Ian, at 06:46 on Sep 09 1996, you wrote to Bill Grimsley...

BG> Not so.  In fact, USRs (and presumably their clones) actually connect low, 
BG> but will shift upwards almost immediately if the line will support it.

IS> Hmmm, that doesn't get around the problem exhibited by some modems (notably 
IS> around here a large rack of Netcomm M34s) in timing out too fast in V.42 
IS> negotiation (thus obtaining non-EC connects, which I seem to remember was 
IS> Michael's problem, no?) because they accept a higher bit error rate as 
IS> worthy of granting (say) a 28800 connect than the calling modem does.

It was my problem too, when calling Paul Edwards' M34F, and to a slightly
lesser extent, his Viper (both Rockwell-based), although in both cases, the
remedy was extremely simple; disabling the Courier's V.42 detect phase
(i.e. forcing a non-negotiated LAPM connect).  One of the fellows who
writes the Courier's code for USR (Lewin's mate, Joe Frankiewicz :)) has
indicated that it's actually a Rockwell problem, but as it's easily
rectified, who cares?

IS> The solution for both my DPX596 and the Motorolas was to set them so that 
IS> they'd go for a lower speed (say one or two 2400bps steps) for a given line 
IS> quality, thus quickly negotiating a lower speed, but getting error 
IS> correction with these modems.  Apart from the link reliability, the overall 
IS> throughput is higher at say 26400, probably even 24000, with EC than at 
IS> 28800 without.

Are your modems incapable of "forcing" non-negotiated LAPM connects, then?

IS> It's possible that the Netcomms have a setting that will either be less 
IS> aggressive about it, or (more likely, being Rockwells) perhaps allow longer 
IS> for EC negotiation - but I've not been able to find out if that's so, as 
IS> yet.

The only thing which comes to mind is the Tx level, which I understand can
be physically adjusted to a specific (higher) level, and I wouldn't be at
all happy using that as a fix for that particular fault anyway.

MR> Does anyone know what S register setting will make a Dynalink train less
MR> aggressively.

BG> There isn't one.  All you can do is try disabling the higher symbol rates 
BG> (say 3429 and 3200, for example) for a lower connect speed.

IS> That surprises me, on a modem of that supposed quality.  Ok, the Rockwells 
IS> don't appear to have any user-accessible BER adjustments either, but still 
IS> ..

Jeeze, Ian, we live in a modern technological society, where everything
these days is done automatically, yet you appear to want to return to the
bad old days of manually setting things?  Bloody hell, is nothing sacred? 
:)

Realistically though, isn't that the whole point of the newer fancy
protocols, such as V.34, where automation is reaching dizzying new heights
(or would have done, had the Dynalink's BTLZ worked properly) ?

IS> Perhaps Lewin can help with this, regarding how the Rockwells can be tuned 

Cross his palm with a MicroSoft product, and I'm sure he'll help.  :)

IS> .. it's my theory (totally by smell) that Netcomm and maybe some other 
IS> modem chip packagers are not testing their modems nearly adequately on the 
IS> less than perfect lines that many of us enjoy, due to distance from 
IS> exchanges, water in the works, bad line connections, etc, with enough other 
IS> makes of modem, thus are setting their BER / line rate selection policy 
IS> toward only better lines.

I suspect the above to be much closer to fact than a theory, too.

IS> If true, this is counter-productive. 

So what else is new, Ian?  Now that V.34 modems can be purchased brand new
in the United States for not much more than US$100 (and around A$150 here),
the manufacturing cost per unit is now so low that it's probably cheaper
for them to ignore production line QA altogether.

IS> Going sooner to a lower rate will achieve higher connectivity with more 
IS> modems on poor lines, without in the least sacrificing connect rates on the 
IS> best lines.  Again, 24000/Arq is a hell of a lot better than 28800/None, 
IS> let alone dropped connects.

Quite so, although it's interesting to recall that when the early V.FC
modems were a bit aggressive during negotiation, everybody complained about
their dropped carriers, yet now that most V.34 modems have gone the other
way (like the USRs), they're all bitching about their lowly 24000 or 26400
connects.  There's just no pleasing some people, is there?

IS> 

Feel better now?  No, I thought not...

Regards, Bill
@EOT:

--- MsgedSQ/2 3.50