Utilities Professional 1-1500/Utilities Professional 1-1500 (1994)(WPD)[!].iso/12511500/var1418.dms/var1418.adf/CVTS.doc

Text File | 1992-09-02 | 45KB | 838 lines

                 The Complete Video Test System
                           Dave Muse
                          Version 1.0
                          March, 1994


INTRODUCTION

     The Complete Video Test System is a collection of twenty test
patterns that you can use to evaluate, troubleshoot, or compare video
equipment.  Many of our tests are unique, and can be found nowhere
else.  This text will explain the tests in clear, non-technical
language, and show you how to use your Complete Video Test System to
make purchasing decisions, adjust your equipment for optimum
performance, or just demonstrate the quality of the system you have.

     It is important to remember that, just as the strength of a chain
is limited by its weakest link, the results of tests performed on your
video system can only be as good as the poorest piece of equipment
that the video signal must travel through. I will point out, as we go
along, what types of problems with your system could result in low
scores on the various tests.

     In addition to helping you evaluate your system, I will be
telling you how to make adjustments to your TV or monitor to improve
it. These adjustments are optional, as you may have to remove the back
of your set to get some of them done.  Many sets have an interlock to
disconnect the power cord when the back is removed. You can defeat
this by removing the end of the cord that is attached to the set's
back and plugging it directly into the chassis (if you observe how the
back goes onto the set, you will see how it connects the power cord to
the chassis).  THIS INTERLOCK IS DESIGNED INTO YOUR SET FOR A REASON!
Working in back of your set with the power on can be dangerous!  Be
careful not to touch anything except the controls we are talking
about, stay away from high-voltage areas (the high voltage cage and
the picture tube), work with one hand only on the set (and keep the
other one away from anything metal), and sit or stand on a dry,
insulated surface while you work (like a cushion or carpet). Last and
most important, don't sue me for your electrocution!  I've just warned
you, and you're on your own!

     CVTS is available as software that runs directly on an AMIGA
computer system, or as a videotape playable on any of several types of
VCRs.  If you have the option, better results will be obtained by
using CVTS directly from your AMIGA system rather than from tape.
Most of the tests, however, will work just fine from tape with no
compromise in function.

USING YOUR SOFTWARE

     When using the software version of CVTS, remember that these
tests are designed for NTSC television systems.  This means you will
need a genlock or NTSC encoder if one is not built into your system.
The black-and-white video output on the AMIGA 500/2000 will work
with a number of the tests (the ones that do not require color), and
in fact I recommend using this output for as many tests as possible,
if your genlock or encoder is not a high quality unit.  The tests you
will need color composite output for are:  COLORBARS, COLOR PURITY,
CHROMA RESOLUTION, CHROMA NOISE, COLOR CRAWL, and the MULTIPLE TEST.
Black-and white output can be used for all the remaining tests, even
when you are looking for the color symptoms mentioned and making the
requested color adjustments.  Users with monochrome only video output
that do not wish to obtain an encoder can purchase a tape to run the
color test patterns.

     If you want to try some of the CVTS tests on your RGB computer
monitor, be my guest.  All you'll prove is how incredibly superior RGB
is to NTSC!

    The tests consist of pictures and animations displayable with any
IFF picture/animation viewer.  An easy way to tell whether a given
test is designed for a TV or a VCR is to look at the icon.  A drawing
of a monitor is represented on some of the icons, and a video tape on
others.  A test that is intended for both will have a drawing of a
monitor with a VCR on top.  One test, the camera resolution chart, has
a camera lens represented on its icon.  This video camera test is the
only one that needs to be printed before use.  The easiest way to
print it is to run Graphicdump (it came with your Workbench, in the
utilities drawer) and display the test.  It could also be loaded into
a program like Deluxe Paint and printed from there.  This test is
specifically designed to work well regardless of the quality of the
printer you have.

GUARANTEE?

     The video tests that make up CVTS are designed for informal use
by both video professionals and non-professionals alike.  CVTS is not
and cannot be designed to the same rigorous technical specifications
as professional test equipment, and results from the former should not
be given the same weight as results from the latter.  I can't
guarantee that you will make correct video purchasing decisions just
because you own CVTS - as much as I hope and think you will.  I can't
guarantee that you won't damage yourself or your video equipment in
the course of using CVTS (please be careful)!  You'll have to assume
the full risks for these things.  So what do I guarantee?  I want you
to be happy with CVTS, so if you've purchased this package or anything
else from me and you're not pleased with it, you may return it for a
refund at any time.


DISTRIBUTION NOTICE

     CVTS is copyrighted.  This package is freely distributable, but
not public domain.  Distribute it as you wish - just keep everything
together, this document included.  Since this version of CVTS is
shareware, you would (hopefully!) feel an obligation to contribute
money to use it.  Instead of a guilt-inducing diatribe from the author
about $$, how about this:  If you like this product and plan to use
it, I hope you will consider purchasing a colorbar viewing filter,
printed manual, videotape, or other product from me.  They will
enhance your copy of CVTS and make it more useful to you.  Purchasing
any product (see the included order form)  will register you and put
you on my mailing list for updates to CVTS.  Yes, new tests and
improvements to existing ones are in the works.  I am also quite
receptive to feedback.  If you spot something wrong, don't curse me -
write me a letter instead!

THE TESTS

     Here is a brief description of each test and instructions for its
use.

COLOR BARS

     Everybody has seen the color bar test pattern before. It's a
standard test signal, used throughout the television industry, for
adjusting video signals. Professional television monitors frequently
have a switch, labelled "blue only" , which is meant to be used with
the color bars for precise adjustment. By using the blue viewing
filter supplied with your kit, you can adjust your TV or monitor the
same way professionals do. Here's how it works:

(1)  Find the control labelled "Color" on your set (it can also be
called "Chroma" or "Color Level").

(2)  Display the Color Bars on your monitor or TV, then look at the
screen through your filter.  CONTINUE TO USE YOUR VIEWER for steps two
through five. The odd-numbered bars should appear blue, while the
even-numbered ones are very dark, almost black. Observe that each blue
bar has a smaller blue box just underneath it.  While adjusting the
COLOR control, compare the smaller boxes of the FIRST and LAST bars to
the larger bars above them. When they are exactly the same brightness,
the COLOR control is set correctly.

(3)  Find the control labelled "Tint" on your set (this control is
sometimes called "Hue" or "Phase").

(4)  While observing the two MIDDLE blue bars, adjust the TINT control
so that the smaller boxes underneath them are the same brightness as
the larger bars.

(5)  If the smaller boxes underneath the FIRST and LAST bars no longer
match the bars above them, then go back to step one and repeat these
adjustments.

(6)  Put down the blue viewer.  Look for three small black bars in the
bottom right quadrant of your screen.  If you can't find them, turn
your brightness up.  When your TV is properly adjusted, the third
(rightmost) of these boxes will be just barely visible - and the other
two, invisible.  Adjust your contrast and brightness controls until
this happens.  The reason you don't want to see the first black box is
that it's actually "blacker" than black.  Normal video material will
never get blacker than the middle box, which is official television
"black".

Your TV or monitor is now perfectly adjusted to display color video.

GREYSCALE

     Providing that you have just set up your monitor using the color
bars, you should be able to distinguish all of the shades of black,
grey, and white in this test pattern.  If not, try adjusting your
CONTRAST and BRIGHTNESS controls (some sets have a PICTURE control
instead of BRIGHTNESS).  Remember that the darkest shade of grey
should be just barely visible apart from the black background.
Observe, as well, whether inappropriate color tints come and go as
these controls are rotated through their ranges.

     If some or all of the shades appear to be tinted with a color, it
may be a sign that the "grey-scale tracking" or "color balance" of
your set is out of adjustment. You can adjust this yourself, but you
will need to look around for the controls in the back of your set. It
may be necessary to remove the back cover to find them.  What you need
is a set of adjustments labelled "RED SCREEN","BLUE SCREEN", and
"GREEN SCREEN". Alternatively, there may be adjustments called "RED
DRIVE (the word GAIN may be used in place of DRIVE),"BLUE DRIVE", and
"GREEN DRIVE".  If BOTH sets of controls are present, remember that
the SCREEN controls should be adjusted while observing the DARKER
shades, or black, and the DRIVE controls adjusted while looking at
LIGHT shades, or white.  If you haven't done something like this
before, be careful, especially if the back of the set is removed.
Don't touch anything inside the set besides the controls you need
(high voltages are present). Use a mirror so you can see the screen,
and adjust the controls slowly.  If turning a knob doesn't produce the
effect you want, turn it back to where it was.  A minor adjustment is
probably all you need, so try to find the RIGHT knob to turn instead
of turning ALL of them!  Obviously, if your picture has a red cast,
you will want to turn down the red control a little bit. If it has a
yellow cast, though, you would need to add some blue. That's because
of the way the three light primary colors mix to make white. Here's a
quick guide:

SINCE THE                       YOU NEED        OR YOU NEED
PICTURE LOOKS TOO:              TO ADD:         TO REMOVE:

RED                         BLUE AND GREEN         RED
BLUE                         RED AND GREEN        BLUE
GREEN                       BLUE AND RED         GREEN
YELLOW                           BLUE          RED AND GREEN
CYAN (BLUISH GREEN)              RED          BLUE AND GREEN
MAGENTA (REDDISH PURPLE)         GREEN         RED AND BLUE

The above adjustments are for tinted shades of grey, white or black -
usually affecting the entire screen at once - not a "splotch" of color
that stays in one spot on the screen. If you have this symptom, or if
a grey square appears colored BUT its identical partner on the other
row is NOT colored or different colored, you have a COLOR PURITY
problem.

