• Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!

    You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!

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

Xyzz

Smash Champion
Joined
Feb 7, 2011
Messages
2,170
Location
Gensokyan Embassy, Munich, Germany
As long as you never make fun of Americans not using the metric system again, we'll call it even.
whoa, whoa, whoa. Hold on a second :D
The musical notation is pretty arbitrary and both systems don't have any objective qualities. The metric system offers super easy conversion between units in comparison to the imperial system. I might be missing something, i'm fairly intoxicated (enough to have taken a cab back home, instead of the car that's right now chilling at the parking lot of the train station ;) ), but that seems to be a somewhat relevant reason.
Then again, I endlessly complain about all commonly accepted systems of writing dates (yyyy-MM-dd master race (make chronological and lexicographical ordering match, because why the hell not)), and nobody listens either :x
 
Last edited:

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
do you think we could get a complete list
if not i could probably ask someone else
Yeah, I can do this. It’s just that there is some ambiguity to what level of difficulty / ridiculousness in setting up I shall account for. Two examples I know of are Fox’s and Ganon’s ledgedashes, who get 16 and 20 frames respectively if they grab the ledge from a very specific animation point (no-fastfall high second jump for Fox, apex of up-B for Ganondorf) and execute the ledgedash differently than how they would do most of the time (after the safest ledgestall they have).

So basically I’m asking: Do you want numbers that account for TAS-like inputs or humanly possible inputs?

@ Xyzz Xyzz : I always use yyyy-mm-dd for my smashfests / tournaments, if you haven’t noticed. Medieval units have their merits, though – with the next lower class being 1/12 instead of 1/10, you have much easier access to 1/4 and 1/3 units. Also, alcohol is the worst drug ever, except for smoking cigarettes maybe.
@ Life Life : add me to the recipient list please, I like talking about these things, but I agree this thread is not a good place. (It would be really nice to be able to off-branche these OT discussions and hide them from the main discussion stream, while still being able to blend them in if you click the starting post. I don’t know of a good way to do this within the two-dimensional plane we use for web interfaces; I especially hate the way reddit attempts to do this.)
 
Last edited:

Tee ay eye

Smash Hero
Joined
Jun 1, 2008
Messages
5,635
Location
AZ
is fthrow WD tipper d-smash guaranteed in any situations when they DI the fthrow properly? i know it works in some cases where fthrow WD f-smash doesn't work, but i don't know if that's guaranteed in any sense.

also, in the OP, could you specify what direction marth is facing on the graphic of throw punishes against puff at 90%? it can be figured out, but it's just a little inconvenient

you're awesome <3
 
Last edited:

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
is fthrow WD tipper d-smash guaranteed in any situations when they DI the fthrow properly? i know it works in some cases where fthrow WD f-smash doesn't work, but i don't know if that's guaranteed in any sense.
The chart clearly illustrates that at 90% damage, fsmash will cover more DI angles than dsmash. For lower damage, say 70%, I can imagine dsmash hitting on downward DI just before Jigglypuff collides with the floor when fsmashed would have whiffed, due to its slower startup. I will not test this specifically, though; it already took very long to make the 90% chart and I’m not going to add another dimension (%) to the equation.
 

Tee ay eye

Smash Hero
Joined
Jun 1, 2008
Messages
5,635
Location
AZ
well, i didn't mean against puff specifically. i meant in a general sense; i was curious to see if you had more info on dsmash in general
 

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
also, in the OP, could you specify what direction marth is facing on the graphic of throw punishes against puff at 90%? it can be figured out, but it's just a little inconvenient
Done.

well, i didn't mean against puff specifically. i meant in a general sense; i was curious to see if you had more info on dsmash in general
Tipper-dsmashing Jigglypuff out of a throw is only possible due to her low weight and falling speed. The only other characters I can think of on which it might work as well are G&W and Pichu, but I don’t believe they are sufficiently relevant to the metagame to justify the effort of testing.
 

ZoSo

Smash Champion
Joined
Jul 7, 2003
Messages
2,885
Location
Melee
Yeah, I can do this. It’s just that there is some ambiguity to what level of difficulty / ridiculousness in setting up I shall account for. Two examples I know of are Fox’s and Ganon’s ledgedashes, who get 16 and 20 frames respectively if they grab the ledge from a very specific animation point (no-fastfall high second jump for Fox, apex of up-B for Ganondorf) and execute the ledgedash differently than how they would do most of the time (after the safest ledgestall they have).

