Gabriel
///////////////////////////////////////////////////////////////////////////////
-------------------------------------------------------------------------------
Gabriel Mark Mains SouthWind Internet Access, Inc.
Department Manager 120 S. Market
SouthWind Technical Support 316-263-7963
-------------------------------------------------------------------------------
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
On Thu, 17 Jun 1999, Bud Bennett wrote:
> Rather than complaining about faulty modem code, I think the more productive
> thing to do is directly address client modem issues and find out just where
> all the problems lie. Unfortunately, I can't give detailed information at this
> time but I think this is a start. I do realize that a PM3 can't be everything
> to everyone, I think the problem is more along the lines of producing modem
> code that is more accommodating to modems that are outside of V.90 spec.
>
> On a side note, Some of our techs say that the PM3's are "Too Aggressive" in
> determining connect speed, (IE. Letting a client modem connect too fast for
> the line conditions.) so they have to manually kick down the top speeds for
> stable connects. This seems to be a reoccurring thing. Is there a way for the
> modem code engineers to make the server side modems behave more conservatively
> when it comes to initial speed handshaking?
>
> All of our PM3A-2T's are running ComOS 3.8.2c2 ... PRI, ESF & B8ZS. I don't
> plan on downgrading.
>
> I would also like to suggest perhaps that Lucent release a white paper
> (Perhaps just specific to Lucent clients?) that
> explains precisely how a chassis negotiates and establishes a V.90 connection
> with a client modem. What the tolerances are, what tones the server side puts
> out and expects from the client modem? Perhaps if the clients can describe the
> problems in better detail by using terms that are a bit more technical, they
> could aid the modem engineers better.
>
> I do honestly hope that the information below is of some use. Perhaps that
> might persuade the Lucent Modem Code engineers to purchase each one of the
> modems below (If they haven't already) and find out precisely where the
> problems lie.
>
> I have had a discussion with our technical support dept and here is what they
> say the majority of the problems are:
>
> Rockwell HCF V.90 code is broken. Won't handshake until you set the
> client
> side to use kFlex only. This is only a workaround.
>
> USR 56k Winmodem Will not handshake properly, slow data xfer when it does
> connect.
>
> AOpen 56k's (FM56-P) Only one version of firmware, doesn't appear to help
> at all.
> Modem doesn't connect, same phasing problem. Will
> not handshake,
> rapid disconnection's (within secs of establishing a
> connection) a
> lot of the time, the modem will just stop responding.
>
> Any Cirrus Logic Modem Handshake/phasing problem. Similar to the AOpens.
> Will not
> connect. (chipset CL-MD56xx typically)
>
> Any ESS modem No documentation available for modem. Have a
> suspected fix, but
> haven't had the opportunity to test it yet. They
> appear to be
> bastardized USR's.
>
>
> -
> To unsubscribe, email 'majordomo@livingston.com' with
> 'unsubscribe portmaster-modems' in the body of the message.
>
-
To unsubscribe, email 'majordomo@livingston.com' with
'unsubscribe portmaster-modems' in the body of the message.