• 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!

Code Decrypt: First Jab Mechanics-WIP; Mario Data Up, IC Next

Joined
May 3, 2009
Messages
7,190
ROB First Jab Mechanics

As I was testing KO percents for ROB(these will come in the near future), I noticed that ROB's jab has moving properties. But not just any movement. They were interesting to me, so I looked into them more. Here are my findings.

Universal Terminology-

Walk:
[collapse=This is debatable, and is only a tentative idea term. It isn’t necessary to understand these data]Walk: Walk is movement produced intentionally by a character. Walk is the general term I am creating to be applied to any movement with an animation specific to it. Hence, craq-walking is not walk because the movement occurs solely with the turnaround animation of the character. Wavedashing is not walk because it only occurs with an idle animation. Stutter-stepping is not walk because the animation is that of an attack.[/collapse]

Displacement: Displacement is an almost universal property of the game. Displacement is the general term I am creating to be applied to any movement of a character or object that is produced without being thrown or by walking(see above collapse if you wish)/without a specific animation. Hence, the movements produced by stutter-stepping, craq-walking, and wavedashing are all displacement.

Poser Point: The area in front of ROB where an opponent hit by ROB’s first jab will suffer from the effects of a Poser Push.

Pull: Pull is a property of ROB’s first jab hit. Pull only affects the opponent when a first jab connects, and pulls them closer to ROB’s Poser Point. Notice that after each consecutive Pull a character is in a less and less desirable position to get hit by ROB’s second jab. This is why the Poser Push and EQ are a part of ROB’s jab mechanics. Pull is a type of displacement.

EQ: EQ is a property of ROB’s first jab hit. EQ only occurs when a first jab connects. It pushes ROB a small distance backwards, and its displacement is less than that of Pull enough so that the opponent eventually reaches ROB’s Poser Point. Note that if it weren’t for EQ, then theoretically, Pull would eventually put the opponent behind ROB because of hurtbubble colliding. This would make two-hit jab combos impossible at close ranges. EQ is a type of displacement.

Troll: Troll only occurs when hitting an opponent that has their back to you, and with the tip of the first jab (test: only the tip?). Troll turns the opponent around and has a greater Pull than that of a jab connecting with a forward-facing opponent. I'm currently working on a theory as to why Troll happens, and I'm thinking it has to do with differing aerial hurt animations. Troll is in reality only an amplified Pull. Troll is a type of displacement.

Error: Occasionally, after ROB connects with the tip of his jab for the first time, he may miss completely with the next first jab, then hit and miss at random. Errors cease to occur after the opponent is pulled into the jab’s No Miss zone. Error does not happen with a backward facing opponent because of Troll. I can’t figure out why Error happens, but I’m thinking it has something to do with hitbubble and slight model distortion, as the Error zone is so small. For simplifying purposes, a glancing blow is recorded as an Error.

No Miss zone: The length of ROB’s jab where Errors do not occur.

Balance: The general term applied to the compiled effects of ROB’s first jab hit, I.e., Push, Troll, and EQ. Balance works toward getting the first jab to be able to connect with the second jab.

Poser Push: A Poser Push is the push an opponent receives when they are hit by ROB’s first jab while they are in his Poser Point. I decided to name this effect a “Poser Push” because it seems like a toned-down version of the knockback dealt by the second of ROB’s jabs, I.e., it is a second jab “poser.” Notice that after a Poser Push a character is in perfect position to get hit by ROB’s second jab.

--

Notes and Zone mappings

First Jab
3%
Hitbubble out: frame 3
Hitbubble end: frame 7
IASA start: frame 20
Animation end: frame 25



-Cyan bubble denotes the approximate area of the Error zone
-Light blue bubble denotes the approximate area of the Troll zone
-Navy blue bubble denotes the approximate area of the Poser Point
-Large purple bubble denotes the approximate area of the No Miss zone

-Decay of the first jab does not seem to affect neither Pull nor Troll in a negative way, or any way for that matter
-Damage seems to affect Pull in a positive way
-EQ seems to adjust automatically to balance with increasing Pull or Troll.

--

What do the numbers mean?

These data provide information on how many first jabs it takes to bring a character into ROB’s Poser Point, on what jab Poser Push is applied, and after the first Poser Push, how many jabs it takes to bring the character back into ROB’s Poser Point and on what jab Poser Push is again applied.

I have also provided character specific average Troll and Error values, and values for a character that is initially facing away from ROB, and a character that is already facing ROB.

--

Notations

To make this much easier to read and not as time-consuming, there are a number of notations I have created. They are as follows, with the notation first and what it means afterwards.

