Project M -- Post if you'd like to offer assistance meleefying chars

[shanus]

I wouldn't trust kupo's data. Although well-intentioned, if it was done the way you said it was (by timing how long it takes for a char to fall to die on hyrule) then that means it was considering both fall speed acceleration and terminal velocity in the figures, which shouldn't necessarily scale equally like that.

Almas' velocity values clearly scales the terminal velocity, if the terminal velocity didn't scale with it, then multiplying it would result in the velocities capping out which is clearly not the case. As the excel method accounts for recalculating the terminal velocity, kupo's total gravity should be correct. In fact, if you look at shell's data for DK (which matches the frames in air exactly, I think it was only 1% off from Kupo's calculations).

So by taking his total gravity measurements, you can simply extrapolate them to determine the new true terminal velocities and jump power (exactly how we were doing it before).

 

[leafgreen386]

That's the thing... almas's code does affect terminal velocity. Meaning the terminal velocity is going to be higher than it should be for a given gravity. That means that it's possible to achieve the same time values with a lower gravity but higher terminal velocity. It'll probably be close to melee, but in most cases there should be a better fit, and since a character's fall acceleration matters more than their terminal velocity when fighting close to the ground, the characters will still be overall slower compared to melee.

 

[shanus]

Alright, new progress:

I need someone to test this code. This code will allow us to take all the new pacs from the next set, and *not* have to modify their ALR by doing the following:

If button Not pressed: 3
Set Frame speed .5x
Else
Generate L-Cancelling graphic

Also, now that I understand the principle of how injectoins work, I will begin drafting up a HAD injection code.

L Canceling v1.2 plus Edge Canceling, Dantarion/Yeroc/Shanus
07060E08 000000F8
08000100 81060E40
000A0200 81060E48
04070100 81060E58
000E0000 81060E60
11001000 81060E68
000F0000 00000000
02010200 80FAF3EC
00090100 81060FE0
00000000 00000001
00000006 00000033
00000000 00000003
00000001 00007530
00000000 00000000
00000000 00000010
00000000 00000000
00000001 FFFA81C0
00000001 00075300
00000001 FFFBE150
00000001 00000000
00000001 00000000
00000001 00000000
00000001 0000AFC8
00000001 00000000
00000001 00000000
00000001 00000000
00000001 00000000
00000001 00000000
00000001 00000000
00000003 00000001
00000002 80FC1C18
00000002 81060E08
06FC1C58 00000008
00090100 81060EF8

Shell, have you managed to find any time to test the gravity ports?

If this code works, we can take the next nightly pac files and dump the gravities into there. Then we will hope to have a HAD injection by Friday or so.

 

[Yeroc]

Alright, optimized. Thanks for doing the bulk of the reworking Shanus

L Canceling v1.2 plus Edge Canceling, Dantarion/Yeroc/Shanus
07060F10 000000F0
00000000 00000001
00000006 00000033
00000000 00000003
00000001 00007530
00000000 00000010
00000000 00000000
00000001 FFFA81C0
00000001 00075300
00000001 FFFBE150
00000001 00000000
00000001 00000000
00000001 00000000
00000001 0000AFC8
00000001 00000000
00000001 00000000
00000001 00000000
00000001 00000000
00000001 00000000
00000001 00000000
00000003 00000001
00000002 80FC1C18
00000002 81060FC0
08000100 81060F10
000A0200 81060F18
04070100 81060F28
000E0000 00000000
11001000 81060F30
000F0000 00000000
02010200 80FAF3EC
00090100 81060FB0
06FC1C58 00000008
00090100 81060FB8

 

[shanus]

Need help compiling this with the appropriate addresses and techniques for subroutines.