So basically I’m asking: Do you want numbers that account for TAS-like inputs or humanly possible inputs?
I'd be interested in both, for what it's worth. However, it seems like it would be pretty tedious and time consuming to test and collect all the data.

Also, I beat Darc again yesterday, and I'm still giving you credit.

You don't have to wait for them to hit you even with that method, though. Isn't slowly tilting to the side and pressing to the notch just as fast as slowly pressing down and then quickly finish pressing down??

Really, why, just curious? I think you should map one input per action, it's less mental load. You're either making an unnecessary decision or are having to train unecessary movements. From physical training theory, maintaining a form of movement takes activation of motor neurons http://en.wikipedia.org/wiki/Motor_unit_recruitment, and is how movements can become automatic and subconscious. When I play the piano, varying up the fingers I use to hit the notes slows my improvement down, and even looking somewhere during the song that I don't usually look messes me up. If you want consistency, you don't want to change your form.

I think it's best to make movements as different as possible, too. You fingers might think (literally) you're doing one thing when you want to do the other. This is partly why I like to wavedash with one trigger and shield with the other. Keeping everything distinct and encapsulated makes movement cohesive.
I just noticed this.

I think you'll find many private teachers and educators agree that it's worthwhile to learn multiple techniques for the same application, multiple ways of executing the same technique, etc. It makes you more flexible and complete. However, I'd probably agree that this is mostly relevant when improvising or in a live performance setting. The former because you need to have a clear channel between conception and execution, which is achieved by practicing as much as you can as many ways as you can, and the latter because live settings introduce variables that you can't account for in practice, so it's important that your technique isn't "fragile" so to speak. A great guitar player named Tuck Andress wrote a few very illuminating articles on this subject and others. Then again, the guitar and the piano present very different challenges to the performer, so your mileage may vary.
 
Last edited:

tm

Smash Ace
Joined
Apr 12, 2012
Messages
819
Location
NWOH
Yeah, I can do this. It’s just that there is some ambiguity to what level of difficulty / ridiculousness in setting up I shall account for. Two examples I know of are Fox’s and Ganon’s ledgedashes, who get 16 and 20 frames respectively if they grab the ledge from a very specific animation point (no-fastfall high second jump for Fox, apex of up-B for Ganondorf) and execute the ledgedash differently than how they would do most of the time (after the safest ledgestall they have).

So basically I’m asking: Do you want numbers that account for TAS-like inputs or humanly possible inputs?
Also very interested (in both) :)
 

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
do you think we could get a complete list
if not i could probably ask someone else
I'd be interested in both, for what it's worth. However, it seems like it would be pretty tedious and time consuming to test and collect all the data.
Also very interested (in both) :)
Grounded actionable ledge intangibility (abb. galint)
╔════════════════╤═╤══╤══╤══╤══╗
║Character ......│E│OL│PL│NL│AI║
╟────────────────┼─┼──┼──┼──┼──╢

║Zelda ..........│N│ 2│..│..│13
║Sheik ..........│Y│1113│..│ 8
║Popo ...........│N│13│..│..│..║
║Nana ...........│N│13│..│..│..║
║Captain Falcon .│Y│10121715
║Young Link .....│N│13│..│..│..║
║Donkey Kong ....│N│ 7│..│..│..║
║Doctor Mario ...│Y│14│..│..│..║
║Falco ..........│Y│1415│..│..║
║Fox ............│Y│1516│..│..║
║Kirby ..........│N│12│..│..│..║
║Bowser .........│N│ 8│..│..│..║
║Link ...........│Y│ 912│..│..║
║Luigi ..........│Y│13│..│..│..║
║Mario ..........│Y│13│..│..│..║
║Marth ..........│N│ 4│..│ 13
║Mewtwo .........│N│ 1│..│..│ 6
║Ness ...........│N│ 4│..│..│..║
║Peach ..........│N│..│..│..│..║
║Pichu ..........│Y│1516│..│..║

║Pikachu ........│Y│13│..│..│..║
║Jigglypuff .....│N│ 9│..│..│..║
║Samus ..........│N│ 4│..│..│11
║Yoshi ..........│N│ 2│..│..│ 8
║Mr. Game&Watch .│N│ 9│..│ 6│..║

║Ganondorf ......│Y│ 4918│..║
║Roy ............│N│ 5│..│..│..║
╟────────────────┼─┼──┼──┼──┼──╢
║Giga Bowser¹ ...│.│..│..│..│..║
║Master Hand² ...│.│..│..│..│..║
║Crazy Hand² ....│.│..│..│..│..║
║Male Wireframe .│Y│ 89│..│..║

