 |
The WAVE Report
Issue #0213------------------4/12/02 |
The WAVE Report archive is available on http://www.wave-report.com
--------------------------------------
0213.1 Hot Topics
0213.2 Story of the Issue
0213.3 3D
0213.4 Semiconductor
--------------------------------------
***CableLabs Issues CableHome 1.0 Specifications
(April 10)
CableLabs has issued specifications for CableHome 1.0, providing
manufacturers with a common approach to distribution of broadband services
inside consumers’ homes. It is anticipated that certification testing
of
CableHome 1.0 products will begin in the latter half of this year.
According to CableLabs, CableHome products will provide
a home networking
solution, including hands-off provisioning of the residential gateway
device, as well as tools that enable the cable operator to assist in
diagnosing and resolving home networking issues. In addition, CableHome
1.0 will provide secure configuration and on-going management of
residential gateways.
CableHome 1.0 also offers firewall management and upgrade
functionality to
help provide a secure home networked environment. Finally, consumers will
have a guarantee that service quality for their PacketCable-based
services, such as voice communications and interactive games, will be
maintained through CableHome products.
As in the specification phase of all of its interoperability
projects,
CableLabs has received leadership from its cable operator members and
support from suppliers to write the CableHome 1.0 interface
specifications, which are freely available on the Internet for
manufacturers to use as a guide to build equipment that would work with
cable-delivered services. The CableHome specifications can be found at
http://www.cablelabs.com/cablehome/specifications.html.
***Caligari Introduces trueSpace
for Adobe Atmosphere
(April 9)
Caligari Corporation, a provider of 3D web authoring technologies,
today
launched a custom version of its trueSpace5 3D authoring software for
Adobe Atmosphere. Currently available as a public beta, Atmosphere is
a
Web tool for authoring, viewing and interacting with virtual 3D worlds.
Caligari trueSpace5 is a 3D authoring tool focused exclusively
on the
needs of web content creators. It combines modeling, rendering and
animation power with a user interface based on direct manipulation of
objects. Used for product design, graphic design, video games, and web
3D-world design, trueSpace5 enables users to do real-time authoring in
a
hardware-accelerated, 3D perspective workspace.
trueSpace for Atmosphere will be available in March 2002 for $199.
***DisplaySearch 2002
by John Latta
US FPD Conference 2002
March 20 - 22, 2002
San Diego, CA
We go to so many conferences a year it is hard to get a
BINGO from an
event. This niche conference on flat panel displays (FPD) did it. It was
a
superb event that combined market analysis from DisplaySearch, who ran
the
conference, with talks from many executives in leading companies in FPDs.
There were 360 attendees at this niche conference. DisplaySearch
began
each session with an extensive set of market data on each application
category for displays. Then companies speak of their view of the
technology or market.
Here are some interesting market statistics from the conference:
Televisions (WW) - 1.7b
Televisions New - 130m
Notebooks (2002) - 30m - all with FPD
PDA/Smart Phone (2001) - 12m
Mobil Phones - 2000 - 437m; 2001 - 364m
LCD Monitors (2001) - 15m
LCD TV (2002) - 1.7m (estimate)
PDP TV (2001) - 200,000
Projection TV (2001) - 4.2m
FPD Market Outlook
There is a common misperception that FPD technology follows
Moore's Law
and nothing could be further from the truth. DisplaySearch has done an
extensive and continuing analysis of the pricing patterns for FPDs. They
call the pattern of pricing a crystal cycle. There is a cyclic
relationship between: supply, demand and pricing, whose period is
approximately 2 years. Yes, over the long term pricing declines but the
pricing effects are significantly modified by this crystal cycle. For
example, in spite of major declines in the prices of large-area panels
we
should expect that the pricing will rise in the next 12 months, and it
has
already begun. This could well lead to rises in the prices of Notebooks
and PC FPD monitors and, in fact, Sony has recently raised its notebook
prices.
This cycle is due, in large part, to the investment cycles
of the
companies manufacturing the panels. Typically they make investment
decisions in good times when supply is low and prices high, and by the
time the fabs come on line there is an oversupply of capacity, which leads
to significant downward price pressure. The demand cycles can be as short
as 6 months, with huge variations in demand in such a short period, while
the fab construction cycle is typically 18 to 27 months. Given the
improvements in process technology, as each new generation of process
equipment comes on line, the amount of capability rises significantly,
only serving to increase the over supply situation when fabs come on line.
A point raised by Ross Young, the president of DisplaySearch,
is that the
impacts of this cycle can be mitigated significantly by making a decision
to invest in fabs when the market conditions are worst not best. He showed
an analysis of the timing of fab construction decisions that indicated
that bottom market timing would result in higher fab utilization and
income.