COLOR PURITY

     In this test, we are determining whether the electron guns are
hitting the right phosphors on the face of the picture tube.  What you
are looking for on each of the colored screens is a small patch of
"wrong" color - a green spot on the white screen, for example. Do not
be alarmed at seeing "streaks" on the colored screens that come and go
(if you are playing back this test from tape).  These are caused by
"noise" and are a normal part of the way tape reproduces colors.  Look
instead for a small-to-medium sized, fuzzy, stationary blotch of
color, probably near an edge or corner of the screen.

     The test consists of several colored screens.  Each screen should
be a solid color from edge to edge.  Is each shade of color consistent
over the entire screen, or does it vary at any spot?  Normally, color
purity has been checked and set at the factory - its adjustment is
located on the neck of the picture tube - and you should not need to
readjust it.  What a color purity problem usually means is that your
picture tube has been subjected to a magnetic field, and may need to
be de-magnetized, or degaussed.  Most newer sets have a "degaussing
coil" built in to take care of this automatically - yours could be
malfunctioning or not be adequate to demagnetize your tube.  See your
local TV repair shop to have your tube degaussed - or borrow a
degaussing coil to do it yourself.  In addition to using this test,
sometimes the GREY SCALE test works better for spotting subtle color
purity problems.

    It is important to keep sources of magnetism away from your TV
screen. Your picture tube can pick up magnetism from a telephone on
top of the set (the bell generates magnetism when it rings), a floppy
drive next to your monitor, a large loudspeaker closer than about two
feet, or even a vacuum cleaner running too close to your TV.

LINEARITY

     This test pattern will show you how accurately your monitor or TV
set can display things like lines, boxes, and circles. The first thing
you should look at is how well centered the picture seems to be. Can
you see about the same amount of each of the four circles in the
corners?  Next, note how level the picture is. If it appears tilted,
the YOKE (the large coil mounted on the neck of the picture tube) may
have been incorrectly installed at the factory, or else has become
loose and has twisted.  Now, look at the large circle. Does it look
perfectly round? If it's squashed or misshapen, that's a linearity
problem.  Observe that the lines appear straight and evenly spaced,
too.  The moving dashed lines will help you spot any sweep
irregularities.  Let your eye follow one of the dashes across the
screen.  Does it's length change as it moves?  Does it appear to
travel up and down a "hill" anywhere?  If you'd like to improve the
situation, look for two controls in the back of your set called VERT
LINEARITY and VERT SIZE. Use your mirror to see the screen while you
adjust them. You will find that these two controls interact with each
other - that is, they both affect the vertical height of the picture -
so you will need to adjust both controls simultaneously.  Use the
VERTICAL LINEARITY adjustment to make the large circle look as round
as possible while turning the VERTICAL SIZE knob to keep the picture
the right height.

     Some older TV sets and monitors have a PINCUSHION control.  Its
purpose is to compensate for the curved screens of some older picture
tubes by "pushing in" the scan lines at the top of the screen.  If the
lines at the top or bottom of your screen ONLY look bowed or wavy,
look for a PINCUSHION control in back of your set.


CONVERGENCE

     Since it takes a mixture of the three primary colors to create
all the colors we know, it probably comes as no surprise to know that
there are three "electron guns" inside your picture tube, one for each
of the primaries. The tricky part of all this is getting the beams
from those three guns to strike the screen in exactly the right
places.  If they don't line up just right, colored fringes will appear
around objects on the screen.

     Display the dot screen and look right in the center of the
screen. Do you see white dots or multi-colored dots? Most sets are
pretty good at getting the dots white in the middle of the screen. Now
look in one of the corners. If you look closely at one of the dots, it
will probably appear to be colored around the edges, or it may have
split completely into two or three colored dots.

     If the convergence of your set is particularly bad (that is, the
dots don't appear white even at a distance) you may wish to adjust the
convergence of your set. There are typically two sets of convergence
adjustments, called STATIC and DYNAMIC. The STATIC adjustments are
little tabs connected to magnets on the neck of the picture tube, and
do not usually need adjustment outside of the factory. They are used
to get the center of the screen lined up.  The DYNAMIC controls are
needed to get the screen converged around the edges, and may need to
be touched up every few years.  These controls may or may not be
clearly marked, so look carefully for a small group of electronic
controls, possibly on a circuit board of their own, near the back of
the set or even on the picture tube neck.  We're talking screwdriver
type adjustments here, so find a small insulated screwdriver and/or
coil adjuster to use. I have no ideal strategy for adjusting these, as
every TV or monitor is a little bit different, and the convergence
adjustments can be difficult. If you would like to experiment, then
use your mirror to see the screen, proceed carefully, and turn each
control slowly to see what effect it has on the picture. If It doesn't
do what you want, turn it back to where it was!

FOCUS

     I have provided a black-on-white dot screen for checking and
adjusting FOCUS, but the same dot screen used in the convergence test
above is also useful - try them both.  Do the dots appear as tiny,
well-defined little rectangles, or as fuzzy round blobs? Many sets
have a FOCUS control which is accessible without removing the back. If
you have such a control, try adjusting it while observing the dot
screen.  You want to make the dots look as small and well-defined as
possible.  If you find it impossible to get both the center of the
screen and the edges to look sharp at the same time, a compromise
setting of the control may be necessary.  Picture brightness often has
an effect on image sharpness as well.  To check this, see whether both
of the dot screens play with equal focus - and try turning the
BRIGHTNESS control up and down as well.  A well-designed and built set
will not lose very much focus as the brightness control is turned up,
but some sets lose quite a bit of their image quality. If yours is in
this category, you may have to choose between a bright picture or a
sharp one!

BLOOM

   Blooming refers to the tendency of the size of the RASTER (the tv
image projected on your picture tube) to increase with brightness.
When you run the BLOOM test, you will see a cross made of two
graduated lines, alternately black on a white background, then white
on black.  To see how much size variation your TV has, you will need a
tape measure.  Measure from point A to B on the black screen, then
again on the white screen.  Divide the larger number by the smaller.
Do the same thing again, measuring from C to D this time.  A good
score on this test will be as close to 1.00 as possible.


WINDOW

     The WINDOW test, despite its simple appearance, presents a
challenge to your video system.  Turn your BRIGHTNESS and CONTRAST (or
PICTURE) control up and down with the window test displayed and look
for problems.  Some things to look out for:
 - Are all the sides of the large white square straight and even, or
are there waves and kinks?
 - Are there streaks or smears running horizontally across the
picture?
 - Is the square white, or tinted with a color?
 - Do you see "ghost" images of the sides of the square (possibly
multiple ghosts, especially with the brightness turned up -- and with
the peaking or sharpness control turned up).
 - Are the square edges sharp or blurry?

     Good results on this test require a solid, flat response from the
electronics in all your video equipment.  A problem could be caused by
several things.  If there is a peaking or sharpness control on your TV
or VCR, make certain it isn't turned up too high.  If you are using a
monitor rather than a TV, make sure that it is terminated properly.
Check for a termination switch on your monitor.  If video "loops
through" your monitor on it's way to another piece of video equipment,
the last video connection in the chain needs to be terminated.

CENTERING

     A quick look at the concentric boxes on your screen should tell
you how well centered your monitor is.  Can you read the same letter
on both the right and left sides of the screen?  Follow the outermost
box all the way around the screen to judge your top/bottom centering.
     You can quickly gauge the amount of overscan your set has, as
well.  The outermost box is labelled "A".  Which letter can you read
at the edge of your screen?  For a more precise overscan measurement,
see the next test.

OVERSCAN

     It's almost impossible to frame a picture without covering a
small part of it behind the frame.  The same thing happens with the
video image displayed on your set.  The image is purposely made just a
little larger than your screen, so that you won't see the edges.
While a small amount of lost image, called overscan, is acceptable, a
good monitor or TV will crop very little off the sides of your
picture.

     To measure overscan in your set, run the test until the expanding
box begins to leave the screen. If you pause the tape (or stop the
animation, if playing from computer), you will be able to read the
percentage of overscan, or lost image in your monitor or TV. A reading
in single digits is acceptable.  This is a good time to check your
centering, as well. If all four sides of the box seem to be leaving
the screen at the same time, then your picture is centered perfectly.

     If you would like to adjust the overscan or centering of your
set, you may be out of luck.  Many TVs or monitors don't have
adjustments for horizontal size or centering. Check for yourself and
see.  Vertical size and centering controls are more common - but
you'll want to be careful about changing the vertical size of the
picture if you can't change the horizontal - squashed or stretched
pictures could result.  If your monitor does have a full complement of
controls (horizontal size, vertical size, horizontal centering,
vertical centering, and vertical linearity), then call up the
LINEARITY test screen and use the centering controls first, then the
size controls. Don't forget that you might need to adjust the vertical
linearity control as well, to keep the circles as round as possible.
Once you get everything right, call up the CENTERING test to fine-tune
the positioning controls.

