Q&A Mechanics & Techniques Discussion

[TheReflexWonder]

I just wanted to point out that not being able to buffer out of a Ledge Climb doesn't stop you from being able to shield frame 32 in your example. I'm not sure that you can't buffer moves during Ledge Climbs (AFAIK only buffering a Ledge Climb is impossible), but let's assume you can't. All that means is the 10 frame window for inputs turns into a 1 frame window. Only glitches like Brawl's dead frame during jump squat prevent you from inputting stuff immediately after an action ends.

Basically the 1 frame window of vulnerability is in fact only 1 frame, not 2.

If that is the case, why did Squirtle and Brawl ZSS get guaranteed Frame 1 Jab1 out of a grab release? I was under the impression that the inability to buffer an option as a grounded grab release victim was why that happened, not that their grab release animations were uniquely 29 frames.

 

[Big O]

If that is the case, why did Squirtle and Brawl ZSS get guaranteed Frame 1 Jab1 out of a grab release? I was under the impression that the inability to buffer an option as a grounded grab release victim was why that happened, not that their grab release animations were uniquely 29 frames.

You had a 1 frame advantage on grab release in Brawl, except on DK, who could act 10 frames earlier than normal, and the Mother kids who acted 10 frames later. Both characters could buffer w/e normally too IIRC. Regardless of how it may or may not have worked in Brawl, Ledge Climb only has 1 frame of vulnerability based on actual in-game testing/results.

 

[icraq]

Do shields continue to go down in strength at their normal decay whilst also taking hits?

I'm asking for situations with Ryu, like with his orange fireball, it has 4 or 5 multi hits on it and traps opponents in shield. Are they only taking shield damage while it's hitting them or is shield decay happening simultaneously?

 

[TheReflexWonder]

I'm pretty sure that as long as the shield is up it is depleting in health, so it would be happening simultaneously.

 

[teluoborg]

I remember reading somewhere that a shield doesn't lose health while in shield hitlag /and hitstun. I'm unable to remember if it's both or just hitlag, and where I did read that.

 

[STiCKYBULL3TZ]

Can someone explain how acting out of a spike bounce works? Sometimes people immediately airdodge or Nair after the bounce. Other times people are stuck in hitstun and open for followups. Is there a window where you can input an action directly after the bounce?

 

[q_e_d]

So right now there appears to be wrong information about the duration of shield stun in Smash 4 (or how it was changed from Brawl).

The oldest thread I found says shield stun was lowered a lot, but apparently the value for Brawl there is wrong (way too high): smashboards.com/threads/shields-in-smash-4.381183/

A reddit thread says the stun is actually very slightly higher in Smash 4, by one frame for some specific damage ranges: reddit.com/r/smashbros/comments/2vh3fj/so_a_mod_on_smashboards_did_some_tests_with/

Yet another smashboards comment however got (in Smash 4) higher shield stun duration than that for low-damage moves, but the same (or similar) for high damage ones: smashboards.com/threads/character-competitive-impressions.367669/page-173#post-18101666

(The ssbwiki page says shield stun was lowered, but doesn't provide any specific values, a formula, or sources: ssbwiki.com/Shieldstun )





So since I don't have the necessary equipment, can anyone test this again and ideally provide some proof (slow frame by frame gifs or similar)?

 

[I speak Spanish too]

I am confused with the exact inputs in this video
https://www.youtube.com/watch?v=Ppw-igC6Uog

 

[ぱみゅ]

It's basically (,), the window for the direction shift can be pretty tight, but it's there.

 

[I speak Spanish too]

It's basically (,), the window for the direction shift can be pretty tight, but it's there.

So Fox trot once, do the fox trot again and hit back on the control stick. Are you sure their is no delay between the second foxtrot and the dash back? If I go to fast I end up perfect pivoting. Maybe its just me.

 

[Fox Is Openly Deceptive]

Can someone explain how acting out of a spike bounce works? Sometimes people immediately airdodge or Nair after the bounce. Other times people are stuck in hitstun and open for followups. Is there a window where you can input an action directly after the bounce?