║Female Wireframe│N│ 5│..│..│..║
║Sandbag³ .......│.│..│..│..│..║
╚════════════════╧═╧═══════════╝


E = ECB manipulation-abuse possible Yes/No (applied for OL only if the execution is managable/realistic; always applied for PL if it increases galint)
OL = Optimal ledgedash (hard to do frame-perfect, but possible)
PL = Perfect ledgedash (TAS-realm – requires a weird setup and/or tricky timing)
NL = No-impact landing (requires a very specific setup usually)
AI = Aerial Interrupt (Captain Falcon’s requires a very specific setup, namely grabbing the ledge out of his up-B apex, as well. Those of Samus and Yoshi do not)

These values are for FD. Other stages, especially YS and FoD, probably have slightly different values due to their ledges not being perfectly even.

This chart is not perfectly accurate. I’m sure there are a few aerial interrupts and lagless landing options that I didn’t find. The ledgedash data itself should be correct, though.

¹Giga Bowser’s ledgehop is too short to rise above the stage.
²Master Hand and Crazy Hand cannot move and thus are unable to grab the ledge.
³Sandbag can move, but grabbing the ledge is beyond his capabilities. (I might work on a Sandbag improvement hack if I find the motivation.)
 
Last edited:

1MachGO

Smash Ace
Joined
Mar 18, 2013
Messages
807
Wow, Marth's ability to abuse ledge invincibility is actually pretty bad compared to most of the cast.
 

ZoSo

Smash Champion
Joined
Jul 7, 2003
Messages
2,885
Location
Melee
Kadano, could you explain the methodology for covering techs on BF platforms with SH uair?
For example, where should I position myself, when should I jump relative to their tech animations, when should I input the uair, and does the timing change depending on which option they picked? When should I input the fastfall to make sure I cover tech in place or no tech? I understand the theory, I'm just having a hard time reducing it to nuts and bolts. Thanks a bunch.
 
Last edited:

Bones0

Smash Legend
Joined
Aug 31, 2005
Messages
11,153
Location
Jarrettsville, MD
To add to ZoSo's discussion, if the opponent does a getup in place right when I land, is it feasible to react with a rising uair or at least a utilt? Or should I just FF uair early and if they guess the right timing then utilt after I land (doesn't seem like there's usually enough time for this)?
 
Last edited:

Tee ay eye

Smash Hero
Joined
Jun 1, 2008
Messages
5,635
Location
AZ
Kadano, could you explain the methodology for covering techs on BF platforms with SH uair?
For example, where should I position myself, when should I jump relative to their tech animations, when should I input the uair, and does the timing change depending on which option they picked? When should I input the fastfall to make sure I cover tech in place or no tech? I understand the theory, I'm just having a hard time reducing it to nuts and bolts. Thanks a bunch.
Techchasing on platforms from below with upair

The timing for jump and upair are the same for all 4 of Fox’s options. The only difference is that you need to fastfall if he techs in place or doesn’t tech to hit him with the tipper hurtbox. This works on all characters, but on Pikachu and Pichu you need to be frame perfect as their tech has longer invincibility. It also works on most stages, but on Battlefield you need to hold left/right as soon as you see them tech-rolling in that direction because its platforms are the widest.
 
Last edited:

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
@ ZoSo ZoSo : Ideally, you should position yourself exactly below the middle of the stage, but it’s better to be slightly before that line than behind it. You should jump ~6-12 frames before your opponent collides with the platform. You should input the uair at the apex of your short hop or a few frames later. Apart from that, everything you need to know is written in the part of the OP TAI kindly quoted.

@ Bones0 Bones0 : If you jump early enough, your victim never gets to the point where it could do a getup. If the situation you are in doesn’t allow for that, then yes, it is possible to react to getup in place and punish it with both tipper uair and utilt. You can also react to grounded rolls by doing a pivot tipper fsmash or a dash short hop aerial. But you need to react more quickly than with early jumps, so I’d say always jump ~10 frames before they hit the ground and do the late uair thing.

