Brenton, at 13:09 on May 19 1996, you wrote to Bill Grimsley...

BG> So much for his main objective, runninng a fast and reliable board.  With
BG> the modem circus now entering its 4th or 5th month, I have to say that
BG> Paul's is the most UNRELIABLE board I've called during that period, by far.

BV> It has become, since he started fucking with it.

It's not just the modems either.  Now his OS is fucked as well.

BV> I'm sure I can remember Paul saying that he was pissed off with Sysops who 
BV> were always screwing with things and breaking them.

Quite likely, although I'm far too polite to have said so myself.  :)

BG> Pretty easy to see now why some sane people choose to pay a premium for
BG> proven BBS-reliable modems, like the $1100 Microcom Deskporte Fast+ (which
BG> is being rebadged by Banksia as its top model), or the $799 USR Courier.

BV> That might be a bit rich nowadays. 

I agree, $1100 these days is outrageous.  And Rod reckons that USR gouge! 

BV> In all honesty I can't see why he didn't look at the Banksia MyFastModem 
BV> V.34, which is a bargain and works.

Possibly because the earlier V.32bis Banksias were notoriously buggy, and
the reputation stuck (not unlike Interlink, in fact).  Banksia also market
an RPI modem, which is not a particularly clever move IMO.

BV> Okay, it uses a Rockwell chipset, but anything local does and this thing 
BV> actually works.

It'd be interesting to see how well it works in a BBS situation, but I
guess it's too much to expect that you'd lend it to Paul for a week or two.
 :)

BV> FWIW, I reckon the only *real* problem with Rockwell modems is that the 
BV> idiots who design stuff with them don't know what they are doing. They take 
BV> the *sample* code provided by Rockwell and use it for production product. 
BV> That's why none of the stuff works. It's all being designed (thrown 
BV> together) by amateurs and people with about as much knowledge as Paul.

That explains the yum-cha nasties, but what about NetComm and Interlink,
both of whom have been making modems for so long now that they should be
more than capable of writing decent controller code?

I've been discussing my LAPM failures to Paul with Joe in the USR echo (he
actually writes the Courier code), and when I asked him who was to blame
for these failures, he stated that they were Rockwell chipset related, but
easily correctable in firmware.  Rod thinks he's a lying wanker (so what
else is new?), but I'd suggest that his only real failing is that he tends
not to publicly slag off at other manufacturers' chipsets.  :) 

Regards, Bill

--- Msgedsq/2 3.20