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!
Options
Tab vs Action Combat Philosphy
beaushinkle
Member
This post should serve as a sort of myth-dispeller about action combat (BDO, New World, TERA) vs tab combat (World of Warcraft, FFXIV) in MMOs.
The short of it is that in action combat, hit validation is performed by comparing the area-of-effect of an ability to the position of relevant opponents. In tab-target combat, hit validation is performed by comparing the area-of-effect of an ability to the position of your target. That's it.
There's a lot of historical baggage associated with these two validation schemes, but at the end, this is all there is. Say, for example, you have an ability like frost nova in World of Warcraft (a tab target game). When you press frost nova, the game knows the radius of frost nova (10 yards), and then freezes everything within 10 yards of the mage. That's an "action" ability.
Similarly, in world of warcraft when a mage presses fire blast, it checks to see if the target is in range (40 yards), and then if it is, then they get fire blasted. That's a "tab target" ability.
Different tab target games handle their hit resolution differently. Some games may want you to also be facing your target. Some require that you have line-of-sight and introduce objects that block line of sight. Some have facing requirements (rogues can only backstab if they are facing their target, their target is within range, and their target is not facing them). Some have cast times, where you need to be within range and within line of sight at the beginning of the cast time and the end of the cast time. Some only require the beginning, some only require the end.
Likewise, action combat games also handle their hit resolution differently. Some games like Dark Souls create moving hitcubes and compare those to hurtcubes like how combat works in a 2d fighter, but in 3d. Other games use simpler distance measuring (like if all of your skills were frost nova).
Separate from these ideas, for historical or traditional reasons, these games also tend to emphasize different skills. I wrote a pretty long writeup about what difficulty actually is, but the TLDR is that when we interact with our mouse and keyboard difficulty only comes in three forms: figuring out the right button/mouse motion to press, pressing complex sequences correctly, and pressing those buttons/sequences with varying degrees of timing leniency.
Tab Target games tend to focus almost entirely on the "which button is the right button" part. The game tends to give you a bunch of information (buffs, cooldowns, debuffs, etc) and a bunch of different (say, 30+) abilities you could press, and make you puzzle out which one is the most important to press right now. Then, once you've chosen, you can't press another button for ~1.5+ seconds. They soften timing requirements by making it so that if you press a button early, it "buffers" the press and so that you can just spam the button you want to do next and not have to worry about timing it.
Action Combat games tend to have less buttons to worry about, and also make the puzzle about which button is the right button less difficult. They tend to have less buffs, cooldowns, debuffs, etc to keep track of. In exchange, they make you perform more sequences and make you care more about timing. For example, maybe if you're barely out of range, your attack will miss in an action combat game, whereas in a tab target game you just wouldn't be allowed to press the button yet. An action combat game might make it so that if two people attack each other at the same time, whichever one's attack comes out first interrupts the other (like a 2d fighting game). They might give players ways to avoid each other's attacks via some sort of dodge roll (like dark souls).
Important to keep in mind is that none of these traditions need to be carried on. Whether or not you have a target doesn't preclude dodge rolling or hit stun. Tab target games don't need to have an auto attack, even though many do.
The short of it is that in action combat, hit validation is performed by comparing the area-of-effect of an ability to the position of relevant opponents. In tab-target combat, hit validation is performed by comparing the area-of-effect of an ability to the position of your target. That's it.
There's a lot of historical baggage associated with these two validation schemes, but at the end, this is all there is. Say, for example, you have an ability like frost nova in World of Warcraft (a tab target game). When you press frost nova, the game knows the radius of frost nova (10 yards), and then freezes everything within 10 yards of the mage. That's an "action" ability.
Similarly, in world of warcraft when a mage presses fire blast, it checks to see if the target is in range (40 yards), and then if it is, then they get fire blasted. That's a "tab target" ability.
Different tab target games handle their hit resolution differently. Some games may want you to also be facing your target. Some require that you have line-of-sight and introduce objects that block line of sight. Some have facing requirements (rogues can only backstab if they are facing their target, their target is within range, and their target is not facing them). Some have cast times, where you need to be within range and within line of sight at the beginning of the cast time and the end of the cast time. Some only require the beginning, some only require the end.
Likewise, action combat games also handle their hit resolution differently. Some games like Dark Souls create moving hitcubes and compare those to hurtcubes like how combat works in a 2d fighter, but in 3d. Other games use simpler distance measuring (like if all of your skills were frost nova).
