PE> Yes, the Spirit is a contributing factor, but it is not
PE> the ONLY modem, GUARANTEED to show the failed-EC problem,
PE> regardless of your claims of 4-years of 100% perfect
PE> connects as supposedly providing the "proof" for that.

BG> From my POV, your Spirit was indeed the only V.32bis
BG> modem which EVER failed to negotiate an EC connect,
BG> regardless of the brand of modem in use at my end.

RS> Proves sweet fuck all about anything at all Bill. You never
RS> even called another non USR at V32bis speeds at anything like
RS> the same frequency to even see if say a Netcon V32bis max
RS> modem would be as bad on a line of the same quality as Pauls.

BG> You forget that I was doing email via Marwick
BG> for a couple of years before becoming a point.

RS> Nope.

BG> I think that it is indeed quite significant
BG> that NONE of those calls ever failed though.

I'm not convinced. None of Brentons or Bobs did with the pair of modems
that fanged your arse. Proves sweet fuck all about anything much except
that your own line may well be worse than either of theirs and thats
what makes the Sportster V32bis/Spirit handshake failure manifest itself.

BG> I'm more than happy to agree that it PROVES nothing at all
BG> scientifically, but the data from which my conclusions are
BG> drawn is extremely limited to say the least.  And from our
BG> PsOV, there's not really a hell of a lot we can do about that.

Corse you can, you can put your brain into gear in the situation
where you dont have the most rigorous evidence available, and
STILL form useful conclusions when the evidence is not so good.
Its still quite possible, tho certainly harder.

BG> He had NetComms and Telebits, and not once
BG> did I see a single failure to negotiate EC.

RS> Proves sweet fuck all about anything much.

BG> Sure, but it still allows me to SURMISE that the problem with failed
BG> EC connects to the Spirit is down to less-than-perfect ROM controller
BG> code, in conjunction with lines which could be considerably better. 

Fraid not, you aint entitled to surmise that at all, and as you
can see from the result when you ended up face down in the mud
when you proclaimed that replacing the Spirit would see it gone
forever, its extremely dangerous to make that claim based on
what is inadequate evidence to substantiate that claim.

BG> Nobody can really PROVE anything, given the limited data available.

Even thats not true. The evidence that neither Brenton or Bob ever
saw it, means there had to be something different about YOUR line
which was producing those handshaking failures when the modem pairs
in use was identical, and even more so when it was seen with more
than one modem at your end to a Spirit, eliminating the possibility
that your particular modem was just defective, tho useable.

And we NOW know that it cant possibly be just dud code in
the Spirit, coz you STILL see the effect even without a Spirit.
PROOF on the question of it being dud code only in a Spirit.

BG> Ditto with David's V.32b USR and NetComm M5 modems.

RS> Ditto, proves sweet fuck all about anything much.

BG> Except that the S2 was severely lacking some decent ROM code, and yes,
BG> I do realise that that still doesn't really PROVE anything at all.

Proves sweet fuck all about anything much at all basically.

BG> Not a single solitary problem.  Not until I called Paul's Spirit.

RS> And Brenton and Bob dont either, so that
RS> proves sweet fuck all about anything much.

BG> Just like the double-send problem, there is still a constant, the Spirit.

And is still seen even with no Spirit. Oops.

RS> Just like your claims about what you say with modem testing at CHH
RS> doesnt either, coz you clearly have a real problem calling it NOW.

BG> Dunno, I'll try again with the Sportie's new EPROM later today,
BG> and post the stats.  So far, it APPEARS to be a bit more solid at the
BG> higher link rate negotiations, but it's way too soon to tell just yet.

Yeah, its clearly quite variable.
@EOT:

---