The State of Tanking in FFXI

[Zagen]

I don't need 20 paragraphs to tell you what's wrong with tanking in this game.

Two Words; Enmity Cap.

Two Words: No Clue.

How fast a job gets to the cap means nothing, how high the cap actually is for any job also means nothing when you consider the fact due to current game mechanics there is no reason to have a dedicated tank. That means stupid 1 shot moves, buffs/items that make any job near invincible or just about as likely to survive X hit/ability as a PLD, etc.

If a monster did 2k damage to all jobs not deemed a tank type and only 500 to the tank type, and Fanatic's, Fool's, and PD were removed then you start to have a reason for a dedicated tank.

[Rustic]

There's a lot that can be done. Give HNM's a "focus" that's generated when someone aggros them. Everyone else in a wide area loses enmity at a greatly reduced rate, meaning it's darn tough to get the HNM unstuck from the target in question. Occasional blasts aimed randomly at other targets on the hate list that aren't #1 is A-OK, especially if the focus is the only one who can interrupt them. In fact, that'd be rather awesome. Focus HNM's will also occasionally attempt to refocus on someone on their hate list- again, interruptable by the focused player.

"*yawn* <targetpc>, you are no longer worthy of my attention. <newtargetpc>, prepare to die!"

(Tank interrupts or in three seconds, new target becomes focus. If interrupted: )

"<oldtargetpc>, you dare! I will double your suffering!"

(Tank keeps focus.)

AoE's that cause worse effects if more targets are in the radius- whether it's damage or other nasty statuses. Reverse 1000 Needles style. Can be targeted on the user or as a ranged AoE. I'm sorry, were you melee zerging? No. Were you riding the enmity cap spamming ranged firepower? Not good either. AoE's that did more damage to everything AROUND the target would be interesting as well, or dealt significantly reduced damage vs. targets equipped with a shield (the larger the shield, the lower the damage)- thereby meaning that twohanded/dual wielding weapon users would take much more punishment than "sword and board" types.

"Occasionally reflects/deflects". Possibly lower evasion rate/defense, but redirects occasional attacks back at players in range. Yes, that was an awesome 5-hit WS. I'm sorry you just hit the other DD with it. And that was a lovely Detonator the Corsair just hit the healer with. The higher the enmity generated by the attack over the enmity cap, the more likely it is to proc a reflect/deflect. Odds increase with enmity levels that the attack is redirected to that target, which would work well with "focus" HNMs (see above). The higher total enmity the HNM has on it's hate list, the more often attacks are bounced. Tends to prefer tank as result. HNM replaces normal attack with reflected/deflected attack at target and briefly halts it's normal attacks to do so. May bypass shadows in process/consume multiple shadows to stop. Cannot reflect/deflect closed skillchain or magic bursted damage.

[Toadie-Odie]

There's a lot that can be done. Give HNM's a "focus" that's generated when someone aggros them. Everyone else in a wide area loses enmity at a greatly reduced rate, meaning it's darn tough to get the HNM unstuck from the target in question. Occasional blasts aimed randomly at other targets on the hate list that aren't #1 is A-OK, especially if the focus is the only one who can interrupt them. In fact, that'd be rather awesome. Focus HNM's will also occasionally attempt to refocus on someone on their hate list- again, interruptable by the focused player.

"*yawn* <targetpc>, you are no longer worthy of my attention. <newtargetpc>, prepare to die!"

(Tank interrupts or in three seconds, new target becomes focus. If interrupted: )

"<oldtargetpc>, you dare! I will double your suffering!"

(Tank keeps focus.)

AoE's that cause worse effects if more targets are in the radius- whether it's damage or other nasty statuses. Reverse 1000 Needles style. Can be targeted on the user or as a ranged AoE. I'm sorry, were you melee zerging? No. Were you riding the enmity cap spamming ranged firepower? Not good either. AoE's that did more damage to everything AROUND the target would be interesting as well, or dealt significantly reduced damage vs. targets equipped with a shield (the larger the shield, the lower the damage)- thereby meaning that twohanded/dual wielding weapon users would take much more punishment than "sword and board" types.

"Occasionally reflects/deflects". Possibly lower evasion rate/defense, but redirects occasional attacks back at players in range. Yes, that was an awesome 5-hit WS. I'm sorry you just hit the other DD with it. And that was a lovely Detonator the Corsair just hit the healer with. The higher the enmity generated by the attack over the enmity cap, the more likely it is to proc a reflect/deflect. Odds increase with enmity levels that the attack is redirected to that target, which would work well with "focus" HNMs (see above). The higher total enmity the HNM has on it's hate list, the more often attacks are bounced. Tends to prefer tank as result. HNM replaces normal attack with reflected/deflected attack at target and briefly halts it's normal attacks to do so. May bypass shadows in process/consume multiple shadows to stop. Cannot reflect/deflect closed skillchain or magic bursted damage.