I'm not entirely sure what the calculations are or why the game does this, but I can tell you that it has nothing to do with a window in which you do the input.
The opponent goes through the same (or very similar) amount of hitstun if they are spiked in the air and act (e.g. use an aerial) in the air before touching the ground, compared to when they are spiked from the ground and then act in the air after that immediate ground bounce. The special one is when they are spiked from the air then bounce off the ground. When the opponent bounces off the ground from the air, their required hitstun is re-calculated; again I'm not sure what goes into this. It seems to depend on their percent, but at lower percents you could be looking at something like a third of the original hitstun, quite possibly less, and then at least part of this hitstun must be counted from the moment they were originally hit, because often you'll find that the opponent is able to act immediately upon hitting the ground, presumably because they've already gone through the required amount of hitstun while getting sent towards the ground. It's certainly not always the case that they'll be able to act immediately though, and this will depend upon what percent they were on and how high up they were hit from. The higher their percent, the more it seems to be the case that time spent in the air before hitting the ground doesn't help as much as time spent in the air upon leaving the ground. So it's not a matter of there being a specific window in which you have to input it. In fact under the right circumstances the action (e.g. an aerial) can be buffered before you even touch the ground. The up-shot is that when getting spiked to the ground from the air, if you're on low percents, your best option will often not be teching it.

 

[Tobi_Whatever]

So Fox trot once, do the fox trot again and hit back on the control stick. Are you sure their is no delay between the second foxtrot and the dash back? If I go to fast I end up perfect pivoting. Maybe its just me.

Imagine the part like the start of a PP. It's basically instant.

 

[zwataketa]

I have a question. Two actually.

One, how do I shield grab? Yes, I know the button combo, but how do I find opportunities to shield grab? Normally my instinct is to punish my opponent (if able) with a forward smash, dash attack, or something else. I can see the effectiveness of shield grabbing, but I need to know how to develop that habit, because dash grabbing only seems to work so much (and I'm not quite sure how the pivot grab works).

Also, I was trying Wii Fit Trainer's Up Throw to Up air combo the other day, and it was working well, most of the time. There were times when the opponent was at very low percentage and still was able to air dodge (not always punish, but nonetheless air dodge). Granted, it was a level 9 cpu, but also it was a Dedede, and Dedede is one of the heavier characters, so whaaaa? I believe I was inputting the commands fast enough ...

 

[I speak Spanish too]

I was looking at soft/hard landing lag frame data and basically wanted to ask about what actions make you go into this landing lag. My best assumption is that if you fastfall to the ground, at any height, you will go into hard landing lag provided you don't use an aerial. If you do not fastfall to the ground you will get soft landing lag which is usually half the amount of frames you would suffer if you would go into hard landing lag, provided you don't use an aerial. If you an aerial while airborne, the autocancel windows and landing lag is specific to that aerial. However, these are just guesses and I really don't know. Pikachu's Nair for example makes you go into a lot of lag if you fast fall(If you fastfall Pikachu's nair you miss the autocancel window) it so maybe the landing lag and the hard landing lag fuse?

Aside from confirming the information above, my biggest question is about Dodge Canceling and Frame Canceled Aerials. So characters, like Falco and Ganondorf can frame cancel their D-air reducing the amount of lag significantly allowing follow-ups. However, when they Frame Cancel their D-air do they go into any landing lag at all? In order to frame cancel your d-air you would have to fastfall the first hit of the D-air, So I assume that the fastfall landing lag(hard landing lag) replaced the d-air landing lag. It's hard to tell if their is any landing lag at all because of the amount of hitlag they go into to.