Variables and what they mean:

BDU- This term will not be used in this research, but it is something you need to know. BDU stands for Brawl distance units. BDU is the official distance measuring system the game uses. One BDU is equal to about one-twentieth of a Stage Builder block. This term was actually coined by Veril as far as I know
#- Any number has the value of one jab, which produces one Pull. These values are therefore simply going to be known as Jab-Pulls. I will give exact Jab-Pull values in BDU eventually, hopefully in the near future. I expect Jab-Pull values in BDU will be character specific due to differing weights, etc.
F- Test/trial for a forward facing character
B- Test/trial for a backward facing character
C- The amount of jabs it takes to get an opponent into ROB’s Poser Point
P- The jab on which Poser Push occurs; always has a value one Jab-Pull higher than C
R- The amount of jabs it takes to put an opponent back into ROB’s Poser Point after they have suffered from a Poser Push
P2- The jab on which Poser Push affects the opponent again; always has a value one Jab-Pull higher than R
Fpp- The number of jabs it takes to get a forward facing opponent into ROB's Poser Point
Bpp- The number of jabs it takes to get a backward facing opponent into ROB's Poser Point
Fpp2- The number of jabs it takes to get a forward facing opponent back into ROB's Poser Point after the first Poser Push
Bpp2- The number of jabs it takes to get a backward facing opponent back into ROB's Poser Point after the first Poser Push
T- Troll
E- Error
N- the number of trials conducted; refers to one forward facing test and one backward facing test
T/N^#- the amount of Trolls or Errors that occurred in a certain trial

A test is finding values once on either a forward or backward facing opponent. A trial is comprised of two tests, one on a forward facing opponent, and one on a backward facing opponent.

--

Formulae

The formulae I used in mining these data are extremely simple. I will provide them for archiving purposes.


To calculate Troll for one test-

T = Fpp - Bpp

Error is simply calculated by counting the amount of times ROB’s jab misses the opponent between after the first successful jab and after the opponent suffers from Poser Push. As I stated before, Error does not happen with a backward facing opponent due to Troll.


Additionally, here are the formulae for calculating average Troll and average Error:

T = (T^1 + T^2 + T^3)/3
E = (E^1 + E^2 + E^3)/3

Please note that it is not necessary to trial test three times. Indeed, trial testing more than three times may harbor more accurate data, but not more accurate by much. Three trials are the standard and still offer exceptionally accurate results.

How will data be presented?

Data will be presented in a format I hope is simple yet gives you everything you need to know about character values. The format is like so:

For average values, it will say AVERAGE beforehand, and for specific tests there will be a collapsed box, with test numbers preceding each data set.

AVERAGE-:character:

Code:
[I]F/B[/I]
[I]C=   ; P=
R-=  ; P2=
T/E=[/I]
Collapse box
--
Next character values


Data

And now, for my data.

AVERAGE-:mario2:

Code:
[I]F[/I]
[B][I]C=[/I][/B] 19; [B][I]P=[/I][/B] 20
[B][I]R=[/I][/B]  5; [B][I]P2=[/I][/B] 6
[B][I]E=[/I][/B] 3.67

[I]B[/I]
[B][I]C=[/I][/B] 16.3; [B][I]P=[/I][/B] 17.3
[B][I]R=[/I][/B] 4.67; [B][I]P2=[/I][/B] 5.67
[B][I]T=[/I][/B] 2.67
[collapse=Mario Trial Data]
Code:
Trial 1: F
Fpp= 18; P= 19
Fpp2= 5; P2= 6; E= 7

Trial 2: B
Fpp= 17; P= 18
Fpp2= 4; P2= 5

Trial 3: F
Fpp= 19; P= 20
Fpp2= 4; P2= 5; E= 3

Trial 4: B
Fpp= 16; P= 17
Fpp2= 5; P2= 6

Trial 5: F
Fpp= 20; P= 21
Fpp2= 6; P2= 7; E= 1

Trial 6: B
Fpp= 16; P= 17
Fpp2= 5; P2= 6
[/collapse]

--

Please tell me if I missed anything or if anything needs any clarification, as this was written in chunks in Word Works, and I might've missed something.

Thank you for taking the time to read this. Make of this data what you will.
 

Sudai

Stuff here
Joined
Feb 14, 2006
Messages
7,026
Location
Baton Rouge, Louisiana
Have fun with this. Hitbubble mod and frame by frame mod = awesome testing. But I don't know how or where you could get that. :ninja:
 

sneakytako

Smash Lord
Joined
Jun 8, 2009
Messages
1,817
Location
Cincinnati OH
Some quick thoughts

  • Displacement should mean when a character moves because another object has come to challenge that space, hence displacing the char. This would apply for pushing, but not stutter step/wavedash motions.
  • Pull and troll can be described as the sweet/sour spots of the Jab. The jab has a trajectory behind ROB, like Tlink's Dsmash.
  • There is a range/% where the first hit of Jab does knockback that's not PoserPush. (Most likely when your opponent is not grounded). Include this data please.
  • This number of jabs data is unusable in-game. In order to use this data in a game, we need percents. If you were to give a relative range for what percents do what kind of knockback during which situations that would be more useful imo.

