It's an interesting question. My understanding has always been that the shielding on printer cables wasn't primarily there to protect the data from external interference. The TTL signals driving a Centronics interface are relatively robust over a few metres. The bigger concern was the opposite direction, preventing the cable itself from becoming an efficient antenna.
A parallel printer cable is switching eight data lines, plus a collection of control and handshake lines, all with fairly fast edges. Without some thought to cable construction, you've essentially built a transmitter. Twisted pairs, alternating grounds and an overall shield all help reduce the electromagnetic energy escaping the cable, which also has the happy side effect of reducing crosstalk between adjacent conductors. That's one of the reasons the cheap ribbon cables often worked surprisingly well over short distances. They weren't ideal from an EMC perspective, but at six feet and with the relatively modest signalling rates of the day, there was enough margin that they generally got away with it. As clock rates increased, and EMC regulations became more stringent, cable construction became much more important. As for the background noise floor, I suspect you're right that it has changed dramatically over the last fifty years. Modern homes are full of switch mode power supplies, LED lighting, solar inverters, battery chargers, Ethernet over power, and countless embedded processors. As an amateur radio operator, I can certainly attest that the RF environment is far noisier than it used to be. That said, I think that's an argument for good EMC engineering rather than against it. Many of the measures that seemed almost excessive forty years ago are part of the reason the spectrum is still usable today. Without them, we'd probably be living in an even noisier world. Kindest regards, Doug Jackson em: [email protected] ph: 0414 986878 On Tue, 14 Jul 2026 at 09:11, <[email protected]> wrote: > Yes, the RS printer cable was just a 26-wire ribbon cable. The high lines > are ground as you would expect so you have ground lines between signal > lines. Not only can the M100 not send out parallel data that fast but the > printers it was used with were slower yet. > > > > Jeff Birt > > > > *From:* M100 <[email protected]> *On Behalf Of *B9 > *Sent:* Monday, July 13, 2026 5:50 PM > *To:* [email protected]; Peter Noeth <[email protected]>; Model 100 > Discussion <[email protected]> > *Subject:* SPAM-LOW: Re: [M100] Ideal length for M100 printer cable? > > > > Radio-Shack's six foot printer cable was unshielded and untwisted? I > wouldn't have expected that to be reliable. > > Am I overly cautious? Or was it perhaps only meant for very low speeds? > Did Radio-Shack at least ground every other wire in the ribbon to reduce > crosstalk? > > --b9 > > P.S. Probably not relevant, but I wonder if the level of background > electromagnetic noise has changed in the last fifty years. > > > > On July 13, 2026 2:33:37 PM PDT, Peter Noeth <[email protected]> wrote: > > Jeff, > > > > If I remember my "70s mini-computer days" correctly, the normal max > length for Centronics Printer cables was 12'. This was with the round > cable, consisting of 18 "twisted wire pairs" (one wire signal the other > wire ground), that was then shielded with an aluminum foil wrap and a bare > ground "drain" wire connecting it the full length to the metal shells of > the connectors. > > > > Doing the same with simple flat "ribbon cable" I would expect that Radio > Shack's 6' length would be the maximum, as they tested functionality with > that length (but not FCC compliance). Radio Shack used that cable for all > of their computers, with the "business machines" usually connected to a > printer on a floor stand, so the 6' length would be appropriate, and for > connecting a "home computer" with an Epson MX-80 size printer on the same > tabletop, the cable would just drape down behind the table. > > > > But for portability with the M100 (or T-102 / T-200) I would think a 3' > cable would be better, especially if you use its original TRP-100 printer > (or some other portable printer of similar size), as it would be more > convenient to roll up the cable to pack in a computer bag. > > > > The one I made to use with my Seiko DPU-414 printer is 34" long. For me > this is long enough to use the computer in my lap while watching T.V. and > have the printer sitting on the end table. > > > > Regards, > > PeterN > > > > On Mon, Jul 13, 2026 at 1:10 PM <[email protected]> wrote: > > > Message: 1 > Date: Sun, 12 Jul 2026 18:56:53 -0500 > From: [email protected] > To: [email protected] > Subject: [M100] Ideal length for M100 printer cable? > Message-ID: <[email protected]> > Content-Type: text/plain; charset="us-ascii" > > Hi all, > > > > I had someone ask about acquiring an M100 printer cable a few weeks ago and > was surprised how much they were going for on eBay. I have an original > cable > which is 6' long, which is a bit too long, I think. I picked up some > Centroincs connectors and just made a 3' one which is fine for sitting the > M100 next to a CGP-115 or larger printer but perhaps not for having the > printer sitting away from the M100. > > > > What do you think the ideal length is? 3', 4', 5', 6' or perhaps just offer > then in either 3' or 6'? > > > > Jeff Birt > > (Hey Birt!) > <big snip> > >
