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

[net1234]

Woe sorry super delayed. I keep getting hit by foxs' shine.

I'm like 95% sure u can sweet spot against a shine or I feel like it would happen way more in high level play. I usually manage to slip under them

 

[Bones0]

2 questions: does marth's ledge hop shield breaker have to be frame perfect?
also is it actually useful?

No, but I don't think it's very useful anyway. Easy way to get CC edgeguarded, and you can't combo from it. If you want to do a risky, gimmicky aerial regrab, pick uair. Use fair regrab if you are interested in actually getting back on the stage.

 

[DJ _ICE]

I just saw Westballz stickywalk to boost grab. Soooooooo cool! Now I must see a frame perfect Marth stickywalk- boost grab.

Also wondering how difficult and how useful that could be.

 

[net1234]

^ someone shud break down how to do that

 

[net1234]

doublepost because wondering: how does holding down R before plugging in your controller affect lightshielding and shield inputs in general

I'm stupid it doesn't work on the 20xx hack pack

 

[oukd]

stickywalk boostgrab looks cooler with sheik imo

for stickywalk just do a moonwalk but then hold diagonal forward-down right before it ends, then do a boostgrab (a > z)

its also not that useful imo but w/e

holding trigger:
if you hold trigger before plugging/during x+y+start reset, you wont have the option to lightshield at all. also youll have to click down for the full shield, analog will do nothing at all

pretty useful if you dont want lightshield stuff on one trigger...also if your controller keeps lightshielding on its own, its useful to do an x+y+start while holding the trigger very lightly. also remember that this means youre forced to click lcancel if you use that trigger for lcancels

 

[Kadano]

I’m sorry, but I don’t have the time and work motivation right now to answer all of your questions. Doing so draws from the same mental motivation pool I need for getting on with my video project, which I feel is more important right now.
Answering in this thread is such a cluttered method of spreading knowledge. It’s too hard even for me to find the posts I wrote on specific subjects quickly. I think I can offer much more condensed, easily indexable content by creating video guides.

Here is the first one of the main series, which is required for understanding follow-up videos: Debug mode introduction

 

[itsbme]

nevermind, im a dummy.

 

[Bones0]

I don't know if anyone was watching Super Smash Sundays last night, but wow I was impressed with Hyprid. He was doing so well and beat some of the pros there. Forgive me, I have not heard of him, but I was impressed. I can't wait for his videos to be uploaded.

Also, forgive me, everyone must be tired of me by now, but what does everyone think of Ken's Marth lately? His edgeguarding I hear has been mediocre, although he did good enough to beat Westballz last night at SSS. If he can fix that he'll be able to win more i'm sure. But anyways, what do you guys think about his play lately, and what good/bad things he does?

Also, i'm moving soon, so for now I won't be actively playing against people, but in a few months i'll be active in netplay again, so expect me to be asking for input.

You might want to post this in PP's discussion thread. This thread is usually more centered around frame data and tech and micro optimization than overall strategy or gameplay footage. Just super saiyan.

 

[SpiderMad]

Aerials apparently lower your overall jump height if done on the first airborne frame, do specials as well? @ Kadano @ Bones0

And how much height does it reduce, would Marth have been able to SH Bair > Aerial if it didn't exist? He can in PM but I can't tell if it's solely because of this not existing or because of this not existing along with the landing detection code being slightly lower

Oh and does light shield come up (or L-cancel) after the analog goes to a 1% threshold or is it actually after like 3% meaning the game can sense your controller's analog for a little bit before it reaches the point but give you the light shield (or L-cancel) action after going past a bit more?

 

[tm]

What are all the methods of pivot ftilting (mechanically speaking)? I'm attempting to get it like this:
initial dash animation -> smash the stick in the opposite direction for 1 frame (turnaround, then reset the stick to neutral) -> tilt the stick slightly and press A (ftilt)
This isn't the -frame perfect- pivot method (because the stick resets to neutral), but it is generally much more consistent. Problem is, it's very hard to do this with ftilt (as opposed to a smash attack, jab, or grab) because after letting go of the joystick you have to move it again, but not too quickly or you will do a smash attack. Any insight on performing this, or ideas for other methods to try?

