Fwd: some useful features
ZXNet echo conference «code.zx»
- --- begin of forward ---
From: Kirill Frolov
Newsgroups: zxnet.code
Subject: some useful features
Date: Sun, 6 Jul 2003 08:07:24 +0000 (UTC)
Press RESET immediately, All!
Some spektrumists asked the question about generation
random numbers and tabular calculation of CRC...
General notes for everyone
presented functions.
----------------------------------
Any functions unless otherwise specified
specifically, do not guarantee safety
any microprocessor registers except
registers I and IY. Features marked
guaranteed to be C compiler safe
ensure the safety of register IX.
All functions unless otherwise noted
specifically, are reentrant,
in other words allow nested or
recursive calls, and do not use
static variables or self-modifying
collapsing code. All variables are highlighted
are on the stack.
Number conversion functions:
------------------------------
htoa
The function writes unsigned
number specified in register A
to the string specified by the HL register
in hexadecimal
Reckoning. 2 will be recorded
digits (leading zero present)
vuet) and the symbol with code 0 is completed
reading a line. Upon completion
function register HL will indicate
call on the last character of the line
ki (terminating 0).
wtoaThe function writes an unsigned
the number specified in the BC register
to the string specified by the HL register
in the decimal number system.
Leading zeros are not recorded.
Up to 6 bytes can be written,
including the symbol with code 0 completed
reading a line. In HL register
returns a pointer to the after-
Day character of the line (code 0).
Functions for working with strings:
------------------------------
strglob
The function performs a comparison
string specified by the HL register with
mask string of the specified region
stroma DE. In case the line
corresponds to the mask flag Z will be
installed.
The mask can consist of any
characters (except the character with
code 0). Some symbols
are interpreted in a special way:
* -- means any number
any symbols.
? -- means any one
symbol.
Math functions:
------------------------------
mul
The function multiplies the unknown
The original 16-bit numbers are given
in registers DE and BC, and placed
stores the result in the DE registers and
H.L. The DE register stores the old
The lowest bits of the result. Function you-
completes in approximately 800 clock cycles
processor. See also descriptionamul function.
amul
The function multiplies unsigned
High numbers specified in registers
A and DE and adds the result to
the contents of the HL register. Function
runs faster than mul,
in about 350 cycles and maybe
be used to access
to array elements. In that case
tea if the dimensions of the elements or
array dimensions are multiples of 2^n
or are constants maybe
be more profitable to use more
other ways to accelerate smart
marriage. In case it is required
faster access to elements
arrays of arbitrary size
element or dimension can
be appropriate to use
Aylifa massifs.
div
The function divides a given number
in the HL register to the number specified
in the BC register, the result is placed
is entered into the HL register, and the remainder
to register DE. In case of division
is written to 0 in the HL register
zero, and in the DE register the value
divisible. The function is running
in about 900 cycles.
sqrt
The function calculates the integer
square root value
from an unsigned number given
nogo in the HL register, the result
returned in register A.
w2bcdThe function converts unsigned
16-bit number in binary
but-decimal format. Argument
is specified in the HL register, the result is
tat is returned in registers
D, E and C. The most significant bits of the result are
they are contained in register D, low
in register C.
randn
The function returns pseudo-cases
tea number in a given range
no. Minimum return value
the random number being specified is 0.
The maximum is always less than the number
specified in the DE register. Result
tat is returned in the HL register.
See also function description
rand.
rand
The function returns unsigned
pseudorandom number in register
H.L. The returned number is
in the range 0..65535. Function
runs a little faster
functions randn. Execution of functions
tion rand in time approximately
equals 3 calls to the mul function,
while runtime
randn functions are comparable to 4
calls to the mul function.
srand
The function sets the generator
torus of random numbers. When installing
vke generator with one value
generated sequence
there will be pseudorandom numbers
always repeat.
Functions for calculating control codes:-------------------------------------
flatcher
The function calculates the values
Fletcher code. In register IX
a pointer to the area is passed
memory, the contents of which
calculation will be made. In re-
size is transferred to the BC hyster
memory areas in bytes.
In registers HL and DE return
16-bit values are available
Fletcher code. It's not too much
good feature for control
data integrity, but works
faster than the CCC calculation functions.
crc16_init
The function initializes the table
for quick counting 16-bit
cyclic excess
code (CEC) with the crc16_fast function.
The de register contains the address
512 byte memory for
saving the calculated table.
The BC register indicates the threshold
giving a CEC polynomial. In telecom
communication protocols using
recommended by the CCITT
value: #1021.
crc16_fast
The function calculates the value
CEC for the memory area from the beginning
address specified in the register
IX and length in bytes specified in
BC register. In the DE register the re-
the address of the preliminary table is given
exactly calculated using
functions crc16_init. In the registerHL indicates the initial value
tion of the Central Election Commission. The result will be calculated
is returned in the HL register.
crc16
The function calculates the value
CEC for the memory area from the beginning
address specified in the register
IX and length in bytes specified in
BC register. In the DE register,
called the generating polynomial
Central Election Commission. In the HL register it is set to
initial value of the CEC. Result
calculation returns to the register
str HL. As a generative
CEC polynomial in telecommunications
cation protocols uses-
this is the CCITT recommended value
#1021.
The result of this function is
the execution is completely similar to
crc16_fast function, however
this function does not use infection
her calculated table and time
her work is much larger.
It is recommended to use the function
tions crc16_init and crc16_fast,
if you need to count
CEC of large data volumes.
- --- beginning of forward ---
From: Kirill Frolov
Newsgroups: zxnet.code
Subject: нетоторые пункциы
Date: Sun, 6 Jul 2003 08:07:24 +0000 (UTC)
[...subject..subject.
ifndef htoa
ifused htoa
; a=0..FF, HL=str[3] -> hl=*0
htoa
push off
rra
rra
rra
rra
call nib2hex
pop off
nib2hex and #0f
cp #0a
sbc a, #69
daily
ld (hl), a
inc hl
ld (hl),
ret
endif
endif
ifndef string globe
ifused string globe
; hl=string, de=mask -> Z
strglob
ld a, (de)
or a
jr nz, strglob_1
cpi ; end
ret
strglob_0
xor a
inc a
ret
strglob_1
ld a, (hl)
or a
jr z, strglob_0 ; not equal
ld a, (de)
cp "?"
jr z, strglob_q
cp "*"
jr z , strglob_a
cp (hl)
ret nz ; not equal
strglob_q
inc hl
inc of
jr strglob
strglob_a
inc hl
push hl
push there
call strglob
pop there
pop hl
ret z
inc of
jr strglob
endif
endif
ifused wtoa
ifndef wtoa
; slightly less than 16-year-old snowflakes; in decimal form, leading zeros are not discarded.
; 5 digits of the number will be written to address (hl) and
; character with code 0 ending the line.
; (C) Kirill Frolov
; *hl = itoa(de) -> hl=pointer to terminating 0.
wtoa
push hl
ex de, hl
call w2bcd
pop hl ; d,e,c=bcd
ld a, d
or a
jr nz, wtoa0
ld a, e
and #f0
jr nz, wtoa1
ld a, e
and #0f
jr nz, wtoa2
ld a, c
and #f0
jr nz, wtoa3
jr wtoa4
wtoa0
or #30
ld (hl), a
inc hl
wtoa1
ld (hl), e
ld a, #33
rrd
inc hl
wtoa2
or #30
ld (hl), a
inc hl
wtoa3
ld (hl), c
ld a, #33
rrd
inc hl
wtoa4
or #30
ld (hl), a
inc hl
ld (hl), 0
ret
endif
endif
ifused rand
; Random number generator, slow
; (C) Kirill Frolov ~3000 takts.
; hl = random(0.. hl = random number 0..65535 ~2300 takts.
rand; randx = randx*1103515245 + 12345)>>16) & 077777)
ld de, (randx+2)
ld bc, #4e6d
call mul
push hl
ld de, (randx)
ld bc, #41c6
call mul
pop de
add hl, de
push hl
ld de, (randx)
ld bc, #4e6d
call mul
ld bc, #3039
add hl, bc
ld (randx), hl
pop hl
adc hl, de
ld (randx+2), hl
ret
endif
ifused srand
; Installing a random number generator
; de = trigger value
srand
ld (randx), de
ret
endif
ifused div
ifndef div
; 16-bit unsigned division function
; (C) Aleksey Malov aka Vivid
; little optimized by Vitaly Vidmirov
; hl = de/bc de = de%bc ~840 takts.
