|
First there was Mini-Me,
now there's the Mini-MCU
by Kyle Nisenson
One line of products we
have been getting a lot of questions about down in the dark recesses
of the Frontier Videoconferencing Evaluation Lab is the Mini-MCU. If
you have not heard of them, some manufacturers have been adding
multipoint capabilities to their codecs. Really, this is not new
news. Sony has had multipoint features inside their high-end systems
for over two years. Now, one company who has received much positive
media attention of late has rolled out similar features, and
multipoint from the codec is suddenly a hot subject again.
Each of the Mini-MCUs on
the market is fully H.320 compliant. Most of the units capable of
multipoint range from $10-$15K. The fact that these systems can now
perform multipoint functions is a testament to how video technology
is evolving. Systems that use to require enormous switches and
complicated electronics now can perform similar functionality in a
unit the size of a shoebox. Granted, the functionality is limited.
Most standard MCU features are missing, for instance users are
limited to continuous presence as the only screen layout, speed
matching is unavailable, and bandwidth is often restricted to two
channels.
If you are a user
considering purchasing such a product, the concern you have should
not be price or features, but responsibility. The reason customers
use service providers is so they don't have to worry about whether
the meeting will work. Most of Frontier's customers are Fortune
1000, with plenty of capital dollars available to purchase their own
MCU. Whether the MCU is $10,000 or $100,000 is irrelevant. The
biggest question is, will the meeting work?
There are several
challenges to managing a multipoint conference. Mid-call failure,
where sites drop off before the end of the call for no obvious
reason, continues to plague videoconferencing. Generally such
failures, labeled as temporary disconnects, are considered a fault
with network reliability. Previous articles have detailed the TD
problem (see the 4th Quarter 1998 NewsLink), but in short, today's
networks were designed for single channel telephone connections, and
functions inherently layered into increase phone reliability
actually can cause problems with multiple channel video calls. Site
scheduling and coordination is also an often-overlooked challenge.
Ensuring availability of video systems and that they are ready to
conference is equally as important as if they work to begin with.
User education and assistance are also critical to the success of
video. Too often conferences fail due to lack of user experience
with equipment that can at times be complicated to use.
The best scenario for using
the Mini-MCU is to equate it to three-way calling on your desk
phone. On-the-fly meetings or simple meetings between workgroups
familiar with using video seem to fit the best for the Mini-MCU.
Features such as address books that can quickly assemble
pre-programmed sites work great for regular meetings. For
complicated meetings with different endpoints, international sites,
or high profile users, a more robust system would generally be a
better choice.
While it is wonderful that
videoconferencing systems continue to evolve with better and better
features, the overall specter of reliability continues to be a
concern. Once again, the industry is trying to run before we can
fully walk. Instead of spending R&D dollars on making the
systems more wiz-bang, I once again will call upon manufacturers the
task to accomplish what the end user really needs and wants: make
video as easy and reliable as the telephone.
About
Kyle Nisenson
back
to top
© 2000 Global
Crossing Conferencing. All Rights Reserved.
|
