Fully agree with everything in that video. When you try to translate it directly to MMOs, you run into netcode problems, unfortunately. Instead, I think it's good to try to re-imagine which parts of single-player action combat are actually satisfying, and how we can port those over to still work well with latency. Do we need precise spacing, cones, startup frames, hitstop, and dodging, or is that a way to model offensive/defensive flow, tempo, proper tool selection, mindgames, and reaction tests?
Precise Spacing: Needed, doesn't matter a lot, but people will absolutely complain, and the game has to build it in anyway.
Attack Cones: Already part of Ashes' main design, no loss in keeping it, impossible to do their intended 'weapon cleaving' without it. Debatable whether we need narrow ones, but here's even more data you might not have.
Hitstop: People seem generally opposed, I don't think I've yet 'explained to someone what the problem with it would be' and they still wanted it, outside of 'big moves occasionally stagger me a little'. I have been treating it as 'around the same amount of time as the recovery animation for the move that does it'. So that it only becomes an offensive mechanic in a coordinated group, and even then is rare. If you mean just the '3-5 seconds animation freeze on the attacker's side on impact', ignore this.
Dodging: Already in, at least a bit. I thought that without it, most people would be upset, and that's where we hit the most confusing part. Lots of people dislike being rooted. I thought this was because they wanted to dodge more attacks. Indication is 'no', that's not the core reason. Many don't care. Those that do care, explicitly care about not only dodging, but attack cones AND the idea that enemies don't target like Tab Target.
I generally feel like dynamism within Ashes' current combat system is lost without dodging, and found no calculable flaws to point out in the 'we should be able to dodge twice but less far' suggestion that was integrated into the Compilation'. Ashes needs 1x 'unlimited no iFrame dodge with a longish recovery' (current ground movement speed makes simply strafing insufficient against 80% of current attacks and abilities, which fluctuates upward with the size of the attack cones) and at least 1 more limited dodge. 2x works on the current Dodge timer without breaking things, and prevents '2v2 staggered burstdown'.
Generally, from all our ingame tests, Ashes' design has these as requirements or they need to overhaul the ability sets of all current classes and meaningfully rethink a considerable number of systems and intentions. Their intent is to have hotbar ability lists be 'manageable' (probably 20 abilities total, also presuming some of those slots will have items in them) and don't intend for every class to be able to learn every ability without a downside (or possibly at all).
Unless these change, most of the requirements you listed have the prerequisites. With the current game flow and speed, even if they rip out Action Combat as a whole, they will still have these problems.
Unfortunately what you would need is a video on 'how netcode works in games'.
Games use different netcode models, and have different ones based on the types of games they are. Some aim for fairness between people of differing ping, some aim for server consistency, some aim for satisfying interactions that work to ensure the client isn't confusing for the player.
These three implementations don't overlap much in the US/EU. In Korea or some parts of Japan, the ping is by nature smaller. Geography matters. Let's assume Ashes 'plans' to try to give us more geographically consistent servers than most games, and since it is already a Western game, we don't have to worry about the usual issues that old MMOs faced on that front.
This implies their optimal model is the last one. Loss of fairness is unfortunate, but surmountable in different ways and the game offers many activities. PvP or not, you can avoid this or just accept it, like... literally anyone who tries to play competitive games on a laggy connection/far from their server, already does.
Server consistency would be a good aim for Ashes if it was going for Action Combat entirely, but Tab or Hybrid with telegraphs doesn't benefit much from this. You already have lots of leniency in terms of what people expect. 'Spend it'.
So that implies the third model is optimal.
This one is done by making it so that the 'startup frames' can be clipped out of certain attacks, but a visual indicator unrelated to the actual animation is required from frame 1. Players who have really good ping have a higher chance of dodging correctly, but you just put the mindgames behind that.
"Make the first move always reactable on 50-80ms ping, but never devastating."
"Make the second move much harder to deal with if you dodged the first."
"Make the player not need to focus as much on the mechanical difficulty of the dodge so that they can focus on the opponent to watch for the second move."
"Give the attacking player options after their first move that put the defender at positional disadvantage if they are watching too hard."
Example:
Mage lines up and throws a Fireball. Their hands flicker an obvious flaming orange from frame 1. You could dodge this fireball. It starts in 5 frames. They got leniency from their soft lock, so if you were in the wrong position on their screen, you're still in danger. Ashes' movement speed already supports this.
If your ping is 80ms, the fireball is 'already moving toward you' by the time you see the flames. So what? Still do something. If your ping is 400ms, call your ISP, but you can just 'not dodge the fireball'. Because you know that the mage is now in a Fireball Recovery animation. And you know where they are. You can attack instead. Suboptimal, maybe, but you still get to play.
Your client can skip some number of frames in the Fireball animation based on the timestamp data it gets from the server. As long as those hands keep flaming for a while, you're still in control.
A player with 50ms ping could easily dodge this Fireball. The key is to make a game where that isn't always the correct thing to do. The Mage needs to be able to 'know that your dodge will probably take you to a certain location', 'watch for the startup of the movement in the direction and know that if they see the startup it won't be rolled back', and 'target their own next ability or movement toward their desired location'.
This is the reason for the Root Motion conclusion in the Compilation. The Mage sees you moving, they can't see exactly where you are facing or where you will dodge to because of latency. They prime to be ready when you lunge or strafe dodge and set their AoE to approximately that location.
Then they see you move. Human reaction time of 240ms means that the dodge itself must have 15 frames of animation recovery. Ping values above 240ms mean the player has to guess a bit since they only see your movement's start 15f late.
So 15f +15f on their side, if the dodge is 34+, they can hit you.
On your end, you could have 240ms ping. That means that you see this as 'I hit dodge', and 15f later the server recognizes that you did that. But the server can 'catch up to you'. It now has 'a clear direction', it can just skip 15f on its side due to timestamps. The server 'puts you where you should be' and then sends the data to your opponent. The server doesn't have to care as much 'what you do when the animation is over' because your animation is going to be another 19f.
So you, after your dodge, decide to strafe back in the direction you came.
If both players have 240ms ping, your opponent 'saw you blip forward' (the netcode difficulty here is determining how much of an animation to skip since which frames are important, must be built into the animation model, which is a processing thing).
On their side, by the time the server info reaches them, it basically looks like:
"Dodge Start Position: Coords, Dodge Start Timestamp: X, Animation Frame Reached: Y, Vector: V3D"
You don't strictly need the animation frame reached, that's the server working to help the client, that load can go wherever.
The Mage's client shows maybe 5 frames of 'player at the Dodge start Position', skips some number of frames, and then we get really close, but it still works. The thing is, the Mage doesn't care. They saw a startup, and if they could interpret it (or not, and guessed well), they targeted the spot where the player will be. They send back 'target location', their own start timestamp, and off we go back to the server, another 15f journey.
But the defending player didn't get out of that AoE in 4 frames by walking. They didn't get out of it by just 'strafing back the other way'.
On their side, they tried. The server cuts 15f off the Mage's AoE animation. The defender's client another 15f. But the key thing that makes this work is that the Mage got to react early enough so that when this all plays out, since that player was 'waiting 19f after the first dodge to be able to strafe', in a specific spot, they know that when they see 'the mage AoE descending on that spot', it's less strange when they get hit despite being a few frames out of the AoE visibility point.
If this isn't acceptable enough even with 15f ping to server on both players, almost nothing will be, short of 'Ashes of Creation releases in Korea'.
A final note for the unfamiliar. Your Ping is based moreso on your ISP and network infrastructure than your distance, in a game with regional servers. Dallas to Boston is 40-50ms (3 frames) along fiber optics. Generally you double that for your expectations (lots of other stuff beyond just the fiber) and 2.5x if rurality starts to factor in (depends on ISP).
@beaushinkle has various links suggesting (and I'll quote from it):
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.
Unfortunately from my perspective, these things are slightly too vague, go against what we've played so far, and multiple parts of the game's Action Combat model, just don't work.
"They give up too much of what an Action Combat game would feel like, for no reason."
It's frustrating to have a skillshot miss some percentage of the time. But 'not having them because of this' is only 'succeeding at not frustrating people' because those people just don't play.
So I've finally had enough time to write this up as a full 'disagreement' with the quoted concept. Tab Targeting has too many limits for a PvP system using the base Ashes has. I finally believe that if Ashes does not implement Hybrid Combat, given their current ideas, they might as well not implement PvP at all with any expectation of competition or skill mattering.
I don't mind PvP that is just the 'is my gear better than yours, am I capable of reading my ability cooldowns and active status effect icons?' in a game where Gearing up is the primary treadmill (i.e. a Themepark) but a soft-power game like Ashes probably needs more agency than that. For the first time, I'm going to say this.
I think Ashes of Creation actually has to have Hybrid Combat, given what we've seen so far.
I played the new world beta. It starts to feel non-competitively laggy after about 15v15. The sieges (50v50) are a complete cluster. They're in the unenviable position of having to support precise moving hitboxes and hurtboxes, which is a lot more computationally complex than aoes/cones/radiuses.
Unfortunately from my perspective, these things are slightly too vague, go against what we've played so far, and multiple parts of the game's Action Combat model, just don't work.
"They give up too much of what an Action Combat game would feel like, for no reason."
It's frustrating to have a skillshot miss some percentage of the time. But 'not having them because of this' is only 'succeeding at not frustrating people' because those people just don't play.
Really what the above situation is trying to accomplish is that you have a low-reward move (fireball) that can bait an option and now the players need to pick between the different options in the small RPS game that each move to different gamestates themselves. This creates interactivity, and I think that's what players crave, not the laggy action combat itself.
Which things were too vague? Which go against what you've played so far or just don't work? Why wouldn't a rogue, for example, be able to trick a tab-combat system by making people target illusions on accident. Why couldn't a Mage put up a wall-of-stone that blocks line of sight?
All of that said, I'm not at all married to tab combat. I just think it's easier to create more physically and intellectually interesting pvp and pve when you don't rely on precise spacing in EU/US network conditions. If your version of precise spacing is stuff like aiming ground targetted AoEs then we're in agreement - those are probably fine. If your version of precise spacing is stuff like dipping into the range of marth's sword to bait him to swing and then dipping back out of range to make him whiff and then dipping back in to punish, then I think we disagree.
So I've finally had enough time to write this up as a full 'disagreement' with the quoted concept. Tab Targeting has too many limits for a PvP system using the base Ashes has. I finally believe that if Ashes does not implement Hybrid Combat, given their current ideas, they might as well not implement PvP at all with any expectation of competition or skill mattering.
I don't mind PvP that is just the 'is my gear better than yours, am I capable of reading my ability cooldowns and active status effect icons?' in a game where Gearing up is the primary treadmill (i.e. a Themepark) but a soft-power game like Ashes probably needs more agency than that. For the first time, I'm going to say this.
I think Ashes of Creation actually has to have Hybrid Combat, given what we've seen so far.
I fully and completely agree that for ashes to work, there needs to be agency. I'm flabbergasted when people seem to be okay with winning or losing PvP with gear/materials staked based on counter-comping and spreadsheet-vs-spreadsheet rather than skill expression.
Why do you think that Hybrid Combat is required for the game to have that amount of agency?
Unfortunately from my perspective, these things are slightly too vague, go against what we've played so far, and multiple parts of the game's Action Combat model, just don't work.
"They give up too much of what an Action Combat game would feel like, for no reason."
It's frustrating to have a skillshot miss some percentage of the time. But 'not having them because of this' is only 'succeeding at not frustrating people' because those people just don't play.
Really what the above situation is trying to accomplish is that you have a low-reward move (fireball) that can bait an option and now the players need to pick between the different options in the small RPS game that each move to different gamestates themselves. This creates interactivity, and I think that's what players crave, not the laggy action combat itself.
Which things were too vague? Which go against what you've played so far or just don't work? Why wouldn't a rogue, for example, be able to trick a tab-combat system by making people target illusions on accident. Why couldn't a Mage put up a wall-of-stone that blocks line of sight?
All of that said, I'm not at all married to tab combat. I just think it's easier to create more physically and intellectually interesting pvp and pve when you don't rely on precise spacing in EU/US network conditions. If your version of precise spacing is stuff like aiming ground targetted AoEs then we're in agreement - those are probably fine. If your version of precise spacing is stuff like dipping into the range of marth's sword to bait him to swing and then dipping back out of range to make him whiff and then dipping back in to punish, then I think we disagree.
So I've finally had enough time to write this up as a full 'disagreement' with the quoted concept. Tab Targeting has too many limits for a PvP system using the base Ashes has. I finally believe that if Ashes does not implement Hybrid Combat, given their current ideas, they might as well not implement PvP at all with any expectation of competition or skill mattering.
I don't mind PvP that is just the 'is my gear better than yours, am I capable of reading my ability cooldowns and active status effect icons?' in a game where Gearing up is the primary treadmill (i.e. a Themepark) but a soft-power game like Ashes probably needs more agency than that. For the first time, I'm going to say this.
I think Ashes of Creation actually has to have Hybrid Combat, given what we've seen so far.
I fully and completely agree that for ashes to work, there needs to be agency. I'm flabbergasted when people seem to be okay with winning or losing PvP with gear/materials staked based on counter-comping and spreadsheet-vs-spreadsheet rather than skill expression.
Why do you think that Hybrid Combat is required for the game to have that amount of agency?
The short answer is 'I played it', but obviously that's not good enough, and I know that.
I'm completely in sync with you about the idea that 'extremely precise targeting of any kind' is not a good plan. The thing I perceive as 'agency' is the precise result of two interactions.
The first is 'that movement is necessary to either fake out an opponent or dodge meaningful attacks'. Not perfectly, not precisely. Just that you generally should have a good reason to consider it. The reason for this is that Ashes intends to have Summoners and those Summoners will have pets that act like Mobs. I'll elaborate on this more if you wish it (even more than I do later in this response), but for now...
The second is that making any movement contains a disadvantage relative to your own ability to launch a counterattack, which you must overcome with your own skill, or in some cases, cannot overcome because it is too large.
So in the example of Marth's sword, we come back to a point where Naraka and Vindictus show us how this works. You move into a position, your opponent makes a melee attack that should hit this position and fails.
In a fighting game, there is hitstop, so a whiff punish happens and you lose control of the situation. In Ashes, there is no hitstop or stagger. So your opponent doesn't care if their first attack whiffs, they only care that their flow into their second attack is correctly targeted. You, in turn, don't actually care if their first attack hit you or not other than 'well it reduced damage'. A Summoner doesn't even get this much. They had to 'position their summons correctly' and hope, because the Summon will generally 'keep attacking'.
For Summoners to be effective, a player's choice has to be 'I can dodge, or I can attack, but usually not both'. If they dodge forever, fine, it distracts the summon, but just drags things out. If they attack, they get hit some amount of the time, and the summon's power gets to work. A Summon generally can't 'decide if to dodge AoE or not', and if they were given this ability, even as a stance, then the speed at which this happens means that now you have just one Archetype with latency issues (technically two because the same thing will happen to Bards).
The problem with any Tab like combat is the stuff we have now.
The Tank ability Javelin, does a pull-in effect on hit. Even if it doesn't have its short CC, this is a big deal. In Tab Target mode, the effectiveness of this move is based on 'can I get line of sight to the target' and 'Will my Acc beat their Evasion'?
Javelin isn't about baiting. Neither is Cleric's Judgement at high level (a relatively strong Movement Speed Down). These abilities need to be Action or Action-lite with a soft lock, or what happens is not 'agency', it is 'My opponent pressed a button to change my agency and I watched to see if it happened'.
If one argues that 'the opponent can react', then you're back to a level of 'netcode' where you might as well have gone all in. I can't use an ability or block in time against Javelin if I couldn't dodge it. Fireball is the example I used because Mages have an obvious "AoE that matters" for the rest of the consideration.
The style of effects, the movement speed, the concept of how melee and procs will work, all of these things would need to be rethought if Hybrid Combat were removed or stripped down to 'ground AoE markers'. And even those fall into the same problem the smaller they get. If melee matters but there is no stagger, then in a fantasy game, magic and similar abilities will have strong effects or the game's (arguably) 4 ability/non-melee based classes are going to be in a bad place. Nothing worse than 'my strong moves are dodgeable AoE but my opponent just melees me down'.
The first is 'that movement is necessary to either fake out an opponent or dodge meaningful attacks'. Not perfectly, not precisely. Just that you generally should have a good reason to consider it.
In a fighting game, there is hitstop, so a whiff punish happens and you lose control of the situation. In Ashes, there is no hitstop or stagger. So your opponent doesn't care if their first attack whiffs, they only care that their flow into their second attack is correctly targeted.
So make them care without stunning them! Make it so that attacking someone applies a "weakened' debuff that you can clear by using a different ability. Your attack doesn't apply 'weakened' if you're already 'weakened'. Now both people still have control over their characters and there's no hitstop, but whiff pushing creates a situation where there is an attacker (the person not weakened) and a defender (the person weakened). It doesn't have to be this specifically, but something like this, that creates a flow of combat.
The Tank ability Javelin, does a pull-in effect on hit. Even if it doesn't have its short CC, this is a big deal. In Tab Target mode, the effectiveness of this move is based on 'can I get line of sight to the target' and 'Will my Acc beat their Evasion'?
