How Big is REALLY Big?
A few days ago, I read that Samsung had just announced that they
have decided on the size of their seventh generation mother-glass
for TFT LCDs. The planned size of these next generation TFT arrays
will be 1,870 x 2,200 mm. For those of us still mired in inches
and feet when thinking about display sizes, that translates to about
74 x 87 inches, or if we speak in the even more familiar diagonal-size
terminology that is approximately a 114 inch diagonal piece of glass.
Wow! Now that should be large enough for even the most ambitious
home entertainment system aficionados.
These numbers were so impressive that I decided to go back and
review the sizes of the other 6 generations. Most of the current
work is being done to refine the manufacturing processes for the
generation 5 and generation 6 factories. At the SID show in Baltimore,
I heard comments from some of our more knowledgeable colleagues
that even the generation 5 plants are still having difficulties
getting all the problems worked out. As I looked in various references,
it became apparent that there was and is no complete agreement on
exactly the size that defines each generation. I suppose that is
to be expected because these choices are somewhat arbitrary. Glass
is easy to cut to about any size one wishes. If the rest of the
production equipment has some flexibility, then there is no absolute
standard necessary. Nevertheless, the numbers that seemed to be
in most common usage were the following. Generation 6 is 1500 x
1650 mm or sometimes 1250 x 1650 mm. The larger of these dimensions
converts to a size of about 60 x 70 inches. Generation 5 most commonly
appears to be specified as 1100 x 1250 mm, and generation 4 is 730
x 920 mm (29" x 36"). Generation 3 (sometimes also noted
as generation 3.5) is 590 x 670 mm (23" x 26"). With a
34" diagonal, even that is not all that miniscule. The discussions
of earlier generations have almost disappeared into the history
books. And I suppose the pioneering researchers working with these
really small pieces of glass did not realize that they were supposed
to call them generation 1 or maybe even generation 0.1.
At this years SID Symposium in Baltimore (May 19 23)
large displays were everywhere. This should be no surprise to readers
of this column because we have been predicting for some time now
that large-screen television is about to become a major growth opportunity
for the display industry. What, however, has caught many in the
display industry by surprise is the recent rapid evolution of large-screen
LCD panels. At an invited talk that I gave a few years ago in Japan,
I mentioned that we should be prepared to see serious competition
between LCDs and plasma panels not too far into the future. Im
quite sure that not many in the audience believed me. Well, that
day has arrived!
At the SID Exhibition, Samsung showed a 46-inch LCD with 720 x
1280 resolution and an even more impressive one with a 54-inch diagonal
and 1080 x 1920 resolution. These displays were placed so that it
was possible to make an easy visual comparison with similar size
plasma panels from the same manufacturer. Im not sure that
I can quantify what I observed, but if I were asked to make a choice,
I would have selected the LCDs over the plasma displays. In fact,
this was the first time that I had to admit that if only the display
were visible to me, from a stationary position I would most likely
not be able to tell whether I was looking at an LCD, a plasma display,
or even a CRT. The display was as perfect as I could observe without
making detailed measurements. Perhaps what attracted me to the LCD
panels was their superb contrast and gray scale. The whites were
clean and the blacks were properly black. The colors were bright
and crisp but not overly saturated. The overall brightness was at
least as high as the comparable plasma panels and perhaps a little
higher. The back lights being used on these panels must be spectacularly
bright. Some warmth could be felt near the front surface.
By a small but significant margin, I would rate these as the most
impressive displays I have ever seen. What makes this statement
even stronger is that for the first time ever I honestly had to
admit that I would no longer be able to say that a CRT is still
better. The only possible quibble could be with the speed of response,
but I had no way to observe that with the program material being
shown. However, I suspect that this would not be enough to dissuade
me from trying to take one of these displays home with me.
At 54 inches, on the 16:9 format diagonal, we are at about the
size that most big-screen television enthusiasts will find exciting
for the typical US home. Eventually, 60-inch displays may also become
interesting as prices continue to come down -- or even perhaps something
as large as 70-inches. But once these become available we will have
covered the significant part of the market. So how big should the
mother-glass eventually become to accommodate these future needs?
Is there a limit to this growth? What will generations 10 and higher
look like? Linear extrapolation is clearly a dangerous way to try
to predict this future. The increases in area from generation 3
to 4, 4 to 5, and 5 to 6 were each by about a factor of two. The
increase from generation 6 to 7 being planned by Samsung is "only"
an increase of about 50%. The size race may be slowing. Perhaps
the next steps will be a change in aspect ratio and the move to
a more continuous production process instead. A logical improvement
would be to put the glass factory at the input end of the TFT factory.
That way the glass can be made in a virtually continuous process
and it can be delivered to the LCD factory with no handling, packaging,
or transportation costs.
In the early part of the 20th century, our parents and/or grandparents
lived through a period of industrial growth symbolized by the competition
for building ever-taller skyscrapers. However, when heights of 100
stories were approached the challenges apparently became greater
than the benefits. Similarly, there was a period of building ever
larger and longer bridges. But these too had their practical limits.
However, that then led to other innovative ways to accommodate our
needs. And so it will be with the scale-up of display technologies.
We may be nearing the limits of growth for single glass panels,
but we will continue to improve the production processes so as to
achieve lower selling prices.
The competition between LCD, plasma, and projection technologies
will likely intensify motivated by desires of survival and
market dominance. Will we end up with a number of successful display
technologies, or will one become dominant maybe too dominant?
It is a little bit too early to make a strong prediction just yet,
but I will predict that in one to two years we will have enough
data to get a much clearer view of how this large-screen competition
will turn out.
I would be most interested to hear your observations on the SID
show in Baltimore or anything else that you may wish to offer on
this topic or others. You may reach me directly from this web-site,
by phone at 425-898-9117 or by fax at 425-898-1727.