home *** CD-ROM | disk | FTP | other *** search
- Path: pathway1.pathcom.com!ts5l5
- From: insystem@pathcom.com (Geoffrey Welsh)
- Newsgroups: comp.dcom.modems
- Subject: Re: 28880 bps with a 386 ???
- Date: 18 Feb 1996 16:37:12 GMT
- Organization: InSystems Technologies Inc.
- Message-ID: <4g7kjo$74j@pathway1.pathcom.com>
- References: <eric-1002962356480001@sobt.accessorl.net> <4g0aad$6vu@news.NetVision.net.il>
- NNTP-Posting-Host: ts5l5.pathcom.com
- X-Newsreader: News Xpress Version 1.0 Beta #4
-
- In article <4g0aad$6vu@news.NetVision.net.il>,
- rsolomon@netvision.net.il (Ron Solomon) wrote:
- >In my personal experience, 28.8 doesn't mix well with 386 for a reason
- >overlooked by previous posters.
- >Most 386's still around, and there's probably enough of them, are pretty old
- >machines, at least by industry standards. Good chance they won't have a
- >16550 UART, if you're half-lucky they'll come with a 16450 ART, which is
- >alot slower. Worse yet, they mostly came with 8250 serial ports, with no
- >buffering; this means a maximum throughput of 19200bps...
-
- I'm terribly sorry to be so undiplomatic about it, but the text of this
- article indicates that the author is not familiar with some of the facts. I'm
- being borderline rude about this, because this is an example of incorrect
- information being spread, and it's long overdue that we killed some of these
- myths.
-
- (1) The 8250 and 16450 UART chips have exactly the same level of receive data
- buffering (i.e., 1 byte).
-
- (2) The use of an 8250 does _not_ limit throughput to 19200 bps; the cause of
- receive buffer overruns is not the UART itself but rather the interrupt
- response time of the CPU, which depends on the CPU, OS, installed hardware,
- system configuration, etc. There is no hard rule about the speed limit on the
- 8250 UART.
-
- Geoffrey Welsh, Developer, InSystems Technologies Inc.: insystem@pathcom.com
- At home: geoff@zswamp.uucp, [xenitec.on.ca|m2xenix.psg.com]!zswamp!geoff
-