Schrijver
| PCM player using SCC
| NYYRIKKI
msx master
Berichten: 1525 |  Geplaatst: 13 Oktober 2007, 18:45 | I think it's time to go to find that SCC cartridge...
| | ARTRAG
msx master
Berichten: 1737 | Geplaatst: 15 Oktober 2007, 16:35 | Before going to line interrupts, can I suggest an experiment ?
Assume one channel for PCM with period Ps and one channel in deformed mode as timer with period Pd;
The Sampling frequency is Fs = 3.5MHz/(Ps+1)/32
The Rotation frequency is Fd = 3.5MHz/(Pd+1)
If relation holds
32*Pd+31 = Ps
We get that each step done by the PCM channel we get an equivalent step in the counter channel
Initialize channel used as counter with numbers from 0 to 31 (put it a volume 0)
Now you can do this loop:
0) do you staff here in less than 32*125us = 4000us (if at Fs=8KHz)
1) test byte #1 of the channel used for counter
2) if it is non zero goto 1
3) LDIR the next 32 samples in the wavetable
4) goto 0
I think that this is the simplest experiment to do to try to isolate
the SCC behaviours we pointed in this tread
after getting this work, I would try to use the counter as pointer
to the PCM segment already reproduced, in order avoid to stop and
wait the end of the current chunk.
BTW 4000us is quite a nice amount of time for the branch of a the
main of a game...
PS
Put Pd = 13 and Ps = 447
this should give about 8KHz of Fs and synchronization between
PCM channel and time channel
| | mohai
msx lover
Berichten: 127 | Geplaatst: 15 Oktober 2007, 23:40 | By the way, do you remember voices played in Pennant Race 2?
A good starting point...  | | ARTRAG
msx master
Berichten: 1737 | Geplaatst: 16 Oktober 2007, 00:13 | I'm very rusty with SCC and mem mappers, why this code (for sjasm) does not work ?
I tried on bluemsx, that detects correctly the output as an scc rom, but do not give a sound
;----------------------------------------------------------------------------
;----------------------------------------------------------------------------
OUTPUT sccplay.rom
org 4000h
dw "BA",START,0,0,0,0,0,0
Bank1 equ 05000h
Bank2 equ 07000h
Bank3 equ 09000h
Bank4 equ 0B000h
;-------------------------------------
; Entry point
;-------------------------------------
START:
ld a,3Fh
LD (Bank3),A
LD A,00011111b ;Enables all SCC channels
LD ($988F),A
Period equ 9
ld hl,Period ; F sample ch1 = 11186Hz
ld (9880h),hl
ld hl,32*Period +31
ld (9886h),hl ; F sample ch4 = F sample ch1/32
ld a,15
ld (988Ah),a ; volume ch1
xor a
ld (988Bh),a ; volume ch2
ld (988Ch),a ; volume ch3
ld (988Dh),a ; volume ch4
ld (988Eh),a ; volume ch5
ld de,$9800
ld hl,sccwaves
ld bc,$20*4
ldir
LD A,11100000b ;Deformation only on 4&5 SCC channels (or not ?)
ld (98E0h),a
.REWIND:
xor a
.LOOP:
inc a
ld (Bank2),a
ld hl, SAMPLE_START
ex af,af'
ld b,0
.page:
push bc
;do your stuff here in less than 32 / F_sample ch1 seconds
.wait:
ld a,(9860h)
or a
jr nz,.wait ; wait for a complete rotation in ch 4
ld de,$9800
ld bc,$20
ldir
pop bc
djnz .page
ex af,af'
cp (SAMPLE_END - SAMPLE_START + #1FFF)/#2000
jr nz,.LOOP
jr .REWIND
ret
sccwaves:
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
;-------------------------------------
; Padding for rom player
;-------------------------------------
DS (#6000 - $)
;-------------------------------------
; Sample data
;-------------------------------------
SAMPLE_START:
incbin "handel3.wav"
SAMPLE_END:
;-------------------------------------
; Padding, align rom image to a power of two.
;-------------------------------------
SAMPLE_LENGTH equ SAMPLE_END - SAMPLE_START
IF (SAMPLE_LENGTH <= #6000)
DS (#6000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #E000)
DS (#E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #1E000)
DS (#1E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #3E000)
DS (#3E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #7E000)
DS (#7E000 - SAMPLE_LENGTH)
ELSE
DS (#FE000 - SAMPLE_LENGTH)
ENDIF
ENDIF
ENDIF
ENDIF
ENDIF
FINISH:
| | NYYRIKKI
msx master
Berichten: 1525 | Geplaatst: 16 Oktober 2007, 06:54 | As your START address points to #4010 I think that you need to select correct slot first to area #8000-#BFFF
(IIRC there is copy/paste ready code in MSX2 technical handbook)
| | ARTRAG
msx master
Berichten: 1737 | Geplaatst: 16 Oktober 2007, 10:01 | Done !