; hl = 0, de = de(dividend) if bc(divisor) == 0.
div
xor a ;change sign
subc ; at the divider
ld c, a
sbc a, b
sub c
ld b, a
ld hl,0 ;reset the new dividend to zero
ld a,d ;move first
rla ;high byte of dividend
;t=18
REPT 8
rl l ;
add hl,bc ;
jr c,$+4 ;8 times
sbc hl,bc ;
rla ; t=8*45=360
ENDR
ld d,a ;st. result byte
ld a,e ;now move
rla ;low byte t=12REPT
adc hl,hl ;
add hl,bc ;
jr c,$+4 ;8 months
sbc hl,bc ;
rla ; t=8*52=4
ENDR
ld e,a ;мл. байт результата
;hl-view of the делеия
;t=4
ex of, hl
ret
endif
endif
ifused mul
ifndef mul
; you can get a 16-year-old boy
; (C) Kirill Frolov
; idea by Vasil Ivanov
; dehl=de*bc ~730h.
mul
ld hl,
ld a, c
ld c, l ; 18
add a, a
jr nc, $+4
add hl, there
adc a, c ; 29
rept
add hl, hl
rla
jr nc, $+4
add hl, there
adc a, c ; 40*7 309
endr
ld c, a
push hl ; 0chl
ld hl,
ld a, b
ld b, l ; 33
add a, a
jr nc, $+4
add hl, there
add a, b ; 29
rept
add hl, hl
rla
jr nc, $+4
add hl, there
add a, b ; 40*7 309
endr
; ahl0 + 0cde = hlde
pop there
ld b, l
ld l, h
ld h, a
ld a, d
add a, b
ld d, a
ld b,
adc hl, bc
ex there, hl ; 60
ret ; 729t.
endif
endif
ifused amul
ifndef amul; 8-bit multiply-sum function
; (C) Kirill Frolov
; hl = hl + a*de ~338 takts.
amul
add a, a
jr nc, $+4
add hl, de
adc a, c ; 39
rept 7
add hl, hl
rla
jr nc, $+4
add hl, de
adc a, c ; 40*7 309
endr
ret
endif
endif
ifused sqrt
ifndef sqrt
; function for calculating the square root of 16-bit numbers
; (C) Aleksey Malov aka Vivid
; a = sqrt(hl) ~470 takts.
sqrt
xor a ;zeroed a ;-)
ld de,64 ;this is required by the algorithm
; t=14
sla h ;take the two leftmost
adc a,a ;argument bit
sla h;
adc a,a ;
jr z,$+4 ;if they are not equal
dec a ;zero, then increase
inc d ;result
; t=39
REPT 3
sla h;follow
adc a,a ;extraction algorithm
sla h ;square root
adc a,a ;"column", only
sla d ;
ld c,d ;3 times
sli c ;
cp c ;
jr c,$+4 ;
subc ;
inc d; t=63*3=189
ENDR
REPT 2
sla l ;we do something similar
adc a,a ;operation with
sla l ;low byte
adc a,a ;argument
sla d ;ld c,d ;2 times
sli c ;
cp c ;
jr c,$+4 ;
subc ;
inc d; t=63*2=126
ENDR
ld h,a ;
or a ;cleared the carry flag
; t=8
sbc hl,de ;now we lack
jr nc,$+3 ;one register d for
add hl,de ;execute the last
ccf ;two cycles, will have to
rl d ;use more
add hl,hl ;slow operations
add hl,hl ;with register pairs
; t=67
sbc hl,de ;and last time
ccf ;perform calculations, not
ld a,d ;restoring hl
adc a,a ;in a-result
; t=27
ret
endif
endif
ifused w2bcd
ifndef w2bcd
; function translates 16-bit unsigned
; binary number to BCD format
; (C) Kirill Frolov
; d,e,c = bin2bcd(hl) ~1330 takts.
w2bcd
ld b, 10h
xor a
ld c, a
ld e, a
ld d, a ; 23
w2bcd0
add hl, hl ; D,E,C=bcd digits
ld a, c
adc a, a
yeah
ld c, a
ld a, e
adc a, a
yeah
ld e, a
ld a, d
adc a, a
yeah
ld d,a
djnz w2bcd0 ; ~83 ~1328
ret
endif
endif
; Attention!; All 16-bit CEC counting functions
; use only shift left!
ifused flatcher
; Fletcher code counting function
; (faster, but less efficient than CRC)
; (C) Kirill Frolov
; ix=data, bc=size,
; -> hl=fletcher code (A), de=(B), E&L=8bit code
flatcher
ldhl,0
ld e, l
ld d, h
flatch0
ld a, b
or c
ex de, hl
ret z
ex de, hl
ld a, (ix)
inc ix
add a, e
ld e, a
jr nc, $+3
inc d
add hl, de
dec bc
jr flatch0
endif
ifused crc16_init
; A function that initializes the CRC table.