There have recently been two such cycles: 1998 and 2001.
Analysis was
shown that these were very similar in terms of the impact in equipment
spending, revenue decline and length of the FPD surplus (excess capacity).
Here are some interesting points from the presentation:
FPDs, for the first time, are expected to exceed CRTs in 2002
(assumed on total $ market size);It is predicted over the period from 2001 to 2006 that LCD
TFTs will show a CAGR of 24% but that OLEDs will show a 110%
rate; andBy 2006 LCD TV will be the second largest market by revenue
behind PC monitors.
Notebook PC Market
Much attention was directed to Notebook PC market because
it provides the
foundation for the TabletPC market and it is also different from the form
factor of the PC.
The characteristics of the Notebook PC panels showed how
quickly this
market has changed.
In 2001 the market shifted from XGA to UXGA panels and at the
same time Q1 2002 was the first time that the average number
of pixels per panel passed 1m;
Another interesting characteristic is the migration to higher
pixel density. Already panels with 100 to 125 ppi are entering
the market with >150ppi predicted to enter in 2004. The higher
the ppi the more paper-like the displays;Also in Q1 2002 the average pixel density passed 100ppi; and
In Q1 2002 the bottom is predicted for notebook pricing. 15"
UXGA panels were at $300 and 12.1" XGA panels at $190. Prices
have already begun to rise form these points;
Some of the most interesting data was presented on the trends
related to
PCs based on Notebooks. These include TabletPCs.
Certainly one of the major changes include larger sizes and
different form factors. Other form factors include HDTV at
1920 X 1080.There are expected to be continuing developments in wider
viewing angle FPDs. I was surprised that this is not to allow
crowds to see the screen but to avoid color shifts in the
display in off perpendicular viewing on these large screens;Significant developments are predicted in what was called the
Rolling, Rotating, Swiveling and Detaching aspects of display
integration into PCs. Concept designs have come from Compaq
and HP that show some of the potential for display
flexibility. Credit was also given to Pace Blade where the
processing is in the display panel and the keyboard is largely
an accessory;Microsoft and its TabletPC concept is felt to likely be a
major spur for innovation in how the display relates to the
PC.
Sony
The head of the Sony VAIO business from Japan gave a very
aggressive talk.
He would not leave a copy of the talk due to its proprietary nature. The
basic premise is that "we are to create markets with our VAIO products."
When they introduced the first VAIO computer in Japan it raised the market
size by 11X. Sony intends to have VAIO notebook computers from Micro size
to the biggest. Wasting no time he pulled out of his bag three new
computers:
VAIO GR
16.1" UXGA display
P4 1.7GHz
Micro Surround Speaker system
5 - 6 hour dual battery lifetime
available this weekend
VAIO UX
14" screen
Ultra thin form factor - 17.1mm thick
5.5 hour battery life
802.11b built in
Japan 3/02 - Coming to the US
VAIO U
Tiny XGA to SVGA display
Smallest and lightest Windows XP computer available
Timing to the market TBD
Sony intends to continue to grow by creating new markets.
Sharp
Sharp has been in and out of the notebook market before.
They presented
the strategy behind the current efforts to reestablish this brand in the
PC notebook market. One of the most interesting aspects is the emphasis
on
very thin computers. This is part of their concept to create a Sheet
Computer concept.
Dell
This was a very interesting talk in that it described the
movement of
products down the maturation curve from leading edge technology to mature
technology. Dell placed the Notebook at the middle of this transition
and
moving to commodity status - which is were Dell plays best. The rest of
the presentation was focused on why the migration of notebooks to China
is
important in this market. The conclusion drawn was that Taiwan had peaked
out.
Some of the manufacturing statistics were very interesting.
A portable
computer factory in China will take 1 m sq. ft. of land and have from
350,000 to 1m sq. ft. of manufacturing space. This is the equivalent of
35
soccer fields. One of the advantages of China is that there is a lot of
space and that this affords great flexibility in the manufacturing layout.
It was concluded that by 2004 93% of the notebook production
would go to
China and Taiwan would have only 7%.
2ndEdison
The president, Chris Bradley, of this project and ID firm,
gave a very
interesting talk. The basis for the talk was conceptual designs which
2ndEdison did for HP. Chris began with the bold prediction that tablets
would reach mainstream within 36 months. However, the basis for this was
interesting and made sense. Key points include:
Stylus-only tablets will remain niche products and not be
mainstream;e-mail remains the driving application and tablet PC will not
be able to avoid the need to have a keyboard to support this.