Separate from these ideas, for historical or traditional reasons, these games also tend to emphasize different skills. I wrote a pretty long writeup about what difficulty actually is, but the TLDR is that when we interact with our mouse and keyboard difficulty only comes in three forms: figuring out the right button/mouse motion to press, pressing complex sequences correctly, and pressing those buttons/sequences with varying degrees of timing leniency.
Tab Target games tend to focus almost entirely on the "which button is the right button" part. The game tends to give you a bunch of information (buffs, cooldowns, debuffs, etc) and a bunch of different (say, 30+) abilities you could press, and make you puzzle out which one is the most important to press right now. Then, once you've chosen, you can't press another button for ~1.5+ seconds. They soften timing requirements by making it so that if you press a button early, it "buffers" the press and so that you can just spam the button you want to do next and not have to worry about timing it.
Action Combat games tend to have less buttons to worry about, and also make the puzzle about which button is the right button less difficult. They tend to have less buffs, cooldowns, debuffs, etc to keep track of. In exchange, they make you perform more sequences and make you care more about timing. For example, maybe if you're barely out of range, your attack will miss in an action combat game, whereas in a tab target game you just wouldn't be allowed to press the button yet. An action combat game might make it so that if two people attack each other at the same time, whichever one's attack comes out first interrupts the other (like a 2d fighting game). They might give players ways to avoid each other's attacks via some sort of dodge roll (like dark souls).
Important to keep in mind is that none of these traditions need to be carried on. Whether or not you have a target doesn't preclude dodge rolling or hit stun. Tab target games don't need to have an auto attack, even though many do.
mmo design essays: http://beaushinkle.xyz/
1
Comments
Difficulty comes from challenge other player poses. I don't quite understand why you focus on this so more. Perhaps the correct term in the context you're describing would be accessibility. As you seem to be talking about input controls and how simple is for user to translate one's intentions into the game world.
The big difference between action combat vs tab combat is due to action combat having more variance in terms of how something can fail. This in turn increased the amplitude between skill floor and skill ceiling. As not something can fail on it's own, but usually opposing party can take counter measures before something lands on them, past the opponent triggered the ability.
In action games it's possible for one player to beat 5 from mechanical skill alone. That isn't the case in tab targeted games for the most part. And for me that the most important aspect of pvp I care about. I want player mastery to take central spot. Not the builds, grinds, gear progression or time sinked, which can still to some extent, but not as much.
You can even have both, tab targeted and "skill-shot" based. When tab targeted combat compliment skill shot moves. By being of lower risk / lower reward type of moves often used to help set up bigger plays for example. But that requires very good PvP design and balance, I don't think AoC currently have talent for to develop.
I would also strongly advocate to not look for pvp designs in other mmo's. As MMO's have shit pvp in terms of mastery from times immemorial and serves to facilitate conflict and social dynamics between player groups by sometimes them having to clash each other characters against each other in rather dull way. It's fun however when lack or requirement of mastery is obfuscated in huge number encounters it serves it's function. After all it's MMO. Perhaps pvp as a thing on it's own doesn't have to be good to do it's intended job.
Personally, I would like to see some innovation in this. But I doubt I will.
The challenge that other players pose shows up as stuff moving on your monitor and sounds coming out of your headphones. You respond to that information by pressing buttons on your keyboard and moving your mouse.
Other players can make it so that you have to solve harder puzzles, make your input sequences more tight and have less leniency for the timing of your key presses.
That's the big difference in many game's implementation of action combat and tab combat, not with the system itself. As I said in the very first line, the only actual difference is how hits are validated.
Again, this is in the implementation and not the system. It's fully possible to design tab-target games that allow you to outplay in 1vX fights, and it's fully possible to design action combat games that have very low mechanical skill caps.
Yeah, but opponents ability to pressure you to hit the buttons correctly is what will ultimately decide the difficulty. Not the method of input. Sure we can talk about theoretical. If inputs are not as accessible, your chance to meet a truly challenging opponent also decreases. The difficulty factor is relative, unless players compete on different input devices / schemes.
And that "only difference" makes and breaks entire ordeal.
Could you provide an example of real time wide skill floor / ceiling game based on tab target? Honestly, I'm not trying to "set up a trap" here. I'm really interested to see a game which actually managed to pull it off. Perhaps there is something worthy to learn from.
There are a lot of action combat based games which are trash too. Yes. Action combat is more difficult to design well. I'm not surprised that some developers would like to avoid it. And perhaps that's a correct call everything considered.
