How to balance L-canceling?

[JOE!]

Move ground moves are much more powerful than a character's aerial variants. Imagine having Mario L-Canceling his Down Smash.





Nothing?
Imagine Brawl MK's Down Smash..... Anything now?

The Knee, Ganondorf in general, Falco's Dair, and PLENTY of strong aerial moves disagree.

 

[gnosis]

Move ground moves are much more powerful than a character's aerial variants. Imagine having Mario L-Canceling his Down Smash.





Nothing?
Imagine Brawl MK's Down Smash..... Anything now?

Check it, we just increase the endlag of every ground move and then l-canceling them brings it back to normal. Also let's make them flash white just like l-cancels in P:M too. More l-cancels = more flashiness = more tech skill = more fun.

 

[1MachGO]

Obnoxious? Why, because you didn't feel like actually answering the question?

Here, I'll ask it again:
Why should only air moves be cancellable? If there was a way to reduce the lag of ground moves by pressing an extra button, would you support its addition?

If you defend L-cancelling as something that adds "depth" to the game, then any answer but "yes" to my question is hypocritical since it would add just as much depth to the game as being able to cancel air moves does.

You are asking a loaded question and are expecting it to prove something.

If you rationalize it, adding a mechanic which allowed you to cancel lag on all moves would add depth to the game. It would be less nuanced than L-Cancelling, but it would technically add more depth.

However, even if it adds depth, it doesn't necessarily mean the addition should receive support; this mod is trying to emulate Melee's gameplay and only incorporating universal techniques which are familiar to the series. Adding a big change like that would interfere with the PMBR's vision by proceeding to alienate some members of the target audience or effect balancing.

 

[BEES]

I think a real argument for L-canceling that you never hear is that melee vets have a hard-programmed itch in their brain to do it automatically for all their aerials, and the thought of not doing so is now uncomfortable. It feels like extra mental energy going to waste. Like for me I literally feel good L-canceling and feel terrible when I don't when I'm playing vBrawl for instance.

I have friends that I want to get into the game but sitting them down to teach them L-canceling was tedious, and they're still going to have to practice for some time to integrate it and learn to SHFFL reliably enough to use it in their playstyle. They're all frustrated by it. I want to keep it now, for a variety of reasons. I think the fact that the game gives you a visual indicator that you canceled your aerial goes a long way to making it easier to learn. The timing for everything was different from Melee at first but because the characters light up when you do it it didn't take long for me to learn the new timing.

I might be in favor of increasing the input window a bit. Maybe 10 frames instead of 7? Then it would really be something that once you learn it and tinker a bit, you've got command over it.

 

[DrinkingFood]

I really don't understand how people can make such a big deal out of this.

L-cancelling is fun, and doesn't break the game, that's all that really matters.

I agree with this
I don't care if this stays or go
sure, it'll make transferring back to melee on occasion much harder once I get used to it
but that's all the issue i'd actually have with removing it
but I also don't see why its inclusion is such a big deal
if you argument is "it's a skill barrier" well, anyone that would be stopped by L-canceling would be stopped by wavedashing, wavelanding, moonwalking (for some characters) dash pivots, ledge hopping, and every other tech we take to be "easy" because they've been a fluid part of our game experience for awhile now. If someone wants to get into the game seriously, I HIGHLY doubt L-canceling is going to stop them. That fact that it's dozens of times easier to know when you're doing it right just makes it that much easier to learn and that much less of roadblock.

 

[PastLink]

not gonna lie, i'd support grounded lag cancels. imagine how much faster the game would be. (this is more a fantasy than anything)

However, even if it adds depth, it doesn't necessarily mean the addition should receive support; this mod is trying to emulate Melee's gameplay and only incorporating universal techniques which are familiar to the series. Adding a big change like that would interfere with the PMBR's vision by proceeding to alienate some members of the target audience or effect balancing.

I'd just like to point out that they went ahead and added the AGT tech, which doesn't exist in any smash series. and there's still brawl techs despite emulating melee. remember this is meant to build upon a foundation. not be a copy. I know you're not exactly trying to say that i'm just pointing it out.

on another note. i've been thinking. what if L-cancel frames weren't universal? each characters have different things that set them apart already such as aerial manueverability, friction, etc. so what if some characters had more lenient or strict timings? and also had less or more lag canceled respectively?

 

[Abeebo]

While the idea of an altered L-cancelling mechanic is intriguing, I am strongly against the auto-cancel direction. As stated before, L-cancelling is more of a movement option rather than a move. An execute rather than an action. If you automate this globally, you are removing a movement option. This will reduce the learning cure, but reduces a player's potential ability, so it's not losing or gaining much in 'depth' of the game.
You may be for L-cancelling. You may disapprove of L-cancelling. Either way, i'm sure you can agree that the skill floor and ceiling WILL be narrower if simply automated. That it is still one less option that ANY player cannot use to their advantage.
What I would not mind is some discussion on what CAN happen DURING or AFTER you nail/miss an L-cancel. Then we would be adding depth. MORE options. What crosses my mind... If you whiff an l-cancel, maybe you can cancel the whiff with a regular shield that has no perfect shield frame IF you hit the shield button again.

 

[Taiko]

It's technically a movement option, but why would you choose not to use that option? Why would you want to leave yourself exposed? The way I see it, it's just a barrier for the sake of being a barrier.

For example, say you could run forever without getting tired. Why would you ever walk? Sure it's an option, but running just makes more sense for getting places faster. However, that's not how it actually works, you can't run forever, and that's why walking is an option.

Now, if it was reworked so there was a penalty for using it, like how someone said slight shield usage, that would make it a more exciting mechanic I think.

 

[GaretHax]

I will reply again, you created a metaphor purposefully suited to make another argument seem invalid by approaching a hypothetical that has and never will exist, with fewer parallels than you seem to think loosely connecting it to the discussion of removing or tweaking a mechanic ALREADY in the game. Your "question" was not a question, nor was it productive. The entire way smash is ARLEADY designed and played makes your statement seem really, really silly. Arguably you could add L-canceling to grounded moves and double their current duration, but even then the metaphor is completely unrelated to this line of discussion... Not to mention how it is impossible to compare and contrast the two. Your post was purposefully inflammatory, off-topic, and drawn from a hypothetical situation that nobody could argue against because it simply doesn't exist, apply, or necessarily make sense; in other words it doesn't contribute anything, which would be what arguing to an absurdity is. Anyway I've spoken my peace already, if it needs to be made to be more mitigatable or what-have-you then so be it, but I feel L-canceling as an execution barrier is lightyears above memorizing pre-fabricated combo's from a list.

 

[Abeebo]

It's technically a movement option, but why would you choose not to use that option? Why would you want to leave yourself exposed? The way I see it, it's just a barrier for the sake of being a barrier.

For example, say you could run forever without getting tired. Why would you ever walk? Sure it's an option, but running just makes more sense for getting places faster. However, that's not how it actually works, you can't run forever, and that's why walking is an option.

Now, if it was reworked so there was a penalty for using it, like how someone said slight shield usage, that would make it a more exciting mechanic I think.

Spare me the analogies, because you shouldn't run in a hospital, stores, school halls, crowded emergencies, etc. Don't forget uneven terrain. You wanna try running across loose gravel cliffs? There's always going to be moments where you're going to need to walk. Though, I understand where you're coming from. Ideally, you'd want to pretty much always L-cancel, but to add a penalty seems a step backwards indeed. If you miss an L-cancel, you get lag and punished. You land the l-cancel, and you get punished. Why? Why do that? Give the l-cancel whiff a different option instead.

 

[Taiko]

True, you'd want to walk in those situations so I guess that wasn't the best way of looking at it. As for a "punishment" I think slight shield drain on use would be a good idea. Then, by using L-cancelling, you'd be effectively trading defense for offense, which would give players a better option of if they want to play offensively or defensively. By my understanding, people dislike defensive characters, and prefer to be offensive, so for the offensive people it wouldn't really be that different, they'd just need to watch their approach more so that they don't run out of shield. It would add more depth in the sense that you'd need to think more about when to use it, rather than simply using it because you can.

