home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.modems
- Path: sparky!uunet!rde!ksmith!keith
- From: keith@ksmith.uucp (Keith Smith)
- Subject: Serial vs. Parallel Modem Interfaces
- Organization: Keith's Computer, Hope Mills, NC
- Date: Mon, 27 Jul 92 22:44:45 GMT
- Message-ID: <1992Jul27.224445.5160@ksmith.uucp>
- References: <clemon.08ii@lemsys.UUCP> <1992Jul22.053652.9090@raven.alaska.edu> <clemon.08kw@lemsys.UUCP>
- Lines: 52
-
- In article <clemon.08kw@lemsys.UUCP> clemon@lemsys.UUCP (Craig Lemon VE3XCL) writes:
- >In article <1992Jul22.053652.9090@raven.alaska.edu> floyd@hayes.ims.alaska.edu (Floyd Davidson) writes:
- >>In article <clemon.08ii@lemsys.UUCP> clemon@lemsys.UUCP (Craig Lemon VE3XCL) writes:
- >>>Lining the bits up single file can
- >>>only go so fast. While we may not have reached this limit with standard
- >>>high-speed modem technology yet, there will be other standards (ISDN
- >>>perhaps) that will.
- >>
- >>ISDN is has a serial interface. And while the fastest serial interface
- >>may not be good enough someday, the fastest one I work with right
- >>now runs well up into the multimegabit range... 45Mbps. And I
- >>hear of others that are faster.
- >
- > OK. Granted, serial links can be fast once you start shielding the
- >path. You can only go so fast on a serial line before the frequency of the
- >bits reaches RF range and cable losses on a normal "twisted pair" become
- >significant. I think the point of the debate centers more around looking
- >for a standard solution to replace RS-232C as opposed to arguing actual
- >serial vs. parallel concepts. For simple things like losses in cabling
-
- Why? Conventional modems will not ever exceed the max bandwidth of the
- dial-up type of phone line anyway. For anyone with the cash there are
- plenty of other off the shelf solutions. As far as twisted pair and
- shielding issues go, Those limits are already higher by a factor of 1000
- over current "state-of-the-art" modem technologies. RS-422, SCSI and
- ethernet, are a few of the existing interfaces that can take you into
- the Mega-bit range. I don't think there will be any non-commercial
- megabit per second modems in the near future. There just isn't enough
- demand. Again Commercial guys may want T3 bandwidth, but for the
- average joe, or even a small business this is simply overkill.
-
-
- >that I mentioned earlier, the frequencies involved in a parallel bus are
- >at least 1/8th what they would be in a serial environment. That is
- >something to consider that is totally separate from servicing the data
- >stream at the terminal.
-
- And the bus requirements are more severe. Cross-talk and cable
- capacitance will change 8/16/32 signals instead of 1, etc ad nauseum.
- There are generally lot's of tradeoffs involved either way. You start
- pushing a WHOLE BUNCH of signals on the same cable that are oscillating
- at a high rate of speed, and you have 8 times the noise headaches to
- contend with.
-
- With TOTAL CPU control parallel is generally cheaper because with an LS
- '373 and a decoder you are in business. 2 '373's for a bi-directional
- port. Net cost in the pennies.
-
- --
- Keith Smith uunet!ksmith!keith 5719 Archer Rd.
- Digital Designs BBS 1-919-423-4216 Hope Mills, NC 28348-2201
- Somewhere in the Styx of North Carolina ...
-