Xref: utzoo comp.dcom.modems:8243 comp.unix.sysv386:4712
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!sdd.hp.com!spool.mu.edu!uunet!wsrcc.com!wolfgang
From: wolfgang@wsrcc.com (Wolfgang S. Rupprecht)
Newsgroups: comp.dcom.modems,comp.unix.sysv386
Subject: Re: Beware the Telebit T18pc
Message-ID: <1991Feb9.022024.10932@wsrcc.com>
Date: 9 Feb 91 02:20:24 GMT
References: <1991Feb7.225401.28138@ingres.Ingres.COM>
Organization: Wolfgang S Rupprecht Computer Consulting, Washington DC.
Lines: 24

>In case you were considering the t18pc, think about this.  I bought one and
>discovered after much wasted efort that the (soldered-on) 8250 UART was
>keeping my throughput at around 70 cps.  The only solution was to replace
>it with another UART (I chose the 16550, of course).  

This brings up a question that has been bugging me about the Telebit
PC bus modems: Why didn't Telebit take this opportunity to design the
UARTS out of the modem?  They could have replaced the two back to back
UARTS with a latch pair, or better yet a FIFO pair and made one
extremely delay tolerant PC bus modem.  Without the serial chips, the
modem really wouldn't care *when* you got around to reading the
character.  The TB's cpu could see that you haven't read the last char
and wouldn't stuff another one on top of the first one. If you waited
too long the receiving TB would just fill up its internal ram buffer.
In PEP mode this could just throttle the sending TB.  If the sending
modem was also a similar PC bus modem, it would just hold off the next
interrupt until it's ram buffer emptied.  No bytes need be lost end to
end.

-wolfgang

-- 
Wolfgang Rupprecht    wolfgang@wsrcc.com (or) uunet!wsrcc!wolfgang
Snail Mail Address:   Box 6524, Alexandria, VA 22306-0524