PE> You're not seriously suggesting you happen to get a good line whenever
PE> I put on a Courier this end, by some miracle of God are you?

BG> Of course not, you goose.

BG> What I am suggesting is that a USR to USR will work better than
BG> a NetComm to USR, on the SAME poor line.  Do you dispute that?

RS> Yes, we dont KNOW its a poor line now that we KNOW the USR
RS> stats are meaningless unless there is a Courier on each end.

BG> In fact, we don't KNOW that at all.

Fraid we do Bill. Or more strictly USR rather than Courier.

BG> You have assumed this to be the case, but I'm no longer convinced.

I havent assumed a damned thing. Its rather unlikely that BOTH an
M34F and Viper are putting out the standard test used to measure
line characteristics a FULL 6db down from what its supposed to be.

In that case the numbers on that graph are meaningless, tho the
shape clearly doesnt change, so thats still useful information.

The only evidence you have that its a 'poor' line is the fact
that the USR/M34F and USR/Viper stats claim the loss is very high.
When its not with USR/Courier, in fact its not too bad at all, you
dont know you have a poor line. Tho you do know its not immaculate.

And if you did some tests with M34F/M34F and a decent connect
every time, thats more evidence that its not poor too.

RS> AND we DO know that WITH a Courier on each end, your line to Paul isnt too 

BG> To be pedantic, the reported levels are different (and better).

That graph is supposed to be of LINE loss, its meant to be independent
of the transmit level. Its clearly telling porkys when its not USR/USR.
Even the Rx level is distinctly dubious coz you can SET the M34F transmit
level and when you do, what your modem claims to receive is MUCH lower
than when you have a USR at the other end claiming to be using the same
transmit level as the M34F is supposed to be using.

Corse its possible that the M34F is telling lies, but since
the Viper tells the SAME lies, that proposition is dubious.

Its far more likely there is some interoperability wart with
the USR when its not a USR on the other end of the line.

BG> Damn, I ran off some stats from Jack's, then left the bloody disk at
BG> Pearl. However, levels to Paul were still quite shitty, and I think
BG> it was only a 26400 connect as well.  Got a 28800/26400 to Dave, with
BG> slightly better line levels reported, but still not as good as I'd hoped.

Yes, it does seem clear that both Jacks like and your home line
arent anything special. And likely actually worse than Pauls.

PE> What am I meant to be proving, exactly?

BG> Dunno now, I've forgotten.  :)

RS> Things are getting just a tad complicated |-)

BG> If you think it was bad before, imagine how confusing things
BG> are after a 5-day break from email.  What a complete bastard!

Yeah, its a problem keeping the detail clear in your head.
@EOT:

---