Debate solved. Bring on the beer.
That difficulty manifests as you seeing your opponent doing stuff (on your screen) that you have to respond to by pressing the correct buttons with the correct timing. Your opponent can typically give you more difficult puzzles, sequences, and timing requirements by correctly solving their own puzzles, sequences, and timing. For example, you have to play faster and more precisely in a fighting game against a player that is playing faster and more precisely. The final layer of difficulty is that you receive data on your screen that you need to respond to. That data can be player-created or game-created.
It doesn't! You can design tab-combat games that are highly interactive, filled with mindgames, execution tests, and quick reactions. See the bottom section of http://beaushinkle.xyz/posts/games-need-yomi
That said, developers for tab-combat games typically don't do this, and so tab gets a bad rep.
I don't think there is a good one =/
WoW 3v3 played at the rank-1 gladiator level like what magnusz explains here gets close on a solving-the-puzzle level, but lacks execution and yomi.
I would love to have MMOs be action based if the whole world had network infrastructure like South Korea. Unfortunately, MMOs targeting europe and the US have to account for west coast players playing with east coast players that have 70+ms connections to each other. They need to be able to put 40+ people like that on the same screen together. here's how that goes wrong
Ye in essence. But while you may encounter a such player, I'm more concerned how often you actually will. As it will form the bulk of what is experienced as difficulty for the player. Even if something can be difficult but barely ever happens, what does it matter?
A bit off topic, but game created difficulty in pvp isn't really a great thing. And it's more about sense of player agency less about the actual system. If player feels that one is fighting the game system more than the actual player on the other side of the screen things kind of break down regardless.
Tab targeted combat constricts player actions to a degree which also risks to create an setting where player feels unfairly constricted. It can also expand the possible actions within same domain, but generally, in my experience it's not really intuitive for people. And hence best served in turn based games where is plenty of room to figure out and think though puzzles. As well as allowing game to provide wide array of different inputs knowing that player will have adequate time to see them through.
I'm more than willing to entertain the possibility. However, hence why I also asked to provide an example. I haven't seen it done. And question is, did no-one though of it or there are some more to it. In example, not all game models translate well into real time combat video games.
Yeah.
Well, generally servers are locally distributed. You need like 3 datacenters in Europe to ensure that everyone here plays at around 50-60 ping. US could probably go away with only 2.
That and Interpid being very determined on providing good networking even in dense player environments. Which I think already proven that they can deliver higher quality in this regard than lets say Blizzard.
Probably the best bet for AoC is to start with rather simplistic pvp model and improve it through the years. And maybe it can go both direction. What I hope is that player mastery of the domain / pvp be rewarded. Sadly that's something we never had in MMO's. And at some point I'm left wondering, perhaps the game as a genre simply can't facilitate that.
In the abstract, all the time! It's more difficult to out dps a player doing their dps rotation correctly, and so by nature of them playing the game well, your task is more difficult. As for how interactive it is, that requires that your game be designed around enabling yomi.
I think there's a couple of things to think about here. The game-created difficulty can come in the form of executing your plan and how physically difficult that is to execute. For instance, in a game like street fighter, doing even a basic fireball as ryu on a keyboard requires that you press "s" down, then press "d" down, and then release "s" and then press "u". It's one of those sequences that I was referring to. It's the game that imposes that requirement, and if you mess the sequence up you don't get what you want. If you do it too slowly, you don't get what you want. MMOs typically don't have sequencing requirements, and I think that's a big missed opportunity.
The second thing to think about is that I'm not referring strictly to PvP. The game can provide these same sorts of input challenges in PvE, either by creating visual puzzles (figure out where to stand to complete raid mechanics), creating reaction tests, dps checks, etc.
Tab targeting is just about hit resolution. That's it! Your ability must have a target that you've pre-selected, and that's how the server determines who gets hit by the ability. You can impose any number of other requirements on top of that, like facing, range, line of sight, travel time, complex sequences, etc. Modern and historic implementations of tab targeting tend to be constricting, but that's not a fault of the hit resolution mechanism, that's the fault of lazy design.
By chance, did you watch the linked magnusz video? I've played *a lot* of games at the highest level over the years (traveled and won super smash bros melee and street fighter 4 tournaments, thousands of hours of tf2, counterstrike, valorant, quake, overwatch, grandmaster in rocket league, and starcraft 2, multi-gladiator in wow, went to MLG orlando for wow in 2008). Modern WoW rank-1 arena pvp has serious depth.