You know... on one hand a part of me says that would be a pissy thing to do - especially the reflected/deflected part - but on the other hand, that kind of thing also sounds like a very interesting and dynamic sort of fight. If they used that we would have to be at the top of our game and really pay attention to what the enemy is doing and adapt accordingly. I think I could live with that.

Edit: somehow I quoted an extra post I didn't intend to that had nothing to do with my post.

[Rustic]

You know... on one hand a part of me says that would be a pissy thing to do - especially the reflected/deflected part - but on the other hand, that kind of thing also sounds like a very interesting and dynamic sort of fight. If they used that we would have to be at the top of our game and really pay attention to what the enemy is doing and adapt accordingly. I think I could live with that.

And that's the point. Enmity matters and it's not a spamfest of aggro-generating hellfire on the target from all and sundry, since said target is just firing off omnidirectional attacks like some sort of Touhou battle at this point. The party who goes berserk on their target invites fratricidal fire, the ones that don't make it more likely the mob directs any pain it can redirect towards the target most likely to tolerate it- the tank.

Special attacks that having a shield benefits more greatly gives a drawback to 2H/dual wielders to take heed of and increases the survival rate (and thereby increases the fight time a bit by reducing melee DPS), along with giving a good reason for other groups NOT to outhate the tank, lest said attacks be directed at vulnerable and oft-un-shield-using mages. (On the other hand, hey...WHM's and RDM's often can, another good for them. Yay!).

Take a special AoE/cone attack that deals, oh, 16% less damage for each shield size in hand- so a size 1 buckler is -16%, while a size 5 like an Aegis is soaking up 80%. Shields already reduce damage as a function of size, so nothing new here- but if it's spammed or used rapidly and often by the mob, it can make it a "if you're in the damage radius, use a shield or get shredded", while shield users can function effectively in the resulting storm of mob DPS.

"Danmaku" should not be the defining trait of FFXI endgame fighting. We should be paying attention to more than how often we can mash the WS macro or spam our damage-dealing/healing button of choice.

[Aeonk]

Two Words: No Clue.

How fast a job gets to the cap means nothing, how high the cap actually is for any job also means nothing when you consider the fact due to current game mechanics there is no reason to have a dedicated tank. That means stupid 1 shot moves, buffs/items that make any job near invincible or just about as likely to survive X hit/ability as a PLD, etc.

If a monster did 2k damage to all jobs not deemed a tank type and only 500 to the tank type, and Fanatic's, Fool's, and PD were removed then you start to have a reason for a dedicated tank.

All that does is create incentive for bringing a tank. But it doesn't help the tank do their job when a single WS from a DD recaps hate.

A DD should be able to do their job without having to hold back to the point of doing nothing in order to make the current enmity cap work. There should be a VERY tangible difference in enmity between an Ukko's Fury with Defender up, and Mighty Strikes Berserk Bloodrage Sekkanoki Ukko's Fury x2 with Light SC. Right now, there's not.

So yes, an overall raise in the enmity cap is a necessary change. But the incentive needs to be there too. So in essence, you're both right.

[Motenten]

I have a huge amount written up on the details and foundations of tanking, but honestly it gets a bit boring, so will just skip a bit.

In the current state of the game, damage mitigation is reasonably well balanced. Plds have shields, nins have shadows, thfs have evasion, mnks have counters, and everyone has PDT. Each of those are suited for some types of mobs, but not as much for othes.

As has been pointed out, there are two problems facing tanking right now: the ability to hold the mob's attention, and the pervasiveness of AOE attacks that make the presence of a tank a moot point.

I've played around with lots of different models for hate mechanics, but Rustic has hit on the basis for an idea that could actually work.

First, separate enmity (CE/VE) into tiers. Not certain of the granularity, so let's just say 1000-point tiers. So, everyone from 9001 to 10000 enmity is on the same tier and, assuming the mob is picking a new target, all of them have equal likelihood of being chosen as the next target.

Second, decrease the granularity of target updates. Instead of near-instantaneous changes every time any action is done (for the good ol' spinning mob effect), the mob will only consider changing targets every ~15 seconds. This means, regardless of enmity changes, if you get the mob's attention, you'll have it for the next 15 seconds (for good or ill). After that, the mob checks to see if there's someone else it should be focusing on instead.

