Skip to content

Jesuszilla/TurtleWax

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TurtleWax for IKEMEN by Kamekaze and Jesuszilla

This system aims to provide a complete command parser and buffering system for the IKEMEN-Go engine. A minimal installation requires turtlewax.zss to be placed in the CommonStates section of config.json, with each character specifying their own buffering.zss file to determine state setting logic.

Available Maps

The following maps are made available to the user. All times are in frames:

Map Name Description Default
Map(tw_ver) TurtleWax version (current version)
Map(tw_debug) Set to nonzero to enable debug print statements. 0
Map(tw_MashPCt) Total buffer time for HHS presses. 10
Map(tw_MashCt) Number of button presses required for a "mashing" command to activate 5
Map(tw_ChargePartition) Set to 1 to enable charge partitioning. 0 (off)
Map(tw_ChargePartitionTime) Time a charge partition should last. 15
Map(tw_D_ChargeTime) Total charge time to hold for a Down charge. 48
Map(tw_D_BufferTime) Total buffer time for Down charge once it's active. 9
Map(tw_U_ChargeTime) Total charge time to hold for an Up charge. 48
Map(tw_U_BufferTime) Total buffer time for Up charge once it's active. 9
Map(tw_B_ChargeTime) Total charge time to hold for a Back charge. 48
Map(tw_B_BufferTime) Total buffer time for Back charge once it's active. 9
Map(tw_F_ChargeTime) Total charge time to hold for a Forward charge. 48
Map(tw_F_BufferTime) Total buffer time for Forward charge once it's active. 9
Map(tw_Dash_input1_BufferTime) Total buffer time for the first input in any dash (FF,BB) command. 9
Map(tw_Dash_input2_BufferTime) Total buffer time for the second input in any dash (FF,BB) command. 6
Map(tw_Input_Delay_frames) Total number of frames to additionally delay all inputs. 0 (no delay)
Map(tw_Buffer_Snk) Set to 1 to treat the character's buffer as an SNK (KOF) style buffer. 0 (off)
Map(tw_Buffer_ChangeStateNo) Set in your buffering.zss file to set move state. Keep at -1 for no change. (dynamic)
Map(tw_MoveStrength) Set in your buffering.zss file to set move strength. (dynamic)
Map(tw_Btn6) Set to 1 to signify that this is a 6-button character. 1 (is 6-button)
Map(tw_Assert_NoBufferTimeDec) Set to nonzero to assert no buffer decrementing. The buffer can still move along, but this can allow for things such as "infinite" buffer time such as SFA2/SFZ2 CC/VC. 0 (only set as-needed)
Map(tw_B_ChargeReady) Indicates that a back charge is ready. 0 (set by system)
Map(tw_D_ChargeReady) Indicates that a down charge is ready. 0 (set by system)
Map(tw_F_ChargeReady) Indicates that a forward charge is ready. 0 (set by system)
Map(tw_U_ChargeReady) Indicates that an up charge is ready. 0 (set by system)

Usage (2.0)

To properly make use of the command stateNo setting system, you must handle the ChangeState logic by setting and reading the Map(tw_Buffer_ChangeStateNo) variable. Once you change to the state that the buffering system sets, it will automatically reset it in +1 if you set it up like the examples have it. An example CNS implementation may look like the example below in your StateDef -1 .cmd (legacy) or .cds (recommended extension) file.

You must also add turtlewax.zss and optionally (if you want the default command maps) turtlewax_common.zss to CommonStates in config.json. Change your CommonCmd to also use common.cds instead of common.cmd.

You will need to create a buffering.zss file for your character(s). You may reference TW Cvs Raiden for an example implementation. When building your buffering.zss, most likely, you will have some commands that are alike such as QCF, HCF, and QCFx2 for instance (they all contain D,DF,F or 236 as a substring). Because doing Lua operations every frame is expensive and there are currently no arrays in ZSS, you, the character author, are now responsible for resetting these. Again, refer to Raiden's buffering.zss for more information on resetting alike commands. You should ideally only reset alike commands with 3 or more elements. A lot of Capcom and to an extent SNK games do this. Observe your game for proper behavior. CvS2 for sure does this, though.

You must add the following to the bottom of your common.const file:

:TW
ih = 9992999

You must also add the following maps to your character's .DEF in order to opt in to the TW input helper:

; Required by TW
[map]
u_tw=1
ih = 9992999

TurtleWax 2.0 uses Capcom vs. SNK 2 3-star defaults for the Capcom commands.

TurtleWax 2.0 is officially Raspberry Pi 4 performant!

;---------------------------------------------------------------------------
; Supers
[State -1, Supers]
type = ChangeState
value = Map(tw_Buffer_ChangeStateNo)
triggerall = Map(tw_Buffer_ChangeStateNo) = [3000,3999]
triggerall = Power >= 1000
triggerall = statetype != A
triggerall = roundstate = 2
trigger1  = ctrl || (StateNo = 100 && AnimElemTime(2) >1) || StateNo = 101 || StateNo = 40 || (StateNo = 52 && Anim = 47 && Time >= 2)
trigger2  = StateNo = 200
trigger3  = StateNo = 210 && AnimElemTime(6) < 0
trigger4  = StateNo = 215 && AnimElemTime(7) < 0
trigger4  = StateNo = 220 && Time < 4
trigger5  = StateNo = 225 && AnimElemTime(6) < 0
trigger6  = StateNo = 230 && AnimElemTime(5) < 0
trigger7  = StateNo = 240 && Time < 4
trigger8  = StateNo = 250 && Time < 4
trigger9  = StateNo = 255 && Time < 4
trigger10 = StateNo = 400
trigger11 = StateNo = 410 && AnimElemTime(6) < 0
trigger12 = StateNo = 430 && AnimElemTime(5) < 0
trigger13 = StateNo = 440 && Time < 4
trigger14 = StateNo = 450 && AnimElemTime(5) < 0
trigger15 = StateNo = 1260 && MoveContact = 1
ignorehitpause = 0