Yeah, so if the server is centrally located, and each player has a 50ms ping, then the information you see from the other player has to travel for 50ms to the server, and then from 50ms from the server to you, making any information you see about that player 100ms old. I really recommend reading http://beaushinkle.xyz/posts/combat-and-netcode and maybe https://www.reddit.com/r/newworldgame/comments/p2c353/after_nw_i_believe_all_future_mmos_must_have/h8kv96v/
Tab combat is easier to define as its basically the requirement of a target to attack or use TT skills.
Action combat is more complex, as there is the simpler type of action combat which only considers a 2 axis plane, the same type used for the implementation of action skills in TT games such as Around Yourself AoE(aura effects included), Frontal cones(and other shapes), Ground spot AoE(ground dots with duration included) and straight thin line skill shots, all those only considering your feets position in the 2 plane axis counting as your hit box. (which is the type i expect AoC to have as there will be quite alot of difference in character sizes).
Then there is the more complex 3 axis plane Action Combat with 3D Hitboxes and full body Hurtboxes which gives the potential to implement crouching or jumping to escape certain attacks/skills and 3D projectile skill shots(like bullets in an FPS)(Works better with character sizes standardized or atleast less discrepancies in sizes).
Aren't we all sinners?
Yeah. Exactly my point why tab targeted game which constricts it's variance in cooldown trading has low floor - ceiling difference. A bot can execute a rotation. Hence the game has to provide a way to disrupt the rotation and offer competition around it if it wants it in the game. Historically it's always been crap design. In early iterations of wow it was based on interrupts. Later on they moved more towards proc based / build up - expenditure rotations introducing more windows for limited player decisions how to vary a rotation.
I'm not saying it's not possible, I just haven't seen it done well.
Not sure what you mean by sequencing here exactly. By your given example it seems that you're referring to restriction of input. Making player harder for one's intentions to map unto the game. Which I find terrible idea as more challenge is put on fighting the system / input device than the actual players and counter acting opponents decisions.
If you mean by using abilities in sequence to get the escalating bonus rewards.. when FF XIV already does that. Fair to say, mastery of a job (class) is way more felt than lets say in WoW and I believe strongly due to these chained actions one has to perform. And how it plays out in whole raid position game.
I however wouldn't call it amazing either. And can't talk about PvP there. It's not really the focus of the game.
Yeah. How exactly not lazy tab target in real time pvp would look like? Which would also be rather intuitive. See action combat is intuitive by design, I press a button - direction -> character moves. It maps onto the muscle memory almost effortlessly. There is a short way between the intention of the player and the game mapping that intentions into the system. I don't think that obscuring that behind some sort of puzzles helps, while you may win on player expression through tab targeting, you may lose by increasing input complexity, forcing players to fight the game system rather the opponent.
Yeah. It's servable. And there are some good parts in it, obviously. Too bad it's still meh as a whole. Some major issues are not related to actual pvp are more to do with gear progression, boosting, pve leakage into pvp balance, no spec bans to counter-act meta imbalances etc. Poor focus on pvp design like 5 second stuns with burst damage on the table, lucky crit roll streaks.
The cooldown trading and forsight of what opponents can do is what an somewhat competent MOBA player will do. Not talking about upper echelons pro play where it's just 4d chess where people track opponents movement within fog of war based on lane priority or what not. So I'm not exactly impressed.
Obviously not every game has to be that and MMO's have more constricted space in terms of combat objectives. I would personally be very happy with what's shown in the video if those decisions the player explains could be reliably counted upon, minus the stupid point and click stuns with no counter-play. That's never the case though.
Frankly I'm not sure how are you going to balance disables in tab targeting. I mean you propose how it's just "hit registration", however there is a huge difference of hit registration which allows counter play or to design abilities facilitating counter play and not having enough room for it. And I would go as far to claim that it makes or breaks the game and entire combat design.
See for example, in games where disables are not avoidable they generally consume all tactics as seen in CRPG "skip a turn mechanics". So either you introduce saving throws (RNG bad for pvp) or have to provide some sort of defense.. question is how are you going to implement it in rather confined design space. Action combat allows for dodges. Simple, intuitive, works. You can even design abilities around positioning / set ups. Etc. While it's possible with tab target, like range, removing "skill-shot" severely limits the options.
Same way HP pool becomes very sensitive metric in tab targeted as well. Good example would be BFA vs Shadowlands in WoW. As players cannot mechanically dodge or at very least it doesn't have that huge of a role for staying power it also cuts into design space. You also cannot design mobile vs immobile specs as much other than general movement speed / gap closers.
