Greetings, glorious adventurers! If you're joining in our Alpha One spot testing, please follow the steps here to see all the latest test info on our forums and Discord!
Aggro/Threat mechanics don't work in PvX
Lodrig
Member
MMORPG's have traditionally used a system of mob AI called Aggro in which a mob will preferentially attack player characters which perform certain actions, primarily that of dealing damage to the mob but later expanded to things like healing friendly characters. This was a simple AI which fit the constraints of early computer processor cycles while producing mob behavior which was better than the most primitive system of just attacking the nearest target or most recent attacker. Such simplistic AI would have allowed trivially easy control of what character a mob attacked making for dull and unchallenging PvE.
Under aggro systems Tanking classes were given special skills which move them up the mob priority list despite a low damage output from the Tank. In essence they break the otherwise effective AI of a mob and force it into inefficiently attacking the character most resistant to that attack. So long as mobs are balanced around this mechanic the gameplay is reasonably compelling mix of puzzle, timing and teamwork and it makes the Tanks role of being a meat-shield far more skillful and enjoyable then it would otherwise be if it was completely passive.
The problem is that these whole dynamic collapses when the opponent is not an AI but instead another player with full awareness of how the game works and full freedom to attack whomever they wish. And yet Tanking IS a viable class in most PvP games because Tanks are built differently in such games so as to make opposing players either rationally choose to attack the tank rather than another target or allow the Tank to create situations in which opponents have no other option but to attack them thus expending offensive potential inefficiently.
AoC would be far better off replicating Tank design from PvP games such as Overwatch and LoL and pairing them with mob AI's that seek to replicate the rational choices that a human would make in combat that cause them to attack Tanks. This will allow Tank gameplay to be consistent from PvE to PvP.
Well established PvP Tank mechanics include targeted pull effects, silence or other enemy DPS reductions, 'vaulting' into the middle of enemy groupings, increased damage potential when surrounded, projectile shields and movement barriers.
Some new methods which might be explored would be 'blinding' type effects on opponents which allow only nearby characters to be seen and attacked. An effect which leashes a melee opponent to remain within a short distance of the Tank. A damage redirection from nearby allies towards the Tank. I'm sure other skills can be created which provide control of enemy damage which would be effective on both players and AI.
Under aggro systems Tanking classes were given special skills which move them up the mob priority list despite a low damage output from the Tank. In essence they break the otherwise effective AI of a mob and force it into inefficiently attacking the character most resistant to that attack. So long as mobs are balanced around this mechanic the gameplay is reasonably compelling mix of puzzle, timing and teamwork and it makes the Tanks role of being a meat-shield far more skillful and enjoyable then it would otherwise be if it was completely passive.
The problem is that these whole dynamic collapses when the opponent is not an AI but instead another player with full awareness of how the game works and full freedom to attack whomever they wish. And yet Tanking IS a viable class in most PvP games because Tanks are built differently in such games so as to make opposing players either rationally choose to attack the tank rather than another target or allow the Tank to create situations in which opponents have no other option but to attack them thus expending offensive potential inefficiently.
AoC would be far better off replicating Tank design from PvP games such as Overwatch and LoL and pairing them with mob AI's that seek to replicate the rational choices that a human would make in combat that cause them to attack Tanks. This will allow Tank gameplay to be consistent from PvE to PvP.
Well established PvP Tank mechanics include targeted pull effects, silence or other enemy DPS reductions, 'vaulting' into the middle of enemy groupings, increased damage potential when surrounded, projectile shields and movement barriers.
Some new methods which might be explored would be 'blinding' type effects on opponents which allow only nearby characters to be seen and attacked. An effect which leashes a melee opponent to remain within a short distance of the Tank. A damage redirection from nearby allies towards the Tank. I'm sure other skills can be created which provide control of enemy damage which would be effective on both players and AI.
3
Comments
However, since it doesnt work in PvP, my thoughts are to turn that threat mechanic in PvP to a forced target instead.
Perhaps make it as simple as taunts working as a forced target status effect on the target, with the strength of the taunt acting as a countdown timer to determine the duration.
This is a much better solution that just dropping threat from PvE, imo. Since Intrepid seem fairly set on the idea of giving tanks a lot of CC (battlefield control is what they say), this seems to me like it would fit in just fine.