My next question is dodge canceling. Most of you are aware of this, but if you buffer an air dodge out of a short hop and perform a specific aerial the air dodge landing lag will be canceled out and replaced. Some character can dodge cancel one of their aerials and land with seemingly no lag, such as Lucas. https://www.youtube.com/watch?v=tcMHm1xBRf0

From the video, it is definitely was faster then SH air dodge, and Lucas doesn't seem to be going into any lag. However, when he lands white smoke appears which I think signifies soft landing lag. Lucas soft landing lag is 2 frames so that is why it appears to be almost instant, but i'm ultimately not sure.

It seems that whenever your airborne, you will go into landing lag no matter what you unless you land in the autocancel window of your aerial, since every special, or aerial, as well as soft/hard landing lag will put you into lag. I'm unaware if any character can SH airdodge and landing in the autocancel window of that aerial. If so they will have 0 landing lag.

 

[Rikkhan]

I been experimentiing with the possibility of using L button as a jump button (R or Z are also an option), and I think I found something interesting, but first let me explain the situation, normally when you press jump + upAir you will end up doing an Up Smash, this problem can be solved by using jump + C stick however it introduces another problem, its beyond awkward to press Y/X + C stick, the workaround its to set a shoulder button to jump so I decided to use L as jump and therefore naming it L-Uair. its not a perfect solution because shoulder buttons have terrible sensitivity, so its pretty hard to do a short jump.

So what its this about? using L-Uair makes easier to do a perfect jump + Uair, since the attack comes faster it reduces your landing lag and also makes the active hitbox comes faster. You can test this going into training mode and picking Fox, doing a standar short jump + Uair will result in fox having landing lag, a perfect short jump + Uair will result in no lag. As a plus L-Uair let you carry the momentum of a dash. Captain Falcon for example is able to hit characters on the ground with a perfect short jump + up Air.

Usages:
So far the characters I found that benefit from this are Pikachu, Mario and Falco and maybe falcon, Diddy and Yoshi. Some examples:

Pikachu:
dash + reverse L-Uair = hits targets on the ground.
dash + reverse L-Uair + HSB = kills anyone around 60%+ near a ledge.
foward throw + reverse L-Uair + HSB = kills anyone around 50%+ near a ledge.

Mario:
dash + L-Uair = hits targets on the ground.
dash + L-Uair + Fair = spike kill around 100%

Falco:
down throw + reverse L-Uair + Uair + back air = 38% damage on almost anyone around 20-50%
down throw + reverse L-Uair + jump + back air = can kill big characters at 70% near the ledge.

Captain Facon
L-Uair = hits targets on the ground.
L-Uair + fast fall + L-Uair + fast fall + L-Uair + jump + Uair = 43% damage, good OoS option.

 

[Pazx]

This requires setting c stick to attack, right?

 

[Rikkhan]

yes it requires c stick to attack, i'll try to upload a video to show how this works.

 

[MajorMajora]

Kind of reminds me of how I use x for grab sometimes.

My main point of confusion is why you can't use x or y to jump and then hold up and A to do the up air?

 

[DunnoBro]

Lots of top players already use a shoulder jump button. Dabuz, Angel Cortez, Ally. (Though ally uses it with smash c-stick so he probably only uses it for jc pivot usmash)

Dabuz and Angel use characters who make liberal use of uairs so that's probably helped them a bit.
They use z-jump though. As do i, it's the only shoulder button you can shff repeatedly comfortably with.

Kind of reminds me of how I use x for grab sometimes.

My main point of confusion is why you can't use x or y to jump and then hold up and A to do the up air?

It'll turn into a jc usmash unless you're frame perfect.

 

[Shadó Chimera / シャドーキメラ]

Autocancelled aerials give 4 frames of landing lag. If you land during an airdodge you recive airdodge lag, but airdodges like most moves can be acted out of before it finishes which is why you can cancel airdodge lag.

The only way to get no landing lag is to land with a special that doesn't cancel when you land, and if you can cancel that move (like Robin/Diddy Neutral B) then you can use that to use specific actions quicker than just landing and using them.

 

[Ffamran]

This is similar to double sticking and some other stuff VaBengal, Tuen, and others figured out. Conversation started around here: http://smashboards.com/threads/mechanics-techniques-discussion.368206/page-34#post-19587915. I think this is what's being described. Not trying to take away anything you figured out, but this should belong in the Mechanics & Techniques Discussion.

