David, at 14:13 on Jan 27 1996, you wrote to Bill Grimsley...

PE> David Begley's Netcomm - all connections at 26400.

BG> That's the initial connect.  What about doing an ATI6 & ATI11 afterwards?

db> Unfortunately for me, ATI6 just dutifully reports, "OK" -
on-line help 
db> doesn't know what it is.  ATI11 reports "ERROR".  In my
case I think the 
db> closest is ATI10 which reports all manner of statistics (not as specific as 
db> yours I see, but more detailed "global usage" stats), most
of which I think 
db> are wrong.

Ah yes, NetComm always seem to do things quite differently.  Even when
compared with other Rockwell-based modems.  I'm not sure that NetComms are
even capable of reporting last call stats anyway.

db> BTEE doesn't support "AfterCall" either so I can't try
that .. hang on, 

A pity.  Many of the later Binkley hacks do though.  Even Paul's.

db> lemme check something .. okay, I've just called Redmond, Washington USA 
db> (ie., Vince Perriello's system) .. twice no less (fucking USR HST shits - 
db> how on earth do you explain a 14,400bps *V.FC* connect with a modem that 
db> doesn't do V.FC?!). 

An incorrect nodelist flag, perhaps?  If the connect string reports V.FC,
I'd be rather inclined to believe it.

db> "The released version of BinkleyTerm (version 2.50)..."

db> Is ancient.  And the Europeans who did the EE version won't release the 
db> source code.  Bastards.  BinkleyTerm 2.59a is still the latest - wide beta 
db> for DOS and OS/2, alpha for Windows NT.

Does it really matter?  I've tried many different hacks here, and all of
them have done the job (albeit some with missing features).

db> Vince doesn't have a "files" listing.  Prick.  Works at Microsoft 
db> apparently.  Or did.

That explains a lot then.  Read Michael Crichton's "Disclosure"
if you get the chance.  Slags off at MS something terrible, he does.  :)

db> To be honest I think Bink is dead.  Seriously, like Msgedsq was.  Vince has 
db> no intention of ever releasing another version.  Maybe he did, one day - 
db> but not anymore.

So what are we supposed to use in its place then?  No matter, the version I
use does the job quite adequately, so I doubt it matters too much anyway.

BG> I'd be very surprised if the modems hadn't shifted up to 28800.

db> I'd be inclined to agree with this:

It happens quite frequently, and I find myself wondering why they don't
just connect at full speed to start with.

db> # 27 Jan 13:45:57 BINK Connect 26400/Lap-M/V.42Bis/V.34
db> [...]
db> + 27 Jan 13:46:05 BINK CPS: 3351 (10053 bytes)  Efficiency: 126%
db> [...]
db> + 27 Jan 13:46:17 BINK CPS: 3080 (33889 bytes)  Efficiency: 116%

db> 10Kb and 33Kb aren't exactly huge files, but they're "large
enough" not to 
db> be sent in a packet or two (thereby screwing up the stats), and they were 
db> already compressed (ZIP) too, so it's not the V.42bis weaving its magic.  

Dunno, I'd not trust the stats for any file under say 100Kb, although it
does appear that those files went through at 28800.  The Efficiency stats
are certainly based on a 28800 connect, at least.

db> Only a 38,400bps serial port.

Gawd, somebody else who likes crippling BTLZ compression...

Regards, Bill

--- Msgedsq/2 3.20