home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!mips!swrinde!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!network.ucsd.edu!qualcom.qualcomm.com!cancun!rdippold
- From: rdippold@cancun.qualcomm.com (Ron Dippold)
- Newsgroups: comp.dcom.modems
- Subject: Re: 16550 vs. 16550AFN
- Message-ID: <rdippold.714353902@cancun>
- Date: 20 Aug 92 23:38:22 GMT
- References: <1992Aug18.221918.1@vax1.umkc.edu> <BtAt51.JD8@news.cso.uiuc.edu>
- Sender: news@qualcomm.com
- Organization: Qualcomm, Inc., San Diego, CA
- Lines: 22
- Nntp-Posting-Host: cancun.qualcomm.com
-
- berger@atropa (Mike Berger) writes:
- >It's not even something you should worry about. The concern is that
- >the modem has to keep up with the maximum data throughput rate. If
- >you use an external modem, then your computer's hardware (your serial
- >port) has to be able to handle the data rate. When you use an
- >internal modem, you eliminate that potential bottleneck.
-
- This makes no sense at all. An internal modem still has to talk to
- the computer, and it does so by having a UART or UART emulator. In
- any case, the communications software still has to service it for
- every character. If the modem doesn't have more than that one
- character buffer, you're going to be interrupting the computer over
- 1500 times a second for a V.32bis/V.42bis connection (about 1700 cps
- with pre-packed data). You still have the bottleneck.
-
- This is one reason to avoid internal versions of all those cheapo
- V.32bis modems if you have any interest in multitasking - most of them
- just emulate a 16450.
-
- --
- Artificial Intelligence: the art of making computers behave like they do
- in the movies.
-