Making javelin tab-target point-and-click is bad. Making it have a miss% based on character stats is worse. Instead, you can make the javelin ability have a wind up animation, and then give players the ability to avoid it. Maybe you can dodge-immune the javelin. To provide layer-2 counterplay, the tank needs to have the ability to feint the javelin, either by winding up and canceling, or winding up and holding the button down to beat players that are trying to dodge-immune it.
The first is 'that movement is necessary to either fake out an opponent or dodge meaningful attacks'. Not perfectly, not precisely. Just that you generally should have a good reason to consider it.
In a fighting game, there is hitstop, so a whiff punish happens and you lose control of the situation. In Ashes, there is no hitstop or stagger. So your opponent doesn't care if their first attack whiffs, they only care that their flow into their second attack is correctly targeted.
So make them care without stunning them! Make it so that attacking someone applies a "weakened' debuff that you can clear by using a different ability. Your attack doesn't apply 'weakened' if you're already 'weakened'. Now both people still have control over their characters and there's no hitstop, but whiff pushing creates a situation where there is an attacker (the person not weakened) and a defender (the person weakened). It doesn't have to be this specifically, but something like this, that creates a flow of combat.
The Tank ability Javelin, does a pull-in effect on hit. Even if it doesn't have its short CC, this is a big deal. In Tab Target mode, the effectiveness of this move is based on 'can I get line of sight to the target' and 'Will my Acc beat their Evasion'?
Making javelin tab-target point-and-click is bad. Making it have a miss% based on character stats is worse. Instead, you can make the javelin ability have a wind up animation, and then give players the ability to avoid it. Maybe you can dodge-immune the javelin. To provide layer-2 counterplay, the tank needs to have the ability to feint the javelin, either by winding up and canceling, or winding up and holding the button down to beat players that are trying to dodge-immune it.
If you would like to engage in the debate on either 'stat based miss chances on abilities' or 'Javelin shouldn't be tab target at all', I can point you at the threads, but note that the opponents of both these concepts have directly quoted the Creative Director's strong implications that these are the intention, specifically as support of their own wish for it to be so.
I have made only overarching/compromise based arguments 'against' those positions. If your stance is stronger, I'm sure they can hear you out.
As for the aspect of making people care about being hit, I don't personally see how this helps if you're not going to be dodging particularly well in the first place, nor was I even trying to imply that there should be a change. My suggestion is based on Vindictus and on combos.
You don't want to get hit by the telegraphed final hit of your opponent's combo because it debuffs you or benefits them in ways I mentioned here, which are based on the concept of weapon procs, as mentioned in that comment.
My overall stance is that Ashes' design already supports everything they intend, and only iterative design and tuning is required now. Any points I make come from 'where they are already', since that's my 'chosen role'. You'll need to talk to some other people to get the more specific 'Acc vs Eva' debates, since 'removing RNG from interactions' is not currently confirmed to be on the table for this game's combat design.
Sorry, my native language is Erlang.
0
Options
DygzMember, Braver of Worlds, Kickstarter, Alpha One
The Tank ability Javelin, does a pull-in effect on hit. Even if it doesn't have its short CC, this is a big deal. In Tab Target mode, the effectiveness of this move is based on 'can I get line of sight to the target' and 'Will my Acc beat their Evasion'?
Javelin isn't about baiting. Neither is Cleric's Judgement at high level (a relatively strong Movement Speed Down). These abilities need to be Action or Action-lite with a soft lock, or what happens is not 'agency', it is 'My opponent pressed a button to change my agency and I watched to see if it happened'.
I disagree.
I should be able to build my character to resist Javelin... which is what I did in Alpha One, and it worked great.
In the Alpha One, I often could not see the prep for Javelin.
If it were Action, Javelin would need a recognizable telegraph along the lines of the original Fireball/Hallowed Ground/Exorcism, so that opponents have the opportunity to attempt a Dodge.
The Tank ability Javelin, does a pull-in effect on hit. Even if it doesn't have its short CC, this is a big deal. In Tab Target mode, the effectiveness of this move is based on 'can I get line of sight to the target' and 'Will my Acc beat their Evasion'?
Javelin isn't about baiting. Neither is Cleric's Judgement at high level (a relatively strong Movement Speed Down). These abilities need to be Action or Action-lite with a soft lock, or what happens is not 'agency', it is 'My opponent pressed a button to change my agency and I watched to see if it happened'.
I disagree.
I should be able to build my character to resist Javelin... which is what I did in Alpha One, and it worked great.
In the Alpha One, I often could not see the prep for Javelin.
If it were Action, Javelin would need a recognizable telegraph along the lines of the original Fireball/Hallowed Ground/Exorcism, so that opponents have the opportunity to attempt a Dodge.
I was trying to figure out a way to summon you without literally giving in to my urge to go 'Dygz, I choose you!' and quoting one of those lines you always react to.
Success!
I have no opinion on your post one way or the other, but I bet @beaushinkle does. I hope it's 'agreement'.
As for the aspect of making people care about being hit, I don't personally see how this helps if you're not going to be dodging particularly well in the first place, nor was I even trying to imply that there should be a change. My suggestion is based on Vindictus and on combos.
Yeah, there would be need to be a long enough tell for the startup of javelin throw and a short enough recovery from your normal non-commital options that if you're paying attention you can outplay. You know all this though, so I won't preach to the choir.
The final hit system is interesting because you're able to effectively extend the startup frames of the "real" threat by a massive amount and use that to hide rollbacks. I like it!
My overall stance is that Ashes' design already supports everything they intend, and only iterative design and tuning is required now. Any points I make come from 'where they are already', since that's my 'chosen role'. You'll need to talk to some other people to get the more specific 'Acc vs Eva' debates, since 'removing RNG from interactions' is not currently confirmed to be on the table for this game's combat design.
I disagree.
I should be able to build my character to resist Javelin... which is what I did in Alpha One, and it worked great.
In the Alpha One, I often could not see the prep for Javelin.
If it were Action, Javelin would need a recognizable telegraph along the lines of the original Fireball/Hallowed Ground/Exorcism, so that opponents have the opportunity to attempt a Dodge.
Yeah, if you want it to be action, it would need to have a telegraph so that it can be dodged. If you want it to have a %chance of not missing, what gameplay does that create?
Players are still going to javelin you when that's their best option, and now one of two things happens:
1) you resist it. they're at relative disadvantage because of circumstances outside of their control, and potentially the outcome of the fight is heavily determined by you happening to resist the javelin.
2) you don't resist it. you're at relative disadvantage because of circumstances outside of your control, and potentially the outcome of the fight is heavily determined by you happening to not resist the javelin.
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net. This same exact situation used to come up in vanilla wow / burning crusade where orcs had a higher chance to resist stuns and that would straight up decide fights.
If I remember correctly I think Intrepid has some talents in their development team that have come from Planetside 2 to help with server stability and netcode. I know this doesn't necessarily fix all of the issues that we have discussed, but I thought it would be nice to mention in case if it helps or impedes certain game mechanics/designs.
0
Options
DygzMember, Braver of Worlds, Kickstarter, Alpha One
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net. This same exact situation used to come up in vanilla wow / burning crusade where orcs had a higher chance to resist stuns and that would straight up decide fights.
Ashes is an RPG, which means I should be able to build my character to counter your character.
It's significantly different than player twitch skill v player twitch skill - which is the purview of other genres.
In an RPG, if you want to resist Stuns, you should choose a race and a class that helps you resist Stuns.
Your opponents should be building characters to maximize Stuns - if they want to maximize Stuns.
Also, keep in mind that Ashes PvP combat is balanced for 8v8 rather than 1v1.
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net. This same exact situation used to come up in vanilla wow / burning crusade where orcs had a higher chance to resist stuns and that would straight up decide fights.
Ashes is an RPG, which means I should be able to build my character to counter your character.
It's significantly different than player twitch skill v player twitch skill - which is the purview of other genres.
In an RPG, if you want to resist Stuns, you should choose a race and a class that helps you resist Stuns.
Your opponents should be building characters to maximize Stuns - if they want to maximize Stuns.
Also, keep in mind that Ashes PvP combat is balanced for 8v8 rather than 1v1.
In an effort to find some common ground, you surely believe that you need to pilot your character, right? As in, you presumably don't want to create a character and then press the "fight" button and see if the character (represented by their stats, level, gear, etc), not the player behind the character (represented by their decision making, mechanical skill, etc) wins a fight.
If you want for player skill to be some part of the outcome, then we're back to why you why you want javelin specifically to be determined by RNG, since the only answer you gave me was a blanket statement.
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net. This same exact situation used to come up in vanilla wow / burning crusade where orcs had a higher chance to resist stuns and that would straight up decide fights.
Ashes is an RPG, which means I should be able to build my character to counter your character.
It's significantly different than player twitch skill v player twitch skill - which is the purview of other genres.
In an RPG, if you want to resist Stuns, you should choose a race and a class that helps you resist Stuns.
Your opponents should be building characters to maximize Stuns - if they want to maximize Stuns.
Also, keep in mind that Ashes PvP combat is balanced for 8v8 rather than 1v1.
In an effort to find some common ground, you surely believe that you need to pilot your character, right? As in, you presumably don't want to create a character and then press the "fight" button and see if the character (represented by their stats, level, gear, etc), not the player behind the character (represented by their decision making, mechanical skill, etc) wins a fight.
If you want for player skill to be some part of the outcome, then we're back to why you why you want javelin specifically to be determined by RNG, since the only answer you gave me was a blanket statement.
This is @Dygz consistent stance across multiple discussions, whether it be 'should Evasion and Acc decide hit chance in Action based attacks' to the linked thread.
It may be best to not discuss that topic in this thread, presuming you prefer to keep this one on a particular track for any post-Livestream feedback.
There'll be a multitude of reasonings in the link provided.
Sorry, my native language is Erlang.
1
Options
DygzMember, Braver of Worlds, Kickstarter, Alpha One
edited August 2021
I am an RPG player, so I don't necessarily want player twitch skill to be some part of the outcome.
Regardless of whether player skill is some part of the outcome, character build needs to be a very significant part of the outcome...and that will include some RNG. Though, I would expect significantly less RNG from Action Combat than from Tab Target.
Decision making is not the same thing as player twitch skill - which is why I specifically said player twitch skill rather than simply player skill.
Yeah, if you want it to be action, it would need to have a telegraph so that it can be dodged. If you want it to have a %chance of not missing, what gameplay does that create?
Players are still going to javelin you when that's their best option, and now one of two things happens:
1) you resist it. they're at relative disadvantage because of circumstances outside of their control, and potentially the outcome of the fight is heavily determined by you happening to resist the javelin.
2) you don't resist it. you're at relative disadvantage because of circumstances outside of your control, and potentially the outcome of the fight is heavily determined by you happening to not resist the javelin.
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net. This same exact situation used to come up in vanilla wow / burning crusade where orcs had a higher chance to resist stuns and that would straight up decide fights.
Escapes from the post main focus but, that's quite a harsh opinion over RNG mate, just out of curiosity, may it be tab target combat or action combat do you believe RNG to be a detriment to an MMORPG combat? Do you believe a MMORPG combat would benefit from having little to none RNG?
Escapes from the post main focus but, that's quite a harsh opinion over RNG mate, just out of curiosity, may it be tab target combat or action combat do you believe RNG to be a detriment to an MMORPG combat? Do you believe a MMORPG combat would benefit from having little to none RNG?
Sure, so maybe some background is in order! MMOs were originally born from table-top rpgs like D&D, which in turn were born from war-gaming. In both of those contexts, they use dice rolls to serve two main purposes: to be simulationist and to provide drama. You can read me elaborate more about this here
The gameplay loop in TTRPGs is as follows: the GM describes a situation and then the players describe an intent (what they want to do) and an approach (how they want to do it). The GM applies this information to the situation and then describes a new situation and the process repeats.
This core gameplay loop can be done without rules and without dice. The players don't know how the GM will decide to apply their intents and approaches, and so there is drama for them. The GM doesn't know what the players will do when presented with a situation, and so there is drama for them. As you do this, and especially if you want to have any form of in-world consistency, the GM will probably feel overwhelmed and fatigued trying to make sure they're consistent in their arbitration. Why was I unable to climb this wall in time when I was able to climb this more difficult wall earlier? Why is this werewolf killing me when I slew a dragon last week? etc.
Eventually, you start to attempt to mathematically model your narrative fiction, and then query the model. You have 18 strength? Then you can deadlift 400lbs.
Then, to add in drama, you can bake in variance via rolling dice. You have 18 strength? You can lift 400lbs 70% of the time, 425lbs 40% of the time, etc. If your rules are well composed, then all of your probabilistic outcomes will also be coherent.
So now you have rules to remove some of the cognitive overhead from a GM trying to simulate a world. You have dice rolling to add volatility to those rules and provide drama.
But when we move from the tabletop to MMO's and especially sophisticated MMO's we can lean more on the game world itself to fairly and accurately arbitrate the world. When we fall, we don't need to take 10d6 random falling damage, we can apply the physics engine, for example. We don't need to have a random encounter against 2d6 kobolds. Instead, the game developers hand-craft dungeons and spawnrates. What used to be approximated via randomness and tables is can now be calculated directly.
The same goes for drama - when before, the players needed the roll-of-the-dice to create at-table drama because combat complexity of a tabletop is relatively low (especially compared to say, a real-life game like tennis), now in modern MMOs, we can get a lot closer. As we raise the skill cap, we can raise how interesting the game is to actually play. We can engage the player with gameplay and create really interesting fights like the rathalos encounter that Azherae dreamed up.
In PvP too, we don't need to lean on randomness. If the devs want to make it so that the same number doesn't always show up when you attack (slight variation in damage), that's totally fine. If they want to make it so that 10% of the time you straight up just resist javelin, I think that's super lazy. Why not make it so that you always get pulled, but your character gets pulled significantly less far (that way you can still build to reduce the effectiveness of javelin)? Rather than a 10% chance to resist stuns, how about a -10% duration of stuns?
The overarching goal here is to make sure that there's enough time for the central limit theorem to do its job. If you're doing a thing less than ~30 times in a fight, it doesn't have any business being affected by RNG, or contests between equally skilled players will often be decided by randomness (who resisted stuns, etc).
DygzMember, Braver of Worlds, Kickstarter, Alpha One
edited August 2021
It doesn't change the need for RNG. You can build your character to have max resistance to Stun - and sometimes the cards are just not in your character's favor, the character effectively rolls a 1 and gets Stunned anyways.
Sometimes shit happens. That's part of the story.
Escapes from the post main focus but, that's quite a harsh opinion over RNG mate, just out of curiosity, may it be tab target combat or action combat do you believe RNG to be a detriment to an MMORPG combat? Do you believe a MMORPG combat would benefit from having little to none RNG?
Sure, so maybe some background is in order! MMOs were originally born from table-top rpgs like D&D, which in turn were born from war-gaming. In both of those contexts, they use dice rolls to serve two main purposes: to be simulationist and to provide drama. You can read me elaborate more about this here
The gameplay loop in TTRPGs is as follows: the GM describes a situation and then the players describe an intent (what they want to do) and an approach (how they want to do it). The GM applies this information to the situation and then describes a new situation and the process repeats.
This core gameplay loop can be done without rules and without dice. The players don't know how the GM will decide to apply their intents and approaches, and so there is drama for them. The GM doesn't know what the players will do when presented with a situation, and so there is drama for them. As you do this, and especially if you want to have any form of in-world consistency, the GM will probably feel overwhelmed and fatigued trying to make sure they're consistent in their arbitration. Why was I unable to climb this wall in time when I was able to climb this more difficult wall earlier? Why is this werewolf killing me when I slew a dragon last week? etc.
Eventually, you start to attempt to mathematically model your narrative fiction, and then query the model. You have 18 strength? Then you can deadlift 400lbs.
Then, to add in drama, you can bake in variance via rolling dice. You have 18 strength? You can lift 400lbs 70% of the time, 425lbs 40% of the time, etc. If your rules are well composed, then all of your probabilistic outcomes will also be coherent.
So now you have rules to remove some of the cognitive overhead from a GM trying to simulate a world. You have dice rolling to add volatility to those rules and provide drama.
But when we move from the tabletop to MMO's and especially sophisticated MMO's we can lean more on the game world itself to fairly and accurately arbitrate the world. When we fall, we don't need to take 10d6 random falling damage, we can apply the physics engine, for example. We don't need to have a random encounter against 2d6 kobolds. Instead, the game developers hand-craft dungeons and spawnrates. What used to be approximated via randomness and tables is can now be calculated directly.
The same goes for drama - when before, the players needed the roll-of-the-dice to create at-table drama because combat complexity of a tabletop is relatively low (especially compared to say, a real-life game like tennis), now in modern MMOs, we can get a lot closer. As we raise the skill cap, we can raise how interesting the game is to actually play. We can engage the player with gameplay and create really interesting fights like the rathalos encounter that Azherae dreamed up.
In PvP too, we don't need to lean on randomness. If the devs want to make it so that the same number doesn't always show up when you attack (slight variation in damage), that's totally fine. If they want to make it so that 10% of the time you straight up just resist javelin, I think that's super lazy. Why not make it so that you always get pulled, but your character gets pulled significantly less far (that way you can still build to reduce the effectiveness of javelin)? Rather than a 10% chance to resist stuns, how about a -10% duration of stuns?
The overarching goal here is to make sure that there's enough time for the central limit theorem to do its job. If you're doing a thing less than ~30 times in a fight, it doesn't have any business being affected by RNG, or contests between equally skilled players will often be decided by randomness (who resisted stuns, etc).
A bit more "context" about the question than expected but still reasonable, even tho RNG in MMORPGs isn't really about the "appeal to tradition" of TTRPGs which are their origins, but about a countermeasure against repetitiveness, predictiveness and staleness in its systems, may it be in the drop systems, gear system [different gear stats(or rolls on those stats)] or the combat system.
A lot of people fails to see the type of Skill RNG in combat provides and incentivises(especially RNG in CCs) which is the Adaptability to deal with unpredictable outcomes and fast judgement by being able to improvise on the fly, and you can still add things like less CC duration on top of that as one system doesn't completely cancels the other, and how meaningful a Hard CC is depends on its duration and on the games TTK.
(stun duration 2 sec and TTK 10 sec? CC = 20% of TTK)
(stun duration 2 sec and TTK 30 sec? CC = 6,666% of TTK)
(Varies depending if on the CC skill cast time, if it does damage or not, and if it does how much damage it does.)(Can consider CC duration reduction or chance to apply).
"Equally skilled players" contests being decided by randomness isn't really a thing, unless in games that purely relies on luck such as dice rolling, the majority of times players excel in one type of skill over the other may it be Puzzle solving skill, Dexterity skill, Tactical skill, for MMORPGs you can add "character building" skill, for RNG you can add Adaptability skill, less measurable but you can add "predicting" skill.
I refuse to believe people would consider this to be "lazy" or "less skillful" in any regard.
but about a countermeasure against repetitiveness, predictiveness and staleness in its systems,
That's more or less because the gameplay is derived from tabletop games that would get stale without forms of randomness. Other genres, like fighting games, are able to provide enough complexity and depth without making the player-to-player interactions lean on randomness.
the Adaptability to deal with unpredictable outcomes and fast judgement by being able to improvise on the fly
I don't know where you are on your own competitive journey, or how seriously you tend to take competition, but if I were to play a game where I knew that my javelin ability had a miss% chance, there isn't improv or unpredictable outcomes going on. The outcomes are predictably that my javelin will work or that it won't. I have a pre-planned strategy that I can execute in either case, because MMOs tend to be simple enough that good players can think 3-4 globals ahead minimum.
"Equally skilled players" contests being decided by randomness isn't really a thing, unless in games that purely relies on luck such as dice rolling
How do you figure? If you have a mirror match between two players with equal gear and roughly equal skill following roughly the same gameplan, and one person's 5 second stun gets resisted and the other's doesn't, that can be the decider 🤷
They're splitting games when they're getting equally lucky (because they're equally matched), so when one person gets more lucky, they win.
That's more or less because the gameplay is derived from tabletop games that would get stale without forms of randomness. Other genres, like fighting games, are able to provide enough complexity and depth without making the player-to-player interactions lean on randomness.
Yes, its certainly derived from its TTRPG origins, which isn't something bad and part of the MMORPGs identity.
It is clear that you have a bias in favor of fighting games and it's rng-less systems(nothing wrong with that), but its important to remember that systems that are better in a type of game may not be as good in another type of game, especially when the difference of games are that big.
I don't know where you are on your own competitive journey, or how seriously you tend to take competition, but if I were to play a game where I knew that my javelin ability had a miss% chance, there isn't improv or unpredictable outcomes going on. The outcomes are predictably that my javelin will work or that it won't. I have a pre-planned strategy that I can execute in either case, because MMOs tend to be simple enough that good players can think 3-4 globals ahead minimum.
I can assure you i have quite a long "competitive journey" and that i take competition quite seriously.
(Not sure why that would be relevantfor the topic but ok).
Even tho there are only 2 outcomes every time to you throw a CC (it will apply or it will not apply) it is still unpredictable because you don't know which one will play out, instead of literally a single predictable answer (IT WILL APPLY) it forces the player to adapt or try to predict which result it will be.
Lets take your javelin example for the 2 situations:
IT WILL APPLY: The javelin will for sure pull the enemy, you can follow up with another CC and if available a third one or maybe a powerfull skill that requires some longer cast time, basically a linear stun lock combo, pretty much something you would find in a fighting game stagger duration.
Will it apply or not? Yes: The javelin will pull the enemy, you can try to follow up with another CC and it may fail and give your opponent an opening against your third skill, so you use the CC skill as follow up for the javelin or do you try to garantee a high damage skill right after it?
Will it apply or not? No: The javelin will not pull the enemy, you gonna need to adapt on the fly, maybe use a gap close? Maybe roll out of your opponent's range?
No matter how i look at it 2 possible outcomes is better than a single predictable one in terms of adaptability skill requirement.
How do you figure? If you have a mirror match between two players with equal gear and roughly equal skill following roughly the same gameplan, and one person's 5 second stun gets resisted and the other's doesn't, that can be the decider 🤷
They're splitting games when they're getting equally lucky (because they're equally matched), so when one person gets more lucky, they win.
Unless the TTK is something like 50-60+ sec 5 sec stun is pretty insane without a CC breaker or some huge duration reductions it would definitely be a match decider.
Other than such insane CC durations, a single CC trade where one fail and other works wouldn't determine the end result of a match(unless if both are ultra glass cannons with ultra low TTK) as there can be more CC interactions, maybe some misses or parries and maybe some lucky critical hits until the end of the fight.
Will it apply or not? No: The javelin will not pull the enemy, you gonna need to adapt on the fly, maybe use a gap close? Maybe roll out of your opponent's range?
Why is what you need to do up in the air? But yes, objectively making people worry about more possibilities raises the skill cap, no argument there. I wrote as much here:
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net.
Players now need to do expected value/variance calculations and plan around both branches (similar to how MTG players have to plan around lines). The tradeoff is that now players can win or lose because they got lucky or unlucky, which while that doesn't reduce the skill cap, it objectively reduces the impact of skill.
Why is what you need to do up in the air? But yes, objectively making people worry about more possibilities raises the skill cap, no argument there. I wrote as much here:
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net.
Players now need to do expected value/variance calculations and plan around both branches (similar to how MTG players have to plan around lines). The tradeoff is that now players can win or lose because they got lucky or unlucky, which while that doesn't reduce the skill cap, it objectively reduces the impact of skill.
If by reduces the impact of skill you mean the other types of skills such as Puzzle solving skill and Dexterity skill then yes, you are correct as the addition of another type of skill reduces the impact of other types of skills.
It seems that we agree on more things than we don't.
BTW, i would like to praise your http://beaushinkle.xyz/ site, it has quite interesting wrightings regarding important game topics.
Thanks for replying in this off-topic discussion.
If by reduces the impact of skill you mean the other types of skills such as Puzzle solving skill and Dexterity skill then yes, you are correct as the addition of another type of skill reduces the impact of other types of skills.
I view handling uncertainty as part of puzzle solving, and if you have to react fast once the uncertainty is resolved (like in a low-gcd game), part of dexterity. More, I was trying to separate the concept of skill-cap and skill-impact. For instance, the skill cap in Magic: The Gathering is extremely high - there's tons of lines of play to think through, game knowledge to know, bluffs to make, etc. But, the winrate for the top players are lower than in a lot of other games, and top players tend to be less dominant, because Magic: The Gathering has a high degree of luck baked into the competition, so we would say that "how good you are" has less of an impact on whether or not you win because of the random elements.
For an easier example of this, consider the following two games: in one game, you try to solve 15 difficult puzzles at the same time as an opponent. When you're done, whoever solved the most puzzles correctly wins. In the other game, you both try to solve 30 difficult puzzles (which increases the skill cap), and then the game looks at how many puzzles you solved vs them. If you solved 25 correct and they solved 15 correct, you have a 25 / (25 + 15) = 62.5% chance of winning. The skill cap has increased, but the skill impact has decreased.
The same thing can happen if you introduce stuff like randomly resisting crowd controls. You add to the skill-cap, but potentially reduce the skill impact (which needs to be looked at on a case-by-case).
BTW, i would like to praise your http://beaushinkle.xyz/ site, it has quite interesting wrightings regarding important game topics.
Thanks for replying in this off-topic discussion.
Thanks for the complement! Feel free to shoot me an email if you wanted to chat about any of the ideas there, or share it with anyone you think would be interested.
Even tho i certainly greatly value the RNG factor as something essential for the MMORPG genre, i still appreciate this writing's value as something more generalized which makes a lot more sense for other genres like fighting games, mobas or FPSs.
Comments
Attack Cones: Already part of Ashes' main design, no loss in keeping it, impossible to do their intended 'weapon cleaving' without it. Debatable whether we need narrow ones, but here's even more data you might not have.
Hitstop: People seem generally opposed, I don't think I've yet 'explained to someone what the problem with it would be' and they still wanted it, outside of 'big moves occasionally stagger me a little'. I have been treating it as 'around the same amount of time as the recovery animation for the move that does it'. So that it only becomes an offensive mechanic in a coordinated group, and even then is rare. If you mean just the '3-5 seconds animation freeze on the attacker's side on impact', ignore this.
Dodging: Already in, at least a bit. I thought that without it, most people would be upset, and that's where we hit the most confusing part. Lots of people dislike being rooted. I thought this was because they wanted to dodge more attacks. Indication is 'no', that's not the core reason. Many don't care. Those that do care, explicitly care about not only dodging, but attack cones AND the idea that enemies don't target like Tab Target.
I generally feel like dynamism within Ashes' current combat system is lost without dodging, and found no calculable flaws to point out in the 'we should be able to dodge twice but less far' suggestion that was integrated into the Compilation'. Ashes needs 1x 'unlimited no iFrame dodge with a longish recovery' (current ground movement speed makes simply strafing insufficient against 80% of current attacks and abilities, which fluctuates upward with the size of the attack cones) and at least 1 more limited dodge. 2x works on the current Dodge timer without breaking things, and prevents '2v2 staggered burstdown'.
Generally, from all our ingame tests, Ashes' design has these as requirements or they need to overhaul the ability sets of all current classes and meaningfully rethink a considerable number of systems and intentions. Their intent is to have hotbar ability lists be 'manageable' (probably 20 abilities total, also presuming some of those slots will have items in them) and don't intend for every class to be able to learn every ability without a downside (or possibly at all).
Unless these change, most of the requirements you listed have the prerequisites. With the current game flow and speed, even if they rip out Action Combat as a whole, they will still have these problems.
Unfortunately what you would need is a video on 'how netcode works in games'.
Games use different netcode models, and have different ones based on the types of games they are. Some aim for fairness between people of differing ping, some aim for server consistency, some aim for satisfying interactions that work to ensure the client isn't confusing for the player.
These three implementations don't overlap much in the US/EU. In Korea or some parts of Japan, the ping is by nature smaller. Geography matters. Let's assume Ashes 'plans' to try to give us more geographically consistent servers than most games, and since it is already a Western game, we don't have to worry about the usual issues that old MMOs faced on that front.
This implies their optimal model is the last one. Loss of fairness is unfortunate, but surmountable in different ways and the game offers many activities. PvP or not, you can avoid this or just accept it, like... literally anyone who tries to play competitive games on a laggy connection/far from their server, already does.
Server consistency would be a good aim for Ashes if it was going for Action Combat entirely, but Tab or Hybrid with telegraphs doesn't benefit much from this. You already have lots of leniency in terms of what people expect. 'Spend it'.
So that implies the third model is optimal.
This one is done by making it so that the 'startup frames' can be clipped out of certain attacks, but a visual indicator unrelated to the actual animation is required from frame 1. Players who have really good ping have a higher chance of dodging correctly, but you just put the mindgames behind that.
"Make the first move always reactable on 50-80ms ping, but never devastating."
"Make the second move much harder to deal with if you dodged the first."
"Make the player not need to focus as much on the mechanical difficulty of the dodge so that they can focus on the opponent to watch for the second move."
"Give the attacking player options after their first move that put the defender at positional disadvantage if they are watching too hard."
Example:
Mage lines up and throws a Fireball. Their hands flicker an obvious flaming orange from frame 1. You could dodge this fireball. It starts in 5 frames. They got leniency from their soft lock, so if you were in the wrong position on their screen, you're still in danger. Ashes' movement speed already supports this.
If your ping is 80ms, the fireball is 'already moving toward you' by the time you see the flames. So what? Still do something. If your ping is 400ms, call your ISP, but you can just 'not dodge the fireball'. Because you know that the mage is now in a Fireball Recovery animation. And you know where they are. You can attack instead. Suboptimal, maybe, but you still get to play.
Your client can skip some number of frames in the Fireball animation based on the timestamp data it gets from the server. As long as those hands keep flaming for a while, you're still in control.
A player with 50ms ping could easily dodge this Fireball. The key is to make a game where that isn't always the correct thing to do. The Mage needs to be able to 'know that your dodge will probably take you to a certain location', 'watch for the startup of the movement in the direction and know that if they see the startup it won't be rolled back', and 'target their own next ability or movement toward their desired location'.
This is the reason for the Root Motion conclusion in the Compilation. The Mage sees you moving, they can't see exactly where you are facing or where you will dodge to because of latency. They prime to be ready when you lunge or strafe dodge and set their AoE to approximately that location.
Then they see you move. Human reaction time of 240ms means that the dodge itself must have 15 frames of animation recovery. Ping values above 240ms mean the player has to guess a bit since they only see your movement's start 15f late.
So 15f +15f on their side, if the dodge is 34+, they can hit you.
On your end, you could have 240ms ping. That means that you see this as 'I hit dodge', and 15f later the server recognizes that you did that. But the server can 'catch up to you'. It now has 'a clear direction', it can just skip 15f on its side due to timestamps. The server 'puts you where you should be' and then sends the data to your opponent. The server doesn't have to care as much 'what you do when the animation is over' because your animation is going to be another 19f.
So you, after your dodge, decide to strafe back in the direction you came.
If both players have 240ms ping, your opponent 'saw you blip forward' (the netcode difficulty here is determining how much of an animation to skip since which frames are important, must be built into the animation model, which is a processing thing).
On their side, by the time the server info reaches them, it basically looks like:
"Dodge Start Position: Coords, Dodge Start Timestamp: X, Animation Frame Reached: Y, Vector: V3D"
You don't strictly need the animation frame reached, that's the server working to help the client, that load can go wherever.
The Mage's client shows maybe 5 frames of 'player at the Dodge start Position', skips some number of frames, and then we get really close, but it still works. The thing is, the Mage doesn't care. They saw a startup, and if they could interpret it (or not, and guessed well), they targeted the spot where the player will be. They send back 'target location', their own start timestamp, and off we go back to the server, another 15f journey.
But the defending player didn't get out of that AoE in 4 frames by walking. They didn't get out of it by just 'strafing back the other way'.
On their side, they tried. The server cuts 15f off the Mage's AoE animation. The defender's client another 15f. But the key thing that makes this work is that the Mage got to react early enough so that when this all plays out, since that player was 'waiting 19f after the first dodge to be able to strafe', in a specific spot, they know that when they see 'the mage AoE descending on that spot', it's less strange when they get hit despite being a few frames out of the AoE visibility point.
If this isn't acceptable enough even with 15f ping to server on both players, almost nothing will be, short of 'Ashes of Creation releases in Korea'.
A final note for the unfamiliar. Your Ping is based moreso on your ISP and network infrastructure than your distance, in a game with regional servers. Dallas to Boston is 40-50ms (3 frames) along fiber optics. Generally you double that for your expectations (lots of other stuff beyond just the fiber) and 2.5x if rurality starts to factor in (depends on ISP).
@beaushinkle has various links suggesting (and I'll quote from it):
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.
Unfortunately from my perspective, these things are slightly too vague, go against what we've played so far, and multiple parts of the game's Action Combat model, just don't work.
"They give up too much of what an Action Combat game would feel like, for no reason."
It's frustrating to have a skillshot miss some percentage of the time. But 'not having them because of this' is only 'succeeding at not frustrating people' because those people just don't play.
So I've finally had enough time to write this up as a full 'disagreement' with the quoted concept. Tab Targeting has too many limits for a PvP system using the base Ashes has. I finally believe that if Ashes does not implement Hybrid Combat, given their current ideas, they might as well not implement PvP at all with any expectation of competition or skill mattering.
I don't mind PvP that is just the 'is my gear better than yours, am I capable of reading my ability cooldowns and active status effect icons?' in a game where Gearing up is the primary treadmill (i.e. a Themepark) but a soft-power game like Ashes probably needs more agency than that. For the first time, I'm going to say this.
I think Ashes of Creation actually has to have Hybrid Combat, given what we've seen so far.
I think it's relevant that in your example, the spacing is around vague areas where someone needs to pre-aim an ability and not something incredibly spacing specific like https://www.clipnotes.io/watch/c5258c47-43ca-4ab2-9e03-060416c1be53?clip=6260e886-7de8-4585-b5fa-ab52914c7aa2
Really what the above situation is trying to accomplish is that you have a low-reward move (fireball) that can bait an option and now the players need to pick between the different options in the small RPS game that each move to different gamestates themselves. This creates interactivity, and I think that's what players crave, not the laggy action combat itself.
