Data Kadano's perfect Marth class -- advanced frame data application

[Bones0]

1. Z-cancelling was in 64, so I find it very hard to believe Sakurai couldn't execute one of the most necessary techniques in his own game. It seems even less likely that he would create such an elaborate KB and hitstun system and not take into account that you can cancel half the lag of your aerials. The fact that you can only cancel half as opposed to the full lag that was possible in 64 demonstrates to me that he knew what amount of lag was appropriate for aerials.

2. Totally agree on this point. If anything about Melee is to be considered a jackpot, it should be wavedashing/wavelanding. While I don't think Melee would be NEARLY as good without it, I still think it'd be a really amazing fighter with tons of depth. I think the overall effect would be a more simplified neutral game, particularly with platforms, but when you compare Melee's current neutral game to a traditional fighter, it's basically footsies on steroids so I definitely don't think it'd be one dimensional or anything. I also think it's reasonable to say that while none of the devs could have fathomed what WDing has become, it seems likely there was a general understanding you could do simple stuff like jump up through a platform and airdodge downwards onto it to land faster. Maybe I'm overestimating the competitive ingenuity of the devs though.

3. Not sure what you are getting at in terms of Fox being easy to SD with. Even today where 20xx is the talk of the town, the difficulty in using spacies is still highly considered in overall gameplay. All characters have to take into account the difficulty of their given options, and I think these risks are accentuated with spacies. The notion that these risks don't apply to really competitive players, let alone the godlike android Smashers we have today, seems naive. Even if a developer was trying to make a hardcore fighter, they'd likely never consider the game would get pushed to a level where players virtually never miss L-cancels, short hops, or even avoid side-Bing off the side of the stage. The fact that such things are infrequent in today's metagame is simply a testament to how insane Smashers are and how long they've been practicing for.

You are probably right that the game is focused on FFA, but I don't think that means 1v1 was ignored entirely. There aren't many exploitable tactics that are prevalent in 1v1 but not FFA. Chain grabbing is the major one, but it only really applies to a small subset of characters and usually only on FD. I believe Sakurai's desire to maintain character personality and limitations on development time are the main factors that prevented Sakurai from achieving a full, well-balanced cast.

While I agree there were many coincidences that helped play a role in Melee's complexity, I don't think that means it would be a subpar game without those coincidences. Even without things like WDing or jab resets or ground teching, the game would still be a deep fighter with tons of room for improvements.

 

[SpiderMad]

I think I recall someone pointing out that the interview where he was talking about Fox, he was referring to Brawl, not that this changes much

 

[Bones0]

Not to kill the current discussion, but Kadano, would you happen to know if there is a difference between smashing the stick during a DD vs. tilting? I am trying to incorporate shield stops during CGs as a more reliable pivot grab. Sometimes it works perfectly and I shield stop grab. Other times however, I roll. I know I researched this before and found the first 3(?) frames of dash will result in a roll even if you press the stick before L/R. However, some of the shield stops I am getting definitely don't look like they are after 4 frames of dashing in the opposite direction. I might just be shielding on the pivot frame (thus eliminating the purpose because I could just grab with the same difficulty), but it feels easier than pivot grabbing.

My current leading theory is releasing the stick allows you to shield during the initial frames of dash, but if that's the case then idk why when I previously tested this that I would ever roll (I was unpausing only holding R, not the stick).

 

[Kadano]

Could you do a few gifs of standard and reverse sweet and sour spots on marth's up b please?

So I suppose you know these images already?:

Only frame 5 has a sweetspot hitbubble. Afterwards, the only difference is the knockback base angle; damage and knockback magnitude are the same.

Reversing and angling takes effect starting on frame 6, so the sweetspot hitbubble is never affected.

Due to interframe hitbubble blending, the hitbubbles are stretched out very much. The one with more outward extension (both on the left and on the right side) is the one with 45° knockback angle.
[Sorry about the graphical dust-spike effect cluttering up the hitbubbles. Last time I tried to remove it, it ended up zeroing Marth’s up-B height gain as well … I’m bad at these things.]

@ Bones0 : I don’t think I understand what it is that you want me to find out.

 

[Metal Gear Salad]

that's awesome, thanks.

is there any way to get an animated reverse up b gif showing max range while still hitting the sweet spots please?

thanks for the stills, not seen that last one before. that's what i'm really interested in seeing, preferably slowed down so we can see the interframe blending.

