - some of their damage (shockwave) is unavoidable and their random hits while enraged / frenzied will hit for a ton hence EH > Avoidance. The abilites of the mobs however also favours abilities like stuns which is a nice "finess" from Blizzards part. Survive a massive strike by stunning, disarming or using a CD.
LK is by far favoured for a EH tank with heavy use of CD's. SR + Melee-hit on a tank with 95K HP or 82K HP is a massive difference in survival. If a shambler beats its tank for 180K damage you've done something wrong normally. The LK damage is just RNG IF the melee and SR tick lines up (which is just poor design), still the tank with greater EH has a greater chance of survival.
We've noticed the survival-rate of the LK MT has greatly improved since the 15% buff. We also noticed a huge survival rate "up" when having me as a warrior tank upstairs on Halion and the Druid downstairs.
2) Lack of EH optimization != avoidance-savvy tank.
3) if even with EH optimization RNG can wipe means that RNG is more king than EH..... EH is supposed to be the solution to RNG.
What I'm questioning is the fact that you need to optimize EH to the last gem. Or, to give a practical example: imagine gemming your red sockets with +agi/+sta instead of full +sta. Do you REALLY think it would make a significant difference?
Raid wipes, not tank deaths. Again, what I question is that spending 10h to optimize a solution out of a problem which happens 1% of the time is useless (even worse: the optimization is not done on real data but on an idealized model).I could try to summarize our tank-deaths on our LK and Halion tries...
It's very difficult to extract data from this statement, since as tries advance people know the combat better, which can make the information completely useless. In the twilight realm there's a lot of movement for healers, in the beginning our tries were very short... (in HM they still are :P)We've noticed the survival-rate of the LK MT has greatly improved since the 15% buff. We also noticed a huge survival rate "up" when having me as a warrior tank upstairs on Halion and the Druid downstairs.
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.