So I don't know. What kind of tab targeted combat you can offer which will be both intuitive to use and allow enough design space to offer variable gameplay and different spec "identity". I imagine having to play a piano with ton of abilities to accommodate the variance which can be achieved with more action oriented combat and perhaps 5-6 spec actions / skills.
I think there are more important things which could be ironed out in pvp other than focusing how to shave 20ms from latency.
The way that I understand how using spells work in Tab Target Combat is like using the default lock on aim in GTA 5, except now you can only shoot your gun when an enemy is within range depending on the gun/spell used. Only the enemy being targeted can take damage unless the player uses an AOE like a rocket launcher. Also damage is consistent no matter where you shoot/cast on the enemy (disregarding gear/equipment). Missing a shot/spell is calculated by numbers and stats determined from gear and equipment, not reticle positioning.
The way that I understand how using spells works in Action Combat is like using a gun in Halo, in this case the Sniper. Shooting/casting spells does not require a target to be in range or locked onto in order to activate. Any enemy in front of the reticle can be damaged. Damage may vary depending on the body part being hit, like headshots doing more damage or insta-killing. Missing shots/spells is determined by reticle positioning, structures or enemies obscuring the intended target, and damage is then determined from gear and equipment.
Do my analogies seem about right here? Let me know what you guys think...
By chance, did you read the yomi article or the netcode article? I outline an example of non-lazy tab target interactive, yomi-enabled combat in both of those posts.
from yomi
Make it so that all three options have the same cast-bar name, about a ~1.0 second cast time, indistinguishable startup animations, and when one cast finishes it interacts with whatever the other player is currently casting. If both players attack each other, they clang and neither take damage. If both players crush, they both take damage. If both players parry, neither takes damage.
Boom! Now you have the foundation for a combat system that is yomi-enabled from the start.
I also expand on this in the netcode post
Precise spacing with latency is going to miss the mark, so don't build a game that relies on it. Instead, build for something that masks the latency (like tab targetting). Build the game around tab targetting concepts - short damage reduction windows on cooldowns, short burst damage windows on cooldowns, baiting enemies into poorly trading offensive for defensives. Add tab-targetting skill-based mechanics, like skill sequences that force players to have a high APM while moving. Add the ability to create/remove line of sight and create/close range gaps. Tax player's tab-targetting abilities with buttons that drop targeting, or create dummy targets (like illusions or totems). Add layers of bluffing/counter bluffing with reflection type abilities. Add reaction-based abilities like interrupts, and reward players for faking and landing those interrupts.
So this concept of "fighting the system/input device" is something other people call "mechanical skill". It would be easier to play Ryu in street fighter if all of the combos or abilities you wanted to do could be performed with a single button press, but that removes tons of the mechanical skill from the game. The difference between how good two different people are at pressing their dedicated fireball keybind with a 200ms buffer so that your next one starts casting immediately is mostly nonexistent. If folks had to press an input sequence, then now there's something to be better at than something else.
To use your moba example, better players are able to chase/kite more effectively than worse players because they're better at animation cancelling and knowing when their next attack is available intuitively. We say those players have "good mechanics", and we're fine with that! The game makes the intention (max dps as i kite away) mechanically difficult, and you're fighting against the input system to accomplish it (click away too early and you don't get your attack. click away too late and you lose distance).
You can implement iframes and dodges in tab target as well! Important to keep in mind: tab target is just about hit validation. You can give players an instant cast ability that gives them 500ms of cc immunity (which is essentially what a dodge roll is). WoW had this for a while via vanish tricking stuns / death coils / sheep as a rogue, but it was on a 3min cooldown. Tab target games are not required to use the "you are not in range" error. They can make it so that attacks whiff against your target if you use them from too far away. They can make it so that your cooldown/mana is used if you finish the cast against an out of range/out of LoS target. They can design abilities that let players create temporary objects that block LoS.
You can make it so that stuns are castable while you're not in range, and they'll be wasted if you misuse them. Now you can bait CC by making them think you'll be in range when they press it, and then dipping back out of range so it misses. There are plenty of things that a tab target game can do to spice up both pvp and pve interaction. We need to get away from the paradigm where two players run up to each other and then slap each other with their dps rotations and use their guaranteed, undodgable ccs until one wins.
By chance, did you read the netcode post? I don't think I talk about reducing latency at all. Rather, the bulk of the post and the included example are about how latency affects different aspects of play, but specifically where you're perceiving where your opponent is vs where they actually are, and how that impacts concepts like spacing, footsies, and whiff punishing (action combat concepts).
This is, more or less, exactly what I'm trying to dispel here. In a tab target vs action combat, the only difference is that one requires that you have a target and the other doesn't. You can add whatever validation on top of that that you want. So, if you want to do a Halo sniper rifle, you can either add that as action validation in a tab-target game (which it can totally support), or you can make it so that you have to have the person you're shooting targetted and in your crosshair.
Further, the idea of only being able to shoot your gun if your gun is within range (and the game blocks you from wasting bullets otherwise) is a common implementation pattern rather than requirement of the system.
You can just watch the first two minutes:
Also, you attributed a passage to me that was not from me.
I feel this is a bit fecetious.
If you cast a tab-targetted AoE - you do the same hit-scans as that of an Action hitbox.
Basically every ability in action combat is either a hitbox (equivalent to tab target AoE) or a projectile.
In general, yes, tab-targetting has 1 pre-selected target which makes things easier to compute.
But you can't say that's the be-all and end-all of the difference.
The debate between AC vs TT isn't about how to do hitscans - it's more a question of UX/UI. TT is far more efficient with target selection than AC, hands down.
in TT you explicitly scroll through predetermined targets
in AC you point a directional vector, and the game does target calculations from there
Even so, your point about auto-attacks still stands.
To me, auto's are like... eating PBnJ with or without the bread. Auto's being the bread that provide a basic layer for your abilities to dance on. It seems weird to eat peanutbutter and jelly on its own - not that you can't.
Nice find! This is more or less exactly the system I was describing, but in an action game instead of a tab-target game. It could easily be ambiguous melee casts instead of animations to make it suitable for tab targeting.
It really is just a question of hit registration. Do you make your attack a have an area of effect (like having melee be a cone), have be targetted with a range (like most tab games), or have a hurtbox and give enemies hitboxes, like in dark souls, new world, or 2d fighting games?
Server complexity wise, having target is the easiest to compute and will hold up the best (because they don't need to loop through every nearby player to check to see if they're in the aoe) in large fights. They only need to check to see if your target is in range. AoE holds up second best, and hitbox-based falls apart in big battles. See http://beaushinkle.xyz/posts/combat-and-netcode for more.
the AC vs TT debate is not focused on the mechanics of hit registration, it's concerning preference for UX/UI.
To clarify, you're being told, by testers and 'people who have been debating this on forums for months' that the aspects you're referring to are already covered, tested and verified in large scale battles, and function effectively at up to about 120ms ping (at last check).
The bulk of discussion on these forums relative to Action vs. Tab isn't 'can this be implemented', it is 'whether or not people will be happy with a given way it is implemented', and 90% of that is related to 'can I use this UI and still be effective'.
There have been a few complaints about 'not being able to hit anything due to lag' from what seems to be a minority, but we expect this because the two suggested methods for dealing with it haven't been showcased (possibly implemented) yet.
Might be arrogance/bias but I don't think we had anyone argue too strongly against the suggested methods, either, so we're moreso 'waiting on the next update' than 'concerned about the philosophy. That is what @maouw is trying to tell you.
EDIT: Reddit User profile data scrape completed, longtime Reddit user, poster, etc that appears to be 'late to forums'. Suggest: Wait for @beaushinkle to get up to speed on forum half of this discussion type. If they were already tracking forums, only then is this level of correction 'valid'.
I see tons of posts here, reddit, and elsewhere that seem to think that tab target games require auto attacks, don't allow for mechanical skill expression, lack yomi, etc. This exists as a sort of document to dispel these myths. If you want to start up a separate conversation about tab vs action UI/UX, feel free!
By chance, did you read the post? I'm referring to specifically to trying to precisely to outspace / whiff punish abilities, which is more-or-less a hallmark of 2d fighters and dark-souls style combat. If you can't do this because the game is either displaying outdated information about your opponent's position (and they'll hit you when it looks like they shouldn't; delay-based netcode), or you're getting rollback netcode visual artifacts (looks like rubberbanding), then it removes a lot of the nuance around spacing based skill expression.
The game doesn't have to care that this is happening, and people will still win and lose, it'll just be less interesting on a competitive level.
Let me clarify.
I too am a fighting game player. I am part of a team of them. I teach people to play them. I know what you are talking about and I have been explaining to others why it is the way it is and how we need to work around it, for a while now.