Does the power-up code solve the issue ?
;----------------------------------------------------------------------------
;----------------------------------------------------------------------------
OUTPUT sccplay.rom
org 4000h
dw "BA",START,0,0,0,0,0,0
Bank1 equ 05000h
Bank2 equ 07000h
Bank3 equ 09000h
Bank4 equ 0B000h
;-------------------------------------
; Entry point
;-------------------------------------
START:
call powerup
ld a,3Fh
LD (Bank3),A
LD A,00011111b ;Enables all SCC channels
LD ($988F),A
Period equ 9
ld hl,Period ; F sample ch1 = 11186Hz
ld (9880h),hl
ld hl,32*Period +31
ld (9886h),hl ; F sample ch4 = F sample ch1/32
ld a,15
ld (988Ah),a ; volume ch1
xor a
ld (988Bh),a ; volume ch2
ld (988Ch),a ; volume ch3
ld (988Dh),a ; volume ch4
ld (988Eh),a ; volume ch5
ld de,$9800
ld hl,sccwaves
ld bc,$20*4
ldir
LD A,11100000b ;Deformation only on 4&5 SCC channels (or not ?)
ld (98E0h),a
.REWIND:
xor a
.LOOP:
inc a
ld (Bank2),a
ld hl, SAMPLE_START
ex af,af'
ld b,0
.page:
push bc
;do your stuff here in less than 32 / F_sample ch1 seconds
.wait:
ld a,(9860h)
or a
jr nz,.wait ; wait for a complete rotation in ch 4
ld de,$9800
ld bc,$20
ldir
pop bc
djnz .page
ex af,af'
cp (SAMPLE_END - SAMPLE_START + #1FFF)/#2000
jr nz,.LOOP
jr .REWIND
ret
sccwaves:
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
;-------------------------------------
; Powerup routine for small model, non-Z180 code.
; set pages and subslot
;-------------------------------------
powerup:
call 0x138
rrca
rrca
and 0x03
ld c,a
ld b,0
ld hl,0xfcc1
add hl,bc
or (hl)
ld b,a
inc hl
inc hl
inc hl
inc hl
ld a,(hl)
and 0x0c
or b
ld h,0x80
call 0x24
ret
;-------------------------------------
; Padding for rom player
;-------------------------------------
DS (#6000 - $)
;-------------------------------------
; Sample data
;-------------------------------------
SAMPLE_START:
incbin "handel3.wav"
SAMPLE_END:
;-------------------------------------
; Padding, align rom image to a power of two.
;-------------------------------------
SAMPLE_LENGTH equ SAMPLE_END - SAMPLE_START
IF (SAMPLE_LENGTH <= #6000)
DS (#6000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #E000)
DS (#E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #1E000)
DS (#1E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #3E000)
DS (#3E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #7E000)
DS (#7E000 - SAMPLE_LENGTH)
ELSE
DS (#FE000 - SAMPLE_LENGTH)
ENDIF
ENDIF
ENDIF
ENDIF
ENDIF
FINISH:
| | mohai
msx lover
Berichten: 127 | Geplaatst: 16 Oktober 2007, 13:35 | This line should say "AB" instead of "BA". Doesn't it?
dw "BA",START,0,0,0,0,0,0
| | NYYRIKKI
msx master
Berichten: 1525 | Geplaatst: 16 Oktober 2007, 13:37 | @ARTRAG: I think it should...
@mohai: No
| | ARTRAG
msx master
Berichten: 1737 | Geplaatst: 16 Oktober 2007, 13:45 | In bluemsx, I get no sound as well.
I think that the issue is due to the lock on the waveforms
(provided that the real HW does this even if bit 7 in def. register
is set).
I tried to reset the deformation reg before dumping the 32 samples
and now I can head some noise now...
 | | ARTRAG
msx master
Berichten: 1737 | Geplaatst: 16 Oktober 2007, 19:06 | I tried to use ch1 as counter and ch4 as player...
noise again.
Nevertheless openmsx and bluemsx sound VERY different.
I think that the deformation register is implemented in
different ways.
(BIFI at http://bifi.msxnet.org/msxnet/tech/scc.html reports
that bit 6 and 7 are used to produce noise. This is
not correct, and maybe one of the two emulators
produces noise instead of the rotating waveforms....)
