PE> If the results of the M34F are consistent, it could well be a good tool.

RS> Presumably you mean if the M34F figures dont change
RS> depending on what modem is on the other end of the line.

PE> That's right.

RS> Yes, that approach has some theoretical validity, BUT we ALREADY know that
RS> the Rx level the M34F claims to have seen with one particular AT command
RS> is dud coz it says 50dbm all the time. Clearly the M34F has problems too.

PE> And we can bin the 50dbm result too.  We can recognize it and bin it.

Yes, but YOU want to just bin ALL the USR data, even when there is
another USR on the other end of the line. For some reason you only
bin SOME of the M34F data, the data with the obvious problem.

ALL I am saying is that when its not practical to actually use test
equipment, and when the USR/USR graphs show a very good correlation
between what we know of the lines and the graph, the best lines being
those virtually next door to exchanges, Poes move further away from
the exchange producing just the effect on the graph you would expect,
increasing loss at the higher frequencys, the fact that Bills line is
clearly quite a bit worse than most, tho not the worst ever seen, and
that correlates well with the cps thruput, particularly when he doesnt
call an immaculate line, the data is useful. Tho clearly when there isnt
a USR on the other end its rather less useful.

And the problem with the M34F is that the only measure of line
quality it provides, the EQM, varys with the carrier rate and symbol
rate, so its not a very useful measure of JUST the line at all.

The short story is that the USR/USR graphs are the best we have.
And they do make considerable sense, particularly when you look
at what happens with the graph when you have two nothing special
lines in the session compared with one immaculate etc.

One thing we really DO now know is that Bills line is about the worst
of you regular callers who have had their lines measured, and that yours
is quite a bit better than his. And that goes quite a long way to explaining
why only he say handshaking failures Sportster V32bis/Spirit and others with
the same Sportster did not. And Jacks line fits too, not exactly brilliant.

There is useful information to be obtained from USR/USR stats and
its mad to just bin it, and there aint nothing as useful from an M34F.
@EOT:

---