Update on the buffered short hop up air thing:

Forget all that junk I said before. It's junk! (OK, knowing there are 3 control stick throws of importance could warrant more research... BUT NOT NOW).

Turns out, this guy had a conversation with @VaBengal on reddit and they made an epic finding: L to jump, tap jump off, c-stick to attack. Do that, then do this:

L + Control Stick UP + C-Stick UP

Do that at once and you get a buffered up air. Apparently throwing the control stick up + jump mimics the double stick effect. This yields buffered short hop up airs for all! This also works out of shield for everyone that does not have a Zair. So, for people like ZSS... well, that's still a mystery.

Shoutouts to the ZSS boards, I love those guys. Credit to the above and not me, ha ha.

 

[I speak Spanish too]

Autocancelled aerials give 4 frames of landing lag. If you land during an airdodge you recive airdodge lag, but airdodges like most moves can be acted out of before it finishes which is why you can cancel airdodge lag.

The only way to get no landing lag is to land with a special that doesn't cancel when you land, and if you can cancel that move (like Robin/Diddy Neutral B) then you can use that to use specific actions quicker than just landing and using them.

Ok, good to know. So in the case of Lucas when he SH buffered Air dodge into back air would that give him autocanceled landing lag? Also, @KuroganeHammer what creates soft/hard landing lag?

EDIT: So now I can see what causes frame cancels and the amount of lag

Thanks for using this!

So, I've done more research since I scrawled that, and my current theory is in this string of tweets: https://twitter.com/Chispshot/status/554845009779122176

Summary: If you fastfall dAir, on Frame 17 of the aerial attack, Falco will attempt to both calculate an air->ground transition and calculate a hitbox->hurtbox collision. The latter takes priority, and Falco clips through the floor, and the landing lag flag is unset. On the next frame, it notices Falco SHOULD have landed, sees no landing lag flag (which was previously set by 170 on Frame 5, and then unset on Frame 17), so it puts Falco into SoftLanding (which is generally 3 frames of lag).

So when frame canceling an aerial, if successful you will go into soft landing lag which is usually 2-3 frames according to KuroganeHammer. An autocanceled aerial will make you go into 4 frames.

So at this point, two of my questions still remains. Aside from frame canceled aerials what else puts you into soft landing lag. As well as what puts you into hard landing lag. Also, when Dodge Canceling for Lucas how many frames of landing lag do you go into then? Idk if that aerial is autocanceled but if it is that would be 4, and if its a hard landing lag in Lucas case it would also be 4 frames of landing lag, however, if its a softlanding it would be 2 frames.

 

[Dr. Tuen]

Ok, good to know. So in the case of Lucas when he SH buffered Air dodge into back air would that give him autocanceled landing lag? Also, @KuroganeHammer what creates soft/hard landing lag?

EDIT: So now I can see what causes frame cancels and the amount of lag


So when frame canceling an aerial, if successful you will go into soft landing lag which is usually 2-3 frames according to KuroganeHammer. An autocanceled aerial will make you go into 4 frames.

So at this point, two of my questions still remains. Aside from frame canceled aerials what else puts you into soft landing lag. As well as what puts you into hard landing lag. Also, when Dodge Canceling for Lucas how many frames of landing lag do you go into then? Idk if that aerial is autocanceled but if it is that would be 4, and if its a hard landing lag in Lucas case it would also be 4 frames of landing lag, however, if its a softlanding it would be 2 frames.

I believe hard and soft landing lags are caused by slow or fast falling into the ground. I would imagine that a frame synced aerial would cause soft landing lag because all aerials stop fast falls (naming pet peeve: frame sync > frame cancel. nothing's being canceled, the landing lag and the hit stun are overlapped).

