Re: Lucent 56k

Marty Likier (marty@livingston.com)
Sat, 26 Jul 1997 16:56:11 -0700

At 03:29 PM 7/26/97 -0400, Jordan Mendelson wrote:
>
>I remember a while back before Rockwell & Lucent merged their work into
>one protocol name that Lucent had announced support for a 45/45 symmentric
>mode. Lucent also said that they had a way to get 56k as opposed to the
>53k that USR was offering.

The 45K upstream was omitted when Lucent went from Vflex.2 to K56flex. I
have heard them talk about doing 45K in the future. As far as obtaining 56K
with some level learning technique, I have never seen it.

>Does anyone know if the modem code for the new modem cards will support
>either of these two options? Also, from what I understand, the 56k
>download speed will result in a slower-than 33.6 upload speed, does anyone
>have a list of upload/download speeds which the K56 protocol supports?

The most common upstream speeds I have seen are 31.2 and 28.8. The uplink
is V.34, so speeds will be familiar.

By the way its probably a good time to get up on my soapbox for a moment.
Let talk 56K connect speeds. Many of you will be installing the new cards,
upgrading ComOS, and experiencing 42-50K connects (CT1 connects are usually
2K less than PRI). However what you might not expect, is that it is the
client modem which dictates the downstream connect speed. The PortMaster 3
could transmit data in excess of 50K all day long, but its the client modem
which is responsible to differentiate the maximum signals levels (at the
local switch), synchronize with the switch, compensate for digital loss.
determine if voice PADs are present, deal with noise, and you guessed it,
determine the downstream speed. From my experience I have seen some 56K
modems negotiate agreesive speeds out of the shout, only to retrain fairly
quickly. Other modems are less agressive at the initial handshake but hang
tougher longer. And as you deploy 56K service, I would suggest that you
peel the onion back a liitle and compare througput and not just what speed
the modems reports. I have heard of cases where a modem connected at a
lower speed, but beat another modem in a throughput test, even though the
losing modem reported a higher connect speed. Just some food for thought.
Off of soapbox.

--
Marty Likier
Product Marketing Mgr.
marty@livingston.com