     Suppose you've adjusted all the overscan out of your set, so that
all the edges of the raster are visible on the screen, and you still
can't get to zero on the overscan test?  This problem has to do with
the BLANKING in your set.  Blanking refers to the period of time the
scanning beam is turned off, so that it can be positioned for the next
line or frame of video.  In many sets, this BLANKING INTERVAL is
longer than absolutely necessary, and so may contribute to the
overscan in your set.

HORIZONTAL RESOLUTION

     This is an important test, and it applies to both your VCR and TV
or monitor. RESOLUTION means sharpness, and what we're measuring is
exactly how sharp an image your video system can display.  There are
separate tests for HORIZONTAL and VERTICAL resolution, as well as one
for CHROMA, or color resolution. We need all these separate tests
because different technical characteristics are responsible for
resolution in each case.

     HORIZONTAL RESOLUTION is typically measured as "Bandwidth" by
engineers. The concept of "Lines" is more easily grasped by consumers,
however, so camcorders, VCRs and monitors usually have their
resolution expressed in lines, that is, the number of vertical lines
that can be fit on the screen while still being able to distinguish
them.  This measurement is sometimes made in more than one way, so I
have included three sets of numbers on the test.  The top set of
numbers represent the traditional way of measuring LINES OF
RESOLUTION, which is the number of black and white lines that would
fit on your screen IF IT WERE SQUARE.  Since it is not square, but
rather rectangular, the bottom set of numbers indicates the number of
lines that will ACTUALLY fit on your screen.  From time to time,
measurements made this way will be quoted in specifications - which is
hardly fair if everyone else uses the traditional measurement.  The
middle set of numbers are for BANDWIDTH, expressed in megahertz.  This
method of measuring resolution is the most popular with engineers.
     The test is for the luminance part of the picture (the
black-and-white portion of the TV signal) SO TURN YOUR COLOR CONTROL
ALL THE WAY DOWN.  How far to the right can you look before the lines
blur together?  The highest number you can get while still being able
to see individual lines is your horizontal resolution.  If the lines
are dim, but you can still see them, it still counts (this means that
your response is "falling off" at these frequencies). Increase your
contrast control and back off the brightness a little bit to improve
your view of the lines.

     A good VHS VCR will only score in the low 200s on this test.  Some
formats (like S-VHS) are capable of scores of up to 400 lines, but
you'll need a good monitor to see it.  Most TV sets are only good for
250 lines or so.  If you have some good quality equipment, you might
be curious why you can see the 315 line grid, but not the 275!  The
reason is complicated, but it has to do with the fact that the part of
the TV signal that carries the color information has a frequency very
close to that of the 275 line grid.  This is why, if you turn up your
color control now, you will see crawling colors all over this part of
the screen.  Your TV is mistaking this grid for color information! If
your TV has a "comb filter", this grid may be a little clearer.  If
your TV or monitor has a special S-VIDEO connection to your VCR, it is
much more likely to be visible. This is why a special monitor is
recommended for S-VHS, HI-8 and similar formats.  If you are running
CVTS from the AMIGA software, you may notice a quality difference on
this test between the color composite output of your genlock or
encoder, and the monochrome output (AMIGA 500/2000) which contains
no color subcarrier (this will cause it to look better).

VERTICAL RESOLUTION

     This is a little tricky to explain, so bear with me. All TV
equipment has exactly the same vertical resolution. This is
necessarily so because this parameter is a part of the standard that
our TV system (the north american TV standard is called NTSC) is based
on. So what is the standard vertical resolution?  Well, this is the
tricky part. There are 525 scan lines making up one frame of video.
Some of these aren't used for video, so there are a little less than
500 left. This would imply a vertical resolution of almost 500 lines.
The catch is that it takes the electron beams in your picture tube two
passes to paint all these lines on the screen.  First the odd-numbered
lines are drawn, then the even-numbered ones.  This even-odd business
(called INTERLACING) makes it very hard to distinguish lines right
next to each other, because very thin lines will appear to flicker
like crazy!

     If you look very closely at the VERTICAL RESOLUTION test pattern,
you probably can see all the lines in all the grids, all the way up to
500. However, it would give you a headache to count them. This is why
the APPARENT vertical resolution of NTSC television is usually quoted
at 300 or so.  Lots of flicker means your equipment is working
perfectly.  Little flicker, or unusual difficulty seeing the lines
could mean that your TV is not interlacing properly, or your VCR is
not playing back both "fields" of the TV picture (mis-tracking, or a
dirty or damaged head are some possible causes).  When using this test
to compare TVs, record the highest number of lines you can EASILY see
(the APPARENT vertical resolution).