Total line count I think is like 91 or 93 lines of code:

Hybrid Air Dodging v.1, Shanus
07060Z08 000000YY    (ADJUST Z AND YY - CTRL F XXXXXX and insert appropriate Mem address)

THIS PART IS FOR INSERTION INTO ACTION 21:
000A0400 XXXXXX00    IF NEUTRAL
000B0400 XXXXXX20        IF NEUTRAL
04000100 XXXXXX40             Change subaction: air dodge
000E0000 00000000    Else
02010200 XXXXXX48        Change Action 19, req = on ground
04000100 XXXXXX58        Change Subaction: Air dodge
12060200 XXXXXX60        Set Float Var: Ra-Float80 = IC Basic 1011
12060200 XXXXXX70        Set Float Var: RA-Float83 = IC basic 1018
120A0100 XXXXXX78        Bit Variable set: RA-Basic85
00070100 CALLSUBR        Call subroutine for the movement (lasts 3 frames)
00020100 XXXXXX80        Asynch timer: 3
06050100 XXXXXX88        Body collision: intangible
00070100 CALLSUBR        Call subroutine for the movement (lasts 3 frames)
00070100 CALLSUBR        Call subroutine for the movement (lasts 3 frames)
00040100 XXXXXX90        Set Loop 16 times
0E080400 XXXXXX98            Set all momentum to 0
00010100 XXXXXXB8            Synch Frame = 1
00050000 00000000        Execute loop
06050100 XXXXXXC0        Body Collision: Normal
0E060100 XXXXXXC8        Disable Horizontal Movement
00010100 XXXXXXD0        Synch Timer: 16 Frames
120B0100 XXXXXXD8        Bit var clear: Ra-Basic[85]
02010200 XXXXXXE0        Change action: 10, req in air
000F0000 00000000    End If

DATA:
Start:
000F0000 00000000
02010200 80FAF3EC
00090100 81060FE0
IF:
00000006 00000007
00000005 000003F5
00000000 00000000
00000005 00000C5C
If2:
00000006 00000007
00000005 000003FB
00000000 00000000
00000005 00000C5D
Change subaction = Air dodge
00000000 00000046
Else
Change Action SpecialFallground, req On Ground
00000000 00000019
00000006 00000003
Change subaction = Air dodge
00000000 00000046
Set RA-Float[80] to equal joystick horizontals (IC-Basic1011)
00000005 000003F3
00000005 21000050
Set RA-Float[83] to equal joystick horizontals (IC-Basic1018)
00000005 000003FA
00000005 21000053
Set RA-Basic[85] to true
00000005 20000055
Asynch Timer = 3 (scalar)
00000001 00012BF20
Body Collision Intangible
00000000 00000002
Set Loop 16 times
00000000 00000010
Set Horiz & Vert momentum to zero
00000000 00000000
00000000 00000000
00000000 00000001
00000000 00000001
Synchronous Timer Frames=1
00000001 0000EA60
Body Collision Normal
00000000 00000000
Disable Horizontal movement
00000000 00000002
Synchronous Timer Frames=16
00000001 000EA600
Bit var clear, Ra-Basic 85 to false
00000005 20000055
Change Action SpecialFall, req In Air
00000000 00000010
00000006 00000004

-------------------------------------------------

SUBROUTINE STUFF:
00040100 CCCCCC00        Set Loop 3 times
0E080400 CCCCCC08            Set all momentum to 0
0E020100 CCCCCC28            Set gravity to 0
0E010200 CCCCCC30            Add momentum X/Y equal to RA Floats
0E010200 CCCCCC30            Add momentum X/Y equal to RA Floats   (RUNNING THIS TWICE, BUT WRITING ONCE)
02010200 CCCCCC50            Change Action 19, req = on ground
00010100 CCCCCC60            Synch Frame = 1
00050000 00000000        Execute loop

Data:
Set Loop 3 times
00000000 00000003
Set Horiz & Vert momentum to zero
00000000 00000000
00000000 00000000
00000000 00000001
00000000 00000001
Set Gravity to zero
00000000 00000001
Add momentum x/y equal to RA floats
00000005 21000050
00000005 21000053
Change Action SpecialFallground, req On Ground
00000000 00000019
00000006 00000003
Synchronous Timer Frames=1
00000001 0000EA60

-----------------------------------------------

THIS PART IS FOR ACTION 19
000A0200 DDDDDD00   If Bit is Set: RA-Basic[85]
08000100 DDDDDD20       Set Air Ground 0-1
04070100 DDDDDD28       Frame speed set at 2.5x
00020100 DDDDDD30       Synch Timer = 25 frames (really 10 frames)
64000000 00000000       Allow Interrupt
000F0000 00000000   End If
120B0100 DDDDDD38   Bit Var clear: RA-Basic[85]=false


Data:
If bit is set, ra basic 85
00000006 00000008
00000005 20000055
Set Air Ground to 0-1
00000000 00000001
Frame Speed
00000001 000249F0
Synch timer
00000001 0016E360
Bit var clear, Ra-Basic 85 to false
00000005 20000055


-----------------------------------------------
End (EDIT ADDRESSES):
00000002 XXXXXX08
06FC1C58 00000008
00090100 XXXXXXE8

 

[Almas]

Be careful with the whole ROB/special fall thing. Unless you found a way to fix that already.

