In the late 1970's, there was a technology in development called "ISDN",  https://en.wikipedia.org/wiki/Integrated_Services_Digital_Network  
(Although, you wouldn't know this from this WIkipedia article, which merely makes reference to a 1988 standard.)    During the 1970's, I was reading the magazines Electronics, and EDN (Electronics Design News), and others, and there were articles and ads for ISDN chips.  Keep in mind that during the 1970's, commonly-available modems worked at speeds of 300 bits/second and 1200 bits/second.  If ISDN had actually been delivered promptly, it would have leapfrogged the 9600, 14.4kbps, 28.8 kbps, and higher speeds. 
 
However, ISDN was probably developed on the Phone Companys' calendar, very slowly, while the rest of electronics (including modems) were operating on a much-faster time frame.
×

The ISDN idea was that a telephone line would have quantity 2, 64 kilobits/second channels, as well as a 16 kilobit/second channel.  

However, one additional good reason that ISDN didn't succeed was simple:  The phone companies would certainly have charged a high per-month fee for ISDN lin es.  (And such a service would have had a box, analogous to a 'modem', into which your computer would have connected.)  The eventual alternative, buying the modems which eventually appeared on the market, was comparatively free.  I think there was a time in the late 1970's when phone companies expressed resentment that their users were employing modems on their phone lines.  

                    Jim Bell


On Wednesday, December 26, 2018, 2:13:41 AM PST, grarpamp <grarpamp@gmail.com> wrote:


[Now using proper Subject tech...]

If you have a line, you can still dial each other
and negotiate up to 33.8kbps v.34bis,
add better software compression (zstd) instead
of depending on v.44, and add encryption algos
on each end. v.92 56k needed an ISP end to work.

Companies like US Robotics and Zoom might still
make v.34bis hardware modems... see USR5637.
Lots of modems on used market.

Full hardware modem with PCM DSP is needed
to do elite first pass random phone scanning that
analyzes the analog instead of depending on
successful second stage "V." negotiation.
Plus you get as bonus all the WAV recordings of:
"Hello... Helloooooo?! WTF!!!" ;-)

Anyone still have that analysis software?


https://en.wikipedia.org/wiki/List_of_ITU-T_V-series_recommendations

You could probably do just as well creating your
own modems with today's DIY hardware platforms
in your local Makerspace. Opensource it on Github.


On 12/23/18, jim bell <jdb10987@yahoo.com> wrote:
> You forgot that in 1992, typical dialup modems worked at 9600 bps.  Now,
> most people have access to 25 megabits/sec Internet.
> I occasionally see people in discussion areas claim that "the U.S.
> Government" was responsible for making "The Internet".I shut that talk down,
> by pointing out "Do you think that The Internet would have 'worked' if a
> person, at home, had to connect up to his ISP at with a 300 bps modem?  1200
> bps?  2400 bps?"I counter by pointing out that the people REALLY responsible
> for a usable Internet were those who developed the 9600 bps, 14,400 bps, and
> 28,800 bps modems.  Rockwell, USR (US Robotics), Hayes, Telebit, and a few
> others.  Had that not existed, it would have been hard to make the Internet
> available to most people.
> From:    https://en.wikipedia.org/wiki/Modem
> "V.32 modems operating at 9600 bit/s were expensive and were only starting
> to enter the market in the early 1990s when V.32bis was
> standardized. Rockwell International's chip division developed a new driver
> chip set incorporating the standard and aggressively priced it. Supra,
> Inc. arranged a short-term exclusivity arrangement with Rockwell, and
> developed the SupraFAXModem 14400based on it. Introduced in January 1992 at
> $399 (or less), it was half the price of the slower V.32 modems already on
> the market. This led to a price war, and by the end of the year V.32 was
> dead, never having been really established, and V.32bis modems were widely
> available for $250.V.32bis was so successful that the older high-speed
> standards had little to recommend them. USR fought back with a 16,800 bit/s
> version of HST, while AT&T introduced a one-off 19,200 bit/s method they
> referred to as V.32ter, but neither non-standard modem sold well."
>
> And:    https://en.wikipedia.org/wiki/Modem
>
> V.34/28.8 kbit/s and 33.6 kbit/s
>
> ×
>
> ×
>
>
> "Any interest in these proprietary improvements was destroyed during the
> lengthy introduction of the 28,800 bit/s V.34 standard. While waiting,
> several companies decided to release hardware and introduced modems they
> referred to as V.FAST. In order to guarantee compatibility with V.34 modems
> once the standard was ratified (1994), the manufacturers were forced to use
> more flexible parts, generally a DSP and microcontroller, as opposed to
> purpose-designed ASIC modem chips.
> "The ITU standard V.34 represents the culmination of the joint efforts. It
> employs the most powerful coding techniques including channel encoding and
> shape encoding. From the mere four bits per symbol (9.6 kbit/s), the new
> standards used the functional equivalent of 6 to 10 bits per symbol, plus
> increasing baud rates from 2,400 to 3,429, to create 14.4, 28.8, and
> 33.6 kbit/s modems. This rate is near the theoretical Shannon limit. When
> calculated, the Shannon capacity of a narrowband line is {\displaystyle
> {\text{bandwidth}}\times \log _{2}(1+P_{u}/P_{n})}, with {\displaystyle
> P_{u}/P_{n}} the (linear) signal-to-noise ratio. Narrowband phone lines have
> a bandwidth of 3,000 Hz so using {\displaystyle P_{u}/P_{n}=1000} (SNR =
> 30 dB), the capacity is approximately 30 kbit/s.[7]