Re: PM3 digital modems SLOWER than analog?

Jaime Bozza (wheelman@nuc.net)
Tue, 8 Apr 1997 11:11:43 -0500 (CDT)

On Tue, 8 Apr 1997, Jordan Mendelson wrote:

> We've seen a lot of posts about latency issues and what not on the PM3,
> but I have yet to see a response from Livingston. With the introduction of
> the new Lucent chip sets, Livingston won't even be using their old modem
> code, so if it WAS the modem code, we should see some sort of difference.

The closest response I received was from MZ about an issue with the
V.42bis dictionary. (PM3 has a 512 byte dictionary, and most - all? -
modems have a 2048 byte dictionary) This would (possibly) account for the
latency issues, and I expect that to be resolved once the new code is out.
We *DO* have an issue of slower download times with one customer (Only
slight), though I've seen other messages (one this morning) about slower
transfers. I'm not sure what this could be, though it still could be
centered around V.42bis.

I would *LIKE* to see something put together about modem problems with the
PM3 and how to resolve them. Since you can't "tweak" the PM3 modems,
we're stuck with tweaking the client modems. The reason we liked the
Couriers is that they just worked, very little tweaking, and pretty much
every modem out there would connect. Since we don't have the same liberty
with the PM3 (We had a V.FAST customer that finally bought a new modem.
<G> Though we warned him months ago this would happen), it would be nice
to be aware of the problems and know what others did to fix them.

I'm just hoping the new modem code is soon. If that fixes the latency
issues, and the transfer speed issues, I'll be happy. As for the 14.4K
problems, we only have a single customer so far. If we start to have more
problems, this will also become something that (I feel) should be
resolved.

Jaime Bozza
Nucleus Communications, Inc.