; (C) Kirill Frolov
; de=table location address (512 bytes),
; bc=polynomial (0x1021 recommended by CCITT).
crc16_init
push ix
ld xl, e
inc d
inc d
ld xh, d
ld e, c
ld d, b
ld c, 0
crc_i0
ld l, 0
ld h, c
dec h
ld b, 8
crc_i1
add hl, hl
jr nc,crc_i2
ld a, l
xor e
ld l, a
ld a, h
xor d
ld h, a
crc_i2 djnz crc_i1
Dec ix
ld (ix+0), l
dec xh
ld (ix+0), h
inc xh
dec c
jr nz, crc_i0
pop ix
ret
endif
ifused crc16_fast; fast (tabular) CRC-16 counting function,
; requires initialization with crc16_init function
; (C) Kirill Frolov
; ix=*data, bc=size,
; de=CRC table, hl=initial value
; on output -> hl=calculated value
crc16_fast
push de
exx
pop de
exx
crc16f0
ld a, b
or c
ret z
ld a, h
xor (ix)
inc ix
exx
ld l, a
ld h, 0
adc hl, de
ld a, (hl)
inc h
exx
xor l
ld h, a
exx
ld a, (hl)
exx
ld l, a
dec bc
jr crc16f0
endif
ifused crc16
; slow CRC-16 counting function
; (C) Kirill Frolov
; ix=*data, bc=size
; de=polynomial (#1021), hl=initial value
; on output -> hl=calculated value
crc16
crc16_0
ld a, b
or c
ret z
ld a, (ix)
inc ix
xor h
ld h, a
ld a, 8
crc16_1
add hl, hl
jr nc, crc16_2
exa
ld a, l
xor e
ld l, a
ld a, h
xor d
ld h, a
exa
crc16_2
dec a
jr nz, crc16_1
dec bc
jr crc16_0
endif
- --- end of forward ---
- --- beginning of forward ---
From: Kirill Frolov
Newsgroups: zxnet.code
Subject: нетоторые пункциы
Date: Sun, 6 Jul 2003 08:07:24 +0000 (UTC)
[...subject..subject.
ifndef htoa
ifused htoa
; a=0..FF, HL=str[3] -> hl=*0
htoa
push off
rra
rra
rra
rra
call nib2hex
pop off
nib2hex and #0f
cp #0a
sbc a, #69
daily
ld (hl), a
inc hl
ld (hl),
ret
endif
endif
ifndef string globe
ifused string globe
; hl=string, de=mask -> Z
strglob
ld a, (de)
or a
jr nz, strglob_1
cpi ; end
ret
strglob_0
xor a
inc a
ret
strglob_1
ld a, (hl)
or a
jr z, strglob_0 ; not equal
ld a, (de)
cp "?"
jr z, strglob_q
cp "*"
jr z , strglob_a
cp (hl)
ret nz ; not equal
strglob_q
inc hl
inc of
jr strglob
strglob_a
inc hl
push hl
push there
call strglob
pop there
pop hl
ret z
inc of
jr strglob
endif
endif
ifused wtoa
ifndef wtoa
; slightly less than 16-year-old snowflakes; in decimal form, leading zeros are not discarded.
; 5 digits of the number will be written to address (hl) and
; character with code 0 ending the line.
; (C) Kirill Frolov
; *hl = itoa(de) -> hl=pointer to terminating 0.
wtoa
push hl
ex de, hl
call w2bcd
pop hl ; d,e,c=bcd
ld a, d
or a
jr nz, wtoa0
ld a, e
and #f0
jr nz, wtoa1
ld a, e
and #0f
jr nz, wtoa2
ld a, c
and #f0
jr nz, wtoa3
jr wtoa4
wtoa0
or #30
ld (hl), a
inc hl
wtoa1
ld (hl), e
ld a, #33
rrd
inc hl
wtoa2
or #30
ld (hl), a
inc hl
wtoa3
ld (hl), c
ld a, #33
rrd
inc hl
wtoa4
or #30
ld (hl), a
inc hl
ld (hl), 0
ret
endif
endif
ifused rand
; Random number generator, slow
; (C) Kirill Frolov ~3000 takts.
