How to detect the SCC?
Simple enough:
- select slot in page 2
- select rom page 63
- check if you can read/write in a wave ram location
- select rom page other than 63 (2 for instance)
- check if the wave ram location is read only
If all succesful, this a slot with an scc.
Repeat this test for all possible slots.
Great, many thanks.
I just finished the code for automatic SCC & SCC+ search for Project Melancholia mapper version. It will scan all slots in search of SCC support, hope it helps:
; --------------------------------------- ENASLT: EQU 0024h EXPTBL: EQU FCC1h SCANSLT: XOR A ; start from slot 0 LD HL,EXPTBL .loop: LD C,(HL) OR C .loop2: EXX LD (CHK_SCC.c_slt),A ; save current slot PUSH AF LD H,80h CALL ENASLT CALL CHK_SCC POP AF EXX ADD A,%00000100 ; increase subslot number LD C,A ; C = new slot + subslot BIT 7,A ; check if slot expanded JR Z,.next ; if not expanded go for next slot AND %00001100 ; check if already checked all subslots LD A,C JR NZ,.loop2 ; still some subslots left .next: INC HL LD A,C INC A AND %00000011 RET Z ; return if all slots & subslots are checked! ; JR .loop CHK_SCC: LD HL,A000h LD DE,BFFEh LD BC,2000h ; C = 00h -> SCC = YES if found XOR A LD (DE),A ; SET ROM + SCC if SCC-I cartridge LD A,(HL) INC A LD (HL),A CP (HL) RET Z ; found RAM -> return LD HL,97FFh LD (HL),L ; activate SCC write (write xx111111) INC HL ; HL = 9800h LD A,(HL) INC A LD (HL),A CP (HL) RET NZ DEC A LD (HL),A .f_SCC: XOR A DEC HL LD (HL),A ; disable write to SCC RAM LD A,B LD (DE),A ; SET ROM + SCC+ if SCC-I present LD HL,B7FFh LD (HL),L ; activate SCC+ write (write 1xxxxxxx) INC HL ; HL = B800h LD A,(HL) INC A LD (HL),A CP (HL) JR NZ,.setSCC DEC A LD (HL),A .f_SCCI: INC C ; C = 1 -> SCC+ = YES .setSCC: XOR A DEC HL LD (HL),A ; disable write to SCC+ RAM LD A,00h ; register "A" will contain slot/subslot with SCC Cartridge .c_slt: EQU $ - 1 ; register "C" will contain "0" for SCC and "1" for SCC+ [...your code here...] RET
You can search for more than one SCC cartridge and choose the "best" one. Also SCC data will remain undamaged after search is finished.
Thanks max_iwamoto, your code works fine ('A' contains the correct slot/subslot of the SCC), but I found a few problems:
1) two " pop af " are missing (one because of the PUSH AF and other because of calling CHK_SCC without returning).
2) I'm not sure if the data of the SCC remains undamaged since I get no sound after running the code. I will inspect what is going on.
S
saccopharynx wrote:
Thanks max_iwamoto, your code works fine ('A' contains the correct slot/subslot of the SCC), but I found a few problems:
1) two " pop af " are missing (one because of the PUSH AF and other because of calling CHK_SCC without returning).
2) I'm not sure if the data of the SCC remains undamaged since I get no sound after running the code. I will inspect what is going on.
S
1) As I mention before, I am not stopping search after I found first available SCC. I am making a table of all available sound cartridges. But if you want exit as soon as you found one, you can indeed do 2 x POP + RET.
2) Technically there is no need to leave wave tables untouched, because most of the time you do check before music player even load them with any data and usually they are contain "00" at the start...
But if you have music re-player for SCC-I and you have found a regular SCC, you need to patch the player to hear any sound. But I could help more if I knew what you trying to achieve.
Max
My pretty simple version:
;------------------------------------------
; SCCDETECT (Made by : NYYRIKKI)
; Input: (None)
; Output:
; Success:
; CF = NC
; SCC SLOT #8000-#BFFF
; A = SlotID
;
; Fail:
; CF = C
; (Random slot on #8000-#BFFF)
; Changes: All registers
;------------------------------------------
SCCDETECT:
LD D,#FF
LD HL,#FCC1
.MAINL:
INC D
LD A,4
CP D
SCF
RET Z
LD A,(HL)
INC HL
AND #80
JR Z,.MAINTST
CALL .SUBTST
JR C,.MAINL
RET
.MAINTST:
LD A,D
CALL .TESTSLOT
JR C,.MAINL
LD A,D
RET
;-----------------------
.SUBTST
LD E,#FC
.SUBLOOP:
LD A,4
ADD A,E
CP 16
SCF
RET Z
LD E,A
OR D
OR #80
LD C,A
CALL .TESTSLOT
JR C,.SUBLOOP
LD A,C
RET
.TESTSLOT:
PUSH BC
PUSH DE
PUSH HL
LD H,#80
CALL #24
LD A,(#9000)
LD D,A
LD A,#3F
LD (#9000),A
LD HL,#9800
LD E,(HL)
XOR A
LD (HL),A
LD A,(HL)
OR A
JR NZ,.NOSCC
DEC A
LD (HL),A
LD A,(HL)
INC A
JR NZ,.NOSCC
LD A,(#9000)
CP #3F
JR Z,.NOSCC
XOR A
JR .EXIT
.NOSCC
LD (HL),E
LD A,D
LD (#9000),A
SCF
.EXIT
POP HL
POP DE
POP BC
RET
Thanks to you both. It is working fine now.
saccopharynx wrote:
Thanks to you both. It is working fine now.