No one will want to write anything more that the most simple
e-mail response, by hand, no matter how good the writing
recognition is;The hottest accessory in the PDA market today is the keyboard;
Thus, a migration strategy from Notebooks to a "tablified"
notebook makes the most sense. Specifically, a notebook which
can become a tablet does not cost that much more and having
such a configuration lowers the OEM risk while raising the
ASP;Such a "tablified" notebook could server as a notebook, a
presentation machine, a tablet, and even a desktop machine.
PDA and Cell Phone
This is the center of the small screen area market. However,
the messages
here were nowhere near as clear as those that came from the Notebook
market. A key reason for this was the continuing integration of the cell
phone with the PDA. Yet, from both a usability, price and product
development standpoint this makes little sense. All of the speakers
questioned how an integrated cell phone with PDA will succeed in the
market. Further, as the market numbers indicate the cell phone market
collapsed last year and the end of the era of the free cell phone has
happened.
LCD Monitors
There is an oxymoron in the LCD flat panel business: profits
from growth.
That is, when there is an oversupply and the market grows significantly
from dramatically lower prices, no company makes any profits. In spite
of
this, the most recent impact of dramatically lower display monitor pricing
has energized the industry to think about its future and the role that
displays can take. The industry has seen the impact of getting close to
making flat panels a commodity item. Thus, much of the discussion in the
LCD monitor session was about - what is the long range view? We summarize
key points.
The LCD monitor business is was about 15m units in 2001 and
LCD panels for notebooks also nearly 30m units. In the tight
times, such as those being experienced now, panel makers and
PC companies would much rather put the emphasis on notebooks
than LCD monitors. The simple reason is price. It is better to
make a notebook sale at $1,500 than a monitor sale for $400.
LG.Philips LCD stated that their corporate goal is to sell
more square inches of glass per person. At the same time, the
displays must meet a WAF - wife acceptance factor, which is
based not only on price but also design and functionality. In
fact, the industry needs to think more about displays as
furniture.With the flexibility in displays and their placement there is
no reason for the PC to be visible. Put it in the closet. All
that is required is the display and a keyboard. A docking
station for a display is out of date. What should be docked is
data not the display.The future of the display, according to Viewsonics, is the
smart wireless monitor. Wireless can be anything in the soup
of standards, however, later in response to a question
Viewsonics stated there is only one viable standard 802.11a
for monitor connectivity.There is a convergence, based on LCD flat panels, with the
monitor and the television. The view is that a LCD flat panel
should be able to connect to any input. (Note that this
directly conflicts with the division between CE monitors which
will employ disabling encryption and PC monitors which will
not. Read further.)Consistent with the thrust to increase the pixel density in
LCDs on notebooks, the same view applies to LCD monitors where
the thrust is to make the display approach film quality.It is expected that the home will also be a source for HD
content and this will only further push the demand for higher
resolution displays. In fact, this can be said to already
happening with the 2, 3, 4 and 5m pixel cameras which cannot
presently be shown anywhere near the original resolution in
digital form in the home.When the monitor goes wireless the usage pattern can undergo a
significant change. A panel may be indoor or outdoor, in the
car or with the PC, or connected to media anywhere.The first wireless monitors will go to market in 2002. It is
expected the additional cost will only be $300 or less. The
market will probably only be about 100,000 units this year and
could go to 1m + in 2003.The interest in wireless connectivity to displays is so strong
that the industry association, VESA, recently held a two day
symposium on digital panel interfaces. It is expected that
MPEG II and 4 will likely play a major role in these
interfaces.
Much of this discussion reminded me of the original CD-ROM
conferences
sponsored by Microsoft in the 1980's. Each year there was the expectation
that " this is the year of CD-ROM," and for many years it was
not. Then
when 1m units were passed the market went crazy. Now when LCD panels,
on
the desktop, have taken off because of favorable pricing, the industry
is
energized. More importantly off on the horizon is the TV market. Once
the
LCD display gets favorable traction in the TV market it has the
opportunity to transform many aspects of the home environment. Literally
there could be a display in every room and the usage goes well beyond
just
television.
Yet, coming out of the blue was a presentation by Silicon
Image. The
thrust was how a digital display interface, DVI, can be adapted for the
home. Thus, the home is moving to be the center of an all digital network
comprising 802.11a and b, IEEE-1394 and DVI. Lurking in the DVI proposal
for the home is HDCP - High-bandwidth Digital Content Protection. This
is
a keyed encryption standard that operates between the source, such as
a
VCR or STB, and the display. The purpose of this is to ensure that the
consumer never gets access to digital data that would move in uncompressed
from the transmitter to the display. No recorders are expected to accept
DVI so the encryption is just to protect the digital data from the
potential of getting intercepted by an unauthorized device.
