http://www.yuko2ch.net/rockman/rockman-11.avi
I found that I can use "DelayTransform" in this place.
It is certain that I can save time in this place.
Perhaps,the probability might be the lowest in a current delay technique. :(
LoadEnemies
1:mistake in the BankNumber by NMI.
2:The object of #$4C is loaded.
DoEnemyAI
3:jsr (#$D006)
The address that doesn't jump is called usually , big mistake occurs.
Because the value of the object becomes #$00-#$3B usually , the processing of AAD1-BD98 is called usually.
The address not intended as AI of # $4C is called this time.
4:MegaMan comes to be buried on the floor and through the wall :P
DoEnemyAI
$AA29> BD E006: LDA ObjectType,X
A=#$4C
$AA2D> A8: TAY
Y=#$$4C
$AA2E> B9 3BAA: LDA EnemyAIaddr+0,Y
A=AAD1+4C=#$06
$AA31> 85 04: STA $04
$04=#$06
$AA33> B9 3CAA: LDA EnemyAIaddr+1,Y
A=AAD2+4C=#$D0
$AA36> 85 05: STA $05
$05=#$D0
$AA38> 6C 0400: JMP ($0004)
JMP (D006)
X:X(XSpeed)
Y:Y(YSpeed)
PRE:Number of instructions of ChangeBank6 from NMI End(frame)
NEXT:Number of instructions of NMI Start from ChangeBank6(frame)
ENMYBNK:When the glitch succeeds, the data of this Bank number is read.
CRNTBNK:It is a value of the Bank when the frame begins.
D582:NMI End
|(PRE instructions)
D89B:ChangeBank6
|(NEXT instructions)
D4A8:NMI Start
Already know,The water current occurs in the IceMan stage.
There is a part where the number of enemies increases.
There is a place where Elec Man shows up.
software reset
The revival point has changed when the screen shifts due to the bug and RockMan dies.
Intense graphic bug
RockMan transform to other characters, and to be able to pass wall.
*Investigation in the future
In a word, it is necessary to look for the useful example by "rm1delaytest.lua" in all stages where the number of instructions(delay) is enough
The result might change even if face turns left or right.
This will become a very serious investigation. :(
It is likely to succeed also in the place where software resets is generated when timing is moved.
A variegated bug occurs because this bug reads the data of a wrong Bank.
If the example found by you and
the place where the number of instructions suffices,
the glitch will occur if we adjust instructions.
Perhaps, for TAS, transformation, water current, and the revival point change might be important.
Default：
$42=01,05,06
d4A8 NMI is caused
d56e A=$42=01,05,06
d571 STA $C001,5,6 ChangeBank1,5,6
d89d STA $C006 A454=29 ChangeBank6
d8A0 $42=06
d99A A454=29 read data of Bank6
DelayWaterCurrent：
$42=05
d89d STA $C006 A454=29 ChangeBank6
d4A8 NMI is caused
d56e A=$42=05
d571 STA $C005 A454=8D ChangeBank5
d8A0 $42=06
d99A A454=8D read data of Bank5
Default:
0001D4A8: //NMI
0001D89B: A9 06 lda #$06
0001D89D: 8D 06 C0 sta $C006 //ChangeBank06
0001D8A0: 85 42 sta $42 //$42=06
DelayWaterCurrent:
$42=05
0001D89B: A9 06 lda #$06
0001D89D: 8D 06 C0 sta $C006 //ChangeBank06
0001D4A8: //NMI
D56E: A5 42 lda $42 //$42=05
D571: 9D 00 C0 sta $C000,x //ChangeBank(#$C000+$42)=ChangeBank05
0001D8A0: 85 42 sta $42 //$42=06
----------------------
LoadEnemies
LoadEnemyNumber
0001D99A: BD 52 A4 lda EnemyDataPointers+0,x
X=#$02,$A454=#$8D,A=#$8D
Bank5($A454=#$8D) and 6($A454=#$29) changed by one frame in this part.
The value of $A454=#$29 is read usually.
But,the data of Bank5($A454=#$8D) is uncommonly read (by lag?).
0001D99D: 65 06 adc CurrentRoomPointer
A=A+$06=99 $06=$0C
0001D99F: 85 06 sta CurrentRoomPointer
$06=99 $07=85
Then,CurrentRoomPointer=($06)=#$8599!
-----------------------
LoadEnemies_Forward
0001D8F0: B1 06 lda (CurrentRoomPointer),y ;load enemy number
$06=#$8599 Y=#$03 $859C=#$2F
There is #$2F in $859C in the icemen stage.
0001D8F5: 20 AD D9 jsr SpawnObject
A=#$2F
-----------------------
SpawnObject
0001D9AD: 85 02 sta $02
$02=#$2F
0001D9E5: A5 02 lda $02 ;object type
0001D9E7: 9D E0 06 sta ObjectType,x
06F1=#$2F
The enemy of Object2F was generated.
-----------------------
DoEnemyAI
$AA29> BD E006: LDA ObjectType,X
AI of Object2F is executed.