Which things were too vague? Which go against what you've played so far or just don't work? Why wouldn't a rogue, for example, be able to trick a tab-combat system by making people target illusions on accident. Why couldn't a Mage put up a wall-of-stone that blocks line of sight?
All of that said, I'm not at all married to tab combat. I just think it's easier to create more physically and intellectually interesting pvp and pve when you don't rely on precise spacing in EU/US network conditions. If your version of precise spacing is stuff like aiming ground targetted AoEs then we're in agreement - those are probably fine. If your version of precise spacing is stuff like dipping into the range of marth's sword to bait him to swing and then dipping back out of range to make him whiff and then dipping back in to punish, then I think we disagree.
I fully and completely agree that for ashes to work, there needs to be agency. I'm flabbergasted when people seem to be okay with winning or losing PvP with gear/materials staked based on counter-comping and spreadsheet-vs-spreadsheet rather than skill expression.
Why do you think that Hybrid Combat is required for the game to have that amount of agency?
The short answer is 'I played it', but obviously that's not good enough, and I know that.
I'm completely in sync with you about the idea that 'extremely precise targeting of any kind' is not a good plan. The thing I perceive as 'agency' is the precise result of two interactions.
The first is 'that movement is necessary to either fake out an opponent or dodge meaningful attacks'. Not perfectly, not precisely. Just that you generally should have a good reason to consider it. The reason for this is that Ashes intends to have Summoners and those Summoners will have pets that act like Mobs. I'll elaborate on this more if you wish it (even more than I do later in this response), but for now...
The second is that making any movement contains a disadvantage relative to your own ability to launch a counterattack, which you must overcome with your own skill, or in some cases, cannot overcome because it is too large.
So in the example of Marth's sword, we come back to a point where Naraka and Vindictus show us how this works. You move into a position, your opponent makes a melee attack that should hit this position and fails.
In a fighting game, there is hitstop, so a whiff punish happens and you lose control of the situation. In Ashes, there is no hitstop or stagger. So your opponent doesn't care if their first attack whiffs, they only care that their flow into their second attack is correctly targeted. You, in turn, don't actually care if their first attack hit you or not other than 'well it reduced damage'. A Summoner doesn't even get this much. They had to 'position their summons correctly' and hope, because the Summon will generally 'keep attacking'.
For Summoners to be effective, a player's choice has to be 'I can dodge, or I can attack, but usually not both'. If they dodge forever, fine, it distracts the summon, but just drags things out. If they attack, they get hit some amount of the time, and the summon's power gets to work. A Summon generally can't 'decide if to dodge AoE or not', and if they were given this ability, even as a stance, then the speed at which this happens means that now you have just one Archetype with latency issues (technically two because the same thing will happen to Bards).
The problem with any Tab like combat is the stuff we have now.
The Tank ability Javelin, does a pull-in effect on hit. Even if it doesn't have its short CC, this is a big deal. In Tab Target mode, the effectiveness of this move is based on 'can I get line of sight to the target' and 'Will my Acc beat their Evasion'?
Javelin isn't about baiting. Neither is Cleric's Judgement at high level (a relatively strong Movement Speed Down). These abilities need to be Action or Action-lite with a soft lock, or what happens is not 'agency', it is 'My opponent pressed a button to change my agency and I watched to see if it happened'.
If one argues that 'the opponent can react', then you're back to a level of 'netcode' where you might as well have gone all in. I can't use an ability or block in time against Javelin if I couldn't dodge it. Fireball is the example I used because Mages have an obvious "AoE that matters" for the rest of the consideration.
The style of effects, the movement speed, the concept of how melee and procs will work, all of these things would need to be rethought if Hybrid Combat were removed or stripped down to 'ground AoE markers'. And even those fall into the same problem the smaller they get. If melee matters but there is no stagger, then in a fantasy game, magic and similar abilities will have strong effects or the game's (arguably) 4 ability/non-melee based classes are going to be in a bad place. Nothing worse than 'my strong moves are dodgeable AoE but my opponent just melees me down'.
Agreed!
So make them care without stunning them! Make it so that attacking someone applies a "weakened' debuff that you can clear by using a different ability. Your attack doesn't apply 'weakened' if you're already 'weakened'. Now both people still have control over their characters and there's no hitstop, but whiff pushing creates a situation where there is an attacker (the person not weakened) and a defender (the person weakened). It doesn't have to be this specifically, but something like this, that creates a flow of combat.
Making javelin tab-target point-and-click is bad. Making it have a miss% based on character stats is worse. Instead, you can make the javelin ability have a wind up animation, and then give players the ability to avoid it. Maybe you can dodge-immune the javelin. To provide layer-2 counterplay, the tank needs to have the ability to feint the javelin, either by winding up and canceling, or winding up and holding the button down to beat players that are trying to dodge-immune it.
If you would like to engage in the debate on either 'stat based miss chances on abilities' or 'Javelin shouldn't be tab target at all', I can point you at the threads, but note that the opponents of both these concepts have directly quoted the Creative Director's strong implications that these are the intention, specifically as support of their own wish for it to be so.
I have made only overarching/compromise based arguments 'against' those positions. If your stance is stronger, I'm sure they can hear you out.
As for the aspect of making people care about being hit, I don't personally see how this helps if you're not going to be dodging particularly well in the first place, nor was I even trying to imply that there should be a change. My suggestion is based on Vindictus and on combos.
You don't want to get hit by the telegraphed final hit of your opponent's combo because it debuffs you or benefits them in ways I mentioned here, which are based on the concept of weapon procs, as mentioned in that comment.
My overall stance is that Ashes' design already supports everything they intend, and only iterative design and tuning is required now. Any points I make come from 'where they are already', since that's my 'chosen role'. You'll need to talk to some other people to get the more specific 'Acc vs Eva' debates, since 'removing RNG from interactions' is not currently confirmed to be on the table for this game's combat design.
I should be able to build my character to resist Javelin... which is what I did in Alpha One, and it worked great.
In the Alpha One, I often could not see the prep for Javelin.
If it were Action, Javelin would need a recognizable telegraph along the lines of the original Fireball/Hallowed Ground/Exorcism, so that opponents have the opportunity to attempt a Dodge.
I was trying to figure out a way to summon you without literally giving in to my urge to go 'Dygz, I choose you!' and quoting one of those lines you always react to.
Success!
I have no opinion on your post one way or the other, but I bet @beaushinkle does. I hope it's 'agreement'.
Yeah, there would be need to be a long enough tell for the startup of javelin throw and a short enough recovery from your normal non-commital options that if you're paying attention you can outplay. You know all this though, so I won't preach to the choir.
The final hit system is interesting because you're able to effectively extend the startup frames of the "real" threat by a massive amount and use that to hide rollbacks. I like it!
💯
Yeah, if you want it to be action, it would need to have a telegraph so that it can be dodged. If you want it to have a %chance of not missing, what gameplay does that create?
Players are still going to javelin you when that's their best option, and now one of two things happens:
1) you resist it. they're at relative disadvantage because of circumstances outside of their control, and potentially the outcome of the fight is heavily determined by you happening to resist the javelin.
2) you don't resist it. you're at relative disadvantage because of circumstances outside of your control, and potentially the outcome of the fight is heavily determined by you happening to not resist the javelin.
Yes, it introduces the skill of playing-around-rng, but overall I think this is skill-reducing in net. This same exact situation used to come up in vanilla wow / burning crusade where orcs had a higher chance to resist stuns and that would straight up decide fights.
It's significantly different than player twitch skill v player twitch skill - which is the purview of other genres.
In an RPG, if you want to resist Stuns, you should choose a race and a class that helps you resist Stuns.
Your opponents should be building characters to maximize Stuns - if they want to maximize Stuns.
Also, keep in mind that Ashes PvP combat is balanced for 8v8 rather than 1v1.
In an effort to find some common ground, you surely believe that you need to pilot your character, right? As in, you presumably don't want to create a character and then press the "fight" button and see if the character (represented by their stats, level, gear, etc), not the player behind the character (represented by their decision making, mechanical skill, etc) wins a fight.
If you want for player skill to be some part of the outcome, then we're back to why you why you want javelin specifically to be determined by RNG, since the only answer you gave me was a blanket statement.
This is @Dygz consistent stance across multiple discussions, whether it be 'should Evasion and Acc decide hit chance in Action based attacks' to the linked thread.
It may be best to not discuss that topic in this thread, presuming you prefer to keep this one on a particular track for any post-Livestream feedback.
There'll be a multitude of reasonings in the link provided.