HDCP works by sending keys between the two connecting devices
to
authenticate their respective security. Once the interface is established,
data can flow. If the source data is to be encrypted but the display is
not a trusted receiver, a message pops up on the screen stating this is
not a valid device to receive this content and nothing happens.
I spoke at length with Silicon Image in their booth and
was flabbergasted
at this proposal that has already gained considerable traction due to
its
strong Hollywood support.
When asked, where does consumer fair use rights fit into this
scheme, I was told to consult my attorney. Can you believe
this! What will consumers think of such a response. Warehouses
will fill with returned product.HDCP has the potential to reek havoc with consumers, stores
and manufactures if one buys a DVI display, say for your
computer or one that does not have the Silicon Image HDCP chip
in it. When it is connected to the DVD, STB or other DVI
output device the display just fails. Thus, back to the store
it goes. Silicon Image used this issue as a reason to argue
that the transition to their proprietary technology must go
very fast. All of the CE and monitors companies need to use
out chips now, was the response.
There will be another whole layer of encryption standards
which are expected to govern media transport within the home.
These are completely separate from HDCP and operate on 1394.
Such issues as enabling zero, one or many copies will be
handled by this transport.Time shifting, such as with PVRs, are not addressed with HDCP
and if PVR functionality is put in the monitor, this creates
problems for the content companies, i.e., Hollywood.The drive behind HDCP are the Hollywood studios. These have
already announced support: spe!, Fox, Disney, Warner Brothers
and Universal. Further, DirectTV and Echostar support it.A key problem is that HDCP splits the digital panel world into
two segments, just as we are seeing trends that the two are
getting closer together (LCD monitor and TV monitor). HDCP has
the potential of creating two worlds in the home - encrypted
and non-encrypted.All of these encryption or DRM schemes have the potential to
convey enormous complexity that the consumer could care less
about. Can you imagine having a discussion in the home on
encryption standards to determine why the audio video
equipment will not work?The prospect that a Tablet PC could play high quality video is
nixed with the intent behind HDCP. An open PC that plays
content from a trusted source would not be acceptable.
The fundamental problem with home encryption, DRM and dual
standards of
use (CP and CE) is that it is impossible for the content industry to think
of all usage scenarios. Thus, when the consumer tries something not
considered by the standards, either the usage fails, is of poor quality
or
Hollywood comes down hard on the usage. Thus, as Hollywood puts increasing
pressure on all aspects of the digital industry, both CE and PC, there
is
no way to avoid potentially high levels of consumer disgust.
Compaq
Due to lower prices for LCD FPDs, beginning in Q3 2001 the demand rose:
4X on 15"
5X on 17"
2 - 3X on 18"
If this momentum is to be maintained towards the adoption
of the LCD
monitor the following prices must be maintained:
15" mid $200
17" mid $300
Prices above this will likely significantly impact consumer demand.
OLED
Key points on this technology included the following:
Barry Young, DisplaySearch - OLED is the perfect display
technology for the TabletPC.DuPont displays, a polymer company, suggested that they may
have active matrix on glass in late 2003 or early 2004. TFT on
plastic could come as early as 2005.Both Universal Display Corp. and CDT (Cambridge Display
Technology) are technology aggregators. That is, beginning
with the technology they originally developed, they are on an
aggressive path to acquire valuable technology to complete the
OLED food chain from materials to final product. They both
openly seek other companies to become a partner. It is a one-
stop IP shop. This generates revenue while the technology is
slow to mature. The down side is that with these deals a
company picks sides. There is no assurance that any one
technology will win out over another. Further, there could be
market segments by type of material and processes used.
Politics, Law and Market in Displays
On the last day of the conference Gary Shapiro, President
of the Consumer
Electronics Association, gave the keynote. The WAVE Report also had a
chance to speak with Gary, and his insight into many of the copy
protection issues was especially keen.
Some of the points that Gary made, both in the conversation
and the
keynote, are outlined below.