 

[tauKhan]

@ Bones0 I tested this and it seems that if you release the stick one frame before pressing shield you don't roll. My method was unreliable, so I would like someone else to confirm it though.

@ Kadano I think Bones0 wants to know how roll input works. So will you do a roll if you release the stick and shield while the buffer window for roll is still active. Roll definitely happens if you release stick and shield on the same frame.

 

[Bones0]

Yeah, I mean I'm basically just looking for anything that will allow me to shield stop grab without rolling and without worrying about how quickly after pivot I shield. Having to wait 3 frames after the pivot frame before shielding seriously hinders my desired purposes for the shield stop in the first place. @ T tauKhan If you are right that the only way to do it is to release the stick and shield at least one frame after, then I may just revert back to JCing my grab because accidentally rolling is way too big of a flub vs. a slightly off spacing with a JC grab.

 

[tauKhan]

@ Bones0 You could also increase your chance of pivot grab by empty pivoting the grab. I know some players do that, but I just jc them.

 

[Kadano]

is there any way to get an animated reverse up b gif showing max range while still hitting the sweet spots please?

Max range of what? Horizontal movement by selecting the forwardmost angle? What do you mean by “hitting the sweet spots”? Please be precise, doing these gifs takes some effort and I don’t want to do something else than you had in mind.

@ Bones0 : You can shield stop immediately during a dash done out of Turn. Even on frame 1-3 of the dash. Shielding for 1 frame is sufficient to get standing grab (and not dash grab).

 

[Metal Gear Salad]

Sorry

I meant 2 gifs of marth standing as far away from another character so that he can just hit his strong-knockback up b and reverse up, respectively.

Also, does the underneath hit box of the reverse up b only come out in the air or does it exist while grounded too? (Basically, is the up b hit box the same grounded as aerial)

Thanks

 

[Kadano]

@ Metal Gear Salad

 

[Metal Gear Salad]

thanks, i didn't realise the regular up b had quite so much range on it. much appreciated kadano, thanks a lot.

 

[tauKhan]

Max range of what? Horizontal movement by selecting the forwardmost angle? What do you mean by “hitting the sweet spots”? Please be precise, doing these gifs takes some effort and I don’t want to do something else than you had in mind.

@ Bones0 : Frames 1-3 of your dash can’t be interrupted with shield, only with roll. Regardless of when you let go of the stick. Frame 4 of your dash is a semi-dead frame where the only affective input is jump. So frame 5 is the first one where your shield can start. Except for when you shield during the pivot / smash turn frame, but as you said, if you aim for that, you can just go for pivot grab.

What I do is that I press both Y and Z, but Y ever-so-slightly earlier. I aim to keep them apart only half a frame or so. This way, I ensure that I get the frame perfect pivot grab most of the time, and if I miss it, I still JC grab instead of dash grab. Does this qualify as an option select? Haha.

Are you sure about this? I'm pretty damned sure that when I do a dash turnaround and shield on the first (or second) frame of the dash out of turn, I cannot roll, but will instead shield even if I hold back when pressing shield. I tested this using pause method and watched at which state of dash animation I was.

However, if you shield on the pivot frame, you will roll if you continue pressing forward.

 

[Kadano]

Are you sure about this? I'm pretty damned sure that when I do a dash turnaround and shield on the first (or second) frame of the dash out of turn, I cannot roll, but will instead shield even if I hold back when pressing shield. I tested this using pause method and watched at which state of dash animation I was.

You are right. I tested for a dash out of Wait, but dashes out of Turn behave differently. I’ll update my post. Thank you!

 

[tauKhan]

@ Bones0 : You can shield stop immediately during a dash done out of Turn. Even on frame 1-3 of the dash. Shielding for 1 frame is sufficient to get standing grab (and not dash grab).

Actually I don't think you can shield on the 1st frame of dash out of turn, since that requires you to have your stick backwards which results in roll. You will have to empty pivot to be able to shield at the same time you would pivot grab. So you can start shielding from frame 2 of turn dash, and don't even have to let your stick back to neutral if you do.

 

[Kadano]

Actually I don't think you can shield on the 1st frame of dash out of turn, since that requires you to have your stick backwards which results in roll. You will have to empty pivot to be able to shield at the same time you would pivot grab. So you can start shielding from frame 2 of turn dash, and don't even have to let your stick back to neutral if you do.

