RB>>> Express? > RPI, no?
 IS>> Fair comment.  You might expand on the problems with RPI a bit?
 HM> > This was actually Hamish's comment let him explain it

 HM> (Sorry, didn't receive your message here Ian.)

Neither did anyone else - your feed is probably fine.

 HM> Well, I've never used one, but the modems are made simpler
 HM> (with no built in error correction or compression). Then
 HM> you use software to perform these more advanced functions.
 HM> The idea is to save a few bucks on manufacturing the thing.
 HM> Of course, you can only use software that supports the RPI
 HM> software-based features. I *think* there's a Windows driver
 HM> which does RPI for all Windows-based software, but for DOS,
 HM> deltaComm, Mustang etc refuse to support it AFAIK.

Thanks for the expansion.  Urgh.

 HM> Also even using MNP 4 you'd (AFAIK) still be using an async
 HM> connection, while with modem based MNP4/V.42 it's 8 bit
 HM> sync. With MNP 4 in software you still have to send the
 HM> start/stop to the modem which will still transmit them.

Yes.  Clayton's error correction - you may as well leave error correction
to the file transfer protocol, which probably adds no more overhead than
LAPM or MNP anyway, given you've lost the 20% advantage synch mode would
give ..

 HM> The money saved isn't worth the headaches to come.

That's probably the idea :(  You know, I bet half the people who buy one of
these things don't even know they're not getting the full quid .. it's all
very well to say 'buyer beware', but whatever they call the TPC these days
should be demanding warning labels be put on the boxes ..

Cheers, Ian

--- MaltEd 1.0.b5