home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.dsp:2797 comp.dcom.modems:18098
- Newsgroups: comp.dsp,comp.dcom.modems
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!udel!intercon!psinntp!wrldlnk!usenet
- From: "Michael Smith" <p00004@psilink.com>
- Subject: Re: DSP code for modem functions?
- In-Reply-To: <1992Dec12.183911.4678@spectrum.CMC.COM>
- Message-ID: <2933435481.7.p00004@psilink.com>
- Sender: usenet@worldlink.com
- Nntp-Posting-Host: 127.0.0.1
- Organization: Performance Systems Int'l
- Date: Mon, 14 Dec 1992 19:16:28 GMT
- X-Mailer: PSILink-DOS (3.3)
- Lines: 24
-
- >DATE: 12 Dec 92 18:39:11 GMT
- >FROM: Lars Poulsen <lars@spectrum.CMC.COM>
- >
- >In article <2933103874.1.p00004@psilink.com>
- > "Michael Smith" <p00004@psilink.com> writes:
- [....]
- >>we need to put together an integrated piece of hardware incorporating
- >>both modem functions and proprietary protocol and network-management
- >>processing
- [....]
- >I also noticed that the device you are describing sounds a lot like "a
- >smart modem". If the modem function is really the least part of your
- >system, I suggest that you simply install a Rockwell-chipset-based modem
- >in a corner of your circuit board.
-
- There's an aspect of the design problem that I neglected to mention.
- Cost-per-port is an important consideration. Ideally, I'd like to have
- one DSP chip handling more than one line. And it would be better if the
- modulation technique were in "soft" (i.e. downloadable) code, as
- insurance against firmware bugs and for purposes of upgrading as new
- techniques are defined.
-
- --Michael Smith
-
-