; hl = random(0.. hl = random number 0..65535 ~2300 takts.
rand; randx = randx*1103515245 + 12345)>>16) & 077777)
ld de, (randx+2)
ld bc, #4e6d
call mul
push hl
ld de, (randx)
ld bc, #41c6
call mul
pop de
add hl, de
push hl
ld de, (randx)
ld bc, #4e6d
call mul
ld bc, #3039
add hl, bc
ld (randx), hl
pop hl
adc hl, de
ld (randx+2), hl
ret
endif
ifused srand
; Installing a random number generator
; de = trigger value
srand
ld (randx), de
ret
endif
ifused div
ifndef div
; 16-bit unsigned division function
; (C) Aleksey Malov aka Vivid
; little optimized by Vitaly Vidmirov
; hl = de/bc de = de%bc ~840 takts.
; hl = 0, de = de(dividend) if bc(divisor) == 0.
div
xor a ;change sign
subc ; at the divider
ld c, a
sbc a, b
sub c
ld b, a
ld hl,0 ;reset the new dividend to zero
ld a,d ;move first
rla ;high byte of dividend
;t=18
REPT 8
rl l ;
add hl,bc ;
jr c,$+4 ;8 times
sbc hl,bc ;
rla ; t=8*45=360
ENDR
ld d,a ;st. result byte
ld a,e ;now move
rla ;low byte t=12REPT
adc hl,hl ;
add hl,bc ;
jr c,$+4 ;8 months
sbc hl,bc ;
rla ; t=8*52=4
ENDR
ld e,a ;мл. байт результата
;hl-view of the делеия
;t=4
ex of, hl
ret
endif
endif
ifused mul
ifndef mul
; you can get a 16-year-old boy
; (C) Kirill Frolov
; idea by Vasil Ivanov
; dehl=de*bc ~730h.
mul
ld hl,
ld a, c
ld c, l ; 18
add a, a
jr nc, $+4
add hl, there
adc a, c ; 29
rept
add hl, hl
rla
jr nc, $+4
add hl, there
adc a, c ; 40*7 309
endr
ld c, a
push hl ; 0chl
ld hl,
ld a, b
ld b, l ; 33
add a, a
jr nc, $+4
add hl, there
add a, b ; 29
rept
add hl, hl
rla
jr nc, $+4
add hl, there
add a, b ; 40*7 309
endr
; ahl0 + 0cde = hlde
pop there
ld b, l
ld l, h
ld h, a
ld a, d
add a, b
ld d, a
ld b,
adc hl, bc
ex there, hl ; 60
ret ; 729t.
endif
endif
ifused amul
ifndef amul; 8-bit multiply-sum function
; (C) Kirill Frolov
; hl = hl + a*de ~338 takts.
amul
add a, a
jr nc, $+4
add hl, de
adc a, c ; 39
rept 7
add hl, hl
rla
jr nc, $+4
add hl, de
adc a, c ; 40*7 309
endr
ret
endif
endif
ifused sqrt
ifndef sqrt
; function for calculating the square root of 16-bit numbers
; (C) Aleksey Malov aka Vivid
; a = sqrt(hl) ~470 takts.
sqrt
xor a ;zeroed a ;-)
ld de,64 ;this is required by the algorithm
; t=14
sla h ;take the two leftmost
adc a,a ;argument bit
sla h;
adc a,a ;
jr z,$+4 ;if they are not equal
dec a ;zero, then increase
inc d ;result
; t=39
REPT 3
sla h;follow
adc a,a ;extraction algorithm
sla h ;square root
adc a,a ;"column", only
sla d ;
ld c,d ;3 times
sli c ;
cp c ;
jr c,$+4 ;
subc ;
inc d ; t=63*3=189
ENDR
REPT 2
sla l ;we do something similar
adc a,a ;operation with
sla l ;low byte
adc a,a ;argument
sla d ;ld c,d ;2 times
sli c ;
cp c ;
jr c,$+4 ;
subc ;
inc d; t=63*2=126
ENDR
ld h,a ;
or a ;cleared the carry flag
; t=8
sbc hl,de ;now we lack
jr nc,$+3 ;one register d for
add hl,de ;execute the last
ccf ;two cycles, will have to
rl d ;use more
add hl,hl ;slow operations
add hl,hl ;with register pairs
; t=67
sbc hl,de ;and last time
ccf ;perform calculations, not
ld a,d ;restoring hl
adc a,a ;in a-result
; t=27
ret
endif
endif
ifused w2bcd
ifndef w2bcd
; function translates 16-bit unsigned
; binary number to BCD format
; (C) Kirill Frolov
; d,e,c = bin2bcd(hl) ~1330 takts.
