using a well optimized algorithm
ZXNet echo conference «code.zx»
From Grigoriev Valery → To All 28 October 2005
Hello Jerry
I think it is possible to transmit at speeds (meaning serial
transmission via the tape recorder port) about 1-2 kb per second. That. 48 KB Spectrum
The whole thing can be filled, you can calculate how long it will take ;)
A specific example: all information is transmitted byte by byte, 10 bits (+2 bits
parity), on the transmitter side there are almost no problems with recognition, on
On the receiver side, apparently 3-5 measurements need to be taken, i.e. 3-5 teams In
A,(#FE) with all the equipment
Specifically:
IN A,(254) ;+11 cycles
ADD A,D ; D=#40, the required bit goes to CY
LD A,E ; E=0
ADD A,B ;+12 clock cycles, B stores the counter of identical measurements = 1
LD B,A ; +4 bars
total 5*27=135 clock cycles (for 5 times checking the transmission of 1 bit)
;at the end, of course, check the level
LD A,B
CP 4; 1 failure is allowed, i.e. 4 identical measurements
JR NC,BIT1 ; got bit = 1
CP 2
JR C,BIT0 ; got bit = 0
ERROR: END
; about 160 clock cycles per bit, transmission frequency = 20 kHz, divide by 10 =
number of bits, it turns out 2 kbytes per second peak capacity if the wire
good and the ports are clean ;) do not make noise on themselves ;))
If there is (as I have in the Baltic) a serial port, everything is simpler, of course
well...
The question remains: has anyone connected 2 Spectrums in this form?
From Grigoriev Valery → To All 28 October 2005
Hello Jerry
I think it is possible to transmit at speeds (meaning serial
transmission via the tape recorder port) about 1-2 kb per second. That. 48 KB Spectrum
The whole thing can be filled, you can calculate how long it will take ;)
A specific example: all information is transmitted byte by byte, 10 bits (+2 bits
parity), on the transmitter side there are almost no problems with recognition, on
On the receiver side, apparently 3-5 measurements need to be taken, i.e. 3-5 teams In
A,(#FE) with all the equipment
Specifically:
IN A,(254) ;+11 cycles
ADD A,D ; D=#40, the required bit goes to CY
LD A,E ; E=0
ADD A,B ;+12 clock cycles, B stores the counter of identical measurements = 1
LD B,A ; +4 bars
total 5*27=135 clock cycles (for 5 times checking the transmission of 1 bit)
;at the end, of course, check the level
LD A,B
CP 4; 1 failure is allowed, i.e. 4 identical measurements
JR NC,BIT1 ; got bit = 1
CP 2
JR C,BIT0 ; got bit = 0
ERROR: END
; about 160 clock cycles per bit, transmission frequency = 20 kHz, divide by 10 =
number of bits, it turns out 2 kbytes per second peak capacity if the wire
good and the ports are clean ;) do not make noise on themselves ;))
If there is (as I have in the Baltic) a serial port, everything is simpler, of course
well...
The question remains: has anyone connected 2 Spectrums in this form?
From Kirill Frolov → To Grigoriev Valery 29 October 2005
Press RESET immediately, Valery Grigoriev!
On Fri, 28 Oct 05 19:34:26 +0400, Grigoriev Valery wrote:
GV> I think it can be transmitted at speeds (meaning serial
GV> transmission via the tape recorder port) about 1-2 kb per second. That. 48 KB Spectrum
GV> can be completely filled, you can calculate how long it will take ;)
Zhavoronkov Sergejj -- 115200 delal. Ja ogranichilsja na urovne 38400.
GV> ; about 160 clock cycles per bit, transmission frequency = 20 kHz, divide by 10 =
GV> number of bits, resulting in 2 kbytes per second peak capacity if
GV> the wire is good and the ports are clean ;) they don’t make noise on themselves ;))
:-O
GV> If you have (as I have in the Baltic) a serial port, everything is simpler
GV> of course... The question remains: has anyone connected 2 in this form
GW> Spectrum?
Spectrum and PC via COM-port.