INTERLACE

     If you think about the the process of INTERLACING, where the
lines of the second field of video are drawn in between the lines of
the first video field, you might start to wonder if anything could go
wrong.  The answer is yes, things can go wrong, and lines of video can
be scanned in wrong places.  In particular, if there are brightness
variations from field to field, the raster can expand or contract -
putting the lines in the wrong place.  The result - fuzzy video.

     When you run the INTERLACE test, you'll see a screen that
flickers like crazy.  Start with the brightness and contrast on your
set turned down as far as possible, while still being able to see the
video.  Superimposed on this flickering pattern are the words "How
well can you read this?", repeated in five places on the screen.
You will notice that this message is white, with a black drop shadow
that just barely falls off to the lower right.  Now turn up the
BRIGHTNESS and CONTRAST (or PICTURE) controls to normal levels, and
observe both the legibility of the five messages, and the positions of
the black drop shadows.  Ideally, all five messages should look the
same, with the drop shadows in the same places - just barely falling
off to the lower right.  Otherwise, your monitor has some trouble
interlacing under the difficult circumstances presented by this test.

CHROMA RESOLUTION

     This test is just like the HORIZONTAL RESOLUTION test except that
we are looking at the resolution in the color portion of the TV
signal.  Make sure that the CHROMA (or COLOR) knob on your TV or
monitor is turned UP to it's normal level.  To score your equipment,
find the closest together set of bars that you can make out, and read
its corresponding number.  Don't expect a high score on this test,
unless you have professional equipment.  This is a good test to use
when trying out VCRs and monitors, as there can often be substantial
differences.  To test a monitor, I recommend playing the test direct
from computer, as your monitor probably has better resolution than
your VCR.

     Why is your chroma resolution so much poorer than your luminance
resolution?  The answer has a lot to do with the designers of our NTSC
TV system (and PAL, as well, a related TV system).  They knew, from
vision research, that humans see much more detail in luminance than
they do in color.  This is why there is much more bandwidth allocated
for brightness information than for color in the TV specifications.
Unfortunately, your TV or VCR will truncate still more of this chroma
information.

LUMINANCE NOISE

     NOISE is what you're looking at when you watch a TV station that
doesn't "come in" very well. Most of us are used to calling this
"snow" because that's what it looks like - a snowstorm in front of the
TV show we want to watch.  All TV signals contain noise, but usually
not in objectionable amounts.  To measure noise objectively,
professional test equipment is required.  However, we have developed
tests that work quite well for comparative measurements of both luma
and chroma noise.  Two different tests are needed because the video
signal has two parts: one carries the brightness, or luminance
information, and the other carries the color, or chrominance
information.  These tests are best used for subjective, comparative
purposes and will not produce meaningful numbers for professional use!

     To use the LUMINANCE NOISE test, first turn the CHROMA (or COLOR)
knob on your set all the way down.  If your VCR or TV has a sharpness
or peaking control, set it at the middle position.  If you have
digital noise reduction, or a "detail enhancer" on your VCR, turn
these gadgets off (they will hurt your score, not help).  Run the
test.  You will see a noise bar, with a significant amount of noise on
the right end.  Now, take note of the amount of noise visible on the
left side of the bar.  Next, look slowly across to the right until you
find the point where the noise seems to be just noticeably worse.  The
number above this spot is your score.

CHROMA NOISE

     The CHROMA NOISE test is just like the LUMINANCE NOISE test
above, except that you should now turn the CHROMA (or COLOR) knob on
your TV back up to its normal position.  Look across the red bar from
left to right until you find the spot where the noise in the color
seems to begin worsening.  Read your score just above this spot.

COLOR CRAWL

     The color TV signal, as we have mentioned before, is a mixture of
luminance (brightness) information and chrominance (color)
information.  An unfortunate side-effect of mixing these two signals
is that they can interfere with each other.  At the places on the
screen where this is happening, an artifact known as COLOR CRAWL or
DOT CRAWL appears.  Run the COLOR CRAWL test pattern, and look at the
left and right sides of the various colored boxes.  You might see a
pattern of dots, "crawling" up the screen just at the borders where
colors change, or, alternatively, it may look more like an ill-defined
crosshatch pattern.  If you see this effect on the left and right
sides of the colored boxes, and nowhere else, your set uses the "trap
filter" method of separating the chrominance and luminance.  If you
see little-to-no effect on the left/right sides, but some chroma crawl
on the tops/bottoms of the colored boxes, your set almost certainly
uses a "comb filter", a vastly better method of separating the
signals.  If you are watching on a set connected by an S-Video
connector to an S-VHS or HI-8 deck, you should be treated to a display
with NO chroma crawl at all.  Now you know why you spent all that
money on your video system!  You could find, as I did, that playing
your test tape on different VCRs, even through the same cheap TV set,
can produce widely differing results on this test - owing to the
quality of the VCR's encoding (the process of recreating the TV signal
from tape).  Don't be dismayed to discover that an expensive monitor
has more chroma crawl than a cheap one - this is probably a trade-off
that comes with the higher resolution of the better monitor.