@ TheCrimsonBlur TheCrimsonBlur : Don’t forget about this part: “requires a very specific setup usually”
17 frames (not 18, that was a mistake on my part because Wait 0 is a forced frame for no-impact land) of grounded actionable intangibility are only possible if the second jump is done out of a sweetspot up-B ledgegrab.
 

hariooo

Smash Apprentice
Joined
Oct 28, 2013
Messages
124
How many frames of leeway do we have on the standard puff throw tippers? Even when I can get one of the options in training mode I almost never get a true combo on the resulting tipper, even if it's happened a few times. Depending on puff's airdodge and jump animations though it might not be 100% necessary though?
 

hectohertz

Smash Ace
Joined
Aug 20, 2006
Messages
800
Location
Brooklyn, NY


Grounded actionable intangibility from the ledge (gaInt)
╔════════════════╤═╤══╤══╤══╤══╗
║Character ......│E│OL│PL│LL│AI║
╟────────────────┼─┼──┼──┼──┼──╢
║Zelda ..........│N│ 2│..│..│..║

║Sheik ..........│Y│ 812│..│..║
║Popo ...........│N│13│..│..│..║
║Nana ...........│N│13│..│..│..║
║Captain Falcon .│Y│10111715
║Young Link .....│N│12│..│..│..║
║Donkey Kong ....│N│ 7│..│..│..║
║Doctor Mario ...│Y│14│..│..│..║
║Falco ..........│Y│1415│..│..║
║Fox ............│Y│1516│..│..║
║Kirby ..........│N│12│..│..│..║
║Bowser .........│N│ 8│..│..│..║
║Link ...........│Y│ 912│..│..║
║Luigi ..........│Y│13│..│..│..║
║Mario ..........│Y│13│..│..│..║
║Marth ..........│N│ 4│..│ 1│..║
║Mewtwo .........│N│ 1│..│..│..║

║Ness ...........│N│ 4│..│..│..║
║Peach ..........│N│..│..│..│..║
║Pichu ..........│Y│1516│..│..║

║Pikachu ........│Y│13│..│..│..║
║Jigglypuff .....│N│ 9│..│..│..║
║Samus ..........│N│ 4│..│..│11
║Yoshi ..........│N│ 2│..│..│ 8
║Mr. Game&Watch .│N│ 9│..│ 6│..║

║Ganondorf ......│.│ 4918│..║
║Roy ............│N│ 5│..│..│..║
╟────────────────┼─┼──┼──┼──┼──╢
║Giga Bowser ....│.│..│..│..│..║
║Master Hand ....│.│..│..│..│..║
║Crazy Hand .....│.│..│..│..│..║
║Male Wireframe .│Y│ 89│..│..║

║Female Wireframe│N│ 5│..│..│..║
║Sandbag ........│.│..│..│..│..║
╚════════════════╧═╧═══════════╝

E = ECB manipulation-abuse possible Yes/No (applied for OL only if the execution is managable/realistic; always applied for PL if it increases gaInt)
OL = Optimal ledgedash (hard to do frame-perfect, but possible)
PL = Perfect ledgedash (TAS-realm – requires a weird setup and/or tricky timing)
LL = Lagless landing aka no-impact landing (requires a very specific setup usually)
AI = Aerial Interrupt

These values are for FD. Other stages, especially YS and FoD, probably have slightly different values.

This chart is not perfectly accurate. I’m sure there are a few aerial interrupts and lagless landing options that I didn’t find. The ledgedash data itself should be correct, though.

first of all, this is amazing, and answers soooo many questions.

when we went over sheik's ledgedashes, you had said that her OL was +11, whereas here its +8. care to clarify? my guess would be that the +8 is off a plank/something without an ECB manipulation, the +11 was off a regular shino stall, and the +12 is off a low shino stall. is that right?
 

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
when we went over sheik's ledgedashes, you had said that her OL was +11, whereas here its +8. care to clarify? my guess would be that the +8 is off a plank/something without an ECB manipulation, the +11 was off a regular shino stall, and the +12 is off a low shino stall. is that right?
For +11 and +12, you are right. +8 is when you only accept notch angles. I just checked for the +11 input value ranges, and I think they are easy enough to count, so I edited it back to +11.
At the jump frame where you have to to input the airdodge for a +11 ledgedash, these control stick values work:


(only the area inside the octagon had its corner values tested; the outside is just a more-or-less approximate extrapolation)
If you go for the notch (225°), you airdodge to your death. However, if you airdodge one frame later, you can use any down-forward angle and get +~5 to +10; +8 if you go for the notch. I thought these 1-frame delayed 225° wavelands were the easiest to execute consistently, so I took them as the optimal ledgedash method.