w2bcd
ld b, 10h
xor a
ld c, a
ld e, a
ld d, a ; 23
w2bcd0
add hl, hl ; D,E,C=bcd digits
ld a, c
adc a, a
yeah
ld c, a
ld a, e
adc a, a
yeah
ld e, a
ld a, d
adc a, a
yeah
ld d,a
djnz w2bcd0 ; ~83 ~1328
ret
endif
endif
; Attention!; All 16-bit CEC counting functions
; use only shift left!
ifused flatcher
; Fletcher code counting function
; (faster, but less efficient than CRC)
; (C) Kirill Frolov
; ix=data, bc=size,
; -> hl=fletcher code (A), de=(B), E&L=8bit code
flatcher
ldhl,0
ld e, l
ld d, h
flatch0
ld a, b
or c
ex de, hl
ret z
ex de, hl
ld a, (ix)
inc ix
add a, e
ld e, a
jr nc, $+3
inc d
add hl, de
dec bc
jr flatch0
endif
ifused crc16_init
; A function that initializes the CRC table.
; (C) Kirill Frolov
; de=table location address (512 bytes),
; bc=polynomial (0x1021 recommended by CCITT).
crc16_init
push ix
ld xl, e
inc d
inc d
ld xh, d
ld e, c
ld d, b
ld c, 0
crc_i0
ld l, 0
ld h, c
dec h
ld b, 8
crc_i1
add hl, hl
jr nc,crc_i2
ld a, l
xor e
ld l, a
ld a, h
xor d
ld h, a
crc_i2 djnz crc_i1
Dec ix
ld (ix+0), l
dec xh
ld (ix+0), h
inc xh
dec c
jr nz, crc_i0
pop ix
ret
endif
ifused crc16_fast; fast (tabular) CRC-16 counting function,
; requires initialization with crc16_init function
; (C) Kirill Frolov
; ix=*data, bc=size,
; de=CRC table, hl=initial value
; on output -> hl=calculated value
crc16_fast
push de
exx
pop de
exx
crc16f0
ld a, b
or c
ret z
ld a, h
xor (ix)
inc ix
exx
ld l, a
ld h, 0
adc hl, de
ld a, (hl)
inc h
exx
xor l
ld h, a
exx
ld a, (hl)
exx
ld l, a
dec bc
jr crc16f0
endif
ifused crc16
; slow CRC-16 counting function
; (C) Kirill Frolov
; ix=*data, bc=size
; de=polynomial (#1021), hl=initial value
; on output -> hl=calculated value
crc16
crc16_0
ld a, b
or c
ret z
ld a, (ix)
inc ix
xor h
ld h, a
ld a, 8
crc16_1
add hl, hl
jr nc, crc16_2
exa
ld a, l
xor e
ld l, a
ld a, h
xor d
ld h, a
exa
crc16_2
dec a
jr nz, crc16_1
dec bc
jr crc16_0
endif
- --- end of forward ---
Hello, Kirill!
July 6, 2003 (it was 12:08 then)
Kirill Frolov in his letter to Kirill Frolov wrote:
KF> Some spectromists asked the question about generation
KF> random numbers and tabular calculation of CRC...
Some spectromists :)
Vega/ex-Style Group.
FIDO: 2:5030/1512 ZXNET: 500:812/19 E-mail: vega56@mail.ru
Hello Kirill!
Sunday, 06 July 2003, at 12:09:19,
Kirill Frolov -> Kirill Frolov.
KF> ; function translates 16-bit unsigned
KF> ; binary number to BCD
KF> ; (C) Kirill Frolov
KF> ; d,e,c = bin2bcd(hl) ~1330 takts.
KF> w2bcd
KF> ld b, 10h
KF> xor a
KF> ld c, a
KF> ld e, a
KF> ld d, a ; 23
KF> w2bcd0
KF> add hl, hl ; D,E,C=bcd digits
KF> ld a, c
KF> adc a, a
KF> daa
KF> ld c, a
KF> ld a, e
KF> adc a, a
KF> daa
KF> ld e, a
KF> ld a, d
KF> adc a, a
KF> daa
KF> ld d, a
KF> djnz w2bcd0 ; ~83 ~1328
KF> retw2bcd ld bc,#1000: ld e,c
w2bcd0 add hl,hl
ld a,c: adc a,a: daa: ld c,a
ld a,e: adc a,a: daa: ld e,a
rl d: djnz w2bcd0
ret
; b=18 ; t=1043
Close the door behind me, Kirill, I'm leaving.