Dark Man Fortress Stage 2
The value of "$3f2" is "0xF9" in this place with the napalm bomb.
And, this value is maintained to the next stage.
Dark Man Fortress Stage 3
And, when you get on the second SuperArrow, the value of "$3f2(0xF9)-0x01=0xF8" is written in "$3f0"(the gravitational velocity!).
As a result, MegaMan falls faster than usually.
In a word, this case is necessary to change "$3f2".
The memory is different because of the SuperArrow used.
$3f3:first SuperArrow
$3f2:second SuperArrow
$3f1:third SuperArrow
The memory will have to be adjusted for the SuperArrow that you use. :)
Enemy x 0
NMI = 1123
MAIN = 3765
TOTAL = 4888
Enemy x 2
NMI = 1123
MAIN = 6078 (+2313) Enemy =+1000
TOTAL = 7201
Enemy x 2 + item1 x 1
NMI = 1123
MAIN = 6588 (+510) Item1=+500
TOTAL = 7711
Enemy x 2 + item1 x 2
NMI = 1123
MAIN = 7076 (+998) Item1=+480
TOTAL = 8199
Enemy x 2 + item1 x 2 + JustScroll
NMI = 1123
MAIN = 7313 (+1235) JustScroll=+300
TOTAL = 8436
Enemy x 2
NMI = 1123
MAIN = 6078
TOTAL = 7201
Enemy x 2 + AirShooter + UsuallyNMIFrame
NMI = 1123
MAIN = 6850 (+772) AirShooter=+750
TOTAL = 7973
Enemy x 2 + AirShooter + MiddleNMIFrame(lot)
NMI = 1340 (+217)
MAIN = 6838
TOTAL = 8178
Enemy x 2 + AirShooter + LargeNMIFrame(less)
NMI = 1593 (+470)
MAIN = 6858
TOTAL = 8451
Enemy x 2 + AirShooter + LargestNMIFrame(very rare)
NMI = 2044 (+921)
MAIN = 6848
TOTAL = 8892
DelayDownScroll demands
TOTAL = 8950-8970
OurTAS
28196frame
NMI =1664(NMI1095 + Sounds)
MAIN =7302(Enemy + Item1 x 2)
TOTAL=8966
28172f NMI = 1723->This frame seems to be good.
28196f NMI = 1095->Our DelayDownScroll
28208f NMI = 1955->This frame seems to be very good!
28556f NMI = 1723
28592f NMI = 2028->very big but too slow.
1:Item 1 is saved.
2:The Arm switch time(Item1->AirShooter) is 40 frames.
3:28208-28196f=12frame
4:Then, we should be about 28 frames faster because of the saving of item1.
1:You use IPSWIN, and apply the IPS patch to the NES file.
2:You copy the file onto PowerPak.(or same tools)
3:You execute NES and you advance in the BombMan stage.
4:You advance to the place of which the small explosion goes out.
5:You face left and walk and push the select button twice.
6:Then, the object appearance occurs because of a wrong bank.
7:You try many times.
The kind of the appearing object is important to prove the writing timing of OldPPU to be correct.
Expectation of results:
*A:FCEUX(OldPPU),Nintendulator,Nestopia
graphic bug and reset, enemy is appear , water current and stage clearing etc..
*B:FCEUX(NewPPU)
The kind of the appearing object or enemy is the same.
*C:Nintendulator(DebugMode)
The enemy who drops the item appears.
*D:Other result of expectation
Anything doesn't appear.
PRE:number of instructions of "NMIEnds->ChangeBank"
NEXT:number of instructions of "ChangeBank-> NMIStarts"
We should adjust this value to "1"
NOWBANK:Bank number of the frame.
"5" is a frame for the investigation.
"6" is a frame for normal game.
The NOWBANK changes like 5.5.6.5.5.6.5.5.6....
Bank6 frame is only recorded in movie.
[6F0]..:ObjectNumbers. It is high of the possibility appearing in the same action and the frame.