+8 ledgedash:


+11 ledgedash (input shown on preceding frame):

+11 ledgedash (input shown on corresponding output frame):

Please, everyone who reads this: Do me the favor and tell me which input-output display option you prefer! (Click the image to view them with gfycat frame-by-frame viewing options)

Edit: Bandwidth issues? Page loading slow? Sorry, but I can’t do anything about these .gifs being ~3MB each until the admins install the gfycat plugins that reduce the filesize to ~1/8th. I could reduce the resolution, but this wouldn’t help that much and detail would be lost. Also, the 256×256 input-output display wouldn’t be accurate any longer.
 
Last edited:

hectohertz

Smash Ace
Joined
Aug 20, 2006
Messages
800
Location
Brooklyn, NY
For +11 and +12, you are right. +8 is when you only accept notch angles. I just checked for the +11 input value ranges, and I think they are easy enough to count, so I edited it back to +11.
At the jump frame where you have to to input the airdodge for a +11 ledgedash, these control stick values work:


(only the area inside the octagon had its corner values tested; the outside is just a more-or-less approximate extrapolation)
If you go for the notch (225°), you airdodge to your death. However, if you airdodge one frame later, you can use any down-forward angle and get +~5 to +10; +8 if you go for the notch. I thought these 1-frame delayed 225° wavelands were the easiest to execute consistently, so I took them as the optimal ledgedash method.

+8 ledgedash:


+11 ledgedash (input shown on preceding frame):

+11 ledgedash (input shown on corresponding output frame):

Please, everyone who reads this: Do me the favor and tell me which input-output display option you prefer! (Click the image to view them with gfycat frame-by-frame viewing options)

Edit: Bandwidth issues? Page loading slow? Sorry, but I can’t do anything about these .gifs being ~3MB each until the admins install the gfycat plugins that reduce the filesize to ~1/8th. I could reduce the resolution, but this wouldn’t help that much and detail would be lost. Also, the 256×256 input-output display wouldn’t be accurate any longer.

holy **** kadano delivers. and i like the one where the inputs are displayed on the preceding frame
 

Chesstiger2612

Smash Lord
Joined
Jun 1, 2013
Messages
1,753
Location
Bonn, Germany
Is there any formula until which amount of knockback (or other factors) you can tech moves with down SDI and "anti-survival" DI? There can be only a few percents between a good option and getting you killed.
 

Bones0

Smash Legend
Joined
Aug 31, 2005
Messages
11,153
Location
Jarrettsville, MD
Is there any formula until which amount of knockback (or other factors) you can tech moves with down SDI and "anti-survival" DI? There can be only a few percents between a good option and getting you killed.
There isn't really a formula since it's based off of spacing. You could probably find the highest techable % for any one move, but even that might differ from character to character based on how their hurtbox contorts during KB. Not sure how that plays a role in one's ability to ground tech launching moves though.
 
Last edited:

6VI6

Smash Apprentice
Joined
Sep 29, 2011
Messages
121


Grounded actionable intangibility from the ledge (gaInt)
╔════════════════╤═╤══╤══╤══╤══╗
║Character ......│E│OL│PL│LL│AI║
╟────────────────┼─┼──┼──┼──┼──╢
║Zelda ..........│N│ 2│..│..│..║

║Sheik ..........│Y│1112│..│..║
║Popo ...........│N│13│..│..│..║
║Nana ...........│N│13│..│..│..║
║Captain Falcon .│Y│10111715
║Young Link .....│N│12│..│..│..║
║Donkey Kong ....│N│ 7│..│..│..║
║Doctor Mario ...│Y│14│..│..│..║
║Falco ..........│Y│1415│..│..║
║Fox ............│Y│1516│..│..║
║Kirby ..........│N│12│..│..│..║
║Bowser .........│N│ 8│..│..│..║
║Link ...........│Y│ 912│..│..║
║Luigi ..........│Y│13│..│..│..║
║Mario ..........│Y│13│..│..│..║
║Marth ..........│N│ 4│..│ 1│..║
║Mewtwo .........│N│ 1│..│..│..║

║Ness ...........│N│ 4│..│..│..║
║Peach ..........│N│..│..│..│..║
║Pichu ..........│Y│1516│..│..║

║Pikachu ........│Y│13│..│..│..║
║Jigglypuff .....│N│ 9│..│..│..║
║Samus ..........│N│ 4│..│..│11
║Yoshi ..........│N│ 2│..│..│ 8
║Mr. Game&Watch .│N│ 9│..│ 6│..║

