56k modems/frame/ptp revisited (LONG) (fwd)

Tue Sep 1 19:40:21 UTC 1998

I noticed several people reciting incorrect information in relation to PCM
modems.  For those who don't know, I worked for Livingston and then Lucent
RABU up 'til February of this year.  I was very involved with issues such
as PCM modems on the PM-3.

Simply put there is NO restriction on the SPEED of PCM modems aside from
what the PSTN can carry.  On a digital line end to end max theoretical is
64K, with RBS and the switch codecs and configurations in use today 56K in
North America.

The misconception is that there is a 53.3K cap on the speed.  This is false.
What the FCC limits is the power (db) level you can place on the line in
the PSTN.  This is to limit crosstalk in copper cable bundles.  This power
limitation does not equal a speed limitation.  This seems to have arisen
from the fact that the first PCM modems - USR X2 units - could not go faster
than 53.3K without violating the FCC power limitations.  All other things
being equal, the more power you can use, the faster you can go.  To cover
their ass USR put the disclaimer on the boxes talking about how X2 was 
capable of 56K, but limited to 53.3K due to the FCC blah, blah.  Many
people read this as the FCC having some cap on allowed speed since they
didn't explain that the trouble was with X2's inability to go faster in
the allowed power band.

Now, K56flex did NOT have this limitation.  Users regularly reported connect
speeds up to 54K, and a few even reported the magical 56K - but that was
very rare indeed.  How?  K56flex was better at using the available power
range (in most, but not all cases) and managed to squeeze out the additional
throughput.

X2 and K56flex both improved with time - X2 users eventually started seeing
54.6K connects, and K56flex saw higher numbers of 54K connects.  But it
doesn't appear that they removed the warnings from the packaging.  And,
now this baffles me, some K56flex modem vendors included the same language
on their packaging.  Which is nutty since 53K isn't a K56flex speed!  It
goes 52K -> 54K, 2K incriments.

Anyway, now we have V.90.  And, if you read user groups for the various 
server HW vendors, as well as USEnet groups, etc, you'll find some people
reporting 54.6K connects with V.90.  Off hand I don't recall seeing anyone
in North America reporting 56K as of yet, which isn't surprising.

Please, don't perpetuate the "FCC limits modem speeds to 53.3K" myth.

-MZ
-- 
<URL:mailto:megazone at megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/>  <URL:http://www.gweep.net/>  Hail Discordia!
 ====================================================================
 ISPF, The Forum for ISPs by ISPs.  October 26-28, 1998, Atlanta, GA.
  Three days of clues, news, and views from the industry's best and
  brightest. http://www.ispf.com/ for information and registration.
 ====================================================================