No company matches the extent to which Disney will go to
protect its property. The Disney position is very clear - no
one will make a copy of anything it produces. The fact that
this conflicts with the fair use principle in the law is
irrelevant.The recent action by the Supreme Court to review the Sony Bono
Copyright Extension Term Act is important. Here is a summary
of the action of the court:At issue is the Sonny Bono Copyright Extension Term Act,
passed in 1998, which extended an author' s copyright by 20
years or until 70 years after death. By extending the
copyright period by an additional 20 years, US law was
brought into conformance with the copyright laws of the
European Union. This case has important implications
concerning the availability of copyrighted material on the
Internet and in digital libraries.The foundation of the monopoly status of arts and sciences is
protected in the constitution:US Constitution, Article 1:
Clause 8: To promote the Progress of Science and useful
Arts, by securing for limited Times to Authors and
Inventors the exclusive Right to their respective
Writings and Discoveries;The fact that the Supreme Court would agree to hear this case
was a complete surprise. Why would the court look at this? It
was speculated that the court was concerned that Congress had
overstepped its bounds with respect to "limited Times." In any
case, many parties have descended on this action by the court
to file "friend of the court" briefs.Why care about something so obscure as copyrights and timing
in this article on display technology? If the Supreme Court
does not support the limits on copyright extension many movies
and properties will go into the public domain, including some
from Disney. An expansive body of previously copyrighted
material could be become available in the public domain.The strident positions of the content producers has distorted
fair use, which many want to totally bypass, and what DRM
should encompass. For example, all government data is open, it
cannot be copyrighted, personal data is not copyrighted and
there are bodies of material coming off copyright. Here is
where a Supreme Court decision could have a major impact.
Gary's question was simple: Does DRM take these factors into
account? By implication the answer is NO. The focus is only on
those just shouting the most - the content producers.What is important here is that the content producers are
seeking to install a whole infrastructure in your home or
even office, which makes encryption for the protection of
content, pervasive. As we saw, even on the display
interfaces discussed above, which totally ignores the scope
of copyright protection, it is being pushed as a solution
to satisfy the limited intent of a few and to the detriment
of consumers.In his keynote Gary made a strong appeal for the audience to
write to Congress to oppose the prospect of a mandatory tuner
requirement. What is this?In the conversion of the broadcasters to digital television
they were grated free spectrum for the period of
conversion. This was to end in 2006 but there is also a
requirement that 85% penetration of digital television
would be reached. However, this conversion is going slowly
and Congress wants to speed it up. One approach is to
require a tuner for HDTV channels in every display (the
definition of what is a display is open to interpretation).
A first response to this is - you must be kidding! The
thinking is that this would have the impact of making
digital television much more pervasive and easier to get.
Note that there is a precedent for this when in the
1950's(?) Congress passed a law mandating that all
televisions be able to receive all stations. This is why
every television set has a UHF tuner section.On Thursday of the week of the conference, legislation was
introduced by Fritz Hollings in the Senate called the Consumer
Broadband and Digital Television Act. This could ultimately
require that all media be encrypted and thus impact computers
and consumer electronics. It is hard to believe that such a
ludicrous legislation would be introduced.In his keynote Gary also articulated a vision for displays
where one was in every room of the house.
PDP
Where is plasma in all of this? Is this not the dominant
display
technology for flat screen industrial and home DTV displays today?
Correct. DisplaySearch estimates the following:
Unit Shipments
2002 - 550,000
2003 - 1,000,000
2004 - 1,600,000
2005 - 2,000,000ASP
2002 - $2,100
2003 - $1,700
2004 - $1,400
2005 - $1,100
To me this really masks the fundamental issues:
High costs of the display due to the BOM for display
electronics; andPhosphor imprinting from even short term exposure to the same
image.
In discussions, DisplaySearch agreed with this assessment.
One point brought out is that when side by side comparisons
are made
between LCD and PDP displays, consumers pick PDP in 90+% of the cases.
The
reason for this is understandable - many of the PDPs are made using the
same phosphors that are in CRT television sets today.
A problem in the market is that the PDP companies are desperately
seeking
to increase volume with the expectation that volume will bring
profitability. However, there is not alignment between selling price and
costs. Prices are falling much faster than costs. Here is where the
display electronics plays a critical role. It is high voltage, uses
components not easily reduced in size and weight such as high voltage
transformers. It is these components that place a bottom of the floor
of
costs for the final display that LCD panels can just blow away.
FAB
This section was entitled: TFT LCD Equipment and Materials
Outlook. Some
highlights include:
LCD TFT Equipment Spending will peak at $7.689b in 2003;
Glass substrate size in 5 - 6th Generation fabs is as large
as:1.5m X 1.6m - nothing less than massive
Some of the equipment was so large, for a recent fab, that it
could only be shipped by ship;Handling large substrates is such an issue that shipping final
panels around can lower final yield, thus, at least one
company has placed its final product production at the end of
the panel fab line;The ability of the LCD TFT fab industry to leverage off of
equipment and experience in semiconductor fab is evaporating
with such large masks, deposition and glass surface handling.