But I do not see any reason why the PvP style Tank abilities would not make for good active Tank gameplay, if anything it should be better because it's more positional and dynamic rather than just pushing yourself up a 1-dimensional threat ranking.
I'm a bit skeptical of trying to make the threat generation of a Tank effect a Player, first off, it's going to remove agency from the opponent is a very severe and frustrating way, even the most aggressive control mechanic I'm imagining give the opponent some counter-play options. Second an amount of threat from one target is only relevant with respect to another targets, aka to create a threat ranking to determine who to attack so all damage is going to need to be tabulated and managed and in a large battle that's going to be nigh unmanageable. Lastly a mob is going to attack someone every chance it gets and generally has unlimited attack potential, but a player, even if there forcibly tabbed over to the Tank can choose to not attack and reserve their mana for when they regain control, that could lead to both sides trying to wait each other out and make for excessively slow combat.
However, probably the biggest point to make, at least imo, is that threat dynamics in a good MMO are not at all one dimensional.
An example of this from a specific game I have played is in relation to hate generation not being static. Perhaps a mob hates fire more than most (treant, perhaps). On such a mob, attacks that deal fire damage generate two or even three times the hate. Or perhaps a mob is robotic in some manner, any healer using nature based heals generates much more hate against these mobs.
As a mechanic, threat/hate/aggro can go REALLY deep.
As to your concerns about players not having counterplays - in every game with deep hate mechanics that I have played, DPS and most healers have access to a de-taunt - an ability that lowers their hate towards the target.
This is a clear counterplay to a taunt.
As for differential hate generation from actions is still one dimensional because there is only one hate ranking list, don't confuse having many different +/- inputs to something with dimensionality.
NiKr, why is a taunted player about to die? Taunts aren't damage and under the proposed forced-tabing the taunted player is still free to retreat, receive heals and presumably use self-targeting abilities or items. If taunting a healer means they can't heal any allies then that's going to make taunts effectively a silence on them and would probably be OP.
In my experience of playing a tank with a pvp taunt, it was used in 3 scenarios. I was either saving my healer from a rogue that wanted to kill him (which then led to assist hits on that rogue to remove him) or I was aggroing the enemy dps to stop them from dpsing, which usually let my own rogues close in on that mage and remove him. The third scenario was just taunting the enemy healer to prevent them from using single target strong heals, but healers are always a target to die so if I was on the healer - he'd be seconds away from dying or at least from my party assist hitting him.
That was in a tab target game that moved your target to the tank every so often, if the aggro debuff was on you. We don't know how Intrepid will design their own taunts, so my experience might not apply here, but my main point was that the combat would not slow down, because the fight would now be 7v8, which requires those 7 players to output more. The same applies when a player dies and needs to be resurrected.
The problem with your suggestion is that it is still relying on a threat list - because that is what people do in PvP. You are constantly assessing and reassessing which target is the most important to attack at any point in time - and that is literally what a threat list a mob in PvE is doing. At any given point in time I have a player that I consider the most important to try and take out, and the player that is next most important, and usually a third after that. That is a threat list.
Threat lists as per PvE content exist as a means of emulating this.
The only difference here is that the mechanic tanks have to put themselves at the top of the list is a taunt. Other players are still assessed via damage output, healing, buffing, debuffing and CC.
At the very least, the point I made above proves that hate lists can be two dimensional, rather than always having to be one.
Having multiple hate lists is but one way of taking a one dimensional thing and making it two dimensional, but not the only way.
This is kind of like looking at something with forward and backwards movement and saying the only way to make it two dimensional is to add up and down movement as well - totally ignoring left and right movement.
That said, with the above suggestions, when you have a multi-mob encounter where different mobs within the same encounter have different modifiers to their hate list - yeah, it's kind of three dimensional because the fight contains multiple individual hate lists that are each functioning differently.
you aggro their dps so that they are hitting you and not your healer. if their dps stops attacking you during the taunt duration, that's fine, you are doing your job...they arent hitting your healer. your healer just gotta be careful with aoes
SWTOR for example has the taunt abilities reduce the damage output of the enemy for the duration. I know some games have used taunt abilities to force opponents to drop targets or even to allow only the tank to be targeted.
We wouldn't have to reinvent anything, just impliment pvp effects to existing abilities.
Be careful about too much CC attached to things though. That can get out of hand and frustrating pretty quickly.