Mike:

Yanno. Have you ever even seen a 56K USRobotics Courier Modem (the
very first ones were 36K and upgraded with a simple 56K flash--upgrade
from software to their internal static memory--when the upgrade
finally became available)

As, there was "controversy" back then if a 56K modem ran on phone
lines would cause interference, "cross talk", etc., it all turned out
to be a ridiculous argument--one akin to the one posed of hooking a
fast modem to a xmitter...

It sold for ~$350.00+ when new (about 1995-1999) and does ALL data
processing within itself (it has a 8086 intel processor onboard), this
includes compaction/de-compaction, error control, line-signal power
adjustments, feeding/pulling data, etc. It virtually "pumps" data to
the computer buffers and virtually "yanks" data from them (I think I
have heard the data screaming at times! grin)

It virtually is a standalone computer with but one function in life,
send and recv data. It is a black box about 6 inches wide, 12 inches
long and under 2 inches high. During the usefulness of 56K modems it
had no equal--that stands to this very day.

It virtually puts NO demand/load on the computers processor, and
insists on doing EVERYTHING itself...

.... one reason it is termed the "v.everything" by USR... it is
software upgradeable with 512,000 bytes of read-only flash memory. It
was a marvel of technology in the data transmission field, it really
still is...

A piece of hardware like that simplifies the project from the very
beginning... you might be lucky enough to find an old one on ebay...
the "sportster" models are NOT ONE-HALF the modem a courier is...

.... when it hits error free transmission of data in the "kludged use"
of it, the data throughput is actually just a hair (few bytes) less
than 57K. It can actually keep logs and display data throughput to
your screen all on its own...

John

"Mike Coslo" wrote in message
...
John Smith wrote:
Mike:

The clock in a ~4GHz computer and DDR memory makes modem data xfr
look incredibly s-l-o-w.... with spaces miles long between marker
bits...

100Mbs nic cards are not even close to a challenge to that clock
speed...


Um, John, just as a simple experiment, what do you get when you
modulate say a 14 MHz signal with that 4 GHz signal?

note: not that you would do that for a live video system, but are
you starting to see my point? What happens at the computer is not
the issue. It is what happens at the frequency we are trying to use.

Computer clock speed is not relevant to the the issue. It is how
much data an HF signal can handle.

There have been a lot of engineers, mathematicians and programmers
working on this problem. If you have a method of doing this, I
*strongly* suggest that you hire a patent attorney, and get to work.
You are gonna be rich, man!

I'm willing to help you with the initial experiments. In fact, in
the interest of the furtherance of Ham radio, science, and mankind,
I have challenged you to produce such a system.

- Mike KB3EIA -

- Mike KB3EIA -