║Ganondorf ......│.│ 4918│..║
║Roy ............│N│ 5│..│..│..║
╟────────────────┼─┼──┼──┼──┼──╢
║Giga Bowser ....│.│..│..│..│..║
║Master Hand ....│.│..│..│..│..║
║Crazy Hand .....│.│..│..│..│..║
║Male Wireframe .│Y│ 89│..│..║

║Female Wireframe│N│ 5│..│..│..║
║Sandbag ........│.│..│..│..│..║
╚════════════════╧═╧═══════════╝

E = ECB manipulation-abuse possible Yes/No (applied for OL only if the execution is managable/realistic; always applied for PL if it increases gaInt)
OL = Optimal ledgedash (hard to do frame-perfect, but possible)
PL = Perfect ledgedash (TAS-realm – requires a weird setup and/or tricky timing)
LL = Lagless landing aka no-impact landing (requires a very specific setup usually)
AI = Aerial Interrupt

These values are for FD. Other stages, especially YS and FoD, probably have slightly different values.

This chart is not perfectly accurate. I’m sure there are a few aerial interrupts and lagless landing options that I didn’t find. The ledgedash data itself should be correct, though.
Falcon has an aerial interrupt? How is that done? I had only done it with Samus and Yoshi previously.
And wtf Pichu's ledgedash is so good
 

SpiderMad

Smash Master
Joined
May 6, 2012
Messages
4,968
The bair changes his model's ECB on frame 1, and that makes him land faster so you can press A again to jab?
 
Last edited:

MookieRah

Kinda Sorta OK at Smash
Joined
Mar 7, 2004
Messages
5,384
Location
Umeå, Sweden
Can someone link me (or tell me where to fine) to a post or something explaining what ECB is? This is the first I've heard of this.
 

Xyzz

Smash Champion
Joined
Feb 7, 2011
Messages
2,170
Location
Gensokyan Embassy, Munich, Germany
The ECB (environment collision box) is that yellow quad (are they called quad in a non 3d-modelling context, too? :D) we can for example see in Kadano's latest post with the falcon getup. It's used to determine the characters position for deciding whether he should pass through a platform, or jumped high enough from the ledge to land on the stage. It's not a static form but changes depending on your last action (that's why we get different ledgedash timings).
 

MookieRah

Kinda Sorta OK at Smash
Joined
Mar 7, 2004
Messages
5,384
Location
Umeå, Sweden
Thanks, it makes complete sense now, and wow, the more you know! Melee is insane.

I take it that Marth's ECB doesn't alter from his attacks much, if at all, then? (I assume the same for M2 as well)
 

SpiderMad

Smash Master
Joined
May 6, 2012
Messages
4,968
@ Kadano Kadano What's your view on Melee's intended design? I don't see how people can say it was designed to be -solely- a party game, when people like us can research just how well designed all their formulas and options/restrictions on the mechanics were so well thought out. I'm also confused as to seriously why there's an illuminati looking triangle inside Dreamland and Shield stun.
 
Last edited:

Bones0

Smash Legend
Joined
Aug 31, 2005
Messages
11,153
Location
Jarrettsville, MD
@ Kadano Kadano What's your view on Melee's intended design? I don't see how people can say it was designed to be -solely- a party game, when people like us can research just how well designed all their formulas and options/restrictions on the mechanics were so well thought out. I'm also confused as to seriously why there's an illuminati looking triangle inside Dreamland and Shield stun.
THANK YOU. I get so upset when people call Melee an "accident" or similar stuff. Like yeah, I understand the game was not designed to be a global eSport, but even saying it wasn't designed for competition is false as clearly demonstrated by the Tournament mode... I have a few goto arguments I use when people disagree with me.

1. PAL
The PAL version of the game wasn't released very long after the NTSC, yet almost all of the changes we see (outside of glitch removal) are in line with the goal of a more balanced cast. Marth's and Falco's spikes, arguably the best two in the game, were nerfed. Fox's recovery and KO power were nerfed. Sheik's dthrow was nerfed. etc. The changes made in the transition to PAL show very clearly that Sakurai understand the balance of the cast and also understood how to change as a little as possible and still influence the core of the character.

