In Broadcast - April 2013 - 83

ISSUE 18 - APRIL 2013 | WWW.INBROADCAST.COM

I N T E R N A T I O N A L

N E W S

InREPLY

Digimetrics Masters File-based QC
Founder and Business Unit Manager for Digimetrics Eric
Carson ﬁelds the InBroadcast questions on ﬁle-based QC,
the need to deliver content to multi-screen environments and
automated loudness normalisation...

InBroadcast: How is your
company changing to meet client
demands?
Eric Carson: Digimetrics has
focused on three speciﬁc areas for
ﬁle-based QC demands: Improving
the QC operator review time and
experience: Aurora QC integration
with our Hydra Player is showing QC
operator time reductions by a factor
of eight, on average, by offering a
very simple method to jump to the
exact frame of a QC event, edit/
change that event (based on the
QC operator’s judgment), and then
jump to the next QC event. We’ve
eliminated the need to rely on a
tool like VLC, which notoriously
prevents accurate ‘go to’ or scrub
points, as well as introduced the
ability for a QC operator to edit or
override the automated QC decision.
Improving overall ‘hit rate’
accuracy: At Digimetrics, we have a
dedicated engineering staff working
constantly on improving our core
expertise, which is visual and
audible disruptions in the ‘baseband’
decoded essence. Every quarter, we
re-tune portions of our algorithms in
a machine learning loop, using
customer-provided
ﬁles that have been
tagged with artefact
locations, allowing
us to reduce false
positives
and
increase the accuracy
o f t h e p ro d u c t .

Performance: We are constantly
pressured to make ﬁle-based QC
run faster, and our commitment
within Aurora is to allow operators to
add any test they want and have the
product run as fast as possible. We
call this commitment ‘guaranteed
QC capacity’, and it works by:
• Having the system monitor
QC load points to ensure that
operators cannot concurrently
run more jobs than the system
can process efﬁciently (Aurora
ensures that regardless of how
many jobs are running, the
performance time of any job
should not affect other jobs);
• Implementing efﬁcient multi-core
CPU utilisation, ensuring that
processes within Aurora, such
as demux, decode and testing are
all properly dealing with multiple
CPU cores and not creating any
slowdown or unwanted issues;
• Utilising nVidia’s CUDA framework
for GPU acceleration using Tesla
and Quadro GPUs for complex
analysis tasks, such as tape hit
and perceptual artefact detection.
Combined, these methods allow us
to QC ﬁle-based HD at real-time or
faster speeds, and we
are currently looking
at pushing this
same performance
envelope into 4K and
H.265 workﬂows for
the future.

InB: What criteria does the market
use to make decisions for your
services?
EC: Currently, loudness is a very
hot topic, and everyone is asking
for QC with automated loudness
correction, which we offer with
our Eos AutoFix line of tools. I
think that within file-based QC,
the QC features that are available
is the number one decision point,
whether it is codec and container
support, distribution proﬁle support
or analysis test support, because
these features must match the
customer’s test plan in order to
be productive in the workflow.
We also see reporting as a big
decision point, where simpliﬁed and
shorter reporting that is easier to
read and interpret results from is
winning out over longer, complex,
multi-level reports. Within reports,
most customers are beginning to
realise that it is important for the
same report to be available to
users, whether they have current
access to the system or not, so
ensuring that the same report
experience is available online
and offline has become critical.
Of course, price is always a
decision point, but there are many
levels of needed QC, and companies
available to serve those price points
distinctly. We haven’t had many
issues where price has disrupted
a conversation for a customer that
needs a particular QC feature set. I
think price issues really come into

play in any market at the common,
lower-end feature sets.
InB: Which territories are showing
particular activity?
EC: File-based QC seems to be
particularly valued in Japan and
Germany, with Canada and the
western United States also showing
particular growth in interest.
InB: Any innovations ahead that
you can discuss?
EC: We’ll be showing our latest
ideas at NAB and particularly,
several performance improvements,
the debut of DPP and Netﬂix testing
proﬁles and native AJA playback in
the Hydra Player.
83
InB: Tell us, what type of customer
is using ﬁle-based QC today?
EC: The integrated solution of Aurora
& Hydra was really designed around
high-volume content aggregation
and distribution facilities, so here
we are talking about cable and IP
VOD operations, post-production
& distribution houses and filebased network operators. We are
just now starting to see traditional
broadcasters become interested
in file-based QC, mainly based
on the need to deliver content to
multi-screen environments and
the need for automated loudness
normalisation.
InB: Thank you for taking timeout
to answer our questions.

Compact stand alone fibre interfaces
Full Broadcast quality and 3G pathological signal performance
Wide range to suit different applications and budgets
Designed for Outside Broadcast and Studio Applications
Best price and performance ratio on the market

www.bluebell.tv

http://www.linkedin.com/pub/dir/?first=Phillip&last=Izzard&search=Go https://twitter.com/inbroadcast http://www.youtube.com/watch?v=qpzEatLsHdE&feature=youtube_gdata_player https://www.facebook.com/InBroadcast http://itunes.apple.com/gb/app/inbroadcast/id468652679?mt=8 https://play.google.com/store/apps/details?id=com.branded.inbroadcast&feature=search_result#?t=W251bGwsMSwyLDEsImNvbS5icmFuZGVkLmluYnJvYWRjYXN0Il0 http://WWW.INBROADCAST.COM http://www.digi-metrics.com/ http://www.digi-metrics.com/ http://www.bluebell.tv http://www.bluebell.tv

In Broadcast - April 2013

Table of Contents for the Digital Edition of In Broadcast - April 2013