Regardless of whether player skill is some part of the outcome, character build needs to be a very significant part of the outcome...and that will include some RNG. Though, I would expect significantly less RNG from Action Combat than from Tab Target.
Decision making is not the same thing as player twitch skill - which is why I specifically said player twitch skill rather than simply player skill.
Escapes from the post main focus but, that's quite a harsh opinion over RNG mate, just out of curiosity, may it be tab target combat or action combat do you believe RNG to be a detriment to an MMORPG combat? Do you believe a MMORPG combat would benefit from having little to none RNG?
Aren't we all sinners?
Sure, so maybe some background is in order! MMOs were originally born from table-top rpgs like D&D, which in turn were born from war-gaming. In both of those contexts, they use dice rolls to serve two main purposes: to be simulationist and to provide drama. You can read me elaborate more about this here
But when we move from the tabletop to MMO's and especially sophisticated MMO's we can lean more on the game world itself to fairly and accurately arbitrate the world. When we fall, we don't need to take 10d6 random falling damage, we can apply the physics engine, for example. We don't need to have a random encounter against 2d6 kobolds. Instead, the game developers hand-craft dungeons and spawnrates. What used to be approximated via randomness and tables is can now be calculated directly.
The same goes for drama - when before, the players needed the roll-of-the-dice to create at-table drama because combat complexity of a tabletop is relatively low (especially compared to say, a real-life game like tennis), now in modern MMOs, we can get a lot closer. As we raise the skill cap, we can raise how interesting the game is to actually play. We can engage the player with gameplay and create really interesting fights like the rathalos encounter that Azherae dreamed up.
In PvP too, we don't need to lean on randomness. If the devs want to make it so that the same number doesn't always show up when you attack (slight variation in damage), that's totally fine. If they want to make it so that 10% of the time you straight up just resist javelin, I think that's super lazy. Why not make it so that you always get pulled, but your character gets pulled significantly less far (that way you can still build to reduce the effectiveness of javelin)? Rather than a 10% chance to resist stuns, how about a -10% duration of stuns?
The overarching goal here is to make sure that there's enough time for the central limit theorem to do its job. If you're doing a thing less than ~30 times in a fight, it doesn't have any business being affected by RNG, or contests between equally skilled players will often be decided by randomness (who resisted stuns, etc).
Sometimes shit happens. That's part of the story.
A bit more "context" about the question than expected but still reasonable, even tho RNG in MMORPGs isn't really about the "appeal to tradition" of TTRPGs which are their origins, but about a countermeasure against repetitiveness, predictiveness and staleness in its systems, may it be in the drop systems, gear system [different gear stats(or rolls on those stats)] or the combat system.
A lot of people fails to see the type of Skill RNG in combat provides and incentivises(especially RNG in CCs) which is the Adaptability to deal with unpredictable outcomes and fast judgement by being able to improvise on the fly, and you can still add things like less CC duration on top of that as one system doesn't completely cancels the other, and how meaningful a Hard CC is depends on its duration and on the games TTK.
(stun duration 2 sec and TTK 10 sec? CC = 20% of TTK)
(stun duration 2 sec and TTK 30 sec? CC = 6,666% of TTK)
(Varies depending if on the CC skill cast time, if it does damage or not, and if it does how much damage it does.)(Can consider CC duration reduction or chance to apply).
"Equally skilled players" contests being decided by randomness isn't really a thing, unless in games that purely relies on luck such as dice rolling, the majority of times players excel in one type of skill over the other may it be Puzzle solving skill, Dexterity skill, Tactical skill, for MMORPGs you can add "character building" skill, for RNG you can add Adaptability skill, less measurable but you can add "predicting" skill.
I refuse to believe people would consider this to be "lazy" or "less skillful" in any regard.
Aren't we all sinners?
That's more or less because the gameplay is derived from tabletop games that would get stale without forms of randomness. Other genres, like fighting games, are able to provide enough complexity and depth without making the player-to-player interactions lean on randomness.
I don't know where you are on your own competitive journey, or how seriously you tend to take competition, but if I were to play a game where I knew that my javelin ability had a miss% chance, there isn't improv or unpredictable outcomes going on. The outcomes are predictably that my javelin will work or that it won't. I have a pre-planned strategy that I can execute in either case, because MMOs tend to be simple enough that good players can think 3-4 globals ahead minimum.
How do you figure? If you have a mirror match between two players with equal gear and roughly equal skill following roughly the same gameplan, and one person's 5 second stun gets resisted and the other's doesn't, that can be the decider 🤷
They're splitting games when they're getting equally lucky (because they're equally matched), so when one person gets more lucky, they win.
Yes, its certainly derived from its TTRPG origins, which isn't something bad and part of the MMORPGs identity.
It is clear that you have a bias in favor of fighting games and it's rng-less systems(nothing wrong with that), but its important to remember that systems that are better in a type of game may not be as good in another type of game, especially when the difference of games are that big.
I can assure you i have quite a long "competitive journey" and that i take competition quite seriously.
(Not sure why that would be relevantfor the topic but ok).
Even tho there are only 2 outcomes every time to you throw a CC (it will apply or it will not apply) it is still unpredictable because you don't know which one will play out, instead of literally a single predictable answer (IT WILL APPLY) it forces the player to adapt or try to predict which result it will be.
Lets take your javelin example for the 2 situations:
IT WILL APPLY: The javelin will for sure pull the enemy, you can follow up with another CC and if available a third one or maybe a powerfull skill that requires some longer cast time, basically a linear stun lock combo, pretty much something you would find in a fighting game stagger duration.
Will it apply or not? Yes: The javelin will pull the enemy, you can try to follow up with another CC and it may fail and give your opponent an opening against your third skill, so you use the CC skill as follow up for the javelin or do you try to garantee a high damage skill right after it?
Will it apply or not? No: The javelin will not pull the enemy, you gonna need to adapt on the fly, maybe use a gap close? Maybe roll out of your opponent's range?
No matter how i look at it 2 possible outcomes is better than a single predictable one in terms of adaptability skill requirement.
Unless the TTK is something like 50-60+ sec 5 sec stun is pretty insane without a CC breaker or some huge duration reductions it would definitely be a match decider.
Other than such insane CC durations, a single CC trade where one fail and other works wouldn't determine the end result of a match(unless if both are ultra glass cannons with ultra low TTK) as there can be more CC interactions, maybe some misses or parries and maybe some lucky critical hits until the end of the fight.
Aren't we all sinners?
Why is what you need to do up in the air? But yes, objectively making people worry about more possibilities raises the skill cap, no argument there. I wrote as much here:
Players now need to do expected value/variance calculations and plan around both branches (similar to how MTG players have to plan around lines). The tradeoff is that now players can win or lose because they got lucky or unlucky, which while that doesn't reduce the skill cap, it objectively reduces the impact of skill.
If by reduces the impact of skill you mean the other types of skills such as Puzzle solving skill and Dexterity skill then yes, you are correct as the addition of another type of skill reduces the impact of other types of skills.
It seems that we agree on more things than we don't.
BTW, i would like to praise your http://beaushinkle.xyz/ site, it has quite interesting wrightings regarding important game topics.
Thanks for replying in this off-topic discussion.
Aren't we all sinners?
I view handling uncertainty as part of puzzle solving, and if you have to react fast once the uncertainty is resolved (like in a low-gcd game), part of dexterity. More, I was trying to separate the concept of skill-cap and skill-impact. For instance, the skill cap in Magic: The Gathering is extremely high - there's tons of lines of play to think through, game knowledge to know, bluffs to make, etc. But, the winrate for the top players are lower than in a lot of other games, and top players tend to be less dominant, because Magic: The Gathering has a high degree of luck baked into the competition, so we would say that "how good you are" has less of an impact on whether or not you win because of the random elements.
For an easier example of this, consider the following two games: in one game, you try to solve 15 difficult puzzles at the same time as an opponent. When you're done, whoever solved the most puzzles correctly wins. In the other game, you both try to solve 30 difficult puzzles (which increases the skill cap), and then the game looks at how many puzzles you solved vs them. If you solved 25 correct and they solved 15 correct, you have a 25 / (25 + 15) = 62.5% chance of winning. The skill cap has increased, but the skill impact has decreased.
The same thing can happen if you introduce stuff like randomly resisting crowd controls. You add to the skill-cap, but potentially reduce the skill impact (which needs to be looked at on a case-by-case).
I think so too!
Thanks for the complement! Feel free to shoot me an email if you wanted to chat about any of the ideas there, or share it with anyone you think would be interested.
Even tho i certainly greatly value the RNG factor as something essential for the MMORPG genre, i still appreciate this writing's value as something more generalized which makes a lot more sense for other genres like fighting games, mobas or FPSs.
Ps: Not lazy btw xD
Aren't we all sinners?