On Wed, 18 Nov 2020 23:20:04 GMT, Russell Marks wrote:
> Lee Osborne <losbo...@gmail.com> wrote:
>> I'm the owner of a (grey case) Spectrum +2, and I'm keen to try
>> communication with other devices via RS232. Obviously, the +2 has a
>> built-in RS232 port, but as far as I can tell, it's not much good for
>> two-way comms with other machines, and is of course hampered by the
>> completely non-standard plug.
> It can be vaguely tolerable if you make sure you use RTS/CTS
> handshaking, and your serial lead is properly wired. It's still pretty
> bad though. I think I got my best results at 2400 baud.
Interesting. So the IF-1 hard/software is that much different to what Sir
Clive and his folks have used for the +2 later? Hard to believe. For my
Microdrive transfers, I have used the IF-1 against an IBM Thinkpad T30
(with a _real_ RS232 port, not the USB-emulation) up to 19'200 baud, no
Details here: https://worldofspectrum.net/pub/sinclair/tools/pc/mdv2img.zip
, also includes how to fabricate a cable supporting RTS/CTS.
>> If I want to play with sending data to and from a PC, and also other
>> old crap I have lying around like my Amstrad NC100, is it best to use
>> the built-in port or the IF1?
> I seem to remember the IF1 port was very slow
Well. The Z80 CPU runs at 3.5 MHz, there is no SIO available and the 16550
UART (with FIFO) was born years after the Spectrum. In that light, serial
communication via software...
0C5A BCHAN-OUT LD B,+0B Counts '11' bits.
0C5C CPL Invert the eight data bits.
0C5D LD C,A C holds the byte to be sent.
0C5E LD A,(IOBORD) Fetch new border colour.
0C61 OUT (+FE),A Change border colour.
0C63 LD A,+EF Reset CTS and select RS232.
0C65 OUT (+EF),A
0C68 OUT (+F7),A Make RXdata have a low signal level.
0C6A LD HL,(BAUD) Fetch timing constant.
0C6D LD D,H Copy into DE register pair.
0C6E LD E,L
0C6F BD-DEL-1 DEC DE Firstly wait (26*(BAUD)) T cycles.
0C70 LD A,D
0C7l OR E
0C72 JR NZ,0C6F,BD-DEL-1
0C74 TEST-DTR LD A,+7F Read port +7FFE (SPACE key).
0C76 IN A,(+FE)
0C78 OR +FE Read port +FEFE (CAPS SHIFT key) only
0C7A IN A,(+FE) if SPACE is being pressed.
0C7C RRA Test bit 0.
0C7D JP NC,0CB4,BRK-INOUT Give an error if BREAK is pressed.
0C80 IN A,(+EF) Read DTR line.
0C82 AND +08 Only bit 3.
0C84 JR Z,0C74,TEST-DTR Jump back until DTR is found high.
0C86 SCF Set carry flag (start bit).
0C87 DI Disable interrupts.
0C88 SER-OUT-L ADC A+00 In fact, shift carry into bit 0 of
0C8A OUT (+F7),A Send this bit over RS232 link.
0C8C LD D,H Copy timing constant into DE.
0C8D LD E,L
0C8E BD-DEL-2 DEC DE Wait 26*(BAUD) T cycles after having
0C8F LD A,D sent each bit.
0C90 OR E
0C9l JR NZ,0C8E,BD-DEL-2
0C93 DEC DE Wait 6 T cycles.
0C94 XOR A Clear A for next pass.
0C95 SRL C Shift the bit to be sent into carry.
0C97 DJNZ 0C88,SER-OUT-L Loop for all 11 bits.
0C99 DI Enable interrupts.
0C9A LD A,+01
0C9C LD C,+EF
0C9E LD B,+EE
0CA0 OUT (+F7),A Make RXdata having a high level.
0CA2 OUT (C),B Turn off RS232.
0CA4 BD-DEL-3 DEC HL Insert a final delay before returning
0CA5 LD A,L via the border-restore routine
0CA6 OR H below.
0CA7 JR NZ,0CA4,BD-DEL-3
... is as good as it probably gets (source code quote and comments courtesy
of Gianluca Carri's "Spectrum Shadow ROM Disassembly") up to 19'200 Baud.
>> Also, as I have the IF1 printer cable, which ends in a 25-pin male
>> plug, is it possible to just get hold of one other cable that will let
>> me connect direct to a PC?
You'll have to download spxfr at
and solder the connection cable mentioned there. Basically you'll need a
female RS232 9-pin connector (PC) and a male RS232 9-pin connector
(Speccy) connected with a shielded 5-pin round cable as follows:
2 - RD 3 - TD
3 - TD 2 - RD
5 - GND 7 - GND (Shield)
7 - RTS 4 - CTS
8 - CTS 5 - RTS
1 - DCD 9 - DTR
>> Am I insane for wondering if it's possible to convert text from the PC
>> to Tasword files???
I would rather do it the other way round (ZX->PC) but my Microdrive
cartridges are all rotten and it's quite hard to forensically
restructure/realign the errorneous sectors. Tasword accepts plain text with
hard formatting. IIRC it's not CR/LF but LF only. Just try it out.
> I think Tasword Two files (at least) were pretty much just a straight
> dump of the 64-column ASCII text with no line breaks. So 64 chars for
> one line, 64 for the next, and so on. I still have a few old Tasword
> files from the 80s (originally from microdrive) and they all seem to
> be like that.
Jup. The files on my tapes (still intact, sortof) were like that.