Samsung
This presentation was an overview of the challenges to make
an acceptable
40" TFT LCD television display. They first began with a list of the
conditions for this market:
Full motion picture compatible faster than 1 f/s
Wide viewing angle with no grey inversion
High brightness > 450 ced/mxm and with color Gamut at least
>72% of NTSC
Lifetime of 50,000 hours
True Black and high contrast ratio
Cost effective
The talk then covered how Samsung addressed each of these
issues. The end
result was the following:
Accomplished - Full motion picture compatible faster than 1
f/s
Accomplished - Wide viewing angle with no grey inversion
Accomplished - High brightness > 450 ced/mxm and with color
Gamut at least >72% of NTSC
Unable to measure this yet - Life time of 50,000 hours
Accomplished - True Black and high contrast ratio
To be determined - Cost effective
These are significant accomplishments and illustrate how
far LCD TFT has
come. It also shows Samsung's determination in this market.
e-ink
e-ink has compelling technology that uses colored particles
to create
electronic paper. They had a number of samples that looked very good.
e-ink has recently extended its partnership with Toppan, in Japan, to
extend the technology and bring it to market. One promise of the
technology is the ability to make flexible displays. What was most
interesting is that the addressing of the electronic ink is done with
a
TFT back panel. Thus, having a TFT partner who builds displays is critical
to the market. For this, e-ink has as a lead Philips. This is another
example of a technology that is compelling, and possibly lower cost, that
is building on the TFT base of both fab and technology.
James Larimer
The last section of the last day was on Human Factors. James
Larimer of
NASA Ames Research Center gave a very interesting talk on matching the
bandwidth between the human and the display. This has important
implications also for compressing the bandwidth to wireless displays.
James stepped elegantly through an analysis that showed
the eye has the
following characteristics:
The central cone of the eye's vision, which is approximately 1
degree in extent, is capable of receiving 11.5Mb/s;The channel capacity of the eye, based on the optic nerve
fibers, is 1.2Gb/s;Big Bertha, the standard for best LCD TFT panels, made by IBM
has the following characteristics:22" diagonal
16:10 aspect ratio
Model IGM T221
Resolution: 3840 X 2400
Display density: 220dpiBig Bertha has the following characteristics:
1 deg FOV at 60Hz - 6.66Mb/s
Entire Display - 13.3Gb/s @ 60Hz
Thus, it is within reach that displays can be as James called
them,
eye-matched. However, as he also stated Big Bertha, when operating at
60Hz, consumes 60 times the bandwidth available on 100BaseT.
To respond to this problem, and also the related wireless
display
transmission problem, he laid out a number of approaches that both tailor
the eyes characteristics to the display panel. In particular, he proposed
that the display panels get smarter and include:
Self-refreshing memory;
Random Addressing for display inputs;
Asynchronous address of display cells;
Implicit pixel shifts to the right and left and up and down
(for scene motion);
Implicit Anamorphic scaling and
Multiresolution reconstruction.
Can you imagine this? Here we are just barely getting TFT
LCD out at
reasonable prices, the number of transistors per pixel in a LTPS panel
is
4 and it is being suggested that major additions be made both in panel
addressing and individual pixel processing. This is some ways away.
LCD Momentum
LCD is the 800lb technology gorilla. Floating around this
conference were
at least 10 different display technologies and some of it very compelling.
OLED is in this bunch. The chances of any of this succeeding, with the
possible long term exception of OLED, are near zero. The first battle
cry
is - that is not fair! This misses the point - LCD has a scale and depth
that is unmatched by any other FPD technology, market and infrastructure.
Consider the following:
In display technology there are only two viable technologies:
CRT and LCD. It has taken LCD technology nearly 30 years to
achieve its strong position and it is the only technology to
successfully compete against CRT.The development of LCD process technology has many parallels
with silicon processes in semiconductor fabrication. That is,
there is so much momentum behind silicon process technology
that any other technology, such as GaAs, only occupies niche
markets. Momentum behind LCD includes the following:The total accumulated experience with the technology and
its understanding cannot compare with any other technology
but CRTs;Skill base of experienced individuals - from fab
technicians to process engineers to research;Invested capital - the huge investment in capital for fabs
forces solutions which use these fabs because the
investments must be written off;Infrastructure that includes supplies to equipment to
distribution and all the elements that support the
technology and its components;However, as we have stated before, to assume that LCD
follows Moore's Law and the scaling of process technology
with silicon or even glass area is a fundamentally mistaken
assumption. Area of a glass panel is a visual and market
issue not a process technology issue.There is another parallel with the semiconductor industry and
LCD. In the semiconductor industry, largely driven by the PC,
there was the need for much higher bandwidth access to memory.
Intel, in its roadmaps, saw this as a major bottleneck to the
advancement of the PC. A clear technology leader was RAMBUS
and Intel made a major push to see that this became the next
memory standard. One might think RAMBUS being like OLED to
provide a context for this analogy. However, RAMBUS was
outside of the mainstream of memory technology that all of the
major DRAM memory suppliers had invested in and had
established IP relationships. RAMBUS came on the scene with
its high licensing fees and higher production costs. In spite
of Intel's support, the attempt by RAMBUS to gain a strong
foothold as the next generation memory technology failed.