;---------------------------------------------------------------------------
; Specials
[State -1, Specials]
type = ChangeState
value = Map(tw_Buffer_ChangeStateNo)
triggerall = !AILevel
triggerall = !IsHelper
triggerall = roundstate = 2
triggerall = Map(tw_Buffer_ChangeStateNo) = 195 || (Map(tw_Buffer_ChangeStateNo) = [1000,2999])
triggerall = statetype != A
trigger1  = ctrl || (StateNo = 100 && AnimElemTime(2) >1) || StateNo = 101 || StateNo = 40 || (StateNo = 52 && Anim = 47 && Time >= 2)
trigger2  = StateNo = 200
trigger3  = StateNo = 210 && AnimElemTime(6) < 0
trigger4  = StateNo = 215 && Time < 4
trigger4  = StateNo = 220 && Time < 4
trigger5  = StateNo = 225 && AnimElemTime(6) < 0
trigger6  = StateNo = 230 && AnimElemTime(5) < 0
trigger7  = StateNo = 240 && Time < 4
trigger8  = StateNo = 250 && Time < 4
trigger9  = StateNo = 255 && Time < 4
trigger10 = StateNo = 400
trigger11 = StateNo = 410 && AnimElemTime(6) < 0
trigger12 = StateNo = 430 && AnimElemTime(5) < 0
trigger13 = StateNo = 440 && Time < 4
trigger14 = StateNo = 450 && Time < 4

;---------------------------------------------------------------------------
; Throw 1
[State -1, Throw 1]
type = ChangeState
value = 800
triggerall = !AILevel
triggerall = (Map(h_Fd) || Map(h_Bk))
triggerall = Map(xa)
triggerall = ctrl
triggerall = stateno != 100
trigger1 = statetype = S
ignorehitpause = 0

;---------------------------------------------------------------------------
; State Correction
; This is to prevent bad things from happening when the state is set to an
; air state immediately upon landing. You can either force it to a ground
; basic or just null it out completely by setting it to -1.
[State -1, MapSet]
type = MapSet
trigger1 = StateType != A && (Map(tw_Buffer_ChangeStateNo) = [600,699])
map = "tw_Buffer_ChangeStateNo"
value = Map(tw_Buffer_ChangeStateNo)-cond(Map(h_Dn),200,400)    ; or -1, depending on your use case
ignorehitpause = 1

;---------------------------------------------------------------------------
; State Correction
; This is to prevent basics from coming out after throws.
[State -1, MapSet]
type = MapSet
trigger1 = stateNo = [800,899]
trigger1 = Time < 3 && (map(tw_Buffer_ChangeStateNo) = [200,799])
trigger2 = stateno = 195
trigger2 = !ctrl && map(tw_Buffer_ChangeStateNo) = 195
map = "tw_Buffer_ChangeStateNo"
value = -1
ignorehitpause = 1

;---------------------------------------------------------------------------
; Basics
[State -1, Basics]
type = ChangeState
value = Map(tw_Buffer_ChangeStateNo)
triggerall = !AILevel
triggerall = !IsHelper
triggerall = roundstate = 2
triggerall = StateNo != [800,899]
triggerall = MoveType != H
trigger1 = Map(tw_Buffer_ChangeStateNo) = [200,299]
trigger2 = Map(tw_Buffer_ChangeStateNo) = [400,499]
trigger3 = Map(tw_Buffer_ChangeStateNo) = [600,699]
ignorehitpause = 0

Debugging

You can reload your characters' buffering.zss files the same as you always would (Shift+F4). This can be really helpful for debugging.

History

Version 2.1

  • Updated to syntax planned for I.K.E.M.E.N-Go 1.0.

Version 2.0

  • Changed to ZSS implementation.
  • Motion map names have been moved to turtlewax_common.zss and now prefixed with tw_. Base directions and buttons are unchanged.
  • Now Raspberry Pi 4 compliant.

Version 1.6

  • Fixed default charge commands.
  • Code optimization

Version 1.5

  • Added input_types array to all move definitions so that the input type can be defined for each motion.
  • tt can now be specified as a single value, or a table of values. Each value in the table must either be an integer value or a function that takes an integer, p (player index) as its sole argument and returns an integer value.
  • tt can now hold up to 31 frames of buffering time.
  • Refactored arrays so that all button inputs are in one array.
  • Added input delay.
  • Added standard Up charging implementation and maps to control the buffer and charge times for it.
  • Added maps to control the buffering times for dash commands.
  • Added StateNo setting system defined by a new file, buffering.lua, required as part of the implementation. Every character should have a buffering.lua file with the command-state mappings defined. Please read buffering.lua.example for more info.
  • Added move strength setting and map for retrieval, Map(tw_MoveStrength). This is set by the "strength" parameter in the move table definition in buffering.lua.

Version 1.0

Initial release

Special Thanks

  • Kamekaze - Ported a lot of the original Deep Buffering code to Lua and re-porting version 1.6 fixes to ZSS, testing, implementation.
  • Jesuszilla - Original author of Deep Buffering, testing, advice & implementation.
  • TTTTTsd - Testing
  • Vans - Original Tiny Buffering system which formed the basis for all good buffering systems in M.U.G.E.N and its derivatives.
  • extravagant - Testing
  • Phantom.of.the.Server - Testing