 

[Yeroc]

The last time we did one of these he just went into regular fall. In fact, nearly every Brawl newcomer, if they had a real special fall or not, went into regular fall afterwards. I ****ing hate the programmers for this game sometimes.

 

[shanus]

The last time we did one of these he just went into regular fall. In fact, nearly every Brawl newcomer, if they had a real special fall or not, went into regular fall afterwards. I ****ing hate the programmers for this game sometimes.

That's because we were silly. We had the code ending the action at 45 frames, whereas their animation actually lasted below 45!

Air Dodge

3-29 / 39 Metaknight
4-29 / 39 Zero Suit Samus
4-29 / 39 Diddy
4-29 / 39 Pit
4-29 / 39 Squirtle
4-29 / 39 Ivysaur
4-29 / 39 Charizard
4-29 / 39 Ike
4-29 / 39 Snake
4-29 / 39 King Dedede
4-29 / 39 Lucario
4-29 / 39 Wario
4-29 / 39 R.O.B
4-29 / 39 Olimar

4-29 / 48 Ice Climbers
4-29 / 48 Ness
4-29 / 48 Sonic
4-29 / 48 Lucas

4-30 / 49 Bowser
4-29 / 49 Mario
4-29 / 49 Donkey Kong
4-29 / 49 Link
4-29 / 49 Samus
4-29 / 49 Kirby
4-29 / 49 Fox
4-29 / 49 Pikachu
4-29 / 49 G&W
4-29 / 49 Luigi
4-29 / 49 Zelda
4-29 / 49 Sheik
4-29 / 49 Falco
4-29 / 49 Yoshi
4-29 / 49 Ganondorf
4-29 / 49 Wolf
4-29 / 49 Toon Link
4-29 / 49 Captain Falcon
4-29 / 49 Jigglypuff

5-30 / 49 Marth
4-19 / 49 Peach

 

[Yeroc]

Here's the preliminary code. ROB, DDD, and for some reason Luigi all don't function. Marth, Ike, and Diddy still have some landing lag issues, and everyone else I tried all work well. And it's only like half as long as the other code. W00t.

HAD 1.3
07060C00 00000308
00000006 00000007
00000005 000003F5
00000000 00000000
00000005 00000C5C
00000006 00000007
00000005 000003FB
00000000 00000000
00000005 00000C5D
00000000 00000046
00000000 00000019
00000006 00000003
00000000 00000046
00000005 000003F3
00000005 21000050
00000005 000003FA
00000005 21000053
00000005 20000055
00000001 0002BF20
00000000 00000002
00000000 00000010
00000000 00000000
00000000 00000000
00000000 00000001
00000000 00000001
00000001 0000EA60
00000000 00000000
00000000 00000002
00000001 000AFC80
00000005 20000055
00000000 00000010
00000006 00000004
00000000 00000003
00000000 00000000
00000000 00000000
00000000 00000001
00000000 00000001
00000000 00000001
00000005 21000050
00000005 21000053
00000000 00000019
00000006 00000003
00000001 0000EA60
00000006 00000008
00000005 20000055
00000000 00000001
00000001 000249F0
00000001 0016E360
00000005 20000055
00000002 81060E78
00000002 80FB18FC
00000002 80FC1CA8
00000002 81060DA8
00000002 81060EC0
00070100 80FB18E4
000A0400 81060C00
000B0400 81060C20
04000100 81060C40
000E0000 00000000
02010200 81060C48
04000100 81060C58
12060200 81060C60
12060200 81060C70
120A0100 81060C80
00070100 81060D80
00020100 81060C88
06050100 81060C90
00070100 81060D80
00070100 81060D80
00040100 81060C98
0E080400 81060CA0
00010100 81060CC0
00050000 00000000
06050100 81060CC8
0E060100 81060CD0
00010100 81060CD8
120B0100 81060CE0
02010200 81060CE8
000F0000 00000000
00090100 81060D88
00040100 81060CF8
0E080400 81060D00
0E020100 81060D20
0E010200 81060D28
0E010200 81060D28
02010200 81060D38
00010100 81060D48
00050000 00000000
00080000 00000000
000A0200 81060D50
08000100 81060D60
04070100 81060D68
00010100 81060D70
64000000 00000000
000F0000 00000000
120B0100 81060D78
02010200 80FAF454
00090100 81060D90
06FB18F4 00000010
00090100 81060D98
00020000 00000000
06FC1CA0 00000008
00090100 81060DA0