Now, together that seems a bit problematic, until you take into account player actions which can change things up. For example, Provoke, in addition to the enmity it provides, instigates an immediate re-evaluation of targets. Puller brings in a mob? Tank vokes, goes up a tier on the enmity list, and forces a new evaluation. Result: mob is now on the tank for the first 15 seconds of the fight, even if the DDs immediately unload with weaponskills. Healer getting too high up on the enmity list and the mob starts heading for her? Provoke can get one of the other players on the same enmity tier a chance of being the target instead.

Various other actions can likewise force a target change. Enmity Douse/Accomplice/Collaborator should do it if used by/on the current target of the mob, the various Bash JAs, etc. Then you could modify various spells like Animus Augeo and Animus Minuo to change how long the mob is targetted on a player. EG: a pld with Augeo might get 20 seconds per evaluation, while a DD with Minuo might get 10 seconds per evaluation. If you want to have fun, add something like dChr/10 as additional +1 second per.

A final change would be the need to adjust how enmity is actually accumulated. I -think- you could get away by simply scaling enmity generated from damage by the fraction of total target HP percentage done, relative to some baseline that goes up over levels. So, a level 75 player fighting a mob with 5000 HP gets a 1:1 scaling; same for a level 99 player vs a mob with, say, 7500 HP. However against a mob with 75,000 HP, each point of damage is only worth 1/10 as much enmity. That should make the JAs with constant enmity values significantly more valuable against high-HP targets.


After that you still need to find some way of better addressing the AOE damage issue. Another idea somewhat similar to Rustic's: use the equivalent of sch's Libra (but for all possible targets, including alliance or multi-alliance) to determine the strength of the AOE attack on any given player.

The tank who does the initial voke obviously starts with 100% hate, and thus will take 100% damage from any given AOE. Meanwhile, the thf that ran into range to try for a TA, but has only poked the mob and is sitting at 3% hate, will hardly take any damage at all since the mob is only barely aware of them and is instead focusing the full strength of their attack on the tank. On the other hand, the DDs that have surpassed the tank in total enmity, even if they aren't the current focus, are now the ones being hit with the full strength of the attack.

[Aeonk]

snip

That's actually a pretty interesting idea, and sounds like it could work pretty well. I'd go for that too.

[Zagen]

All that does is create incentive for bringing a tank. But it doesn't help the tank do their job when a single WS from a DD recaps hate.

A DD should be able to do their job without having to hold back to the point of doing nothing in order to make the current enmity cap work. There should be a VERY tangible difference in enmity between an Ukko's Fury with Defender up, and Mighty Strikes Berserk Bloodrage Sekkanoki Ukko's Fury x2 with Light SC. Right now, there's not.

So yes, an overall raise in the enmity cap is a necessary change. But the incentive needs to be there too. So in essence, you're both right.

Raising the caps doesn't do anything if the DD aren't losing enmity faster than they regain it, currently the only way to do that is get hit with your back turned or not casting.

Making tank jobs take less damage than others and lowering the enmity gain of non-tank jobs or speeding up the enmity loss is a viable option but then it makes tank jobs over powered like SE doesn't want.

[SpankWustler]

A change I'd like to see is just for more jobs to receive more relevant ways of reducing enmity, if changes are ever made that make enmity and tanking really relevant again. Lots of melee have excellent ways to mitigate damage or even turn being a monster's target into a way to deal more damage, but ways to lower enmity are far more rare.

"Sit there, do less or do nothing, think about the poor life choices that led you to play FFXI, double-down on your shame by using this opportunity to browse Redtube," is a very dull and one-note strategy for controlling enmity gain and should be an option of last resort.

Some such changes have even been mentioned by the Development Bros then seemingly passed over, such as Super Jump always erasing Enmity for a bro behind the dragoon or a Ranger accumulating less enmity when shooting from behind a monster. Adding more and more abilities like these would allow players to still be productive while applying strategies to manage enmity.

Also, the same changes that made enmity and tanking relevant again would hopefully make current abilities and spells such as Accomplice or Animus Minuo relevant to people who hit things with things.

Second, decrease the granularity of target updates. Instead of near-instantaneous changes every time any action is done (for the good ol' spinning mob effect), the mob will only consider changing targets every ~15 seconds. This means, regardless of enmity changes, if you get the mob's attention, you'll have it for the next 15 seconds (for good or ill). After that, the mob checks to see if there's someone else it should be focusing on instead.

This would be super-helpful for all the position-based abilities in FFXI that can currently be very grating to use, too.

[Ziyyigo-Tipyigo]

"Sit there, do less or do nothing, think about the poor life choices that led you to play FFXI, double-down on your shame by using this opportunity to browse Redtube,"

You do dat, you go to da box, ya know, two minutes by yourself and... you feel shame, ya know? And den you get free.