2. Tech timings
One thing I think gets overlooked a lot is how well designed the teching system is. On average, a character's techrolls or TIP are just out of range of human reaction time. I don't think it's a coincidence that teching in place is just barely within the range of the average human reaction time. I have no trouble believing that a lesser developer would have chosen much more arbitrary timings for techs and would have ultimately led to either super hard reads because you can't cover multiple options with a combination of reaction and prediction, or players would always be able to react to all of their opponent's options easily.

3. Everything barely combos
Even when you have DI thrown into the equation, the majority of combos (excluding ones with grounded spikes which were clearly designed to combo better) are reliable, but really hard to link. I think this is why even so many years after the game has evolved, there is still a high prevalence of juggles and pseudo-combos (instances where you aren't being true comboed but can't do much, if anything anyway). I have to imagine the KB and hitstun calculations were designed meticulously, and if you need an example of how important it is, just look at the combos in 64 or lack thereof in barlw.

4. Grabs are really good
I think people sometimes get a little annoyed grabs are as good as they are. DA Dave's infamous "it's all about the grab game" commentary over Ken vs. Mew2King kind of hints towards this, but I would hope most people realize the significance of having good grabs. When grabs lead to good damage/setups, it tilts the game in favor of aggression. Especially with some of the shields in Melee which seem really safe and abusable (Sheik, Peach), you need to prevent people from spamming shield all game. Melee has the traditional fighting game RPS of shield > attacks > grabs > shield, and buffing the grab game downplays shielding, which keeps things more active and interesting than passive and predictable.

5. "But what about the bad characters?"
Like I said above, just because Melee was well-designed does not mean it was designed for super serious competition. While I think most of the actual fighting elements are designed for competitive environments, certain characters, stages, and of course items were designed with a more party mentality. I have a hard time believing the same guy that made all the NTSC to PAL adjustments somehow didn't realize Pichu, Kirby, and Bowser were weaker than normal. It seems much more likely that Pichu was just a character added for extra challenge in single player. He is fast and agile and actually has decent KO moves, so to me he is the perfect handicap character that requires the tightest gameplay to get the full potential vs. someone like Ganon who is much easier for a newbie to pick up and smash attack his way through classic mode. Kirby has the copy the ability, so the novelty factor is pretty clear there as well. As far as Bowser and other low tiers, it seems like Sakurai just preferred to include characters with actual personality and unique traits at the expense of balance. He'd rather have Bowser be pretty awful but feel like Bowser should feel than water him down by an overall speeding up and weakening of his attacks.



Well, that's my uncalled-for rant for today. Hopefully it didn't come off too much like I'm ****-riding Sakurai, but he did make Melee after all. Even barlw can't make me lose respect for the guy, and that's saying something.
 

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
@ Kadano Kadano What's your view on Melee's intended design? I don't see how people can say it was designed to be -solely- a party game, when people like us can research just how well designed all their formulas and options/restrictions on the mechanics were so well thought out. I'm also confused as to seriously why there's an illuminati looking triangle inside Dreamland and Shield stun.
My opinion is that Sakurai went to extreme lengths to create the perfect game and by comparison didn’t put much thought in things like “am I catering to pros or to casuals?“. He wanted to make a flawless and balanced game, and except for a few exceptions – I’m pretty sure he would have disabled Wobbling if he had found it – he was successful.

Adding to what Bones0 wrote, whenever I load up develop mode and research things, I’m amazed by how well-engineered everything is and how many weird quirks that still feel great there are (tilt turns vs. smash turns, for example, which kind of nerfs wavedash-turn edgehogs, which would be a bit too quick and easy otherwise).

Also, I got my first Wii two days ago, it had a Brawl disc inside. I played a bit for video quality testings before I softmodded the Wii, and Brawl is even worse than I remembered it. This horrible feeling when you try to move around and this ridiculous syrup-like air characters get stuck within when they are thrown up … *shudders*

I feel like the creation of Brawl was like this: Sakurai was still exhausted from creating Melee, and Nintendo told him that they needed a new Smash game for the Wii console. Sakurai was confused, because he already created the perfect Smash game that could also be played on Wii consoles. So he thought to himself, “to make things fun, this time I will give my best to make the worst gameplay mechanics possible, just to see if people will still like it”.
Or something like that. I think that after Melee, he had fulfilled all of his ambitions for creating something great, so all that was left was to mess around.
 
Last edited:
Joined
Aug 6, 2008
Messages
19,345
Then, one gets reminded of characters like Pichu and Roy who just seem like awful design oversights.