 

[1MachGO]

I'd just like to point out that they went ahead and added the AGT tech, which doesn't exist in any smash series. and there's still brawl techs despite emulating melee. remember this is meant to build upon a foundation. not be a copy. I know you're not exactly trying to say that i'm just pointing it out.

Actually, believe it or not, but AGT can be performed in Melee, too.

 

[The_NZA]

The best argument I'm seeing is shield tilting, but I'm still not sold on it being something significant enough to cause top players to miss l cancels often. Has anyone actually done any research on by how many frames a fully tilted shield can adjust l cancel timing?

It also makes technical manuevers more taxing. Such as doing DJC effectively with Ness. There are many heights you can do it at, and having an L-cancel in place makes it more cerebral (and not just reflexive). I sort of like that.

 

[Burnsy]

There are many heights you can do it at, and having an L-cancel in place makes it more cerebral (and not just reflexive). I sort of like that.

As someone who extensively plays DJC characters, I strongly disagree with this notion. Its no different than if you are running off a platform (or jumping onto one), doing an aerial, and l-cancelling. The height doesn't affect me because it is a mindless reaction/response to having done an aerial and being close to landing. I dont have to "think" about lcanceling just because it wasn't following a SHFF or FH or part of some other drilled muscle memory, if that's what you're saying.

Nobody has answered my question from before, perhaps I should restate it clearer: does anyone have QUANTITATIVE evidence that might shed light on "by how many frames can sheild tilting actually affect the opponent's l-cancel timing?" If its only 1-2 frames out of the 6 frame l-cancel window, I think its a bit harder to argue this particular point.

 

[Chzrm3]

L-canceling is cool because it serves as the perfect "gateway tech" for new players. If you're trying to teach your friends stuff, it's really, really hard to show them wave-dashing and have them immediately apply that to their game. Even things like DI can be pretty abstract, and they might just say "oh yeah, I already do that" when they don't, haha.

But L-canceling is something fairly easy that makes a ton of sense to anyone who's playing Smash. When they see how much faster it makes them, they're gonna be really excited, and they're gonna wanna master it. It also gets them to begin to accept that there are harder, more cryptic things hiding in the game, and that mastering these tricks will help them improve in really interesting ways.

If it was removed from the game, I think the skill curve would actually be much sharper for newer players. It'd be like "practice your spacing, reads, and other fundamentals... alright now who's ready to moonwalk?!!??!" and their minds all explode.

Don't think of it from the perspective of a Melee vet. We're all going to L-cancel anyway - if it was removed tomorrow, we'd still L-cancel every time we landed. It's baked into our minds forever, hahaha. (Seriously, who else has been air-dodging constantly in turbo because of that? I CAN'T MAKE IT STOP). Think of it from the perspective of a player new to the competitive scene, and what shape their growth and experience is going to take.

The L-cancel might seem like a non-option in a competitive environment, because it is (there's never a reason not to do it), but it's a really important option for growing players, and because of that alone, I think it should stay.

 

[standardtoaster]

i'm sure it wouldn't alter the timing by more than a few frames like you said. it wouldn't make it significant like hitting a shield tilted all the way up making you have to wait 10 frames or something

 

[GP&B]

Well, the whole idea is causing hitlag at different points and this also varies with the amount of hitlag any aerial induces on hit. Fox's Dair has relatively low hitlag for a multihitter compared to Mario's Dair so the effect is not very significant while a meaty aerial like Falcon or Ganondorf's Dair causes quite a lot making it much easier to miss the L-cancel after connecting at varying heights. If I had to whip out a simplified theory algorithm, it'd be something like:

Hitlag modifier + Character shield size +/- Shield Tilt = Frame Difference before landing.