Thus, even in memory, the established business and technology
foundation is hard to shake. We see many parallels, with the
much less mature FPD markets and technology. LCD FPD are like
DRAM.There is an oversupply of qualified suppliers. This is a
market share issue. With all of the companies being Asian, the
mantra is market share in spite of all else, including
profits. The view is that this ultimately will be a massive
market on the scale of semiconductors and televisions. Those
that dominate it, when commodity status is reached, will be
the long term winners - like Intel. No end of sacrifice will
be exercised to gain a dominant market position.We are entering the 6th generation of LCD fabs, with the
latest generation being done in Japan by Sharp, and with
commensurate huge investments. But these companies are
competing with Korean and Taiwanese companies with equal
commitments to market domination and fab investment. Thus,
where there is a down turn in the market, in large part due to
oversupply, this is seen only as a short term profit hiccup.
From the buyer side and the consumer perspective it is just
great. The vast majority have no idea who made the LCD panel
and the technology that underlies it. The market forces,
which shape pricing, are unknown to the average buyer - all
that counts is price.Technology refinement fills holes in LCD weaknesses. With the
momentum outlined above there is great incentive to apply huge
resources to tuning the technology to fit the problem. Samsung
gave a very interesting presentation on 40" LCD panels for
home television. The list of weaknesses of LCD in this market
were near depressing- poor color, blurring, lifetime and
suitability for movie display. Yet, Samsung took that as a
challenge and showed how they had addressed the issues (see
below). On multiple times we have seen where LCD has the net
advantage in a market because companies are willing to invest
the resources to refine and advance LCD technology to make it
work.All of this results in the requirement that any new display
technology must have an overwhelming advantage in an area
which LCD fails, a proposition very hard to fulfill. Consider
the following:LCD Strengths
Size and the ability to create sizes from 1" to 40"+
Portability
Density - displayed pixels
Color
Viewing Angle
Varied ApplicationsLCD Weaknesses
Flexibility - Actually Rigidity
Thickness - Not extremely thin
Cost - Market drives to lower and lower costsOne might put slow cycle time, grey to grey, but this is being
addressed to meet the needs of the television market. It is
also interesting to note that many of the weaknesses are those
that make OLEDs so attractive. However, what OLED lacks is the
momentum behind its process technology. Due to the
requirements for vacuum application of the emitters, small or
large molecule, OLEDs cannot gain from the scale of the LCD
process equipment development. In fact, LCD has gained from
the development of process equipment, up to a point, from
semiconductor fab.
Implications
LCD technology did not begin as a disruptive technology
but a demand
technology. The driving market for FPD technology was portable computers
and from this LCD was the winner. It could be said that LCD was the winner
that made the vision of the original Compaq Portable and Osborne I a
reality. One might respond that LCD are disruptive technologies against
CRT based televisions but the problem here is that the costs of LCDs,
and
virtually every other FPD technology, rendered them uncompetitive and
thus
not viable against low cost CRTs. The value of the portable computer
market was that FPD screens were essential and that buyers would pay a
hefty premium for lightweight portability. Thus, the market dynamics were
in place that provided a foundation for the development and refinement
of
the FPD market. LCDs are moving on to the display monitor market, due
to
favorable pricing, and off on the horizon is the television market. One
wonders is there a demand market for OLEDs? Could it be clothing. The
problem is that for most clothing the ASP is too low.
The size of the display market means that companies and
investors will
make huge bets in the gamble for the chance to succeed. Certainly OLED
is
a good case in point but we also have FED and PDP. Even PDP has achieved
its own measure of success in both the industrial and consumer markets.
However, long term it is doomed.
In technology, the only government blessed monopolies are
patents. This
has created the formation of what might be called "Patent Clubs"
where
they are very evident in OLED. This includes:
Eastman Kodak
- Small molecule
CDT
- Polymer
UDT
- Phosphorescent
Dupont
- Polymer
Yet these clubs have no assurance of success. Each of the
clubs seek to
provide a measure of comfort for its members by creating pilot fabs and
possibly even a few early products that have a tough time competing with
LCDs. The basic problem is one where without the massive scale of LCD
production it is very expensive to work process technologies, the basic
material technologies and competitive product creation against the 800lb
gorilla.
There is another factor also in the favor of LCD. Many of
the display
technologies rely on addressing individual display elements. Further,
some
of the pixel elements look like capacitors, which means that current is
the driver for the display element. Certainly OLEDs are the best example
of this. When this is the case, the well developed active matrix
technology from LCD is key; including LTPS as the process technology.
