13

I find the RS-232 standard fascinating as it's one of the standards in computing that's survived - and is still used - for well over 50 years. The only major development has been a change in voltage from 12 V to 3.3 V in most applications, but with 12 V still available and being used.

The Wikipedia article on RS-232 tells us the standard was developed in 1960. The Bell 103 modem, the first one (acoustically coupled), was released in 1962.

That article also says the very first devices were "electromechanical typewriters" - which makes sense. The famous ASR-33 was released in 1963, but there were models beforehand. The Bell 103 couldn't have then been the first RS-232 device. So what was the first?

12
  • 1
    Note, there was a Bell 101 modem, commercialized in 1959 out of work on SAGE. It was 110 baud, which matches KSR/ASR 33 speeds. I can find nothing about it. It predates EIA-232 as an approved standard, but I suppose it could have implemented a protocol which was subsequently standardized.
    – dave
    Commented Sep 5, 2020 at 1:08
  • 6
    Given that RS-232 is basically just the traditional teletype current loop translated to voltage levels plus some control signals, it may be very hard to narrow down the first device that was /practically/ compatible with RS-232. It's relatively trivial (even with 1960s electronics) to convert between the two. The first RS-232 devices may have been a retrofit or custom option of some kind.
    – RETRAC
    Commented Sep 5, 2020 at 6:05
  • 3
    That question cannot be answered in a definitive way. It shows a clear misunderstanding on how standards used to evolve: RS-232 was designed and standardized after a significant number of devices implemented serial comms and it tried to align these implementations towards a common set of parameters. These devices obviously can't have been designed after the standard, because it didn't exist yet . (Typical chicken-and-egg-question). .....
    – tofro
    Commented Oct 8, 2020 at 8:56
  • 5
    BTW: There's never been a "change in voltage" of RS-232. Devices that use LVTTL for serial transmissions are simply not RS-232.
    – tofro
    Commented Oct 8, 2020 at 9:40
  • 1
    @tofro: The RS-232 standard provides specifications for the voltage which a conforming device must output onto a cable, the voltages which a conforming device must recognize as valid when received on a cable, and the allowable amount of noise that may be introduced by a conforming cable. If one regards a transmitter and cable as a unit, it would be possible for a device that outputs +/- 5 volts over a low-noise cable to behave as a conforming combination of a transmitter and cable even if the transmitter, individually, would not be conforming. This isn't the same as the polarity-reversed...
    – supercat
    Commented Feb 25, 2021 at 20:30

1 Answer 1

8

What was the very first device to have an RS-232 serial port

The question is a bit misleading, especially when tying it to 1960.

RS-232 is a standard. Standards are usually made after something has been invented and used first. Their purpose is usually to either unify a varying landscape or turning de facto usage into standard (*1). Think of bolt dimension, tube sizes or screw driver heads.

RS-232 is such a standard codifying and unifying existing industry usage and terminology.

As such there is no real 'first' that can be pointed out without doubt. It's kind like a quantum problem. It's like the cat in the box. prior to opening the box, noone knows if the cat is alive or dead. Or in this case, there can't be no device adhering to RS-232 before it got recommended, but as soon as it was published, a lot of existing devices were compliant and instantly became RS-232 devices, or not ... Getting a common classification is the whole purpose of such a standard

--

Serial communication using asynchronous is way older. The beginning of what after several steps became RS-232 was Baudot's "Système de télégraphie rapide" which in many ways did forshade upcoming information age - for example by being the first to use a binary representation for encoding and transmission (*2). Like telegraph before it used a current loop for the signals.

Development evolved over the next 100+ years. While Baudot's first machine was semi synchronous (the operator had to type the next character every time the keyboard got release from the previous character, later became truly asynchronous.

While original current was around 60 mA, it got lowered to 20 mA when the 7 bit teletypes got introduced. All still asynchronous over a single line.

With emerging switching and amplifying voltage driven communication became a thing. Various companies developed a great range of interfaces/devices, using different voltages and additional signals.

EIA generated RS232 to unify these into using the same name, workings and meanings.

That article also says the very first devices were "electromechanical typewriters" - which makes sense. The famous ASR-33 was released in 1963, but there were models beforehand. The Bell 103 couldn't have then been the first RS-232 device. So what was the first?

Hard to say, as RS-232 covers a wide range of devices already in existence in 1960. There is no minimum requirement for being either duplex or using any of the control signals. A single pair carrying a word synchronous transmission guarded by a start and a stop bit will be RS232 compliant. So essentially a (Baudot) TTY interface changing current loop to voltage level would fit - and they have been around from the 1870s - for sure including the early days of computing.


*1 - Not to be confused with (interface) specifications made up by companies/inventors/committees to define something new to be implemented later - these may as well become a standard at some point.

*2 - All the way to time multiplexed transmission.

12
  • 3
    Why is this being voted down Commented Sep 5, 2020 at 7:55
  • 5
    Because it’s a long, meandering post that doesn’t actually answer the question? It might as well point to the Big Bang as the origin of the RS-232 standard. Commented Sep 5, 2020 at 8:17
  • 5
    @user3840170 Good point, so lets use 'Big Bang as generic answer and close everything. After all, RS232 wasn't anything specific new, but just the codification of what happened before for the purpose of industry wide standardisation to make devices interchangeable. Asking for the first device and standard compilance in this context is shrouded in a kind of quantum state. One can only measure either, device usage or standard definition. For historical purpose guess you really want to dig a bit into Teletypes first and Baudot's inventions.
    – Raffzahn
    Commented Sep 5, 2020 at 8:24
  • 4
    I think, though, that we can treat the question as asking "what was the first device to use the communication protocol that subsequently became standardized as RS-232?".
    – dave
    Commented Sep 5, 2020 at 15:37
  • 1
    @another-dave True, but at what level? After all, essentially everything in RS232 past the basic serial wire is optional. And using an asynchronous protocol, using word synchron transmission guarded by a start and stop bit over a single pair is already what a Baudot teleprinter did in the 1870s.
    – Raffzahn
    Commented Sep 7, 2020 at 22:07

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .