MasterCorev3 Thread!

[Dantarion]

http://opensa.dantarion.com/s4/mastercore3/

This is my data resource for Smash4
It includes raw dumps created by parsing through the patch files, as well as diffs that attempt to isolate changes between versions.

It currently has data for....

All characters and all articles (projectile, luma, etc)
params (some are linked to subactions and displayed inline, but thers also a seperate dump of unknown data too)
Basic display of real in-game frame number estimates
diffs for the most recent patch

TODO

anchor navigation
more labels
dropdown navigation in the header
Advanced display of real in-game frame number (FrameSpeedMods, Loops, etc)
Thinkaman's labels

I need feedback. I will update this thread when I add new stuff.

 

[TheReflexWonder]

I wanted to look at it to explain a weird phenomenon found with Falco D-Air that people can't seem to replicate. From what I can "read" of your data dump, this shouldn't be possible:

http://opensa.dantarion.com/s4/mastercore3/dump/144/falco

06C - ATTACKAIRLW - DE3CDC95
params: {1, 0, IASA?=52, 0, 0, 0}

AsynchronousTimer(4.000000)#frame 5
?SetBit(0x2100000d)
AsynchronousTimer(16.000000)#frame 17
Hitbox_B738EABD(ID=0x0, Part=0x0, Bone=0x5, Damage=13.000000, Angle=0x11d, KBG=0x50, FKB=0x0, BKB=0xa, Size=4.200000, X=3.500000, Y=0.000000, Z=-1.000000, Effect=0x0, Trip=0x0, Hitlag=2.000000, SDI=1.000000, Clang=0x1, ?=0x1, SFXLevel=0x0, SFXType=0x2, Ground/Air=0x2, Direct/Indirect=0x2, 0x1, 0x6)
Hitbox_B738EABD(ID=0x1, Part=0x0, Bone=0x5, Damage=13.000000, Angle=0x50, KBG=0x50, FKB=0x0, BKB=0xa, Size=4.200000, X=3.500000, Y=0.000000, Z=-1.000000, Effect=0x0, Trip=0x0, Hitlag=2.000000, SDI=1.000000, Clang=0x1, ?=0x1, SFXLevel=0x0, SFXType=0x2, Ground/Air=0x2, Direct/Indirect=0x1, 0x1, 0x6)
AsynchronousTimer(20.000000)#frame 21
Hitbox_B738EABD(ID=0x0, Part=0x0, Bone=0x5, Damage=8.000000, Angle=0x169, KBG=0x5a, FKB=0x0, BKB=0x14, Size=5.300000, X=3.500000, Y=0.000000, Z=-1.000000, Effect=0x0, Trip=0x0, Hitlag=1.000000, SDI=1.000000, Clang=0x1, ?=0x1, SFXLevel=0x0, SFXType=0x1, Ground/Air=0x2, Direct/Indirect=0x3, 0x1, 0x6)
AsynchronousTimer(31.000000)#frame 32
RemoveAllHitboxes()
AsynchronousTimer(38.000000)#frame 39
unk_17232732(0x2100000d)
End()

I get that the ID1 hitbox doesn't go away when the sourspot appears, but the sourspot is on the same bone with the same x/y/z coordinates and is bigger. All hitboxes appear to have the same ground/air hitting potential. Since ID0 has priority over ID1, I can't imagine how the 13% D-Air hitbox could hit while the sourspot is out. What am I missing here?

 

[icraq]

Ryu's data is kinda wonky, I'm assuming because it didn't parse correctly because of his extra moves, so I believe I found most of his subactions, they're posted in http://smashboards.com/threads/ryus-frame-data.408822/ down near the bottom. I'm also very confused about his fair's faf and how the fsm applies to it.

Anyways I hope this helps somehow. You've been a great help to me.

 

[pikazz]

my biggest consern, does it also show the data for Normal Items and Projectile Items? (not projectiles like Lucarios Aura Sphere or Pikachus Thunder Jolt but all the Projectile Items like Link Bombs, Peach Turnips and Jr Mecha Koopa)

 

[Dantarion]

Ryu's data is kinda wonky, I'm assuming because it didn't parse correctly because of his extra moves, so I believe I found most of his subactions, they're posted in http://smashboards.com/threads/ryus-frame-data.408822/ down near the bottom. I'm also very confused about his fair's faf and how the fsm applies to it.

Anyways I hope this helps somehow. You've been a great help to me.