No rush, and I appreciate any insight.

 

[Kadano]

Aerials apparently lower your overall jump height if done on the first airborne frame, do specials as well? @ Kadano @ Bones0

And how much height does it reduce, would Marth have been able to SH Bair > Aerial if it didn't exist? He can in PM but I can't tell if it's solely because of this not existing or because of this not existing along with the landing detection code being slightly lower

Oh and does light shield come up (or L-cancel) after the analog goes to a 1% threshold or is it actually after like 3% meaning the game can sense your controller's analog for a little bit before it reaches the point but give you the light shield (or L-cancel) action after going past a bit more?

Specials cause instant gravity as well, yes.
Speaking of removal is kind of incorrect, imho. The anomaly is that empty jumps (JumpF and JumpB ) don’t apply gravity on their first frame. All attacks do so, though, and that’s what results in the comparatively lower jump apex.

I don’t care about PM and what’s possible in it, so if you expect an answer to your wonderings, well, you won’t get it from me.

I think I remember analog shoulder slider ranges go from 0-127 and the threshold below which you don’t lightshield is 75.

What are all the methods of pivot ftilting (mechanically speaking)? I'm attempting to get it like this:
initial dash animation -> smash the stick in the opposite direction for 1 frame (turnaround, then reset the stick to neutral) -> tilt the stick slightly and press A (ftilt)
This isn't the -frame perfect- pivot method (because the stick resets to neutral), but it is generally much more consistent. Problem is, it's very hard to do this with ftilt (as opposed to a smash attack, jab, or grab) because after letting go of the joystick you have to move it again, but not too quickly or you will do a smash attack. Any insight on performing this, or ideas for other methods to try?

No rush, and I appreciate any insight.

Oh, answering this question is surprisingly easy because I tested this stuff just a month ago. I practiced pivot dtilts for 20 minutes and at some point, two out of ten attempts were successful while six produced a pivot ftilt. That got me curious.

It turned out that you can do an ftilt with the same control stick coordinate (henceforth CSC) that triggered the smash turn, so it is feasible to do a frame perfect pivot ftilt. The input range is probably more forgiving on worn out stickboxes, but I haven’t done an IOD¹ for it yet.

As long as notch modding is possible and not banned ∧ prosecuted at the tournaments you want to attend, it’s imho far superior to using worn out stickboxes.

¹Input-output display; those 256×256 diagrams I sometimes post that map out the entire control stick

 

[SpiderMad]

I think I remember analog shoulder slider ranges go from 0-127 and the threshold below which you don’t lightshield is 75.

What threshold is L-cancelling? Does lightshield become a full shield at 127 or sooner around 120 and it's already max?

 

[Kadano]

What threshold is L-cancelling? Does lightshield become a full shield at 127 or sooner around 120 and it's already max?

75 and ~105, I think. I’m not willing to do all of this again (I’m sure I’ve posted about this already, but I don’t know where exactly), so if you want the exact values, just do it on your own with Dolphin.

By the way, I got a new Melee Mechanics episode up (knockback and hitstun). It took me forever to finally get into Avisynth layer and animate usage, and I rushed in the end, so the timing is probably pretty bad. But I just wanted to get it up so I can carry on with the other videos until I get to the in-depth stuff.
http://youtu.be/c93Fwem-ngw

 

[SpiderMad]

@ Kadano
Why didn't they put the threshold lower for L-cancelling? Why didn't they give L-cancelling a fail window? Why didn't they put the Power shield window at 105 when you get a full shield? Did you ever make a thorough post on your thoughts of what Project M is doing that you don't like? Like I think I recall you saying you didn't like the RAR mechanic making turnaround WD too easy or something.

That video was good, you had the chance though after the Marth one to demonstrate the fast fall effect to people where Spacies are in long hitstun from their low weight but don't fly that far from the knockback due to their fast fall. You could use Fox+Falco+Samus+Jiggs all get hit
And then there's Hitlag/Shieldlag/SDI/CC which you've often talked about but not yet in video form, I'm guessing that's part of the more in-depth stuff you have in mind

 

[Bones0]

@ Kadano
Why didn't they put the threshold lower for L-cancelling? Why didn't they give L-cancelling a fail window? Why didn't they put the Power shield window at 105 when you get a full shield? Did you ever make a thorough post on your thoughts of what Project M is doing that you don't like? Like I think I recall you saying you didn't like the RAR mechanic making turnaround WD too easy or something.