This goes without accounting for fall speed and whether or not fastfalling was used. The point made here is that when hitlag is induced is a pretty important factor depending on the move used. Obviously, there are no real numbers as this is to get an idea of how shield tilt inducing hitlag earlier can alter L-cancel timing. If I had to throw out universal numbers, fast fallers would have a difference of 3-4 frames while floaties could be anywhere around 5-7 frames. It's nothing huge but enough to account for in a way, but that can be argued. I'm curiously interested in seeing TAS demonstrations of this though now to get real data.

EDIT: Eh, thinking over this I didn't really consider shields tilted away or closer to the character. In retrospect, it's probably not even that significant in respect as hitlag would still clue you into canceling on reaction.

 

[1MachGO]

IMO, one of the most interesting aspects of L-Cancelling is how it interacts with reaction times. People like to think of L-Cancelling as "mindless", but the truth is, the only habitual part of L-Cancelling is that you immediately think to do it after performing an aerial. Players become tied up with a rhythmic thought process as opposed to focusing on new information.

To elaborate, if a Fox player attacked with a nair, in a perfect world, they would grab if the nair hit and shine>grab if they were blocked. However, since the Fox player would have "I need to L-Cancel" in the back of their mind instead of "what is my opponent going to do?", flow chart plays such as these would be seldom seen.

 

[DarkStarStorm]

50% Lag reduction is good. If you opt not to use it... there's absolutely no advantage to it, meaning you better L-Cancel everything you can, right?


what kind of "punishment" for L-canceling could've been done to make it balanced? Having a limit of uses over a small ammount of time? decreasing the Lag redution after every use? Maybe that again, but with a restoration system like shields have? Dealing a small bit of damage to yourself by using it? Decreasing a bit of your shield for using it?

i know this would change a lot of things in Melee and PM, but just the same, what would be fairer to do with it, if it HAD to have some sort of drawback?

I would rather that L-canceling stay as it is, trying to change melee and stuff relating tends to get people's hackles up.

 

[The_NZA]

There is absolutely NO reason to punish people for using advanced techniques. Especially for something EVERYONE can use. That would just repeat PM's mistake in taking away the Ice Climbers chain grab infinites, or in fact ALL mods removing or nerfing chain grabs. And it would be repeating Sakuri's insistence in making smash into a casual game.

okay, this is obviously an overreaction and does not represent most of the anti-change l cancel voices.

 

[DarkStarStorm]

okay, this is obviously an overreaction and does not represent most of the anti-change l cancel voices.

Sorry, there was no reason for me to overreact like that. I didn't mean for it to sound like I was angry, I really wasn't. I was just trying to get a point across.

 

[TobiasXK]

this doesn't really have to do with "balancing" L-canceling or removing it wholecloth, since i think both of those are really silly notions; there's nothing imbalanced about L-canceling—it's just uninteresting—and flatly reducing all aerial landing lag by half, though it would remove a highly artificial barrier to entry into competitive Smash, is equally uninteresting.

but i do think that there are more interesting potential systems for lag-canceling in a way that fits with Melee/P:M's technical character. aerials in Melee already have frame ranges where aerials will auto-cancel on landfall, usually some small window before hitboxes appear and after they disappear. and back in the pre-Brawl days of "what advanced techniques would you add/remove to SSBR?", there was a decent amount of discussion of how that idea could be expanded to produce aerials that simply had differing amounts of lag based on when landfall occurred during designated ranges of the animation frames.

this wouldn't be simple automatic l-canceling; rather, frame ranges could be decided on a per-move basis related to the ideal uses of that attack. for instance, something like Captain Falcon's u-air has basically two ideal SHFFL uses: one is the instant u-air (which autocancels without fastfalling in Melee and can be autocanceled with a very brief fastfall as well, though it's usually just SHFFLed) and the other is the very late u-air that hits out in front and is used for horizontal carry. so in this advanced auto-cancel schema, those two timings would be rewarded by having the least landlag frames towards the beginning just after the frontmost hitbox and towards the end as the hitbox neared its end. and the median frames could be divided to make a bracketed scale of landlag times in some way.

i think a system like that could be really cool because it incentivizes really specific attack and fastfall timings and doesn't really change the fact that the biggest execution factors in continuing a link are the attack timings and moving on the first actionable frame. in addition, it actually does allow for lightshielding and shieldtilting and unexpected platform landing to potentially throw off cancel timings in some instances (because even if it happens sometimes in Melee, it's rare and is purely player error, since you can pretty easily alter your L-cancel timing reactively and be successful; in this system it could actually be forced on you in rare but potentially significant situations).

