When Red is red, aka Colour Bar

[Davade]

One of the most annoying thing for me about editing is the inconsistency of colours appearing on the television.

I know that most television has different calibration and can never be accurate in a way, hence it differ from television to television. But how do I minimise the difference? Which make it RED is RED!

I know that in the AVID, there is a vectorscope that sort of measure the accuracy of colour but I don't know how to use it? Does anybody know how to work it?

Furthermore what is the purpose of colour bar???

Here is and excellent article posted on the AVID-L by Jeff Sengpiel
http://archives.itg.uiuc.edu/avid/2003-02/msg01317.htm
who has contributed many more such useful articles such as this one on Waveform monitors http://archives.itg.uiuc.edu/avid/2003-03/msg00172.htm
and this one one Vectorscopes http://archives.itg.uiuc.edu/avid/2003-02/msg01355.htm

lots of good stuff on "the -L" try serching the archives at
http://www.avid.com/community/mailli...sagesearch.asp
or
http://archives.itg.uiuc.edu/avid/

and here is another link about colour bars
http://www.videouniversity.com/tvbars2.htm

and remember, no one like rejection - learn to love your scopes.

-------------------------------------------------------------
General Concepts

Why Test Video Signals?

Working in video, the quality of your final product depends on many factors.
Scripting, casting, directing, and many other ingredients add together to
make (or break) a production. But even when all the creative efforts are
perfectly balanced, poor technical quality -- and the resulting poor picture
quality -- can obscure all the hard work and skill invested.

This holds true in any facility from the largest production house or network
studio, to corporate studios, and to the smallest freelance production shop.
The difference is, the larger facilities have a technical staff devoted to
keeping equipment performance and picture quality in peak form. Smaller
commercial, business, and industrial facilities have smaller staffs.
Sometimes a staff of one!

Whether you're a video veteran or the newest intern, it behooves you to know
something more about video quality than what you see on a picture monitor.
The goal of this booklet is to help you acquire additional technical
knowledge, and be more comfortable and capable with video testing.

You must, however, exercise caution and common sense. Be sure to heed all
warnings printed on equipment covers. And never remove any equipment casings
or panels unless you are qualified for internal servicing of that equipment.
The information in this booklet is intended only for use in external
monitoring and adjustment of user accessible controls on video equipment.

A television picture is conveyed by an electrical signal (Figure 1-1). This
video signal is carried from one place to another by cables (coax) or by
radio-frequency (RF) waves. Along the way, it must pass through various
pieces of equipment such as video tape machines, switchers, character
generators, special effects generators, and transmitters. Any of this
equipment can change or distort the signal in undesirable ways.



Since picture quality is largely determined by signal quality, it's
important to detect and correct any signal distortions. The signal has to be
right before the picture can be right.

Many video facilities rely heavily on picture monitors for quality checks at
various stages of production and distribution. A picture monitor, after all,
displays the final product, the picture. It's a quick and easy means of
ensuring that there are no obvious picture impairments.

But picture monitors do not tell the whole story. In fact, relying solely on
picture monitors for video quality checks can be an open invitation to
disaster.

First of all, not every picture impairment is obvious on a picture monitor.
Minor problems are easily overlooked. Some cannot be seen at all. For
example, a marginal video signal can still produce a seemingly "good"
picture on a forgiving monitor. This can produce a false sense of security
as signal degradations accumulate through various production stages. The end
result can be a nasty surprise that can lead to costly remakes and missed
deadlines.

To avoid such surprises, you need to look at more than pictures. You need to
look at the video signals that convey the picture information through the
various devices and interconnecting cables of a video system. Specialized
instruments have been developed to process and display these signals for
analysis.

A waveform monitor is an instrument used to measure luminance or picture
brightness as well as a high frequency color signal called chrominance. An
instrument called a vectorscope is required for quality control of video
chrominance, especially in more complex systems.

When properly used, these tools allow you to spot signal problems before
they become picture problems. You can make certain that the signals are
optimal, rather than marginal. Plus, with regular video system testing, you
can catch minor degradations and performance trends that indicate a need for
system maintenance. This allows you to avoid costly failures, both in
equipment and in program production or distribution.

Test Methodology

There are two somewhat differing ideas used in testing video. Our approach
is to test characteristics of a specific signal to ensure that it meets
certain artistic or technical requirements. Or, we may test the
characteristics of individual pieces of equipment or several pieces in the
video signal path to determine that signal distortions introduced by this
equipment are acceptable -- or, at least, minimized.