That video was good, you had the chance though after the Marth one to demonstrate the fast fall effect to people where Spacies are in long hitstun from their low weight but don't fly that far from the knockback due to their fast fall. You could use Fox+Falco+Samus+Jiggs all get hit
And then there's Hitlag/Shieldlag/SDI/CC which you've often talked about but not yet in video form, I'm guessing that's part of the more in-depth stuff you have in mind

They probably didn't give L-cancelling a fail window because some people aerial with Z and there are a some situations where you would be shielding and then L-cancelling soon after. I'm glad for it because with a fail window stuff like jumping OoS with an aerial and landing on a plat or shield drop aerials would be impossible to L-cancel. Mashing the triggers to L-cancel to account for whiff vs. hitlag timings seems impossible anyway because you would inevitably be shielding by accident after whiffs.

 

[SpiderMad]

@ Bones0 is there any kind of buffer for get-up attacks, where you get hit out of something and hit A and then it makes you do a get-up attack even though you felt the input was a bit before

 

[Bones0]

@ Bones0 is there any kind of buffer for get-up attacks, where you get hit out of something and hit A and then it makes you do a get-up attack even though you felt the input was a bit before

I'm almost certain there is, but I don't know the specifics of it or how it works. I think I have had a lot less issues with accidentally doing GUAs after getting knocked down than most people so I never really dwelled on the buffer.


@ Kadano The KB video was perfect! You should make a thread for your series in the Melee Videos forum to start spreading it around. I'm sure it will catch on in no time, especially with a few more vids on common stuff like DI or teching.

 

[Kadano]

@ Kadano
Why didn't they put the threshold lower for L-cancelling? Why didn't they give L-cancelling a fail window? Why didn't they put the Power shield window at 105 when you get a full shield? Did you ever make a thorough post on your thoughts of what Project M is doing that you don't like? Like I think I recall you saying you didn't like the RAR mechanic making turnaround WD too easy or something.

You see, I never spent much time with PM. When I try it out at a friends’ setup, I just notice that the animations don’t look near as beautiful as in Melee, the voice effects are hilariously bad, the name tags have this weird property of wiggling up and down when you dash dance, etc. And there is no debug mode to test out stuff easily. Not to mention that it just doesn’t feel as smooth to me.
So, I simply don’t care much about PM. I’m a Melee guy.

That video was good, you had the chance though after the Marth one to demonstrate the fast fall effect to people where Spacies are in long hitstun from their low weight but don't fly that far from the knockback due to their fast fall. You could use Fox+Falco+Samus+Jiggs all get hit
And then there's Hitlag/Shieldlag/SDI/CC which you've often talked about but not yet in video form, I'm guessing that's part of the more in-depth stuff you have in mind

The video was about knockback and hitstun. I didn’t feel gravity, hitlag and DI belonged in there, so I’ll make separate videos about them.

@ Kadano The KB video was perfect! You should make a thread for your series in the Melee Videos forum to start spreading it around. I'm sure it will catch on in no time, especially with a few more vids on common stuff like DI or teching.

Thank you! Yeah, I’ve thought of doing that, but I decided to make the thread when more main line videos are finished.

 

[hectohertz]

Specials cause instant gravity as well, yes.
Speaking of removal is kind of incorrect, imho. The anomaly is that empty jumps (JumpF and JumpB ) don’t apply gravity on their first frame. All attacks do so, though, and that’s what results in the comparatively lower jump apex.

I don’t care about PM and what’s possible in it, so if you expect an answer to your wonderings, well, you won’t get it from me.

I think I remember analog shoulder slider ranges go from 0-127 and the threshold below which you don’t lightshield is 75.

Oh, answering this question is surprisingly easy because I tested this stuff just a month ago. I practiced pivot dtilts for 20 minutes and at some point, two out of ten attempts were successful while six produced a pivot ftilt. That got me curious.