No problem.
Here is screenshot from openMSX with 8 x SCC/SCC-I cartridges attached 
I tried to insert image directly to this post, but it didn't work properly. So this is the direct link:
https://docs.google.com/file/d/0B1kyb4fqWeQ9eGpZTWRMczI3SVU/...
Sorry for resurrecting this post but the subject is related.
Is there any reason for this code to not work on bluemsx when an SCC-I is plugged ?
The SCC-I should start in SCC mode at boot, never the less, on bluemsx it is not properly detected.
;------------------------------------------------------------
; SCC-search v1.0
; by Alwin Henseler
; using method described in bulletin # 18 MSX-club Enschede
; input: none
; output: B=slot that contains SCC (=255 if no SCC found)
enaslt: equ #0024
exptbl: equ #fcc1
slttbl: equ #fcc5
begin:
MAP #c000
in a,(#a8) ; read prim. slotregister
rra
rra
rra
rra
and %00000011 ; A = prim.slot page 2
ld b,0
ld c,a
ld hl,exptbl
add hl,bc
bit 7,(hl) ; page 2-slot expanded ?
jr z,scctest
ld hl,slttbl
add hl,bc
ld a,(hl) ; A = sec.sel.reg. of page 2-slot
rra
rra
and %00001100 ; bit 1/2 = sec.slot page 2
or c
set 7,a ; compose sec.slot-code
scctest: push af ; save page 2-slot on the stack
ld a,(exptbl) ; 1st slot to test
testslot: push af ; save test-slot on the stack
ld h,#80
call enaslt ; switch slot-to-test in 8000-bfffh
ld hl,#9000
ld b,(hl) ; save contents of address 9000h
ld (hl),#3f ; activate SCC (if present)
ld h,#9c ; address of SCC-register mirrors
ld de,#9800 ; 9800h = address of SCC-registers
testreg: ld a,(de)
ld c,a ; save contents of address 98xxh
ld a,(hl) ; read byte from address 9cxxh
cpl ; and invert it
ld (de),a ; write inverted byte to 98xxh
cp (hl) ; same value on 9cxxh ?
ld a,c
ld (de),a ; restore value on 98xxh
jr nz,nextslot ; unequal -> no SCC -> continue search
inc hl
inc de ; next test-addresses
bit 7,l ; 128 addresses (registers) tested ?
jr z,testreg ; no -> repeat mirror-test
ld a,b
ld (#9000),a ; restore value on 9000h
pop bc ; retrieve slotcode (=SCC-slot) from stack
jr done ; SCC found, restore page 2-slot & return
nextslot: ld a,b
ld (#9000),a ; restore value on 9000h
pop bc ; retrieve slotcode from stack
bit 7,b ; test-slot = sec.slot ?
jr z,nextprim
ld a,b
add a,4 ; increase sec.slotnumber
bit 4,a ; sec.slot = 4 ?
jr z,testslot
nextprim: ld a,b
and %00000011
cp 3 ; prim.slot = 3 ?
jr z,noscc
inc a ; increase prim.slotnumber
ld d,0
ld e,a
ld hl,exptbl
add hl,de
or (hl) ; combine slot-expansion with slotcode
jr testslot
noscc: ld b,255 ; code for no SCC
done: pop af ; retrieve page 2-slot from stack
push bc
ld h,#80
call enaslt ; restore original page 2-slot
pop bc
ei
ret
end:
endmap
; -------------------------------------------------------------
; ====================
; Initialization
; ====================
SCCINIT
ld hl,begin
ld de,0C000H
ld bc,end-begin+1
ldir
call 0C000H
ld a,b
ld (SCC),a
ret
Needless to say it seems a bluemsx relate bug, as it works fine on openmsx.
Sorry to unearth this old topic, but I'd like to make some first steps with the SCC soundchip.
If I use the SCCDETECT routine written by Nyryikki a few messages above, how you go on using the SCC afterwards ?
Do you have to switch to the SCC slot at every access ?
Something like :
scc_on: call SCCDETECT ret c ; SCC not found ld (slot),a ret scc_write: ld a,(slot) ld h,80h call ENASLT (then write to SCC registers)
Is that correct ?