54
The point of all this technical information is ultimately to help you to make high quality video: video
that looks good and serves the purpose for which it is made. But how do we know if the video is of high
quality? And what does that really mean?
There are certain technical standards that video must meet simply in order to be viewable on a
monitor. And beyond that fairly cut and dried realm is the area of aesthetics. Which leaves us with the notion
that most discussions of video quality are relative, in the sense that there is a context, a purpose to which
definitions of quality level relate. For example, a video that looks acceptable on an inexpensive television set
might look absolutely horrible on a high end video monitor. Is this because the expensive monitor cannot
display the video properly? No, probably not. In fact, the reverse is true: the inexpensive monitor doesn’t
show enough of the signal to reveal how bad it actually is or where the problems actually lie. The context in
this case is the display, the quality of the monitor.
As video makers, we have to produce to the quality specifications our clients demand, what is
appropriate for the project, what they can afford. If the clients are ourselves, then the quality level has to be
better than the weakest link in the viewing chain - now or in the future - but not so good that the cost of
producing prohibits the production.
One important way, perhaps the most important way, to make sure that your video really looks as
good on other monitors as it does on yours is to monitor the video with a properly calibrated monitor of as
professional-level quality as you can afford. See below for some methods to calibrate monitors. And it goes
without saying that you should have dependable, high quality audio monitoring as well, either through
headphones or speakers you can trust.
A second way to be sure that your video looks as good as it can is to use, if available, monitoring
test equipment (waveform monitor, vector scope) in parallel with the well-calibrated monitor mentioned
above.
Monitor Calibration
(procedures, test patterns/bars)
The following technique was suggested by an old broadcast engineer who began working in video
when tape was two inches wide and scene changes were made with razor blades and tape, Fades and
dissolves did not exist. One had to walk up 6 flights of stairs to the studio and back, often carrying a 65-
pound camera. Test instruments were expensive and often unavailable. Network engineers had to find a
convenient, simple, reliable way to calibrate monitors that could be done in the field. (You could always
recognize one of these fellows by the Wrattten 47B (dark blue) they kept in their wallets or in their
mechanical pencil pocket protectors, along with their little screwdrivers.)
To calibrate a monitor is to adjust it so that it displays colors that are the same as a standard. That
standard, and a major aid to calibration, is called color bars: a pattern of colored strips (and in some cases
gray scale strips) of very specific colors, arranged in a very specific way.
To do the following calibration procedure, you will need a source of standard color bars. There are
several possible places to get this: your camera may generate bars; perhaps your black burst generator puts
ouT-Bars; you could use a graphics program on your computer to display an image of standard color bars.
There are numerous places on the Internet to download standard color bars if you don’t have them; do an
Internet search for “color bars” and take your pick.
The following procedure will be described using SMPTE bars, but EBU bars will work for most of this
technique as well. As you will see, they bars are different in format, in part because of the differences in
