PDP-8/e

[Jay Jaeger]

I did that sort of thing for my PDP-8/L, where the reader run drove the
RS-232 "CTS" control signal and wrote a "C" program to do simple TTY
emulation in DeSmet C back in the day.

That code would not run in Windows of course, but it wouldn't be all
that difficult for someone with a C programming background to move it to
Windows under gnucc, or even Microsoft C++ or C#.

On 12/8/2018 1:10 AM, Rod G8DGR via cctalk wrote:
> I’m sure that would work  but I only have an 8650  110 baud only card
> Rod
> 
> 
> Sent from Mail for Windows 110 baud 
> 
> From: Bob Rosenbloom via cctalk
> Sent: 08 December 2018 03:41
> To: cctalk at classiccmp.org
> Subject: Re: PDP-8/e
> 
> On 12/7/2018 7:01 PM, Pete Turnbull via cctalk wrote:
>> On 07/12/2018 17:44, Jon Elson via cctalk wrote:
>>> On 12/07/2018 11:22 AM, systems_glitch via cctalk wrote:
>>>> Indeed, unless you need character pacing.
>>>>
>>>>
>>> Actually, with the correct settings of the serial port (xon/xoff or 
>>> CTS pin) the serial port driver should do this, too, so cat would work.
>>
>> A PDP-8/E doesn't have a CTS pin and the loaders don't support 
>> XON/XOFF, though.
>>
> The PDP-8 needs to control the serial CTS function. This was called 
> reader-run when using a Teletype machine. FOCAL won't load without it.
> You can modify the serial card (mine was an M8655) to support the 
> function. Here's what I did:
> 
> Cleaned up from Aaron Nabil's and Lyle Bickley's write up.
> 
>   Hack the M8655 to support reader-run by mapping it to RS-232 hardware 
> flow control.
> 
> 1. Cut the trace leading from Pin 1 of E54 (a 7400).  This is the input 
> that clears the Reader Run FF when a new character starts to come in.
> 
> 2. Jumper from Pin 1/E54 to Pin 3/E38, a spare gate on a 7400 that we 
> are going to use an inverter.
> 
> 3. Tie Pin 1 and Pin 2 of E38 together, and run them to Pin 20 of E19, 
> the UART.
>      This supplies the signal to the reader-run FF that tells it that 
> it's got an incoming character and to de-assert the reader-run line.
>      Normally this is tied to the current loop receiver, we've just 
> moved it to the UART so any received data will clear the FF.
> 
> 4. Cut a ground traces on 4 of E50, a 1488 RS-232 transmitter. This is 
> what would normally supply the continuously asserted RTS (and DTR) signal.
> 
> 5. Jumper from pin 7 of E39, a 7474 flip-flop to pins 4 of E50. E39 is 
> the "reader-run flip-flop".  Now RTS follows the reader run signal.
> 
> Bob
>