What I am telling you is, that in all that time of doing so, of debate, clarifying information for players, talking about these aspects, etc, we have not generally had a difficulty in discussion of how to implement Action Combat in ways that would help deal with it.
What we have had is effectively:
"Ok guys so after much discussion this is how most of us believe Intrepid should implement Action Combat to make the game skillful, challenging, and get around most of the obvious netcode problems."
"Most people agree one way or another eventually, a subset of people argue against the very idea of basic principles required to achieve it, disliking either Action Combat or 'having to do certain things'."
You do not need to convince anyone or possibly even explain what the model should be like. We have a model. We technically have three models.
Disagreements occur because some people are arguing against the very idea of things like 'soft lock targeting', 'multi hit combo systems', 'using WASD-attacks instead of split body movement to help the server and players see predictable hitboxes', 'projectile behaviour', 'hurtboxes on large monsters and behavioural interactions intended to enable Tab Target players'.
So it may be harder to get traction on your posts, since it's 'stuff that has been covered already'. The people you need to convince are not 'those who have concerns about implementation', it's 'those who don't care about yomi', 'those who don't care about netcode', 'those who don't want different attack buttons because it makes the game too Actiony'.
Ahhhhh, thanks a ton! How much headway have ya'll made in the getting people to care about yomi, getting people to care about netcode, and getting people to care about adding in mechanical difficulty?
That of course depends entirely on who you are talking to...
The last 'most basic Compilation' from the oldest thread was:
1. Sideways and forward animations for the Q/LMB to add variety and evasion ability, along with chase options, while either neutral Q/LMB or S+Q/LMB prevents any forward movement while attacking.
2. A Brace option that reduces damage and blocks CC unless the enemy has 'Overwhelm' on their ability. Some stamina cost here.
3. Possibly a small stamina cost for some versions of those directional Q abilities so we can have movement but not eternally.
4. Gated strings with different attack animations (we already have a combo attack like this, but some posters have asked for more variety)
5. Limitations on the usability of certain skills when the enemy is out of its attack cone, regardless of your Tab Target.
6. 'Sticky reticle' behaviour to enhance targeting, possibly assisting with the Tab Targeting itself, but not requiring it, to make the transitions between Tab and Action easier
7. Ability to dodge twice before Dodge Cooldown fully triggers (Dodge retains a stamina cost). E.g. Dodge once and it 'recharges your first dodge' but you have another you can always use. Dodge twice and now you can't dodge again until recharged.
8. Some telegraphs of the attack vector or cone of abilities (especially those that can be varied through augments)
9. A stamina cost tied to jumping (perhaps only when weapons drawn).
Recent further discussions have been based mainly around the fact that the Split Body Combat has some definite support, a concern due to its tendency to make the hit resolution more difficult relative to the Sideways (WASD) option. If you want to know how that one went (probably the most relevant to you) you can check that one here.
I'm sure you've seen much of this feedback since you track the game closely, but maybe not this half.
Other discussions recently have been more focused on hitbox/hurtbox interactions between mobs and players so that PvE isn't terrible while still maintaining PvP and even Tab Target PvP as viable. Ongoing.
So as of now, multiple people are in support of using more 'set movements' and brace/blocking to help with netcode issues, given that the current gameplay seems to be server authenticated (this may change, or may just have been implemented so well that we can't tell the issue, even in Sieges). Attack animations allowing you to proc things on the third or 4th hit which is itself a more committed attack, to allow for whiff punishes and similar.
Soft-lock/sticky Reticle, some arguments have been given for how to allow this to retain some RNG elements, others have spoken out against any RNG for Accuracy and Evasion, debate ongoing relative to iFrames, but these are minor. The current tilt is 'away from iFrames' which would also help a bit with the netcode concerns, and I personally have suggested 'factoring how precise one's targeting is relative to the reticle' to determine how much Accuracy and Evasion multipliers to use, basically 'fuzzy hurtboxes' in total implementation, to remove some of the 'hit or not hit' lag/rollback artifacts.
So overall, 'how it's going' is best understood by reading some or all of the 'Melee Combat Feedback Post 7/30 Livesetream' thread. Let us/Intrepid know where you stand.
Your willingness to improve without complaint when presented with a minor technical deficiency is admirable.
Neither of those talk about what I'm really trying to get at, which is dispelling common notions that you see even in this thread, like "The way that I understand how using spells work in Tab Target Combat is like using the default lock on aim in GTA 5, except now you can only shoot your gun when an enemy is within range depending on the gun/spell used."
