Well - maybe there in sh*t anyways...
From DSL-specialist Dave Burstein's site:
"Correction: VDSL 'Crossover' Not Close
mode and interoperability disappointing VDSL is working
well for highspeed service from the basement, deploying
rapidly in Japan. From the neighborhood node, DT has 3M
homes passed and AT&T over a million, with both intending
to rapidly expand. So the niche will expand, but I have
been unable to find a carrier regularly using VDSL gear for
customers over 5,000 feet, despite a VDSL standard designed
for that choice.
Yet another carrier has told me DSL Primes enthusiasm
for an early switch to all VDSL2 service is premature.
'Id never get 72 ports on a line card with VDSL2, and
the power problems are hard to solve,' their DSLAM
designer tells me. Density, heat, power are proving
very resistant to improvements. Prices are staying
high, 2 or 3 times the $6 or less ADSL2+ chips can go
for. TI isnt competing for the high end, and several
other announced chips have not come to market.
From 5,000 to 12,000 feet, the goal was equal
performance. The actual VDSL chips were considerably
slower (up to 20%) to customers at those ranges. Only
one profile (8b) allows full downstream 20.5 dBm
power; all profiles reduce power by several dB and
result in less maximum performance.
'Deployable interop will take more time to settle,' an
chip engineer laments. 'All vendors are upgrading VDSL2
firmware rapidly. Production designs are changing
rapidly.' Carriers deploying now must use modem and
DSLAM chips from the same vendor, and may be locked
into that vendor for several years to maintain
compatibility. The DSL Forum and UNH are working on the
My prediction, '20% of dollar volume will be VDSL by
the end of 2006, I believe,' is almost surely wrong.
Currently, only deployments under 5,000 feet are going
to VDSL. Those with a substantial portion of lines over
5,000 feet are going for ADSL2+."