CAMERA RESOLUTION

     If you have a video camera, you can perform horizontal and
vertical resolution tests on it using this chart.  First, you'll need
to have it printed.  To print from Workbench, open the printer
graphics section of preferences and set your ASPECT to VERTICAL and
SHADE to BLACK AND WHITE.  Now run GRAPHICDUMP (this program is in the
utilities drawer of your workbench) and immediately display the CAMERA
RESOLUTION chart using the IFF viewer of your choice.  Alternatively,
you may print from a program such as Deluxe Paint.  When your printer
is done, compare the finished chart to the one on the screen.  It
doesn't need to look perfect - as long as all the lines seem to be
present, it's fine.

    To measure the horizontal resolution of your camera, set your
camera on a tripod and tack the chart on the wall as straight and
level as you can.  Get the camera as close and perfectly centered to
the chart as you can, while still being able to focus.  Then, zoom all
the way in on the exact center of the pattern and focus there.  Now,
slowly zoom out, watching the pattern of lines.  When they have
blurred together so that you can no longer make out individual lines,
stop zooming out.  Look to either edge of the frame.  The number you
see there is your resolution.  Measuring vertical resolution is simply
a matter of turning the chart sideways, and repeating the above
instructions.

     It's possible that your camera won't be able to focus close
enough to the chart.  One solution is to print the chart larger, if
possible.  Otherwise, use the MACRO setting on your camera lens to
focus closer.  Since this will preclude the use of the zoom lens,
simply focus on the resolution brackets one at a time, until you find
your camera's limit.

DROPOUTS

    This test is designed for comparing video tapes.  If you were to
try to choose the best brand of video tape by running the CVTS tests
on them (the appropriate tests are: LUMINANCE NOISE, CHROMA NOISE,
CHROMINANCE RESOLUTION, and LUMINANCE RESOLUTION), it isn't likely
that you'll spot any differences at all.  Video tape is made to such
high specifications that any performance differences are going to be
extremely minor - unless you're comparing ultra-cheap made in China
dollar-store special tape to extra-high-grade super-VHS stock.

     The one area where you can often find substantial differences
between tapes is in DROPOUT COUNT.  A dropout is a small area of tape
where a signal won't record.  This could be caused by a speck of dust,
a small wrinkle, or an uneven distribution of oxide.  A dropout, on
playback, looks like a small - or occasionally large - horizontal
white streak on the picture.

     What this extremely simple test does is give you a chance to
count dropouts.  To use it, record it on the tape of your choice in
two spots - at the very beginning of the tape, and again in the middle
of the same tape.  Next, play back the sections of tape you have just
recorded, and get ready to count the number of white streaks you see.
The test consists of a countdown, followed by exactly one minute of
black.  When you see the black, begin counting dropouts.  When the
screen says stop, write down the number of dropouts you saw.
Typically, there will be many more dropouts at the beginning of the
tape than in the middle.  To get the best idea of the quality of the
brand you are testing, test several of their tapes bought at different
times, including some that have been used for a year or two.

    If you can't spot ANY dropouts at all, your VCR may have a DROPOUT
COMPENSATOR, a circuit that "fills in" dropouts so you won't see them.
Use a different VCR for your tape tests.


THE MULTIPLE TEST PATTERN

     You will recognize elements of many of the tests discussed above
in this test.  When using this pattern to evaluate a TV set or
monitor, first set the COLOR and TINT controls using the color bars,
then use the grey scale to set the BRIGHTNESS (or PICTURE control) and
CONTRAST.  Now, look at the grey scale... are the shades of grey
accurate, or tinted with a color?  Look at the dots around the edges
of the picture... how sharp are they?  Does the sharpness, color tint,
or picture size change when you turn the BRIGHTNESS or PICTURE control
up and down?  How many dots look mis-converged?  Examine the
transitions between colors on the color bars for dot crawl.  Look at
the large circle... is it round, or is it oval or misshapen?  Does the
picture appear level, and centered?  How much overscan does this set
have (count the number of dot rows you see - there are two columns on
each side and four rows on the top and bottom).  You can use the two
resolution tests if you like, but to be fair, if you can't run the
test directly from your computer, you'll have to use a SUPER-VHS or
HI-8 version of this tape to test monitors.  A plain old VHS machine
will always have poorer resolution than a TV or monitor.  I don't
recommend using the noise tests at all when comparing TVs, for the
same reason.  Any non-professional format VCR will create more noise
than a typical TV set.  If the score on one of these tests looks way
out of line, though, the TV could have a problem.