 

[shanus]

Awesome job yeroc. I'll look into if I can figure out why. If not i can install it manually for each of them, which isnt hard, barring D3.

I saw in the IRC that the new L cancelling code doesn't shift the Allow interrupt time which is too bad. At least the old Lcancelling injection worked.

 

[shanus]

Double post, but some other minor changes.

We should look into adding the following:



Under action 7F:
--Change Action E: req=animation end-- (Already There)
Add in the following to this statement:

02040200 ZZZZZZ00    Additional requirement, Bit is Not Set: RA-Basic[85]
02010200 ZZZZZZ10    Change Action: 10, req=animation end
02040200 ZZZZZZ20    Additional requirement, Bit is Set: RA-Basic[85]
02010200 ZZZZZZ30    Change Action: 19, req=On Ground
02040200 ZZZZZZ40    Additional requirement: Bit is set: RA-Basic[85]

Data line:
Additional Req: Bit is NOT set - RA Basic 85
00000006 80000008
00000005 20000055
Change Action SpecialFall, req Animation End
00000000 00000010
00000006 00000001
Additional Req: Bit is set - RA Basic 85
00000006 00000008 
00000005 20000055
Change Action SpecialFallGroundlag, req On Ground
00000000 00000019
00000006 00000003
Additional Req: Bit is set - RA Basic 85
00000006 00000008 
00000005 20000055

This would make it so if you grapple during your MAD, you enter special fall. Or if you MAD and grapple into the ground, you still get 10 frames of landing lag.

Under Actions, 10, 44, and 45, the first line should be:
Bit variable clear: RA-Basic[85]

120B0100 XXXXXXD8        Bit var clear: Ra-Basic[85]

We already have it in memory quite a few times, but yeah...

This would make sure that if you are in special fall OR hit out of your air dodge before the bit clears, it will allow you to drop items again during a BAD or make sure your grapples don't enter special fall again.


Thoughts?


Also, should we make a trivial injection to enable JC-Grabs and Power shield reflecting?

 

[Shell]

(Yes please!)

If you could implement the old PS-Reflect idea of PSing and pressing B (?) to reflect at the cost of 1.5-2x shield damage that'd be ****. **** is censored?! Boobs? That'd be boobs just doesn't have the same ring...

Initial JC-grab ftw.

 

[shanus]

Alright, so here are the current needs of the HAD code:

for Marth & Ike and & Diddy
we just need to copy in subaction 33 changes
for Ness Yoshi and Lucas
we just need to copy in the whole code
and for D3, fix his subaction 46
and for ROB, Brawlbox in some animatoins for him

Specifically for ROB:
Subaction 38 is StepFall, I want to use that animation to give him graphics for his special fall with subactions 26-28; For landlag (33), I'd like to give him an animation of dair, 66

Then we need to add in the flashing effect for him during special fall, and make sure he can grab the ledge during it.

 

[Shell]

I couldn't resist trying out the HAD code v1.3 and here's my feedback:

-The landing lag for a WD is much more than 10 frames. I couldn't get Frame Advance to cooperate, but I'm guessing it's around 40ish.

-If you WD off of a ledge, you'll continue as if on an invisible platform until the AD ends, and then fall. AKA the Wiley E. Coyote Effect.

-Some characters special landing animation looks really awkward for WDing... check out Boozer and DK, for example.