Roy having all sorts of flimsy attacks on contact. Marth's Fair non-tip still does more damage than Roy's tipper. But, this is common across many of their moves. Or, how some attacks seem to be missing there full animation like his Dash attack timing seems like a mistake. Or why does Pichu even exist as it does? Everything about it Pikachu does better expect pichu is more adorable.
 

Kadano

Magical Express
Joined
Feb 26, 2009
Messages
2,160
Location
Vienna, Austria
Maybe Sakurai was forced to add Roy and Pichu and he made them really bad as a kind of retaliation? At least I would have reacted like this in his position.
 
Last edited:

hectohertz

Smash Ace
Joined
Aug 20, 2006
Messages
800
Location
Brooklyn, NY
there's an interview somewhere where sakurai talks about the development of melee where he says that:
- the original "finished" build of melee lacked the clones (doc, falco, roy, pichu, ganon, yl [notice how they are all on the edge of the CSS])
- nintendo decided there weren't enough characters / to push the games release date back, sending them back to the lab to make more characters (fairly quickly). thus the clones were born.

so that means there's a parallel universe in which there's no falco....
and in which kage never beat mango
and armada never beat hbox
...weird
 

1MachGO

Smash Ace
Joined
Mar 18, 2013
Messages
807
@ Bones0 Bones0 while I do believe Melee was designed with some competitive orientation and there are quotes from Sakurai to suggest that, I think you are giving him way more credit than he deserves.

1.Barely combo'ing
For starters, there is L-Cancelling. I simply cannot believe that Sakurai, or anyone in his staff for that matter, were able to consistently perform it. Considering how L-Cancelling aerials unlocks the grand majority of the combo game, I doubt Sakurai had balanced his game with half landing lag, doubled it, and added L-Cancelling (especially when you consider his rough 18 month development time). The whole "barely combos" thing is likely a result of how L-Cancelling ironically undid Sakurai's methodology. Knockback was probably designed in tandem with lag to ensure there weren't any additional advantages after landing an attack. This is what 90% of the ftilts, fsmashes, and non l-cancelled aerials in the game do. The addition of L-Cancelling then disrupts this balance because it changes the ratio of lag and knockback/hitstun to be in favor of the attacker. Voila, things now barely combo.

Certain tilts, jabs, dash attacks, and grabs, on the other hand, were definitely designed to combo. However, I would say most of their windows are very lenient.

2. Wavedashing
We all know it isn't a glitch but we do know that the game wasn't designed around it. Despite meeting the definition of a "physics exploit", wavedashing and wavelanding close the gap between aerial and ground movement in innumerable ways. Having that implication in the realm of a platformer is honestly so genius that I am not surprised it was an accident. The consequences it has on shielding as well as Fox and Falco's shines is also proof that Sakurai never foresaw its potential.

3. Bad Characters and Character Balance
Sakurai has gone on record saying that Fox is balanced around the concept of being easy to SD with while possessing "other skills." If the downfall of your character is dependent on a player's likelihood to kill himself, you clearly have specific views in terms of how far your players will take the game or what direction they will take it in.

To me, it seems pretty obvious that Melee was never balanced for 1v1 play. While I agree that Sakurai was willing to sacrifice character balance for identity, I think even the intern could have realized that Bowser would've never succeeded in a 1v1 setting. The game is clearly advertised as a 4 player free for all and the characters seem balanced for this kind of play. In a free for all, mobility and speed are important, but there is also a larger emphasis on individual attack strength and durability. When considering balance in this respect, its easy to see how Sakurai overlooked Sheik's dthrow in NTSC or never felt that Bowser's fsmash or Ganon's warlock punch were unusable. The game becomes unbalanced the moment we try to make it competitive because its clear it was meant to be a free for all game.

Now I will reiterate, I do not think Melee was void of competitive orientation. However, much of its success (and lack there of) is a culmination of many oversights, coincidences, and sheer will of the competitive community. I also agree with Kadano that Sakurai was probably trying to make the best game he could. Making a balanced, 1v1, itemless platform fighter was probably on his lower end of priorities. Ironically, smash 4 looks less free-for-all oriented than Melee is (though it will probably never be as deep)
 

Metal Gear Salad

Smash Apprentice
Joined
Dec 3, 2013
Messages
78
Location
Peterborough, UK
Hi Kadano, massive fan of your work in this thread.

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

It'd be interesting to see it in slo mo as well cos I understand his sword swivvles round on the reverse. That'd be helpful anyway.

I appreciate it's a lot to ask for but thanks for your time if you can.
 
Last edited:
Top Bottom