NOTES ON USING THE TESTS

     When comparing VCRs, the important tests are HORIZONTAL
RESOLUTION, CHROMA RESOLUTION, LUMINANCE NOISE, CHROMA NOISE, WINDOW,
and COLOR CRAWL.  Using the same good-quality TV or monitor each time
will give you the fairest result.

     When comparing TVs or monitors, all the tests are valid, though
the NOISE tests have been designed principally for VCRs - a typical TV
by itself (running the tests directly from computer) will score at the
top of the scale.

     Don't be led to believe that just because you've set your monitor
up to the CVTS colorbars, all TV programming will now appear in
perfectly adjusted color.  This may turn out to be the case, by the
way, but if it's not, it's because you don't have the colorbars
associated with each and every tape or TV show, as in the professional
video production world.  Some programmers (your local cable access
channel, for example) may be lazy about implementing good technical
standards.  Having your personal equipment properly set up does
nothing to assure that the signal entering it has been processed with
care.

HINTS ON SHOPPING FOR A TV

     You know what it's like.  Fifty-seven television sets, all
displaying the same TV show.  Should you buy the set with the picture
that looks best to your eyes?

     Believe it or not, sometimes the set with the worst-looking
picture is the best one.  If the program displayed on all those TV
sets contains NOISE (a VHS movie or a snowy cable channel, for
instance), then the set with the best resolution will be displaying
the most noise!  There's a relationship between noise and sharpness -
they go together.  The situation is analogous to playing a noisy old
78 RPM record through a set of expensive stereo speakers.  This record
might sound better played through an old, low-fidelity speaker that
doesn't reproduce every scratch and pop with annoying clarity.  You
wouldn't use an old record like this to buy a stereo system, would
you?  Don't forget that most sets have a SHARPNESS control.  You can
turn down the sharpness when watching those fuzzy VHS tapes, to get
rid of the noise.  If your set has the extra resolution, however,
you'll really appreciate it on high-quality noise-free program
material.

     If you can record the CVTS tests on tape and take it along, you
will find out a lot about the set you want to buy (you will awe the
sales staff, too!).  You may want to narrow down the field a bit
first, though, so here are some ways to analyze the sets quickly.

     Watch the picture as the program fades to and from black between
commercials.  Does it seem to change size, focus or color? (Bad
signs.)  Compare the pictures of several adjacent TVs to find and
eliminate the ones with too much overscan.  Look for dot crawl on the
sharp edges of strongly colored objects.  If you see it on the TOP and
BOTTOM edges more than on the SIDE edges, this is a sign of a COMB
FILTER (this is good).  Check for straight, level lines and round
circles whenever these things appear on the screen.  Next, look around
the edges of sharp bright objects (white lettering, for example)
whenever you can.  Are these edges clean, or are there "wrinkles",
"ghosts", double edges or unnatural looking white borders?  Look in
the corners of the screen for colored fringes around objects - a sign
of poor convergence.  Finally, try not to overlook a set that may have
been simply misadjusted by a salesperson.  You may even want to help
him/her out by properly setting it up.


ORDER FORM

The following items are available from Dave Muse Video.

You may print this form and simply circle what you want - or write on
a separate sheet of paper.

I am working on a list of benchmarks for various pieces of video
equipment - I will include this with every order.

COLORBARS VIEWING FILTER
Used in conjunction with the colorbars test signal for adjusting
monitors and TVs. - $5.00

COMPLETE VIDEO TEST SYSTEM - TAPE, DISKS AND MANUAL

The CVTS tests recorded on tape and ready to use for evaluating
monitors, TVs, and VCRs.  Includes printed manual, colorbars viewing
filter, camera resolution chart and the latest Amiga computer version
on disk as well.  Additional help screens explaining each test are
recorded on the tape.
VHS - $28.00
SUPER VHS - $34.00
3/4" - $36.00
Other formats - Write first

COMPLETE VIDEO TEST SYSTEM - DISKS AND MANUAL
The latest version of CVTS on disk plus a printed manual, camera
resolution chart and colorbars viewing filter. The manual contains
additional instructions and diagrams to make using the tests even
easier. - $16.00



Please circle the prices of the items you want above, and total here:

                                                  $________
Shipping: Add $2.00 per product ordered
(except for the viewing filter - no addl. charge.)$________

MI residents have to include sales tax:           $________

Add everything up here:                           $________


Mail this and any other correspondence to:

Dave Muse
22443 Leewright
Southfield, MI
48034