Yeah, im gonna have to take a look at ryu, he has all kinda weird stuff.

my biggest consern, does it also show the data for Normal Items and Projectile Items? (not projectiles like Lucarios Aura Sphere or Pikachus Thunder Jolt but all the Projectile Items like Link Bombs, Peach Turnips and Jr Mecha Koopa)

I have diffs but not dumps of those. Ill fix that sometime this week.

 

[Shaya]

Perhaps you should sort the order of what we see by whether or not it has "game" particular data. Then I guess by any IASA greater than 0?
(long time hopeful)

 

[Zage]

Perhaps you should sort the order of what we see by whether or not it has "game" particular data. Then I guess by any IASA greater than 0?
(long time hopeful)

What does a value of 0 mean? Some new changes include a new IASA value of 0.

 

[Shaya]

What I mean is that things that say IASA=0 are probably things we don't care about. But otherwise it's perhaps useful

It would mean that there is no 'restriction' while being in that animation. Like walking for example.

 

[Linkshot]

Cleaning up my list of labels, will edit this post with a spoiler of them when they're done.

Alright, what I know and hypothesize are both done. Ton of blank spots on things I can't even fathom yet.

Spoiler: CHANGED PARAMETERS IN 1.1.0

FLOATING POINTS

Bowser Jr.
[17][2] Clown Cannon Initial Damage
[17][3] Clown Cannon Max Damage

Dark Pit
[19][1] ???Silver Bow Base Damage???
[19][4] ???Guiding Bow Base Damage???
[19][7] ???Piercing Bow Base Damage???

Link
[19][0] ???Bow Initial Damage Modifier???
[19][1] ???Bow Max Charge Damage Modifier???

Little Mac
[12][2] ???Rising Uppercut Helpless State Fall Speed???
[12][7] ???Rising Smash Helpless State Fall Speed???
[12][79] ???Giga Mac Rising Uppercut Helpless State Fall Speed???

Lucas
[20][4]

Lucina
[14][1] Dolphin Slash Landing Lag
[14][10] Crescent Slash Landing Lag
[14][19] Dolphin Jump Landing Lag

Marth
[14][1] Dolphin Slash Landing Lag
[14][10] Crescent Slash Landing Lag
[14][19] Dolphin Jump Landing Lag

Mii Swordfighter
[14][44]
[15][12]
[16][24]
[16][25]
[16][26]
[16][27]
[16][52]
[16][53]
[16][54]
[16][55]
[16][84]
[16][85]
[16][86]
[16][87]
[17][24]
[17][26]
[17][52]
[17][54]
[17][84]
[17][86]
[18][44]
[18][45]
[18][46]
[18][47]
[18][52]
[18][53]
[18][54]
[18][55]
[18][62]
[18][63]
[18][64]
[18][65]
[18][66]
[18][67]
[18][68]
[18][69]
[18][70]
[18][71]
[18][72]
[18][73]
[18][74]
[18][75]
[18][80]
[18][81]
[18][82]
[18][83]
[18][84]
[18][85]
[18][86]
[18][87]
[21][1] ???Skyward Slash Dash Speed???
[21][3] ???Aerial Skyward Slash Dash Speed???

Robin
[14][2] Elwind First Gust Aerial Control
[14][3] Elwind Second Gust Aerial Control
[14][9] Elwind Helpless State Aerial Control
[14][29] Gliding Elwind First Gust Aerial Control
[14][30] Gliding Elwind Second Gust Aerial Control

Shulk
[12][40] ???Back Slash Horizontal Speed???
[12][81] ???Back Slash Leap Horizontal Speed???
[12][122] ???Back Slash Charge Horizontal Speed???

Wii Fit Trainer
[12][1] Sun Salutation Healing
[12][9] Sweeping Sun Salutation Healing
[13][5]
[13][6]
[13][7]
[13][8]
[13][9]
[13][10]
[13][11]
[13][12]
[13][13]
[13][14]
[13][15]
[13][16]
[13][17]
[13][18]
[13][19]
[13][20]
[13][21]
[13][22]
[13][23]
[15][14]
[15][15] Deep Breathing Damage Modifier
[15][58]
[15][80]
[19][20]
[19][28]
[19][29]
[19][68]
[19][106]
[19][107]
[19][146]

I definitely encourage people to just copy and fill that in with corrections and new data.

 

[Dantarion]

Don't bother marking the [0][xxx] ones, they are all clearly linked to subactions, and now appear in the diffs next to each subaction

 

[Linkshot]

Edited the spoiler to only show the floating points (or non-subactions), also changed Link's [19][0] and [19][1] hypotheses, since those numbers seem way to early in [19] to be the Power variation.