 

[DarkStarStorm]

this doesn't really have to do with "balancing" L-canceling or removing it wholecloth, since i think both of those are really silly notions; there's nothing imbalanced about L-canceling—it's just uninteresting—and flatly reducing all aerial landing lag by half, though it would remove a highly artificial barrier to entry into competitive Smash, is equally uninteresting.

but i do think that there are more interesting potential systems for lag-canceling in a way that fits with Melee/P:M's technical character. aerials in Melee already have frame ranges where aerials will auto-cancel on landfall, usually some small window before hitboxes appear and after they disappear. and back in the pre-Brawl days of "what advanced techniques would you add/remove to SSBR?", there was a decent amount of discussion of how that idea could be expanded to produce aerials that simply had differing amounts of lag based on when landfall occurred during designated ranges of the animation frames.

this wouldn't be simple automatic l-canceling; rather, frame ranges could be decided on a per-move basis related to the ideal uses of that attack. for instance, something like Captain Falcon's u-air has basically two ideal SHFFL uses: one is the instant u-air (which autocancels without fastfalling in Melee and can be autocanceled with a very brief fastfall as well, though it's usually just SHFFLed) and the other is the very late u-air that hits out in front and is used for horizontal carry. so in this advanced auto-cancel schema, those two timings would be rewarded by having the least landlag frames towards the beginning just after the frontmost hitbox and towards the end as the hitbox neared its end. and the median frames could be divided to make a bracketed scale of landlag times in some way.

i think a system like that could be really cool because it incentivizes really specific attack and fastfall timings and doesn't really change the fact that the biggest execution factors in continuing a link are the attack timings and moving on the first actionable frame. in addition, it actually does allow for lightshielding and shieldtilting and unexpected platform landing to potentially throw off cancel timings in some instances (because even if it happens sometimes in Melee, it's rare and is purely player error, since you can pretty easily alter your L-cancel timing reactively and be successful; in this system it could actually be forced on you in rare but potentially significant situations).

In Project M who would you say the best character to practice SHFFLs would be?

 

[PastLink]

Actually, believe it or not, but AGT can be performed in Melee, too.

really? i thought i remember hearing there was a time when people were all hyped about atgs... i suppose i misinterpreted why there was hype. ok well i still stand by the fact that PM should develop it's own unique techs.

 

[Lethalx]

Came back to this thread and it's still dumb. If you want auto L-Cancel play B+

 

[Player -0]

In Project M who would you say the best character to practice SHFFLs would be?

Your main.
________________________________________________
I think L-Canceling is kind of like learning to walk, first you try and figure it out; do you stand on one foot or the other? Once you figure it out then you can do it no problem but there are times where you screw up and trip or something such.
After learning to walk you can learn to run (or hobble as a baby). Learning to walk helped you figure out running (or in smash terms other technical things like SHFFL and WD'ing).

Either way it makes you not just throw out stuff lazily while playing, you have to L-cancel and actually play. Say you were playing Fox, if you were Shine -> SHFFL Nair -> Shine and repeating it, then you have a chance to screw up and accidently FH. Having the L-Cancel input being shield makes it so people who missed the attack button airdodge which lengthens the time for punishment; it also throws off the person's timing.

What I meant by "it makes you not just throw out stuff lazily while playing" I meant autocanceling kind of like MK's fair in brawl, you could just SH Fair and have practically no lag. Yes you still had to pay attention to the opponent but it made the game less exciting.