2F,AF:object of DelayWaterCurrent
5D,DD:objcet of DelayWarpDeatch
75,F5:object of DelayStageClear
The DelayStageClear glitch will succeed if we can adjust "NEXT" to "1" by 4071 frame.
If "NEXT" becomes "1" by the same frame(4071frame and NOWBANK=6)... (Though The load instructions is insufficient in this example.)
In a word, this.
1:FCEUX(NewPPU)
A decided enemy appears.
2:FCEUX(OldPPU) and Nestopia and Nintendulator(not use debugger)
Various objects appear.
3Emulators,Almost the same result.
Perhaps, BombStageClear seems to become possible as for a real machine :P
Default Rockman(J):
A9 06 8D 06 C0 85 12 A5 8F F0 03 20 02 DB
lda #$06
sta $C006
sta CurrentBank
lda ZigZagFireStatus
beq +
jsr UpdateZigZagFire
Patch HackROM:
A5 16 A0 06 29 10 F0 02 A0 05 8C 06 C0 EA
lda $16
ldy #$06
and #$04
beq +
ldy #$05
+
sty $C006
nop
D943:B1 06 LDA ($06),Y @ $200F = #$07
D947:C8 INY
D948:B1 06 LDA ($06),Y @ $2010 = #$07
D94C:C8 INY
D94D:B1 06 LDA ($06),Y @ $2011 = #$07
D94F:C8 INY
Memory of HexEditor
D943:
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
2000 91 1E 07 00 F8 07 62 07 91 1E 07 00 F8 07 62 07
2010 91 1E 07 00 F8 07 62 07 91 1E 07 00 F8 07 62 07
2020 91 1E 07 00 F8 07 62 07 91 1E 07 00 F8 07 62 07
D948,D94D:
other value
D943:B1 06 LDA ($06),Y @ $200F = #$06
D947:C8 INY
D948:B1 06 LDA ($06),Y @ $2010 = #$10
D94C:C8 INY
D94D:B1 06 LDA ($06),Y @ $2011 = #$10
D94F:C8 INY
Memory of HexEditor
D943:
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
2000 90 1E 10 00 F8 02 6A 06 90 1E 10 00 F8 02 6A 06
2010 90 1E 10 00 F8 02 6A 06 90 1E 10 00 F8 02 6A 06
2020 90 1E 10 00 F8 02 6A 06 90 1E 10 00 F8 02 6A 06
D948:
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
2000 90 1E 10 00 F8 02 4A 04 90 1E 10 00 F8 02 4A 04
2010 90 1E 10 00 F8 02 4A 04 90 1E 10 00 F8 02 4A 04
2020 90 1E 10 00 F8 02 4A 04 90 1E 10 00 F8 02 4A 04
D94D:
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
2000 90 1E 10 00 F8 02 42 04 90 1E 10 00 F8 02 42 04
2010 90 1E 10 00 F8 02 42 04 90 1E 10 00 F8 02 42 04
2020 90 1E 10 00 F8 02 42 04 90 1E 10 00 F8 02 42 04
D943:B1 06 LDA ($06),Y @ $200F = #$FF
D947:C8 INY
D948:B1 06 LDA ($06),Y @ $2010 = #$FF
D94C:C8 INY
D94D:B1 06 LDA ($06),Y @ $2011 = #$FF
D94F:C8 INY
Memory of disasemmbly
Always:
00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
2000 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
2010 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
2020 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF
I had the BANK number change ROM and made a mistake. .
As a result, Nestopia shows the same result as OldPPU of FCEUX.
The enemy who appeared well with NewPPU did not appear either.
Next, I will use Nintendulator to test. :)
http://www.yuko2ch.net/rockman/rockman-13-clear.avi
The probability might be the highest low.
Perhaps, it is not possible to use it in all stages.
It should be a stage to which the enemy changes since the bank is changed.
It might be possible in BombmanStage, Wily1, and Wily2.