Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: (ASCEND) Digital line quality issues

On Tuesday, August 11, 1998 23:43 Marcel Brown <marcel@marcelbrown.com>
wrote:


>Just a quick, curious question. I feel pretty comfortable telling my
>customers that since we have digital PRI trunks instead of regular analog
>lines coming into our equipment, our end is free from line noise. Is there
>any way that digital lines can experience noise (other than noise that
>originates from the analog lines between the customer's house and the
>central office)?
[snip]

I may not totally 100% correct on how I explain this, as its from
bits and pieces of information.

We pull ISDN PRI's into our MAX 4000 boxes ... and therefore
get the full gambit of ISDN benefits, which includes features
over and above what gets delivered on plain 'ole T1 circuits.

But both T1 and ISDN are digitally delivered circuits ...
and they have acceptable bit rates that ensure that if
errors do happen (to the bit stream) its very low (and infrequent).

As compared to taking an analogue line (POTS) into a modem
(or bank of modems) - a world of difference.  Plus: you're never
going to provide v.90 style connections with anything BUT a
digital connection on your end.

Here's the kicker:
In researching some "performance" problems that we have
had - noting that this is somewhat subjective, in the sense
that we get connections that are "slower" than expected ...
we found the following:

When a user makes that v.90 dial-up connection to you,
quite possibly the call is originated on one CO ... and
then crosses one or more trunks to get to the switch
that finally provides the call via your (nice, clean)
ISDN-PRI circuit.

We did some tests ... from various locations that were
one or more CO hops away, where we found that calls
into our Max boxes were "slower" than calls to the
local uunet/MSN numbers, or other local ISP's
(running the 3Com stuff).  Of course - that was unacceptable.
In the end - I believe that what some of our problem was
was due to the types of trunks that tie switches together.
Something about some different types and encoding,
where at least one setup was not favourable to the
bandwidth required by a good v.90 connection.

Therefore when people called in, going over these
"undesirable" CO trunks (between switches?!)
the modem would negotiate a slower baud rate connection.
We've got a great CLEC who's done a lot of work to
clean this kind of thing up (and for the most part,
its not even their problem, but the ILEC).

Hope this helps.
Bob McCormick


++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>
Follow-Ups: