home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!stanford.edu!bcm!lib!oac.hsc.uth.tmc.edu!jmaynard
- From: jmaynard@oac.hsc.uth.tmc.edu (Jay Maynard)
- Newsgroups: comp.os.os2.misc
- Subject: Re: How to increase speed of communications in OS/2 2.0
- Keywords: os2 waffle 165
- Message-ID: <8149@lib.tmc.edu>
- Date: 16 Dec 1992 18:59:37 GMT
- References: <sk6DVB3w165w@vicuna.ocunix.on.ca> <4ewqVB16w165w@vicuna.ocunix.on.ca> <brown.198.724388877@chinchilla.ir.ucf.edu>
- Sender: usenet@lib.tmc.edu
- Organization: UT Health Science Center Houston
- Lines: 21
- Nntp-Posting-Host: oac.hsc.uth.tmc.edu
-
- In article <brown.198.724388877@chinchilla.ir.ucf.edu> brown@chinchilla.ir.ucf.edu (Bill Brown) writes:
- >This is probably your problem. Although the 16550AN has a FIFO buffer (a 16
- >byte queue to prevent character overrun under high speed serial rates), it's
- >completely broke. The 16550AFN is the fixed version of this chip.
-
- Sorry. The 16550 (non-A, non-AF) was the broken one. (The N is National
- Semiconductor's package designator for a plastic DIP.) The 16550A fixed the
- FIFO so that it works. I'm using two of them continuously under heavy loading,
- and the FIFOs work fine. The AF fixed an obscure bug, but the info I have is
- that there's no practical difference in normal use.
-
- While I'm here, it's time to reiterate that the 16550A/AF is a pin-for-pin
- replacement for the 8250 and 16450. Occasionally, folks will hear that there
- are special support chips required for the 16550. Hogwash. Those support chips
- turn out to be 1488/1489 RS-232 line receiver/drivers, and are completely
- compatible with the 16550. Thousands of folks have replaced their 16450s and
- 8250s with 16550s and had no problems at all.
- --
- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can
- jmaynard@oac.hsc.uth.tmc.edu | adequately be explained by stupidity.
- "Brought to you by the letters O, S, and by the number 2." -- Mike Levis
-