OT: just a suggestion

[Joseph Gwinn]

On Dec 26, 2015, krw wrote (in article):

Here is the missing bottom of my reply:

All these modems take an incoming bit stream and convert it to a series of line symbols on the wire (or fiber or radio wave). These line symbols are designed to be easy to tell apart, even after having been distorted and corrupted a bit on the line, so the receiving modem can tell with very high reliability which line symbol was sent, thus recovering the original bit stream.

Because only a known set of line symbols are sent, and nothing in between, the receiver can assume that the intended line symbol is that one that is closest to one of these known line symbols.

Design of such symbol sets is a career, and there is a large literature, but a good place to start is any college textbook on data communications.

Joe Gwinn

[Leon]

Actually in the late 70's early 80's I ordered parts from GM/GMPD using a thermal heat print terminal/computer terminal. With a keyboard and a dual cassette deck on top I would communicate with our inventory control service provider, Reynolds and Reynolds. I would run a routine from that terminal to generate a stock order, get the results, and record them on a cassette, they looked like the common cassette. From there I flipped a series of toggle switches and typed in a phone number to GMPD's computer and their computer communicated with my terminal and read the cassette. The response of their computer was sent back and recorded on the other cassette on my terminal. And finally my terminal read the cassette and answer from GMPD and printed the results on that terminals thermal heat print paper.

[Markem]

The V.90/92 modems used phase shift key, detecting a phase shift in an analog carrier. Only offering a 0 or a 1 whether the carrier is phase shifted or not. DSL is also PSK.

[-MIKE-]

Thanks for that info, John. It doesn't happen all the time and your explanation tells me why. However, it happens way too often to tolerate. I suppose one day the technology will improve and the issue will disappear. Until then, however, I'll stick to my "land line" for important calls when I'm at home.

[Larry Blanchard]

I started playing with microcomputers with 2 8" floppies - then 5.25" - then 3.5". Each time either the capacity or the speed went down. I knew they couldn't last :-).

[Spalted Walt]

The 1541 was too rich for my blood back then, I was stuck with the slower, less expensive, but very reliable 1530 Datasette!

I still remember spending many hours typing in page upon page of raw machine language from 'Compute!'s Gazette' magazine and saving it to the 1530.

's_Gazette#/media/File:ComputesGazetteProgramPage.jpg

[clare]

I've got ups on the cable modem, the router, the VOIP appliance, the cordless phone base and bothDTV boxes - as well as both computers.

[clare]

I've got 4 in use. One on my wife's computer, one on mine, one on the network /phone devices, and one on each DTV box.

Then I've got another 2 spares sitting in a cabinet.

[clare]

The problem is he needs a REAL UPS - not some crappy unit like an APC Back-UPS. Use a Smart-Ups from the same company, or a real UPS like a Powerware Elite (dual conversion) and the loads don't even know the UPS is there.

Absolute minimum protection is a "line interactive" unit that has buck/boost votage regulation if you are not going to spring for dual conversion.

[clare]

V 90 was strictly analog. Any "modem" is analog because it modulates and demodulates (MODulatorDEModulator) the signal at each end. Modulation is the hallmark of an alalog signal.(modulation also described, as I did in my "simplified" reply as varying frequency and intensity.

A digital signal does not use a "modem" It uses a "codec" (coder decoder) which converts a binary chain of data to "tokens" and transmits those "tokens" digitally across the network, and then at the other end "decodes" the "tokens" to a digital binary data stream again. All computers today are "digital devices", working with and understanding ONLY binary digital data.

JPEG, MP3, MP4, GIF, TIFF etc are all compression algorithms that define what tokens are used and how to code and decode the binary data for transmission.

The errors and retries in digital communications are the result of the "error correcting" or "error checking" built into the communication protocol - if a token arrives damaged, it is rejected and a resend is requested.