No. When you press L/R and still hold forward (or backward, if you go from the initial orientation), you won’t roll unless you hold forward for another frame.

If you press and hold L/R on the same frame as the pivot input, you will first turn and shield on the frame afterwards. “Pivot shield grabs” are really easy, it seems.

 

[tauKhan]

Thanks, so on the pivot frame shield out prioritizes roll, I didn't think of that. Also is the buffer window for roll input longer for forward roll? It seems that when I input shield on 4th frame of dash (I pause when 3rd frame of it is being shown on the screen) I get a roll, but when dashing back I get shield frame 3 but no roll on frame 4.

 

[Xyzz]

This whole issue just serves to remind me that melee is stupid :D

 

[Kadano]

Thanks, so on the pivot frame shield out prioritizes roll, I didn't think of that. Also is the buffer window for roll input longer for forward roll? It seems that when I input shield on 4th frame of dash (I pause when 3rd frame of it is being shown on the screen) I get a roll, but when dashing back I get shield frame 3 but no roll on frame 4.

I tried to replicate what you observed, but I haven’t been able to get a forward roll at frame 4 of the dash (regardless of whether it was done out of Turn or Wait) by holding R and forward. I always stayed in the shield animation.

 

[tauKhan]

Perhaps there is 1 more frame than I thought in the dash animation. I inputted shield when foxes dash was on the 1 frame before his other feet lifts off the ground and got roll. Oops, I think there is 1 frame less in the dash than I thought, and the one where fox lifts his feet is 3rd one.

 

[Bones0]

No. When you press L/R and still hold forward (or backward, if you go from the initial orientation), you won’t roll unless you hold forward for another frame.

If you press and hold L/R on the same frame as the pivot input, you will first turn and shield on the frame afterwards. “Pivot shield grabs” are really easy, it seems.

Why are they easy? Don't you have to be frame perfect the same way you would have to be with a normal pivot grab? Or is it a 2-frame window because you can input shield the same frame as the pivot input as well as the frame of the pivot output (<-- hey, look at that lol)?

 

[tauKhan]

They are easy because you have 1 extra frame to release the stick. You can't roll on the pivot frame since shield takes priority. If you don't want to move at all then you have to be frame perfect with the shield input ofc.

 

[Kadano]

Why are they easy? Don't you have to be frame perfect the same way you would have to be with a normal pivot grab? Or is it a 2-frame window because you can input shield the same frame as the pivot input as well as the frame of the pivot output (<-- hey, look at that lol)?

For pivot frames, your inputs need to be like this:

Frame 1 (last frame of the first dash)
Smash the stick backwards → frame 2: pivot / turn frame
Press Z or L/R+A → frame 3: grab starts

If you press grab one frame too early, you grab in the wrong direction. If you grab 1 frame too late, you do a dash grab.

My pseudo option select method has slightly better leniency:

Frame 1 (last frame of the first dash)
Smash the stick backwards → frame 2: pivot / turn frame
Press Y and, with minimal delay, Z → frame 3: grab starts

If you grab 1 frame too late, you do a JC grab (with considerable sliding).

Now to the shield stop pivot grabbing method:

Frame 1 (last frame of the first dash)
Smash the stick backwards → frame 2: pivot / turn frame
Press and hold L or R → frame 3: shield, your movement stops completely¹
Press A or Z on any consecutive frame

You can delay both the shield and the grab input as much as you want.

At the lowest percent range where standing regrab will miss against a no-DI Fox (24% after uthrow), you have 6 leniency frames if you do a pivot grab. You can easily use them to, say, dash away 2 frames more than necessary, shield 2 frames later than possible and press L/R+A or Z 2 frames later than possible and it will still be a true combo.


¹There are exceptions, but they won’t happen in situations where you’d want to do a pivot grab.

 

[Bones0]

Nevermind, that was already answered...

Thanks guys.

 

[Kadano]

@pewpewyou recently tweeted about Shield Breaker doing shield damage equal to 50% of max health unless fully charged. Of course it’s a bit more complicated than that:

Max shield health is 60.
Edit: An old Magus post says that shield health is variable and depends on shield level. So don’t rely on this table.

 

[Bones0]

@ Kadano
Broken image.