 

[I speak Spanish too]

Yeah, this also my guess. But, I have no way to prove it aside from the fact that when I empty hop land, then fast fall and land I feel like I can act faster. It's most notable with Mewtwo since his soft landing lag is 2 frames but his hard landing lag is 5. I guess I can safely assume that my guess is true. Only thing I'm unsure about now is the lag from dodge canceling. I can imagine it's a soft landing though because you are not fastfalling.https://youtu.be/tcMHm1xBRf0?t=200

 

[Rikkhan]

So this a little example on falcon, the first two hits are imperfect Uairs the third is perfect, you can do a perfect Uair without recurring to shoulder buttons but imo is harder and you risk to doing a Usmash or whiffing.

 

[Raijinken]

I can't give precise data myself (I get it all by hearsay), but it is worth noting that your shield drops faster if it takes a hit first. Shielding then dropping without blocking is slower.

 

[David Viran]

I can't give precise data myself (I get it all by hearsay), but it is worth noting that your shield drops faster if it takes a hit first. Shielding then dropping without blocking is slower.

Actually it is slower.

http://smashboards.com/threads/mechanics-techniques-discussion.368206/page-30#post-19312998

 

[Raijinken]

Actually it is slower.

http://smashboards.com/threads/mechanics-techniques-discussion.368206/page-30#post-19312998

There's that hearsay stuff playing tricks on me again. This is why I should stick to non-mechanics theory.

 

[I speak Spanish too]

So this a little example on falcon, the first two hits are imperfect Uairs the third is perfect, you can do a perfect Uair without recurring to shoulder buttons but imo is harder and you risk to doing a Usmash or whiffing.

We talked about this a few pages back and ulitmately it just depends on how consistently you can do it and what's most comfortable. I personally set B to Jump since It's so close to c-stick all i have to do is slide in the respective direction. However, it's difficult to do this when trying to buffer FH aerials. But, I'm willing to get used to it because it's so comfortable. It would be cool if scuf could give the gamecube controller and sick new upgrade.

 

[Rikkhan]

This is similar to double sticking and some other stuff VaBengal, Tuen, and others figured out. Conversation started around here: http://smashboards.com/threads/mechanics-techniques-discussion.368206/page-34#post-19587915. I think this is what's being described. Not trying to take away anything you figured out, but this should belong in the Mechanics & Techniques Discussion.

Yep this exactly what I was talking about, actually using L for full jumps and double sticking for short hops seem to be the best option.

I think is interesting that a lot of combos and approches are unlocked when you do frame perfect Uairs, falcon can combo UAir into knee very easily or pikachu that can combo HSB from a grab.

 

[I speak Spanish too]

This could be a very groudbreaking technique but I can't understand it well.
https://www.youtube.com/watch?v=ET1PZwOJ6mU&index=18&list=PLFXOgPi6_N5Q1AR3Z9l549HBoZzpBAz0h
Basically, what they are saying is that with this Option Select in a falling aerial on shield situation, which happens alot, you don't have to guess what they will do and can always punish. The inputs confuse me though.


https://www.youtube.com/watch?v=o4Ncyx31pPM&list=PLFXOgPi6_N5Q1AR3Z9l549HBoZzpBAz0h&index=19
However, i dont understand this video.

 

[TheAnomaly]

This could be a very groudbreaking technique but I can't understand it well.
https://www.youtube.com/watch?v=ET1PZwOJ6mU&index=18&list=PLFXOgPi6_N5Q1AR3Z9l549HBoZzpBAz0h
Basically, what they are saying is that with this Option Select in a falling aerial on shield situation, which happens alot, you don't have to guess what they will do and can always punish. The inputs confuse me though.


https://www.youtube.com/watch?v=o4Ncyx31pPM&list=PLFXOgPi6_N5Q1AR3Z9l549HBoZzpBAz0h&index=19
However, i dont understand this video.

The second video is the same principle but you are choosing to make the inputs in such a way that a roll/dodge comes out after the attempted Perfect Shield instead of grab should you not get the perfect shield into counterattack option. They then buffer their next option out of the end of the roll/dodge frames to make a very quick counterattack depending on the situation.

 

[Ffamran]

Has anyone figured out what causes knockback for Luma, Duck Hunt's Can, and other stuff like say, Pac-Man's Hydrant? I think that Hydrant's just "health-based". Indigo Jeans said that Luma takes knockback after a certain knockback threshold is incurred, but nobody's figured out what that threshold is.

 

[Lavani]

Just think of Luma as a standard fighter. The more damage it's taken the farther it flies, and if the attack isn't strong enough to send it tumbling it flinches instead.

 

[Ffamran]

Just think of Luma as a standard fighter. The more damage it's taken the farther it flies, and if the attack isn't strong enough to send it tumbling it flinches instead.

I know that, but what about knockback? Two Ftilts from Falco don't do anything. Only the sixth one will, but that's a dead Luma. The fourth Dtilt sends Luma into tumble. Both moves do 9%, but Ftilt only has knockback growth of 100 while Dtilt has 35 base and 88 growth. I think this is the same case with Side Smash and Bair too. A late Side Smash can send Luma tumbling with 1 hit, but Bair takes two hits before Luma is sent tumbling. Late Side Smash has 20 base and 76 growth while Bair only has 130 growth.

Edit: Wrong about Side Smash, but apparently landing Fair and the last hit of Nair can cause it to tumble. Landing Fair does 3% and has 50 base and 160 growth. Last hit of Nair does 4% and has 49 base and 130 growth.

 

[DanGR]

I'm posting here so I don't clutter the patch notes thread with my questions.

@ A_Kae
I don't understand this hitlag formula you posted in the old patch thread.
http://smashboards.com/threads/luca...nity-patch-notes.406319/page-31#post-19501103

((d/2.6)+5)*h
d = Attack Damage before staling.
h = Hitlag Modifier

If I plug in the data for the second hit of Rosalina's downsmash into this formula...

Rosalina second hit of down-smash:
1.5 hitlag modifier
9% damage

((d/2.6)+5)*h
((9/2.6)+5)*1.5
=
12.69 frames of hitlag?
It seems very high. Is that correct?

----

Also of note, it was found recently that if an attack's hitlag multiplier is above 1.0, the total hitlag of the attack is divided by 1.25.
http://smashboards.com/threads/tour...nity-patch-notes.412130/page-29#post-20031057

This reduces the hitlag of Rosalina's down-smash by 2.54 frames, down to 10.15 frames. Is that also correct?

 

[A_Kae]

I'm posting here so I don't clutter the patch notes thread with my questions.

@ A_Kae
I don't understand this hitlag formula you posted in the old patch thread.
http://smashboards.com/threads/luca...nity-patch-notes.406319/page-31#post-19501103



If I plug in the data for the second hit of Rosalina's downsmash into this formula...

Rosalina second hit of down-smash:
1.5 hitlag modifier
9% damage

((d/2.6)+5)*h
((9/2.6)+5)*1.5
=
12.69 frames of hitlag?
It seems very high. Is that correct?

----

Also of note, it was found recently that if an attack's hitlag multiplier is above 1.0, the total hitlag of the attack is reduced by 1.25.
http://smashboards.com/threads/tour...nity-patch-notes.412130/page-29#post-20031057

This reduces the hitlag of Rosalina's down-smash by 2.54 frames, down to 10.15 frames. Is that also correct?

That's all correct, yes.

I should say that decimals are ignored for hitlag, so 12.69 is just 12, and 10.15 is just 10. Also, the hitlag reduction doesn't go below a 1.0 modifier, so if the modifier is 1.05, instead of being .84, it's 1.0.

 

[Ulevo]

How do stale move effects work in this game compared to Brawl?

 

[TheReflexWonder]

Far as I know, it's the exact same. A basic Google search seems to suggest the same.

It had a much greater effect on KO potential in Brawl because DI was more useful and because momentum canceling existed.

 

[busken]

Can you jump out of an extended dash dance?