In analog communication there is no way to error check and error correct, so instead of getting data interruption you get data corruption - instead of getting pixellation and "digital dropout" you get "snow" "hiss" and other forms of "corruption" in the signal.

In Digital communications you get latency and dropouts.

[clare]

Some casette data can be digital. The cassette tapes used on early computer systems like the RadioShack COCO (Color Computer) were digital data stored on an analog medium. There were only 2 tones recorded on the tape. If the frequency was between 2 levels it was a

1, and between two other totally different frequencies, it was a 0. Just like RS232, 2 voltage ranges -3 to -25 and +3 to +25 - anything between -3 and +3 was "noise"

A high quality tape duplicator could duplicate the digital tapes reliably, but trying to do it with a "dubbing" deck was pretty much destined to fail. When we had the "coco club" we would get programs published in "hot coco" magazine, and one member would key the program into his computer and save it to a cassette tape, then I would run off the tape on the duplicator so each member got a copy. I used a TELEX Copyette 4 head duplicator that we also used to make copies of sermon tapes at church for shut-ins who could not get to church.

(now it's all recorded digitally and put on the church website)

There was no ambiguity. So it was (binary) digital data. Any music or voice or other non-binary data on a cassette tape is analog.

Then of course, there is DAT (Digital Audio Tape) but that's a different subject.

[clare]

There were lots of tape backup units in the early hard drive years that used (basically) an audio cassette type tape for digital data backup. Everything up untill the advent of DAT was basically an analog (audio) tape system using modem technology to put the digital data on the tape using analog technology.

When the moved to helical scan technology the lines started to blurr a bit more and the data became more digital in it's actual storage - similar to magneto-resistive recording on "magneto optical drives" and die sublimation on optical drives. - which were getting awfully close to the digital technology used for CD ROM and DVD in that there was no "modulation" used any more.

[krw]

The alternative is for others to pay for cell coverage for your area, which doesn't have enough customers to pay the freight. It's part of the cost of living where others don't.

[clare]

It's been tried. It kinda half works - but 99%accuracy in digital communication without error correction is virtually useless - and the best results were well jnder 99%. (If I remember correctly we achieved better than 80% -)so we went to "plan b" with the Telex tape duplicator and "sneakernet" to distribute the programs.

[clare]

The Reynolds and Reynolds and ADP systems used modems to send and recieve the data - which implemented error checking and correction.

A lot of those early modems used "accoustic couplers" - you placed the handset of the phone onto the coupler and it "talked" to the phone to transmit the modulated signals. I think that worked all the way up to about 200 baud. - mabee 300. That was about the limit for data transmission on the audio tape too.

[krw]

Sure, DirectTV's bird is at about 47-degrees, so it's fairly high. As I said, there is nothing around.

Fortunately, the networks put their recent shows online, as well. We've been using them to catch up (and stay current) with the outages we've been having. The gym we go to has Netflix. Seems to have some pretty good movies, though they tend to play the same ones over and over (and over...).

[krw]

Our computers are laptops, so no UPS needed. I rarely use even my laptop, anymore. I have a tablet that I do all things Internet on. It's a lot easier to carry around than even a laptop (remember Compaq luggables?!).

[krw]

Good advice. UPSs tend to be pretty dirty power-wise, too, though I can't see why a DVR would care. A UPS *shouldn't* allow brown-outs, either, but I guess you get what you pay for - sometimes.

[clare]

Actually both speed and capacity went UP. The first 5.25 disks I used were somethink like 80kb Shugart hard sectored, then 360kb soft sectored, then up to DSDD 1.2mb. The 3.5" disks started at 280 (which never went anywhere) then 360 single sided, 720 double sided, and 1.44mb DSDD (or HD).

Then came the Flopticals, LS120 and LS240 3.5 disks with 21, 120, and

240MB capacity.and a few other nonstandard (Proprietary) systems that were ineffective to varying degrees.

[clare]

It is if you are transmitting binary code. Your baud rate would be abyssimal though!!!!