 

[lokt]

How do you do platform cancel fairs that you see in tas videos where they still end up on the platform? Would it be possible to do consistently?
@ Kadano

 

[1MachGO]

1. Z-cancelling was in 64, so I find it very hard to believe Sakurai couldn't execute one of the most necessary techniques in his own game. It seems even less likely that he would create such an elaborate KB and hitstun system and not take into account that you can cancel half the lag of your aerials. The fact that you can only cancel half as opposed to the full lag that was possible in 64 demonstrates to me that he knew what amount of lag was appropriate for aerials.

2. Totally agree on this point. If anything about Melee is to be considered a jackpot, it should be wavedashing/wavelanding. While I don't think Melee would be NEARLY as good without it, I still think it'd be a really amazing fighter with tons of depth. I think the overall effect would be a more simplified neutral game, particularly with platforms, but when you compare Melee's current neutral game to a traditional fighter, it's basically footsies on steroids so I definitely don't think it'd be one dimensional or anything. I also think it's reasonable to say that while none of the devs could have fathomed what WDing has become, it seems likely there was a general understanding you could do simple stuff like jump up through a platform and airdodge downwards onto it to land faster. Maybe I'm overestimating the competitive ingenuity of the devs though.

3. Not sure what you are getting at in terms of Fox being easy to SD with. Even today where 20xx is the talk of the town, the difficulty in using spacies is still highly considered in overall gameplay. All characters have to take into account the difficulty of their given options, and I think these risks are accentuated with spacies. The notion that these risks don't apply to really competitive players, let alone the godlike android Smashers we have today, seems naive. Even if a developer was trying to make a hardcore fighter, they'd likely never consider the game would get pushed to a level where players virtually never miss L-cancels, short hops, or even avoid side-Bing off the side of the stage. The fact that such things are infrequent in today's metagame is simply a testament to how insane Smashers are and how long they've been practicing for.

You are probably right that the game is focused on FFA, but I don't think that means 1v1 was ignored entirely. There aren't many exploitable tactics that are prevalent in 1v1 but not FFA. Chain grabbing is the major one, but it only really applies to a small subset of characters and usually only on FD. I believe Sakurai's desire to maintain character personality and limitations on development time are the main factors that prevented Sakurai from achieving a full, well-balanced cast.

While I agree there were many coincidences that helped play a role in Melee's complexity, I don't think that means it would be a subpar game without those coincidences. Even without things like WDing or jab resets or ground teching, the game would still be a deep fighter with tons of room for improvements.

In regards to 1, It is hard to say which one of us is right because we would be speculating at this point. We may never know what implications Sakurai felt L-Cancelling/Z-Cancelling had in smash bros. For all we know, he could have intended it to be a "neat trick to show your friends" or simply a competitive-oriented technique. I guess I would be more inclined to believe the latter if the game's AI was able to do it or if the technique itself was more widely advertised to western audiences. As for the changes from Z-Cancelling to L-Cancelling, this could have also been a logical decision rather than a competitive one. Speed and power are the de facto attributes that are balanced against each other in fighting games. Z-Cancelling would totally nullify that relationship whereas L-Cancelling keeps it intact (though its still disruptive, I can't see something like Falcon's triple knee combos on certain characters to be something he intended).

And with 3, I may have worded my stance poorly. My point in citing Sakurai's design philosophy with Fox (http://www.polygon.com/2012/12/12/3...ling-twitter-whiners-and-oh-yes-the-new-smash) is that anyone with that philosophy is clearly not a competitive-versed designer. I do agree that most developers have little idea on just how far players will take their game, but using the lack of player skill (especially in a game with simple controls) as a method of balance is laughable. Granted, he was regarding to Brawl Fox, but that method of balance was clearly being applied in Melee.

With that said, I pretty much agree with everything else in your post; especially your last two sentences.

 

[crispfish]

can someone explain running powershield


when i am in my dash animation my shield is behind my head so how can i ps a laser in front of me? especially a low shl

 

[Bones0]

can someone explain running powershield


when i am in my dash animation my shield is behind my head so how can i ps a laser in front of me? especially a low shl

Dash back instead of forward. #derp

 

[crispfish]

Dash back instead of forward. #derp

omfg thanks so much for this
foxes lasers were killing me so i needed to punish them so now im powershielding them
also this sets up for a great position for marth to be in

with my back turned i can hit my opponent with marths amazing auto cancel back airs

 

[crispfish]

but yo seriously is dash powershield a one frame window cuz i wanna approach with it
arc help me your godlike

 

[AustinRC]

I mean, there's no real threat from Fox's laser so you don't need to power shield it. It's mostly only useful vs Falco's laser and other projectiles as well. There are other set ups after the power shield I could go into though.