Interesting data though, you should also test Up-tilt if you like these 'suction' properties :D.
 

stingers

Smash Obsessed
Joined
Oct 21, 2006
Messages
26,796
Location
Raleigh, NC
basically this means that robs jab1 has reverse knockback and similar properties to peach and pikachus dtilts which make them move backwards (but far less pronounced)?

cool stuff
 

Silhouette

Smash Lord
Joined
Apr 16, 2007
Messages
1,830
Location
Springfield/Athens, OH
Yeah cool stuff

and I agree with tako, it would also be interesting to see data like this for his U-tilt since it has potential 'Locks' on characters.
 
Joined
May 3, 2009
Messages
7,190
Have fun with this. Hitbubble mod and frame by frame mod = awesome testing. But I don't know how or where you could get that. :ninja:
http://www.smashboards.com/showthread.php?t=272304

There's actually a frame advance hack included in BBrawl's standard release, and all you've got to do to use it with vanilla is run it through codemanager to deactivate the rest of the codeset.

As for a hitbubble mod....;)

Some quick thoughts

  • Displacement should mean when a character moves because another object has come to challenge that space, hence displacing the char. This would apply for pushing, but not stutter step/wavedash motions.
  • Pull and troll can be described as the sweet/sour spots of the Jab. The jab has a trajectory behind ROB, like Tlink's Dsmash.
  • There is a range/% where the first hit of Jab does knockback that's not PoserPush. (Most likely when your opponent is not grounded). Include this data please.
  • This number of jabs data is unusable in-game. In order to use this data in a game, we need percents. If you were to give a relative range for what percents do what kind of knockback during which situations that would be more useful imo.

Interesting data though, you should also test Up-tilt if you like these 'suction' properties :D.
Displacement: Yeah, you summed up what I was thinking. You worded it much better though. I was also toying with the idea of seperating Displacement into Passive displacement, which would be what you were describing(character frame collision), and Active Displacement, which would be stuff like Wavedashing.

Pull and Troll: I don't think they are in fact spots, seeing as they don't deal any more/less damage, but they might be classified as such because of their differing KB. I also didn't know ROB's first jab had range behind ROB. :O

I'll look into the non-Poser Push knockback

I'm not really following you. The jab doesn't really deal KB, and the amount of distance Pull is increased by through percents is almost negligable. If you could sort of elaborate on it, I'd see what I could do...

basically this means that robs jab1 has reverse knockback and similar properties to peach and pikachus dtilts which make them move backwards (but far less pronounced)?

cool stuff
Yep, pretty much. though I'll admit, that this is pretty much useless in the heat of a match. The only thing that would really be usefull would be Poser Push alone, because it can lead into more options other than the second jab.

Yeah cool stuff

and I agree with tako, it would also be interesting to see data like this for his U-tilt since it has potential 'Locks' on characters.
UTilt doesn't really "lock", per say, unless the opponent has terrible DI, but the hitboxes at the side do pull the opponent higher above ROB to get hit by the second hitbox. If you could elaborate as well as to what kind of data you'd like to see, I'd gladly do it. :D
 

sneakytako

Smash Lord
Joined
Jun 8, 2009
Messages
1,817
Location
Cincinnati OH
I'm not really following you. The jab doesn't really deal KB, and the amount of distance Pull is increased by through percents is almost negligable. If you could sort of elaborate on it, I'd see what I could do...
For example, when you air release ness, you can follow up with a Jab>Grab. The jab has KB away from ROB when he hits an aerial foe if I'm not mistaken. I think I've seen this on the ground too, although it might be due to strange di...
 
Joined
May 3, 2009
Messages
7,190
Ah, I think I see what you mean. Do you want the rough amount of KB dealt at certain percents on aerial and grounded opponents?
 

sneakytako

Smash Lord
Joined
Jun 8, 2009
Messages
1,817
Location
Cincinnati OH
KB should not be quantified but rather more general and relative, like short and long, with a video :D. Also, Up-tilt may not lock, but maybe a combination of Jab/uptilit will lock? If you could master the range of the jab, this could be a very interesting mindgame.
 

sneakytako

Smash Lord
Joined
Jun 8, 2009
Messages
1,817
Location
Cincinnati OH
I was practicing some Jab locks, and it seemed to apply here.

Jabbing when the opponents are facing you will lock them, but it will not when they are facing away. Also the knockback seem to differ whether they are face down/up. Will post when I know more.
 
Top Bottom