An example of the first case might be monitoring the output of a video
source (camera, character generator, etc.) to ensure it is not producing
signals that exceed black or peak white signal limits. As an example of the
second case, we may wish to ensure the overall gain of a record/playback
system is correct, i.e., a recording made with a standard amplitude video
signal at the machine's input will produce standard amplitude video at its
output during playback.

In the first case (specific signal) we'll need test equipment that enables
observation of the signal and knowledge of what the appropriate limits or
characteristics are. The second case (equipment or system testing) is more
general and is assumed in the following discussion.

The usual method of evaluating video equipment is with a well-defined,
highly stable test signal having known characteristics, such as a color bars
signal.

Video testing is based on this simple principle of applying a known test
signal to the video system or equipment input and observing the signal at
the output. Any distortion or impairment caused by the system is observed
and measured on the output signal. If there are distortions, the equipment
is adjusted to eliminate or minimize them. The point is, if the system can
pass the test signal from input to output with little or no distortion, it
can cleanly pass picture signals as well.

The signals necessary for such testing are obtained from a test signal
generator. This instrument produces a set of precise video signals with
carefully defined and controlled characteristics. Each signal is ideal for
verifying one or more specific attributes of the video system under test.
For all practical purposes, these test signals are "perfect" signals.

Other instruments such as waveform monitors, vectorscopes, combinations of
waveform and vector monitors, or specialized video measurement sets are used
to evaluate the test signal at the output of the path under test. As an
example, Figure 1-2 shows a waveform monitor display of a color bars signal.
This display is also called a waveform -- it is actually a graph of the
changing voltage of the signal (plotted vertically) and time (plotted
horizontally). Calibrated scales on the waveform monitor's screen allow the
various amplitude (voltage) and time parameters of the waveform to be
measured. Other test signals and their related instrumentation and displays
are discussed in the following sections.





Test signal generators and signal evaluation instruments are available in a
wide variety of models. These can range from simple production-oriented
instruments to highly sophisticated engineering instruments. Waveform
monitors, vectorscopes, video test sets, and other specialized equipment to
display and/or evaluate the signal are also available in a wide variety of
configurations. The following sections will acquaint you with some of these
tools and with methods of using them to enhance your effectiveness in
maintaining video quality.

Connecting and Terminating Instruments

The Tektronix instruments discussed in this booklet, and most others, have
rear panel BNC connectors. For many video tests, you only need to use one
connector on each instrument. This is the connector marked TEST SIGNAL on
the signal generator and the connector marked CH A INPUT on the waveform
monitor or vectorscope.

Notice there are actually two "loop-through" CH A connectors on both the
waveform monitor and vectorscope (Figure 1-1). If you feed the signal in one
side of these inputs and out the other, the signal will pass through the
instrument unaffected. This type of loop-through input lets you connect the
same signal to more than one instrument.

For example, after running the test signal from the signal generator through
the system under test and from there to a waveform monitor, you can then
loop the signal through the waveform monitor into a vectorscope. Using the
same method, you can also loop through the vectorscope to a picture monitor.
This connection method allows you to look at the same test signal on all
three instrument displays (Figure 1-3). The order in which the instruments
are connected doesn't matter -- if the connecting cables are short.



Figure 1-3. Loop-through inputs on instruments allow the same video source
to be viewed simultaneously in waveform, vector and picture forms.

Coaxial cable does have signal loss -- the signal's amplitude decreases as
it progresses down the cable. For runs of a few feet, this decrease in
amplitude is typically 1% or less and is usually ignored. For longer cable
lengths, or when precision measurements are being made, the loss must be
taken into account or corrected with a compensating distribution amplifier.
Not only is coax lossy, the loss is a function of signal frequency -- higher
video frequencies are attenuated more than the low frequencies.

Also, not all cables are well shielded and signals may cross- talk from
outside the cable into the video path inside. In other words, the
interconnecting cables themselves may be introducing signal distortions.

While small, flexible cables are convenient to handle, they are not without
technical cost. Consideration should be given to using larger,
double-shielded cables in long or critical runs. The techniques in this
booklet can be used to evaluate the distortions introduced by various
lengths and/or quality of cable.

While on the subject of cables and interconnections, always remember to
properly terminate each signal path. If a signal path is left "open" at the
end -- such as a high impedance loop-through with nothing connected on one
side -- several problems can result.