It turned out that you can do an ftilt with the same control stick coordinate (henceforth CSC) that triggered the smash turn, so it is feasible to do a frame perfect pivot ftilt. The input range is probably more forgiving on worn out stickboxes, but I haven’t done an IOD¹ for it yet.

As long as notch modding is possible and not banned ∧ prosecuted at the tournaments you want to attend, it’s imho far superior to using worn out stickboxes.

¹Input-output display; those 256×256 diagrams I sometimes post that map out the entire control stick

can you expand on this pivot ftilt explanation a little bit more? what do you mean by you can do a pivot ftilt with the same CSCs that triggered the smash turn?

i currently do pivot ftilts by doing empty pivots (using smash turns) --> back to neutral --> ftilt, but as the one who originally asked the question pointed out, its slow. would love a better method :D

also the new vids are looking awesome! can't wait to see more

 

[Kadano]

can you expand on this pivot ftilt explanation a little bit more? what do you mean by you can do a pivot ftilt with the same CSCs that triggered the smash turn?

i currently do pivot ftilts by doing empty pivots (using smash turns) --> back to neutral --> ftilt, but as the one who originally asked the question pointed out, its slow. would love a better method :D

also the new vids are looking awesome! can't wait to see more

I’ll explain the details in a video, it’s a much better medium than these forum posts. My speed at putting them out is constantly increasing; I think I’ll be at one video every two days soon. (I have enough knowledge / ideas for at least 100 videos.)

Currently working on “Hitlag and DI”, by the way. It’s taking a bit longer because it encompasses so much.

 

[tm]

I’ll explain the details in a video, it’s a much better medium than these forum posts. My speed at putting them out is constantly increasing; I think I’ll be at one video every two days soon. (I have enough knowledge / ideas for at least 100 videos.)

Currently working on “Hitlag and DI”, by the way. It’s taking a bit longer because it encompasses so much.

I'm extremely excited for all of these.

 

[Kadano]

I'm extremely excited for all of these.

Just finished “Hitlag and directional influence”: http://youtu.be/7RP3sbS7Dm0

 

[DJ _ICE]

Awesome, I've been looking forward to this. Such a nice and clear explanation. I've wondered though, what if your control stick input for trajectory DI was past perpendicular? Such as the attack angle being 0° and the control stick is held to 100° or even 180°?

Also no music? The other video's music was pretty chill and easy

 

[Kadano]

Awesome, I've been looking forward to this. Such a nice and clear explanation. I've wondered though, what if your control stick input for trajectory DI was past perpendicular? Such as the attack angle being 0° and the control stick is held to 100° or even 180°?

That will result in about 16° deviation, just like 80° in your example. I had hoped the illustration at 3:20 would be enough to see that, but apparently it’s not as obvious as I thought.

Also no music? The other video's music was pretty chill and easy

Youtube removed the track due to copyright. I would exchange it for a different track from my PC, but it seems Youtube offers no such option, so I settled for the first track of Youtube’s audio library that seemed somewhat decent. It should play by now.

 

[Bones0]

I never have to explain how DI works ever again... This is a momentous occasion!

 

[SpiderMad]

Youtube removed the track due to copyright. I would exchange it for a different track from my PC, but it seems Youtube offers no such option, so I settled for the first track of Youtube’s audio library that seemed somewhat decent. It should play by now.

They removed it? I've only seen that happen back around 2010, all my music on my vids usually says it's licensed but it doesn't remove it. This could be because of your country policy now that I think of it.

Also what do you plan to do once your famous and get 1k subs?

 

[Signia]

I like to short hop and waveland a lot, but I've been thinking -- is it better to just fast fall and dash?

btw shared that DI video on reddit

 

[oukd]

depends

backwards:
reverse waveland is better if you want to maintain the direction you're facing (ie waveland back dtilt)
sometimes its more important to retain a high-reaction state, and ff dash away is probably better in that instance

forwards:
forward waveland i think lets you reset to neutral stance earlier than ff dash forward
but ff dash forward lets you act earlier if necessary, but only out of initial dash and not from neutral stance
(not sure about this one. im lacking in frame knowledge)

wavelands will affect your momentum if you want to dash/jump/etc out of them

also, i THINK:
out of waveland youre actionable after 10 frames
out of empty land youre actionable after 4 frames

lastly i think distance is an important comparative factor but idk how to quantify that

EDIT:

really important bit, you have instant movement out of a waveland
on empty land you do not (meaning youll be stationary for 4 frames)

 

[Mahie]

The fact that wavelands are laggy is the reason why sometimes it's better to Foxtrot forward.