;----------------------------------------------------------------------------
;----------------------------------------------------------------------------
OUTPUT sccplay.rom
org 4000h
dw "BA",START,0,0,0,0,0,0
Bank1 equ 05000h
Bank2 equ 07000h
Bank3 equ 09000h
Bank4 equ 0B000h
;-------------------------------------
; Entry point
;-------------------------------------
START:
call powerup
ld a,3Fh
LD (Bank3),A
LD A,00001000b ;Enables all SCC channels
LD ($988F),A
Period equ 9
ld hl,Period ; F sample ch4 = 11186Hz
ld (9886h),hl
ld (9888h),hl
ld hl,32*Period +31
ld (9880h),hl ; F sample ch1 = F sample ch1/32
ld (9882h),hl ; F sample ch2 = F sample ch1/32
ld (9884h),hl ; F sample ch2 = F sample ch1/32
xor a
ld (988Ah),a ; volume ch1
ld (988Bh),a ; volume ch2
ld (988Ch),a ; volume ch3
ld a,7
ld (988Dh),a ; volume ch4
xor a
ld (988Eh),a ; volume ch5
ld de,$9800
ld hl,sccwaves
ld bc,$20*4
ldir
LD A,11000000b ; WAVEFORM only on 4&5 SCC channels (or not ??)
ld (98E0h),a
.REWIND:
xor a
.LOOP:
inc a
ld (Bank2),a
ld hl, SAMPLE_START
ex af,af'
ld b,0
.page:
push bc
;;;;;;;;;;;
;do your stuff here in less than 32 / F_sample ch1 seconds
ld b,50
.inner
ld a,r
out(99h),a
ld a,128+7
out(99h),a
djnz .inner
xor a
out(99h),a
ld a,128+7
out(99h),a
;;;;;;;;;;;
.wait:
ld a,(9800h)
or a
jr nz,.wait ; wait for a complete rotation in ch 1
ld (98E0h),a ; No deformation (unlock waveforms)
ld de,9860h
ld bc,$20
ldir
LD A,11000000b ; WAVEFORM only on 4&5 SCC channels (or not ??)
ld (98E0h),a
pop bc
djnz .page
ex af,af'
cp (SAMPLE_END - SAMPLE_START + #1FFF)/#2000
jr nz,.LOOP
jr .REWIND
ret
sccwaves:
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
db 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,25,27,28,29,30,31
;-------------------------------------
; Powerup routine for small model, non-Z180 code.
; set pages and subslot
;-------------------------------------
powerup:
call 0x138
rrca
rrca
and 0x03
ld c,a
ld b,0
ld hl,0xfcc1
add hl,bc
or (hl)
ld b,a
inc hl
inc hl
inc hl
inc hl
ld a,(hl)
and 0x0c
or b
ld h,0x80
call 0x24
ret
;-------------------------------------
; Padding for rom player
;-------------------------------------
DS (#6000 - $)
;-------------------------------------
; Sample data
;-------------------------------------
SAMPLE_START:
incbin "handel3.wav"
SAMPLE_END:
;-------------------------------------
; Padding, align rom image to a power of two.
;-------------------------------------
SAMPLE_LENGTH equ SAMPLE_END - SAMPLE_START
IF (SAMPLE_LENGTH <= #6000)
DS (#6000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #E000)
DS (#E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #1E000)
DS (#1E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #3E000)
DS (#3E000 - SAMPLE_LENGTH)
ELSE
IF (SAMPLE_LENGTH <= #7E000)
DS (#7E000 - SAMPLE_LENGTH)
ELSE
DS (#FE000 - SAMPLE_LENGTH)
ENDIF
ENDIF
ENDIF
ENDIF
ENDIF
FINISH:
| | manuel
msx guru
Berichten: 3531 | Geplaatst: 16 Oktober 2007, 22:18 | ARTRAG, you could also just look at the source code of those emulators and find out what exactly the code is doing  | | ARTRAG
msx master
Berichten: 1737 | Geplaatst: 17 Oktober 2007, 08:43 | !!!
in the emulators it seems that even setting bit 7 in 98E0h, ALL the channels continue their rotation
This makes all the channels not useful for playing PCM samples.
If this detail reflects the real HW and correctly implemented by emulators,
SCC cannot be used in this way for PCM audio.
BTW if emulation is incorrect I cannot go further, as I do not have any real HW.
I give up, anyone with real HW willing to try? I can send him my files
The ideal HW for test is a flash cartridge with SCC and konami mem mapper.
Can anyone tell me if such an HW exist and is sold somewhere ?
| | Huey
msx professional
Berichten: 630 | Geplaatst: 17 Oktober 2007, 08:56 | Google is your friend
linky | | nikodr
msx addict
Berichten: 491 | Geplaatst: 17 Oktober 2007, 11:12 | Artrag i have a flash cartridge with scc and konami memory mapper (manuel Pazos's Megaflash),if you want any help my email is mperdemas at yahoo dot gr
| | manuel
msx guru
Berichten: 3531 | Geplaatst: 17 Oktober 2007, 12:19 | | |
| |
| |