-----------------------
AI_Object2F
$BA79> A9 30: LDA #$30
The water current is generated by AI of Object2F.
-----------------------
$8C->$4B->$0A->$C9(item1 up)->
$88->$47->$06->$C5(item1 up)->
$84->$43->$02->$C1(item1 up)->
$80->$3F->$FE(item1 up)
$BD->$7C->$3B->$FA(item1 up)
$B9->$78->$37->$F6(item1 up)
$B5->$74->$33->$F2(item1 up)
$04C4 is affected by A,Q,M,1
$04C3 is affected by A,B,Q,M,1
$04C2 is affected by A,B,Q,M,C,1,2,3
$FF～$81: Item1 rises after 4frame
$81～$41: Item1 rises after 3frame
$40～$00: Item1 rises after 2frame.
$FF～$81:rise:4-4-.....3-4-4..
$FF～$81:rise:3-4-.....3-4-4..
$40～$00:rise:2-4-.....3-4-4..
EECD:SEC //set carry flag
EECE:LDA $04C0,X @ $04C4 = $67 //$04C4: It is a related value in coordinates of item1 or other arms.
EED1:SBC $0660,X @ $0664 = $41 //When the boundary of $0 is exceeded, clear carry flag.
//$0664: It is fixed with $41.
EED4:STA $04C0,X @ $04C4 = $26
EED7:LDA $04A0,X @ $04A4 = $5F //$04A4: Y coordinates of item 1
EEDA:SBC $0640,X @ $0644 = $00 //When the carry flag is cleared, the value of A is subtracted.(-1)
EEDD:STA $04A4,X @ $04A4 = $5F
EEE0:CMP #$F0 //Judgment outside screen
EEE2:BCC $EEE7
EEE4:JMP $EF6A
Usually Glitch
1:NG NG
2:NG OK(use NMI)
Usually Glitch
1:Always NG
2:NG OK(use NMI)
Usually Glitch
1:NG NG
2:OK NG(use NMI)
Usually Glitch
1:NG OK(use NMI)
2:OK
Usually Glitch
1:OK NG(use NMI)
2:NG OK(use NMI)
1:$829E or $8287 calles $C7A1
2:$C7A5 calles SwitchBank
3:SwicthBank
4:NMI occuerd.(C004 ≤ PC ≤ C01B)
5:SoundCode returns $1C etc..
6:RegisterY(=$FF) is mad then (AND $82D0,Y) is not zero.
7:The scroll direction is filled.
8:"The next screen" become "previous screen".
QuickmanNoSpoon.fcm
Playing Airman before Heatman is obsolete even if the screen scroll in Airman could be performed with M-Buster, is this correct?
And finalfighter when you say "$15 was not changed" do you mean in that particular video; or that it is impossible to change $15?
0:$15 is set.
1:$8EE9 or $8FAB calles $C7A1
2:$C7A5 calles SwitchBank($C000)
3:SwicthBank
4:NMI occuerd.(C004 ≤ PC ≤ C01B)
5:SoundCode returns $0B
6:$14 is set.($14 is mad)
7:$14 and $15 is read
1:$8EE9 or $8FAB calles $C7A1
2:$C7A5 calles SwitchBank($C000)
3:SwicthBank
4:NMI occuerd.(C004 ≤ PC ≤ C01B)
5:SoundCode returns $0B
6:$14(=$F5) is mad.
1:$15 is cannot change.
2:The scroll judgment processing($C7A1) is necessary as well as the delay scroll for this tech.
3:In this case, a small value such as $0B is good though the sound code returns $1C at the delay scrolling.
4:Because the room number in horizontal scroll left end is changed,
there is a possibility that we can quickly go to the revival middle point.
5:The change in beginning of stage is useless.
6:There is no place in which this tech can save the time. ;)
1:$829E or $8287 calles $C7A1
2:$C7A5 calles SwitchBank
3:SwicthBank
4:NMI occuerd.(C004 ≤ PC ≤ C01B)
5:SoundCode returns $1C etc..
6:RegisterY(=$FF) is mad then (AND $82D0,Y) is not zero.
7:The scroll direction is filled.
As a result of delay downscrolling, Y becomes "#$FF".
$82A1:98 TYA A:00 X:00 Y:FF 1:Y is copied to A.
$82A2:A4 37 LDY $0037 = #$03 A:FF X:00 Y:FF 2:$0037 is loaded into Y. (in downscrolling , $0037=#$03)
$82A4:39 D0 82 AND $82D0,Y @ $82D3 = #$40 A:FF X:00 Y:03 3:A=($82D3=#$40)AND(A=#$FF)
$82A7:F0 12 BEQ $82BB A:40 X:00 Y:03 4:If A is 0, goto this branch(not scrolling)
$82A9:20 39 8F JSR $8F39 A:40 X:00 Y:03 5:If A is not 0,the scroll processing begins.If there are $40 bit mask ,
delay downscrolling succeeds.