home *** CD-ROM | disk | FTP | other *** search
- Path: dlmlap.supra.com!dan
- From: dan@supra.com (Dan Moore)
- Newsgroups: comp.dcom.modems
- Subject: Re: Supra will offer upgrade to 33.6!
- Date: Thu, 22 Feb 1996 15:19:23 GMT
- Organization: Supra Corporation
- Message-ID: <dan.877.312C897B@supra.com>
- References: <4g3t66$26um@hopi.gate.net> <312172b0.302209@uchinews.uchicago.edu> <4ftaot$rpc@shellx.best.com> <31226a7d.63749347@uc <4g932g$npk@detroit.freenet.org> <Dn1MIK.15H@freenet.carleton.ca>
- NNTP-Posting-Host: dlmlap.supra.com
- X-Newsreader: Trumpet for Windows [Version 1.0 Rev B.7]
-
- In article <Dn1MIK.15H@freenet.carleton.ca> an171@FreeNet.Carleton.CA (Anthony Hill) writes:
- >Bill Beogelein (ag775@detroit.freenet.org) writes:
- >> Where can I find a detailed list of those 2 things:
- >> 1) More v.34 options that the Supra doesn't (yet) have.
- > 64-State trellis decoding, assymetric symbol rates and carrier
- >frequancy and the 2749 optional symbol rate. That's all I know of off the
- >top of my head, although I'm sure there are plenty of others (there are a
- >whole bunch of optional features for v.34, Supra has some of them, but
- >certainly not all of them). FWIW though, I wouldn't count on any of these
- >features being added with any new Rockwell datapumps.
-
- Asymmetric carrier frequencies are supported by the Rockwell
- datapumps. The 2749 symbol rate is not supported, but I've always wondered
- how much use it really is. IMHO there isn't enough resolution in the line
- probe to determine when enough bandwidth is available for the 2749 symbol rate
- but not for the 2800 symbol rate.
-
- If people want to know what is or is not going to be in any
- possible future Rockwell datapump they should contact Rockwell. Any info
- other people may have about something that may or may not exist is covered
- by NDA. (In fact this whole message is probably under some NDA or another, so
- I really shouldn't be typing any of this...)
-
- >> 2) Bugs in the regular v.34 datapump.
- > There is a bit of a problem with the way that it handles the 3429
- >symbol rate, although most non-Rockwell companies worked around this
- >problem long ago (since Rockwell can't fix problems in their datapumps
- >already in the field). Again, there are almsot certainly other things
-
- Technically there isn't a problem with the 3429 symbol rate, it
- works correctly during data transfer. There is a glitch (great high tech
- term) in the transmit level when phase 4 ends and data transfer starts when a
- 3429 symbol rate is being used. Rockwell's receiver implementation is immune
- to this glitch as are some other implementations. There are some
- implementations that have real problems due to the glitch. The glitch is
- fixable in firmware, a partial fix was in Supra's 1.000-16 firmware and a
- complete fix was in the 1.200-02 firmware. (Many Rockwell datapump issues
- can be corrected from firmware without changing the datapump. Things are not
- as inflexible as everyone seems to believe.)
-
- >(IMO there's no such thing as a bug-free modem, espeically not with v.34).
-
- I'm still waiting to find a perfect Bell 103 modem... ;-)
-
-
- --
- Dan Moore
- Supra
-
-
-