 

[TheReflexWonder]

Lots of question marks on Little Mac's thing. Is it his fall speed during SpecialFall?

I wish it were landing lag. :<

 

[Linkshot]

Lots of question marks on Little Mac's thing. Is it his fall speed during SpecialFall?

I wish it were landing lag. :<

Anything with question marks are things people have brought up, check out with the variables, but haven't gotten to confirm.

The actual values are

Changed fighter_param_vl_littlemac.bin[12][2] from 1.79999995232 to 1.89999997616
Changed fighter_param_vl_littlemac.bin[12][7] from 3.09999990463 to 3.40000009537
Changed fighter_param_vl_littlemac.bin[12][79] from 3.0 to 3.09999990463

It could very well be the descent before the SpecialFall

 

[icraq]

I was digging through Ryu's stuff some more.

So the only changes were to subactions 22B and 22C which are for Down-B Focus Attack, I don't really know what their function is exactly but it's weird.

Basically the only effective change I see is it's causing ?SetBit(0x21000013) to happen 1 frame later, on frame 11, for both entries. I don't know why there's two entries for focus attack. I just know the damage in there matches focus attack, 12, 10 and 17.

Any ideas?

Also here's his subactions I've found:

Spoiler

air dodge subaction 4E

spot dodge subaction 4B

both rolls are 4E 4C

dtilt1 (subaction 219)
dtilt2 (subaction 21A)

utilt1 (subaction 217)
utilt2 (subaction 218)

dash attack (subaction 056)

ftilt 1 (subaction 215)
ftilt 2 (subaction 216)


fsmash (subaction 05F)
dsmash (subaction 065)
usmash (subaction 063)

fair (subaction 069)
nair (subaction 068)
bait (subaction 06A)
uair (subaction 06B)
dair (subaction 06C)

grounded shoryuken B version subaction 223
grounded shoryuken true inputted version subaction 224

aerial shoryuken B version subaction 227
aerial shoryuken true inputted version subaction 228

focus attack something or another 22B and 22C

 

[TheReflexWonder]

I was looking at the data dump for stuff to theorize about and saw differences between Wario D-Air in Brawl and Smash 4 in terms of hitbox size.

Assuming bones and animations are the same between Brawl and Smash 4 animations, would it be possible to create reasonable approximations of Brawl veterans' hitboxes with PSA/BrawlBox?

I tried to do it with Smash 4 Wario D-Air, and it seems accurate based on your data dump, but I don't know if there are inherent scale differences getting in the way or anything.

Spoiler

My D-Air example ignores the ground-only hitbox in ID 1 because it's in the same spot as the air-only one on his head.

Spoiler

 

[Dantarion]

We may be better off just looking at the Sm4sh models and animations and making a hitbox viewer that way.

 

[TheReflexWonder]

I've heard that people have the ability to mod Smash 4 now. Does that mean we have access to the individual models and animations, then, even if we don't have a good way to look at or modify them yet? Is a hitbox viewer something that could happen in the near future?

 

[Dantarion]

"People" = me and soneek.
Model and animation format isn't fully known yet, but we have all the files.
I want to integrate a hitbox viewer into the game itself, but its a long way off.

Basically whats going to be happening soon is that I am going to be developing a way to test patches on games, much like Riivolution or Bean and Chadderz's Caffine.

Right now the issue is that we have very spotty debugging tools. Basically, this entire process is YEARS accelerated past the point where we were this many years into Brawl's lifetime. We have a moveset editor before we have a way to test the files produced by the moveset editor!

Its going to be interesting to see where this takes the scene in the next 6 months. While my focus this time is going to be on adding customization and competitive based features to the game, and not making something new, I am excited to be working on more low level things like exception handlers, asm patching, etc, because thats what got me where I am now, and thats the kinda stuff I liked to do. Reflex, you know that I barely do anything on PM anymore, and thats mostly because this kinda work is no longer needed for PM.

 

[Ffamran]

You're probably working on it, but checking Falco's data and it seems like his Nair's last hit and Fair's last hit are missing. I remember seeing them before you overhauled the Master Core.

 

[drakeirving]

unk_F13BFE8D seems to set a flag for a bone to become partially invincible. First parameter being the bone, second likely being the mode. Relatedly, these moves generally call unk_CEDC237E(0x0) after, which I can only assume deactivates partial invincibility on all bones.

Also, I think some hitbox params are mislabeled. BKB and KGB seem to be reversed on at least Hitbox_2988D50F and Hitbox_14FCC7E4; either that, or Hitbox_B738EABD is the one with them reversed and everyone's been writing down the wrong information for months.