-I also looked at the L-canceling 1.2 code and something's not working, as I think you might've already pointed out. The animation is constantly played at .5x speed, and it seems like the interrupt might be faster or not, I can't tell.

-Shanus, I dunno what those gravity test .pacs from your 11-16 folder were, but the hops and jumps are all super high. Were those Kupo's values?

 

[shanus]

which characters, shell?

yeroc only reported the bugs on landing lag with marth, ike, and diddy.

I havent tested it yet, so i know nothing of its failures.


Physics i didnt est yet, those were kupos. I'll use our SH and FHs, then

 

[shanus]

Double post update on the physics:

This excel should work better then:
http://www.mediafire.com/?y1iwoimmyzu

 

[Shell]

which characters, shell?

yeroc only reported the bugs on landing lag with marth, ike, and diddy.

I havent tested it yet, so i know nothing of its failures.


Physics i didnt est yet, those were kupos. I'll use our SH and FHs, then

There seemed to be ~40 frames of landing lag and the Wiley E. Coyote Effect(TM) for every character. I can double check that the code c/p'd correctly and make a fresh .gct later, but that's what I saw.

Also, I'm 99% positive that "Terminal Velocity" is, in fact, terminal velocity. With the square function added in, DK feels about right.

 

[shanus]

There seemed to be ~40 frames of landing lag and the Wiley E. Coyote Effect(TM) for every character. I can double check that the code c/p'd correctly and make a fresh .gct later, but that's what I saw.

Also, I'm 99% positive that "Terminal Velocity" is, in fact, terminal velocity. With the square function added in, DK feels about right.

There is also the chance Yeroc cp'd the wrong version of the code?

I'll need to crosscheck it later when i leave work.

 

[Dantarion]

Ill do JC grabs sometime today or something.
Where the hell in memory are you guys putting these subroutines :D

 

[shanus]

I figured out the endlag bug which also caused te coyote bug.

HAD 1.3
07060C00 00000308
00000006 00000007
00000005 000003F5
00000000 00000000
00000005 00000C5C
00000006 00000007
00000005 000003FB
00000000 00000000
00000005 00000C5D
00000000 00000046
00000000 00000019
00000006 00000003
00000000 00000046
00000005 000003F3
00000005 21000050
00000005 000003FA
00000005 21000053
00000005 20000055
00000001 0002BF20
00000000 00000002
00000000 00000010
00000000 00000000
00000000 00000000
00000000 00000001
00000000 00000001
00000001 0000EA60
00000000 00000000
00000000 00000002
00000001 000AFC80
00000005 20000055
00000000 00000010
00000006 00000004
00000000 00000003
00000000 00000000
00000000 00000000
00000000 00000001
00000000 00000001
00000000 00000001
00000005 21000050
00000005 21000053
00000000 00000019
00000006 00000003
00000001 0000EA60
00000006 00000008
00000005 20000055
00000000 00000001
00000001 000249F0
00000001 000927C0
00000005 20000055
00000002 81060E78
00000002 80FB18FC
00000002 80FC1CA8
00000002 81060DA8
00000002 81060EC0
00070100 80FB18E4
000A0400 81060C00
000B0400 81060C20
04000100 81060C40
000E0000 00000000
02010200 81060C48
04000100 81060C58
12060200 81060C60
12060200 81060C70
120A0100 81060C80
00070100 81060D80
00020100 81060C88
06050100 81060C90
00070100 81060D80
00070100 81060D80
00040100 81060C98
0E080400 81060CA0
00010100 81060CC0
00050000 00000000
06050100 81060CC8
0E060100 81060CD0
00010100 81060CD8
120B0100 81060CE0
02010200 81060CE8
000F0000 00000000
00090100 81060D88
00040100 81060CF8
0E080400 81060D00
0E020100 81060D20
0E010200 81060D28
0E010200 81060D28
02010200 81060D38
00010100 81060D48
00050000 00000000
00080000 00000000
000A0200 81060D50
08000100 81060D60
04070100 81060D68
00010100 81060D70
64000000 00000000
000F0000 00000000
120B0100 81060D78
02010200 80FAF454
00090100 81060D90
06FB18F4 00000010
00090100 81060D98
00020000 00000000
06FC1CA0 00000008
00090100 81060DA0

KNOWN ISSUES:
Marth, Ike, and Diddy groundlag is not speeding up. We need to take the code from subaction 33 from this file: http://shanemulliganphotography.blo...gold/Melee2.0/11-15-09/GravityPort/FitFox.pac

and place it in the pacs for them from the lastest nightly.