Most obvious will be a change in amplitude of the signal on that path. With
an open termination the amplitude will be higher than expected (usually
about twice amplitude, but actually depending on the signal source
impedance). With a double termination, such as will result if an internally
terminated loop-through and an external terminator are used on the same
path, the amplitude will be decreased. A double termination will often
result in a two-thirds amplitude signal, again depending on source
impedances.

Amplitude is not the only effect of misterminations. Don't be tempted to
make up for improper termination by adjusting the signal amplitude.
Misterminations also introduce problems with frequency response (amplitude
becomes a function of signal frequency) and with differing response
depending on location along the signal path. Use the correct terminating
resistor on the end of each coaxial path.

There are three ways a signal path can be properly terminated:

* Some instruments have a single-connector input with a built-in
terminator. You can connect such an instrument "as is" at the end of a
signal path, but you can't loop the signal through it.

* Some instruments have a two-connector input with a built-in terminator
and a switch for selecting loop-through or terminating connection. You can
connect such an instrument anywhere in the signal path, but you must set the
switch correctly for the intended use (either Hi-Z for loop-through or 75
ohm for end-of-line termination).

* Some instruments, such as the 1720 and 1730, have two-connector inputs
with no built-in terminators. You can connect such an instrument anywhere in
the signal path. But, if you connect it at the end of the signal path, you
must attach a 75 ohm terminator to the unused connector.

-Jeff Sengpiel
---
---
Find answers to your questions in the archives:
http://www.avid.com/community/mailli...sagesearch.asp

For messages before May 2002, visit the archive at http://archives.itg.uiuc.edu/avid/

[Matthew]

Red on a SVGA monitor and red on a PAL TV screen will never look the same. Different technology.

Internal vectorscopes are only so good. They are a guide, no more.

How do you know the signal leaving the Avid is any good/legal? You can't really rely on an uncalibrated vectorscope that only reads one line of picture (at least, in offline Avids that all you get it seems)

And ... have you lined up your monitor correctly? How many people know how to do that these days? If its not a quality monitor properly calibrated, your eyes will tell you nothing.

[Davade]

The setup at my old office doesn't have any properly calibrated monitor. What they have done is use different brands of television monitor as viewing monitor. That's the reason why I have so much problem when it goes to the network. They are complaining that it is too dark or too saturated and there is no way to know what is the standard.

After long period of time, it boils down to experience of guessing what is acceptable for the network. Hence the reason why I posted this thread.

And thank you Shilby for the reply and it is very helpful.

[rachelw]

Guessing gets you knowwhere, neither does the brand of one monitor versus another. The important thing is to have a REAL vectorscope or multiscope. With this you can not only calibrate your monitors, but check signal going into and coming out of VTRs, racks, your AVID, etc. And using one is very, very easy - once you've had the right person explain it to you. But they're pretty easy to figure out on a basic level. Depends on what type you've got or choose to get.

The article posted here says all that needs to be said - and essentially you can't really ensure quality without being able to check your signal across the path it travels to being put on tape, as well as looked at; and you need to have a source for bars/test signals that can be put through all equipment, so that you have a uniform test standard against which to calibrate your gear. This can be a bit on the expensive side, but it's the only way to calibrate all of your equipment, as well as checking for oddities such as intermittent hum, which can cause odd disturbances of vision that resemble waves of vibrating lines of luminance across the screen - this appears as a sort of vibrating sawtooth pattern on the baseline of the video signal, and is very common in setups involving routing through complex paths and patchbays, as well as poorly installed equipment.

If you want to get a serious grasp on technical callibration, hang out with engineers; they really do rock when it comes to installation and maintenance issues. Also people who tech check programs for purchase and transmission, and use scopes to check the general quality of vision and audio (with a multiscope you can easliy detect phase errors and the like). It's much easier to decide whether or not to reject material when you can just take one look at the colour bars and first few minutes of program, and decide whether there's any real need to view the rest...so chatting with the people who reject the material could be very helpful.

Also, it helps a great deal to have the same bars/signal generator to lay test signal at the front of your masters and bed tapes, so that you can check back your recorded material as well; you can be more certain of the standard of quality when you double-check your work in this way, and also it lets the receiving station see what your levels are really like; if you're using some computer generated bars and pumping it through your system without using those same bars as a reference to calibrate the rest of your gear you can find yourself with a standard that means nothing at all. Very much like a vectorscope in an AVID or FCP or whatever. By all means, if you are concerned about luminance or blacks and have no alternative, use the thing - but otherwise route to your waveform and a calibrated monitor - which in a perfect world would be right next to your AVID - and check it out properly.

