home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.modems
- Path: netcom.com!marvinl
- From: marvinl@netcom.com (Marvin Lichtenthal)
- Subject: Re: IBM Mwave 28.8 and Multitechs
- Message-ID: <marvinlDKnAM7.JJF@netcom.com>
- Organization: NETCOM On-line Communication Services (408 261-4700 guest)
- X-Newsreader: TIN [version 1.2 PL1]
- References: <4bpgc4$c5i@news1.exit109.com> <nathan-2912952111390001@nathan.southern.co.nz> <4c6ec4$lfv@jeeves.usfca.edu>
- Date: Thu, 4 Jan 1996 07:10:55 GMT
- Sender: marvinl@netcom.netcom.com
-
- I have a response from the OS2BBS Cforum on MWAVE and thought others
- would be interested:
- ===========================================================================
-
- Marvin, I don't know about the AT&T revision levels (AT&T chipset
- inside MultiTech), but I can say that one of the newer revision
- levels has a violation in it that caused problems with previous
- levels of the Mwave V.34 (found on the Aptivas). We have since
- worked around AT&T's violation. Strangely enough, the older
- AT&T revisions did not have this violation!! We discovered this
- many months ago when an IBM site upgraded their home terminal
- program MultiTech modems, and someone who had been using
- the Mwave V.34 software suddenly couldn't connect. I then called
- this number and after some debugging, discovered the violation, and
- we then created a workaround.
-
- I can confirm that the failure mechanism reported here is precisely
- what happens with our older microcode when it hits the AT&T violation.
-
- I have heard rumor of some type of customer service software update
- coming for Aptivas, but unless the customer can confirm this with
- the PC Company, than it is just a rumor. Such a CSD to the machine
- would most likely include a newer V.34 microcode level with
- the work around to this violation.
-
- This begs the question why Mwave would have trouble with the
- violation and not others. The facts are simple: V.34 modems
- are incredibly complex and each modem design is unique. One modem
- design may be sensitive to one violation while another may not
- be. I have consistently seen this in all V.34 modem brands, i.e.,
- varying degrees of sensitivity to various violations in other
- brands. Any V.34 modem that can be updated with software, though,
- will likely have workarounds to other modems' violations.
- We must have a dozen workarounds at this point for other
- modems' violations of the V.34 recommendation. With new
- V.34 designs coming out as time goes on, this situation is
- not stable even now, but it gets better because BBSes are
- pretty much installed now with their 28.8 modem support.
-
- The one known Mwave V.34 violation that existed half a year ago
- was ONLY uncovered when Telebit V.34 modems came on line. All
- other V.34s were not bothered by a a couple percent timing
- tolerance we were off by (by mistake). As soon as we started
- debugging with Telebit, we found the problem and fixed it.
-
- Hope this helps. You are welcome to share this with the folks
- posting.
-
- Malcolm Ware - Team Mwave (thousands of miles from home)
-
- ========================================================
-
-
-
-
-
-
-
-
-
-
-
-
-
- --
- Marvin Lichtenthal
- marvinl@netcom.com
-