Ness, Yoshi, & Lucas do not perform HAD at all. We need to take the above file, and copy pasta in the HAD code in its entirely (subaction 46, 33, and appropriate routine)

D3, Luigi & ROB crash. D3 appears to be because he has no interrupt lines in subaction 46, will need to explore it.



We need to figure out these few stragglers issues, but we also need to do the following:

Take Cape's latest nightly pac files: http://www.filefront.com/14949387/GSH1118.rar

Modify Landlag according to OP, and import physics from here: http://www.mediafire.com/?y1iwoimmyzu

 

[Shell]

What were the problems with using the fall 1 or 2 lag animations, again? DK literally falls on his back and shakes the screen for a split second with every wavedash when using special fall.

Also, any ideas on what's wrong the L-cancel 1.2 code? Edit: I was using the .pacs from the 11/16 folder, including my own DK .pac.

 

[shanus]

What were the problems with using the fall 1 or 2 lag animations, again? DK literally falls on his back and shakes the screen for a split second with every wavedash when using special fall.

Also, any ideas on what's wrong the L-cancel 1.2 code? Edit: I was using the .pacs from the 11/16 folder, including my own DK .pac.

The main issue is the duration of the move. However, things seem to behave differently with fighter.pac injections

In the old pac code, i sped up the landing lag animations by 2.5x, and then set a synchronous timer at 25 frames followed by an allow interrupt to give a 10 frame lag window which did not look awkward.

For some reason, when injected into fighter.pac, the speed up only effected the animation, and not the synch timer. This is why you were in a laggy state forever, and slid off the level and could not fall. By changing the timer to 10 frames from 25, it fixed all that.

I do not know why the inverse L-cancelling code (.5x instead of 2x) does not work, but we will just need to port over the landing lag to 2x values from the nightly which we can do while editting the gravities. Not too bad.


I'm going to work on fixing the pacs and begin gravitiy ports tomorrow.


The end is in sight!

 

[shanus]

I have currently ported ALRS+gravity for:

Fox
Ness (should now have a working HAD - untested)
Yoshi (should now have a working HAD - untested)
Lucas (should now have a working HAD - untested)
Ike (should have working specialfall landlag now - untested)
Marth (should have working specialfall landlag now - untested)
Diddy (should have working specialfall landlag now - untested)


If the above all work, that leaves D3, Luigi, and ROB to fix.

Then, finish porting landing lags for L-cancelling and new physics.


Download for testing:
http://shanemulliganphotography.blogsite.org:6111/~shane/Neargold/Melee2.0/11-18-09/


ALL CHARACTERS ASIDE THOSE MENTIONED ABOVE WILL FEEL SLOW DUE TO NON-PORTED PHYSICS AND ALL

 

[Shell]

Ike, Marth, & Diddy have the slow / long landing lag bug and are unable to go off ledges at all (no coyote effect).

Fox / Ness / Yoshi / Lucas worked pretty well, although Ness's hop and jump heights are reversed. Also, Diddy's barrels quickly reach a terminal velocity faster than his own peake fall speed and he dies. Marth's grav settings are much too high, he's unable to SHDF or recover. Ike can't recover for ****, definitely going to need that float while charging side-b and/or do an up-b after. Otherwise his grav looks mad fun.

We played some fox dittos, but he feels strangely unwieldy. Everyone needs a lot more friction, especially fox. Wavedashes are too long and imprecise, leaving us whizzing past each other. There's also something else that's not as precise about the controls vs. Melee that I can't quite put my finger on. Something about initial accelerations and decelerations, I think.

Sorry I can't be more help.

edit: if we really can't use special lag 1 / 2 for the landing animation, we might have to replace the actual animations of the problem characters (DK, Mario, etc) with their land lag 1 / 2 animations.

 

[shanus]

I'm wondering if we should up the air control on all characters to match the faster pace of the gameplay?