And once you've gotten used to calibrated gear and checking levels, you eventually will become accustomed to what is and is not technically acceptible. Another hint is that 100% white and 100% black give everybody the poohs - often this is used in credits - use a soft white and a not quite black and you'll fare better. That goes for all graphics in telly land, especially in compositing and animation, where often extreme highlights and shadows against one another can cause digital artifacting of the worst kind; as can certain bright colour combinations like bright blues and oranges, etc.

And this is how I procrastinate...I ought to be editing the newsletter....doh!

I'm already kicking myself for this TIP as I type, but as I'm aware how expensive scopes can be for a small shop.... here goes;

AVID TIP: - applying a colour effect to the top most layer will legalise your video levels to meet transmission standards.

here's the link to the avid help page that will absolve you of all responsibility



REASONS TO IGNORE THIS TIP: -
1: whites will clamp i.e. your subtle puffy white clouds will become as flat as an English sky (no offence DAZ).

2: blacks will clamp, all those nice shadows - gone, flat, black, featureless blob. (a bit like my last home brew)


4: Do you know what "limiting levels to 7.5 and 100 IRE." means ? ( don't be scared, you'd be suprised how many people don't...like me when I first started editing)

3: ... and here come the cranky old onliner in me... "rules is rules Jacko"
Years back, myth has it, whites greater than 1 volt (100%) used to cause major damage to microwave and satellite TX gear. While I'm prepared to be corrected on this, what I am very certain of is something that Rachel touched on regarding 100% white giving "everybody the poohs"

RANT MODE ON-
I live in an area of Brisbane that is 10 km from Mt Coottha (network Transmission points)

I also have a small mountain range between me and the TX towers

Every time a white super appears on screen on the TV set at home that has been transmitted with its IRE value over legal, the audio buzzes. That's right, vision quality effects audio, yin yang, its all connected.

If you are the editor responsible for "finishing" it is imperative that the product you produce meets all industry standards for this reason: -

YOU CANNOT CONTROL THE CALIBRATION OF THE RECIEVING VIDEO MONITOR.

The best we can do is make sure the program is in a near perfect state for playout.

Everybody has been to a movie where the projectionist has fallen asleep and audio has gone south and focus blurs.
Well, the filmmaker can't help that, but if the DP shot it soft, the Cam Assist got the shutter angle wrong and the CC didn't care. Foley put horse hooves where ducks were splashing, then you really have a reason to giggle and demand your dough back.

There's this bloke on the AVID-L who reckons that one day soon he'll be able to build a million dollar linear suite with stuff off E-bay for under 10 grand.
He may be getting close to his wish, 4 channel ADO $200 chyron $1000 GVG edit controller $1000 etc .... but he's still a long way off.
He'll still need tech monitors and scopes...and so do the rest of us, and always will, because we are the manufactures of vision, NOT the receivers.

RANT MODE OFF.

Davade, its a pity we're on opposite sides of the island, because as Rachel said it's so much easier to understand this gear if you can get a demo from someone who's got their head around this stuff. I'd love to help if i could.

mmmmm maybe time for some local mentor volunteering.....

WaZ.

[Davade]

Thank yoo so much WaZ, if i need you, I know who to call...After all these discussion, I think it voice down to one very critical issue that I would like to raise.

Most small production tends to ignore this simple step, the maintence of equipments and the importance of the calibration of equipments. These small setup not only don't have the required equipment to do the testing, most of the time don't even have a qualified technicians that does the setup in the first place. Most of the time is a matter of trial and error.

Hence, most of the time, the responsibility for maintaining the standard of picture quality falls into the hands of the editor. Has the job scope of the editor expands further? aka techncian/editor

[Daz]

"YOU CANNOT CONTROL THE CALIBRATION OF THE RECIEVING VIDEO MONITOR"

Reminded me of a story told to me by a VT shift leader at the ABC, who used to be a TV repair man...

He arrives at the house of Mr & Mrs D......, whose reception is poor. Checks the outside aerial. Seems ok. Thinks it might be the connection.

Opens the back of the TV set - to be hit by the god awful stench of rotting meat. Blinking back tears of revulsion, he sees the maggot-ridden remains of a sausage - jammed on to the aerial connection (which has slipped off the coax). He turns to Mr D......., who explains: "Well, when I touched that wire the picture seemed to get better, so we thought it might have been the meat! But the sausage didn't do the trick, so we called you..."

[Neil Ryan]

I wouldn't worry about all this, as I believe red is out this winter ...




tee hee!