60882

神经网络昆斯林的新闻组分类2006

Path: cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!fs7.ece.cmu.edu!europa.eng.gtefsd.com!howland.reston.ans.net!usc!sdd.hp.com!nigel.msen.com!fmsrl7!glang
From: [e-mail] (Gordon Lang)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: Laplink serial & parallel cables
Date: 22 Apr 1993 03:46:42 GMT
Organization: Ford Motor Company Research Laboratory
Lines: 24
Distribution: world
Message-ID: <[e-mail]>
References: <[e-mail]>
NNTP-Posting-Host: slee01.srl.ford.com
X-Newsreader: Tin 1.1 PL5

Serial cables:

There are only three output signals: TD, DTR, and RTS.
There are five inputs signals: RD, DSR, DCD, CTS, and RI.

There are many differnt null modem requiremts as dictated by the software.
The IBM BIOS requires asserts DTR and RTS and then waits for both DSR and
CTS before sending.  When a BIOS recieve call is made, it asserts DTR and
waits for DSR to become true (times out after a while and returns with an
error if DSR never becomes true).  It requires CTS to be true for sending
AND recieving.  Most communications packages bypass this and replace it
with their own protocol.  The key is that each comm. package could very
well have different requirements.

My favorite cable works in many cases: short RTS to CTS at each end, but
also run RTS through to DCD at the opposite end.  TD runs through to RD
and DTR runs through to DSR from both DTE's and of course SG goes through
to SG.

I have never had trouble with this null modem even though I have used it
with a comm. package that was expecting RTS to go to CTS instead of DCD.
The advantage of this cable is that it also works with the IBM BIOS.

Gordon Lang