 

[GP&B]

really? i thought i remember hearing there was a time when people were all hyped about atgs... i suppose i misinterpreted why there was hype. ok well i still stand by the fact that PM should develop it's own unique techs.

It's doable in both but I believe there are certain properties in PM that allow AGT to cover more distance. Back before Demo 1.0, AGT used to carry all air dodge momentum, which thankfully got changed before release because it would be crazy OP then.

 

[PastLink]

It's doable in both but I believe there are certain properties in PM that allow AGT to cover more distance. Back before Demo 1.0, AGT used to carry all air dodge momentum, which thankfully got changed before release because it would be crazy OP then.

aah i see, thanks for clearing that up then.

 

[Sixth-Sense]

lol this thread reminds me of the melee legends thread haha

btw for all of you who want auto-l cancel, PM 3.0 released this super duper fun mode called "turbo" it basically l-cancels everything! even ground attacks and animations itself!

 

[gnosis]

Does shield hitlag actually affect l-cancel timing? Think of it like this - it's adding X frames to when you would land regardless of when you hit the shield. You are going to land on frame Z where Y is your normal short hop time and X is hitlag from the shield regardless of when you actually hit the shield. Hit it right when you leave the ground, right at the top, right before you touch the ground - doesn't matter. X+Y is always going to equal Z. Am I misunderstanding something here?

So shield hitlag shouldn't throw off l-cancel timing... assuming you hit your fastfall on the first possible frame still. And this is what shield tilting actually messes up that can affect l-cancel timing. You can't fast fall during hitlag, so if they tilt it up so you hit at the apex of your jump when you would normally be putting in your down input, it won't go through and you'll fall slower and maybe miss the l-cancel.

BUT, even with that, you can just spam L. It's not teching, you don't get punished for a 'missed' l-cancel timing.

So the whole 'shield tilting to throw off l-cancel timing makes it deeper' thing is kind of moot with how the system actually works currently. Shield tilting is still good though, because even if they do still hit their fast fall and l-cancel, you still take the blow sooner, start your shield stun sooner, but they don't land and recover any sooner, so you get a little boost in terms of frame advantage.

 

[Vigilante]

We balance the game around L-cancelling existing, so there's no need to change the mechanic at all, methinks.

 

[Sixth-Sense]

We balance the game around L-cancelling existing, so there's no need to change the mechanic at all, methinks.

the discussion is now over, l-cancel will be there forever

 

[Shell]

As extra food for thought iirc Peach and Kirby both have aerials with land lags that allow them to dodge some attacks/grabs, and this dodge is more effective when not abbreviated by L-canceling. Also 3.0 Ivy has a nifty wind push on Synthesis Start which gains a lot of value by allowing Ivy to throw off opponent's L-canceling timing and capitalizing on the full landing lag. These are both nuances that are lost by Auto L-canceling.

 

[Yashichi]

What if when you L-cancelled with different buttons it made you flash different colors than just white? How ill would that be

 

[JOE!]

the discussion is now over, l-cancel will be there forever

However, is it balanced around the action of doing it or the reward? If it is balanced under the assumption of good play more or less halving landing lag... Then there's not much difference between that a.d balancing around lag that is half of what we normally see.

 

[Sixth-Sense]

However, is it balanced around the action of doing it or the reward? If it is balanced under the assumption of good play more or less halving landing lag... Then there's not much difference between that a.d balancing around lag that is half of what we normally see.

wrong person haha

 

[JOE!]

Damn phone, lol

 

[W¡ndy]

People keep talking about lightsheilding.
but what if you aren't using gamecube controller

 

[Vigilante]

In tournament play, we assume people will know how to L-Cancel. Not taking into the mechanic into account will screw over the entire balance ecosystem.

When you balance a character, you must try to look at it a every possible angles, and also, try to anticipate what the character will play like at the highest possible level. If L-canceling is advantageous and fairly easy to learn over time, then of course, we will assume that it will be used.

 

[Zef]

please no god not again