 

[crispfish]

I mean, there's no real threat from Fox's laser so you don't need to power shield it. It's mostly only useful vs Falco's laser and other projectiles as well. There are other set ups after the power shield I could go into though.

i was trolling bones cuz he was trolling me

my legit question is how hard is powershielding from your dash animation
ive done it but is it 1 frame

i can get normal powershields often

 

[Kadano]

Don’t troll in this thread. Do it again and I will ignore all of your requests and report the trolling instead.

Low laser powershield window is 0-2 frames if you are standing (depends on horizontal position), 2+ frames if you are dashing away and 0 frames if you dash towards.

 

[Elyssa Xey Hexen]

I am interested in gaining a bit of background info regarding crouch canceling. I suppose the first question is when can crouch canceling occur? Does it happen only in a neutral animation of holding down (crouching) or can it occur even when in the middle of another animation such as the tale end of a tech-in-place? Similarly, what affects the "strength" of a characters crouch canceling. It seems linked to faster falling characters such as Falco/Fox despite there "weight"? By strength I mean how lower percentage one can crouch cancel before they can no longer get a desired affect out of it.

Finally, what about a hitbox factors into the negation of a crouch cancel. It would seem as if moves with high WKB cannot get a desired affect ouf of crouch canceling compared to other properties. While on the other hand CC is often disrupted by very strong attacks. However, I am not certain if this is due to damage, BKB.

My interest in this comes from primarily determining what moves are most effective "counters" against CC. Very often people will simply try to tank the hit from Marth at the ledge with CC and retaliate at lower percentages. I believe that Marth's tipper Dair is most effective at this due to what I just described above. But, I am not certain if it would be the same across all characters. Plus, added bonus of being able to attack from underneath a person's shield rather than from above when getting on stage.

 

[Kadano]

@ E Elyssa Xey Hexen : http://smashboards.com/threads/rest-and-shield-break-punishing-strategies.206995/page-2#post-6019525

True CC isn’t that useful as you can simply shield instead. If you are in lag, you can’t CC, instead you use ASDI / SDI to force a ground collision. https://www.youtube.com/watch?v=Ei-iOIN32i4

 

[1MachGO]

My interest in this comes from primarily determining what moves are most effective "counters" against CC. Very often people will simply try to tank the hit from Marth at the ledge with CC and retaliate at lower percentages. I believe that Marth's tipper Dair is most effective at this due to what I just described above. But, I am not certain if it would be the same across all characters. Plus, added bonus of being able to attack from underneath a person's shield rather than from above when getting on stage.

If I am reading this right, you want a CC counter for when Marth is returning from the ledge? Well I can tell you that dair isn't the answer since its extremely easy to punish with shieldgrab if you can't cross up with it (also, by the time you are able to mobilize dair you would certainly be coming in on top/in front of their shield, not under it. Marth's uair is his best option for attacking underneath the opponent from the ledge). I guess your best option in this situation would be to space ledge-hop fair regrabs until they give you space.

 

[tauKhan]

What is the frame window for pivot powershielding falco lasers? From what I can tell, powershielding freezes your character in to the position he was before shielding, and hurtboxes might be better aligned during the pivot frame. I expect the window to be 0-1 frames if you shield on turn frame since I think your speed matters most for the window. The second question is do you get enough speed on the first few frames of dash after turn to extend ps window?

I have a small correction to a post:

The pseudo-jump flag is refreshed during the first frame of every landing animation. Because no-impact-landing and B and ↓B have no distinct landing animations, landing with them does not refresh the flag.

Landing during a B does refresh the pseudo-jump flag, but the landing lag of any aerial attack doesn't refresh it. I wouldn't have brought this up, since this isn't that significant, but the post is linked to the OP and is the only information I found on the boards about refreshing of B.

 

[Bones0]

i was trolling bones cuz he was trolling me

my legit question is how hard is powershielding from your dash animation
ive done it but is it 1 frame

i can get normal powershields often

I wasn't trolling... You really should dash back to PS low lasers.