We can either give them higher ability to weave in the air and/or faster ability to stop themselves in the air.


Also, boost the friction?

Sad panda @ that method not fixing marth, diddy, and ike

 

[shanus]

I'm thinking we should try and make the switch to Almas' new gravity code, and then disable these recoveries from being effected by gravity.

Edit: If the code works?

 

[shanus]

Not sure if this will be visible to people other than my account, but i ported additional pacs and made a few more changes. You will need to compile your own GCT though.

http://www.mediafire.com/?w231wjzj3wz

 

[leafgreen386]

Well, either raising stopping ability or increasing initial acceleration will increase weaving capacity. The notable difference is that increasing initial acceleration buffs a character's recovery. With higher gravs, this may not be such a bad thing if done in moderation, and it could really improve the feel of the game.

 

[shanus]

Here is a version with everyone's gravity and ALRs ported not including Squirtle, Charizard, and Ice Climbers.

Also, ROB, D3, and Luigi *might* not cause crashes in this one. ROB should be sporting his own special fall animation now

http://www.mediafire.com/?u3ewyyyczl4

Check the physics excel file for the precise changes.

 

[shanus]

New build up

All characters except for ICs have gravity ported. ROB now works perfectly.

Still havent figured out D3,Luigi, marth, ike or diddy though.

Rest of the changes = EPIC

http://shanemulliganphotography.blogsite.org:6111/~shane/Neargold/Melee2.0/11-20-09/

 

[JCaesar]

ROB works perfectly? mmmmfffffff!

 

[shanus]

yeah, just dont try and recover as sonic, link, or toon link yet lmao

 

[Shell]

I dunno if you saw my message, but you might try CAMing "if previous AD, current landlagspecial, replace w/ Landlag 2" and see how that works. Might get some coyote effects, but we'll see. I'm at the library, but I might be able to test some stuff later.

 

[shanus]

I dunno if you saw my message, but you might try CAMing "if previous AD, current landlagspecial, replace w/ Landlag 2" and see how that works. Might get some coyote effects, but we'll see. I'm at the library, but I might be able to test some stuff later.

That doesnt work. There is no actionID for LandingHeavy.

As i said before, its landing 1 or landing 2 dependent on your velocity all encapsulated within the same actionID.

 

[shanus]

New updates:

http://shanemulliganphotography.blogsite.org:6111/~shane/Neargold/Melee2.0/11-21-09/

Sonic / Link / others recoveries should work better now with the higher gravity.

Crude fix for Diddy/marth/ike's wavedash lag.


This leaves only D3 and Luigi as problems!


testers, go gogogo!

 

[Shell]

That doesnt work. There is no actionID for LandingHeavy.

As i said before, its landing 1 or landing 2 dependent on your velocity all encapsulated within the same actionID.

I misunderstood this point before, merely thinking that because there were two separate IDs in PSA they must be distinct, but lacked a way to change to them for some strange reason. Don't mind me.

 

[shanus]

luigi now works.

Only remaining problem is D3.

Need to raise SH height of sonic.

Add additional height to sonic/link upB

 

[Veril]

Ganon/Samus/Lucas have boner inducing wavedashes. I'm having so much fun with them! The melee stage select screen is all sorts of messed up. Is that just me?

Lucas's zapjump is way weaker, that makes me sad.

Some wavedashes seem excessively long. Bowser's wavedash landings make me laugh ;p

 

[CT Chia]

any vids by any chance?

 

[shanus]

Ganon/Samus/Lucas have boner inducing wavedashes. I'm having so much fun with them! The melee stage select screen is all sorts of messed up. Is that just me?

Lucas's zapjump is way weaker, that makes me sad.

Some wavedashes seem excessively long. Bowser's wavedash landings make me laugh ;p

i was messing with phantomwings stage code thats why

im working on a way to improve it though.

The reason many characters WD's feel too long is it seems to count all of the WD's as an equialvent to a perfect wavedash.

I have a solution in mind which shouldnt be too difficult to program in where it decelerates a bit at later frame counts to make it require higher precision.


BTW luigi's WD is HILARIOUS


@Chibo, no vids yet. I'm hoping to make a trailer at some point this week to unveil the hype!