 

[hariooo]

unless you're a pivot god, the choice between waveland and dash after sh fair is the same as the choice between wavedash and dash while standing for the same reasons.

 

[BTmoney]

okay so if you're watching KoC4 and you saw the vgbc down angled ftilt you're also wondering if that works.
(the gimr vs crimson blur set)

@ Kadano , can spacies beat a Marth's sweet spotted up B with their down angled forward tilt? Can they hit that without getting hit or can they force a trade? How good can it be vs. a perfect up B sweet spot and a pretty good one?

 

[Bones0]

okay so if you're watching KoC4 and you saw the vgbc down angled ftilt you're also wondering if that works.
(the gimr vs crimson blur set)

Kadano, can spacies beat a Marth's sweet spotted up B with their down angled forward tilt? Can they hit that without getting hit or can they force a trade?

Marth can definitely curl to hit Falco or just outspace the ftilt in the first place. Crimson just didn't sweetspot horizontally, and I wouldn't be that surprised if a vertical sweetspot alone would avoid LA ftilt (but I don't think it does).

 

[BTmoney]

Marth can definitely curl to hit Falco or just outspace the ftilt in the first place. Crimson just didn't sweetspot horizontally, and I wouldn't be that surprised if a vertical sweetspot alone would avoid LA ftilt (but I don't think it does).

yeah I thought the same thing but real-time that sweetspot attempt didn't look too off (looked like 80% of sweet spot attempts). maybe it's anti scrub stuff, you know things you do that you know shouldn't work lol. I would like to know how legit it is it's probably like a 6 or 7 out of 10 option

 

[Bones0]

yeah I thought the same thing but real-time that sweetspot attempt didn't look too off (looked like 80% of sweet spot attempts). maybe it's anti scrub stuff, you know things you do that you know shouldn't work lol. I would like to know how legit it is it's probably like a 6 or 7 out of 10 option

I mean, iirc Crimson was legit touching the stage, so it couldn't have been THAT close to a sweetspot. I don't see any reason why Falco wouldn't just dsmash instead, but maybe it has a good enough cooldown in order to get your shield up before Marth LH aerials.

 

[Tero.]

Answering in this thread is such a cluttered method of spreading knowledge. It’s too hard even for me to find the posts I wrote on specific subjects quickly. I think I can offer much more condensed, easily indexable content by creating video guides.

You should get an actual blog (wordpress or blogger for example) where you can collect all your information. By adding tags it would make it much easier for you and others to find what you're looking for.

 

[Chesstiger2612]

@ Kadano
If a grabbox collides tangential with a hurtbox, does it grab or not?

 

[SpiderMad]

@ Kadano Did you ever make gif comparison of how insta-gravity from a first frame airborne aerial (Fair) lowers your height compared to a normal height SH height/2nd frame aerial (Fair)?

 

[ZoSo]

Hey Kadano! Long time no see. I understand you're busy with other projects and this is kind of vague, but if you could talk a little about SDIing out of Fox uairs with Marth I'd be super appreciative. I find it extremely difficult or impossible against certain players so I was wondering if you could talk about which direction to SDI in different situations and the optimal way to do so in various non TAS conditions. Thanks!

 

[bossyjellyfish]

Hey Kadano! Long time no see. I understand you're busy with other projects and this is kind of vague, but if you could talk a little about SDIing out of Fox uairs with Marth I'd be super appreciative. I find it extremely difficult or impossible against certain players so I was wondering if you could talk about which direction to SDI in different situations and the optimal way to do so in various non TAS conditions. Thanks!

https://www.youtube.com/watch?v=7RP3sbS7Dm0#t=86 this could be potentially helpful