I also didn't see any real form of community consensus around how to make sure that combat is intellectually stimulating (yomi, depth), and physically interesting (high mechanical ceiling), though I'm sure you know where to look better than I do.
I didn't figure you'd want to slog through all of this one. In either case, if that was the conclusion, I'm glad that I pointed you to read them instead of just asserting anything. I'm sure my own bias about 'what people understand or don't' came in.
The other part you're missing is the threads from the Alpha Test forum itself, which aren't all 'stuff non-testers can't see', but rather because they could be that, there are some. I don't think it will affect your final conclusion though.
So, please proceed, I'll hope for your efforts at consensus to be fruitful. In case it helps, here's yet another data compilation that you may be able to use to get a general idea of... something.
It seems like your stated goal was to compile/summarize what the community wants out of a combat system - any indication of progress on that front? Have any of the devs contacted you or engaged? if not, how do we get that part to happen?
Maybe we can reach out to some of the folks that are doing content creation or something and get some press.
As of the timing of the main conclusions of those threads, the general perception of players seemed to be 'it's Alpha One, let's wait and see the Combat Revamp'.
It is unclear whether or not the Combat Revamp was 'delivered' as Intrepid intended it. What we got was the 7/30 Livestream related Split Body Combat option, hence the reference to it and my personal concern over the test and the 'strong support' the change received.
As of now, I personally am waiting for the next Development Livestream on this Friday to see what, if any modifications have been made to Combat. By the nature of the 'stance I took during Compilation' I neither expect nor 'want' any direct Dev contact, as the topic is still 'sensitive' and it's easier for Intrepid when people aren't complaining directly at them about every little aspect.
Therefore, if I may suggest, please add anything you have in mind, and wait until Friday. The reasoning is simply as mentioned before. Those who explicitly want Action Combat have managed to pare down most of their disagreements to something sort-of workable, (with the Split body vs Root Motion being the big issue now), so any discussion against the main models tends to be 'I don't want to need to play Action'.
The Content Creator scene is something I don't have full view of yet, due to time. The ones I'm aware of tend to be either 'don't care', 'play Mage', 'think Intrepid should maybe give up on Action Combat', but I only know about 8 of them in terms of 'watched their content seriously', so far.
The current flow for Action/Hybrid Combat fans is just 'design and clarify as much as we possibly can' just in case Intrepid needs the feedback or 'help' in terms of definitions, but as you saw from the Combat Discussion thread, there's a lot of information that needs to be covered for players. Your links do a lot of this, it's just that you haven't got much of a response yet from many 'people who don't like the model, don't fully understand, or don't agree'.
One could carry on therefore assuming 'well that means if they implement it, it should be fine', but as noted, there are other threads you can't see.
As for the other aspect I forgot to clarify (this might be repetition), we've tested exactly how most things work in the Alpha-1 client, so, for example, projectile based skills act as Projectiles even when Tab Targeted, leading to the somewhat funny situation of a Light Spear attempting to home in on a target and ending up circling them due to the usual arc resolution issue.
Skills work in Action Targeting mode fairly naturally, but attack cones are not all tuned yet. So, given that, is there anything you'd want to note for this thread? I believe we've been holding off on 'discussion of targeting' because Melee attack cones are currently 180+ degrees in most cases, and any changes of these will constitute a significant change that will likely cause another debate.
I'd assume that your position would be 'no, that cone is too big, it doesn't allow for outplay or mind games', but I have learned not to assume anything.
I was thinking it would be nice to have one of the more journalism-style content creators essentially interview us, and we can talk about what makes for meaningful combat, how raid design / challenge works (sort of like when you were describing your rathalos encounter), the differences between starfox and touhou (from your analogy), etc. That, of course, requires that they're interested!
I think that cone would be too big in a korean game (where they can design for 3ms pings), but for a game optimized for US and EU, they can keep the 180 cone and just give us other ways to interact. In fighting games, we get yomi situations out of frame advantage, and hitboxes that are unreactably fast. We can freeze-frame two characters at two different spacings and be like "these are all of Sol's options and these are all of May's options, and this is the matrix of how they interact".
In laggy MMO's that visual information is always super inaccurate, and so it gets extremely difficult to have tight spacing battles like you have in 2d fighters. Rather than try to re-create that, I'd rather them just try to figure out different ways to have yomi and mechanical skill expression other than tight spacing and frame advantage (like the attack/focus/parry system in naraka that Ugoogee linked).