In
multiple talks we have heard of TFT being used as the underlying
addressing technology. This is a good news bad/news story. The good news
is that this leverages existing technology including the LCD process
technology, the bad news is that the alternate technology begins with
a
cost foundation that uses components of the technology that is being
competed against.
The continuing desire to dominate the LCD market leads to
absurd fab
decisions. For example under construction are various generation 5.5 and
6
LCD fabs. What is a generation in fabs? By most measures it is the size
of
the glass sheet handled, in much the same was as the size of the wafer
handled in semiconductor fab. Look at this data for the latest fabs under
construction:
LG
- 1000mm X 1200mm
Samsung
- 1100mm X 1250mm
Sharp
- 1500mm X 1800
Where are glass sheet standards? They do not exist. Thus,
there is this
mentality that a few extra mm will allow one company to have a scale
advantage over another. This is directly related to the maturing of the
industry. That is, competition is focusing on the entering glass size
and
not on process technology efficiencies as the primary means of gaining
advantage in the market. These inefficiencies result in many custom panel
sizes, different process technologies per plant and a delay in the
development of high volume, industry wide focus, on process improvements.
***InterSense Announces New Wireless Tracking Technology
(April 10)
InterSense, a provider of motion tracking products that
enable users to
interact with 3D computer-generated environments, is proud to introduce
wireless MiniTrax devices for the IS-900 Motion Tracking Systems.
Intersense developed the new wireless devices to be plug compatible with
existing IS-900 VET & VWT systems.
Designed to augment the current IS-900 systems, the devices
are connected
to a small, belt clip module containing a rechargeable battery and RF
transmitter/receiver circuitry. As with the original MiniTrax devices,
the
new wireless implementation is ergonomically designed for use with
existing data gloves and stereo glasses.
Wireless MiniTrax Technical Features
- Spread spectrum 2.4 Gigahertz RF technology for interference
free
operation
- MiniTrax Head & Hand tethered to small (3.7 inches
x 2.4 inches x 1.0
inches) belt worn RF transmitter/battery module
- Wireless Wand with integrated RF transmitter/battery for tetherless use
- Rechargeable lithium ion battery provides 6 to 8 hours
of continuous
operation
- Existing MiniTrax devices are upgradable to wireless implementation
- IS-900 systems can support simultaneous tracking of four
wireless
devices over large area
- Wireless technology approved by FCC (USA), IC (Canada)
and ETSI/SE
(Europe)
- Wireless Head & Hand Trackers available Q2 2002, Wand
& Stylus available
Q3 2002
***ASIC Market Takes a Turn for the Better in 2002
(April 8)
According to the latest report from iSuppli Corporation,
after a sales
slide in 2001, the Application Specific Integrated Circuit (ASIC) market
in 2002 is beginning to turn the corner, setting the stage for long-term
growth.
The report, titled WHITHER GATE ARRAYS, OR WILL GATE ARRAYS
WITHER?,
states that worldwide ASIC sales fell by 28 percent in 2001. The sales
slide was prompted by the downturn in the electronics industry in general,
and particularly in communications equipment, which depressed demand for
semi- custom chips, including gate-array and cell-based ASICs. The report
states that the year 2002 will bring a return to growth, although that
growth will be marginal at best, with sales rising a scant five percent.
As shown in the table below, ASIC growth will kick into
high gear in 2003,
and the market will fully recover by 2004, when revenue will finally
exceed the peak level of 2000.
Worldwide ASIC Revenue and Growth Rate Forecast 2001-
2006:
Year ASIC Revenue, Annual Growth
$Millions Rate
2001 $12,905 --
2002 $13,551 5.0%
2003 $16,125 19.0%
2004 $18,705 16.0%
2005 $21,070 12.6%
2006 $21,702 3.0%
Despite the return to growth, ASIC suppliers face new challenges
that
could diminish their sales growth rates over the next few years. The
greatest challenge may arise from the emerging fabless ASIC companies,
which are expected to mount increasing competition to the established
players. The increasing use of standard product solutions also may become
an obstacle to growth for the entrenched ASIC vendors.
While ASICs overall will enjoy sales growth, revenues for
gate arrays are
projected to continue their decline. Gate array sales will decline from
about ten percent of total ASIC revenues, or $1.34 billion, in 2001, to
less than four percent of ASIC revenues, or less than $1.2 billion, by
2006. This will occur despite efforts by some gate array vendors to
provide easy migration paths from more costly Field Programmable Gate
Array devices.
WHITHER GATE ARRAYS, OR WILL GATE ARRAYS WITHER?, from the
iSuppli
Specialized Semiconductors Practice, is a 20-page quarterly report that
provides a comprehensive analysis of the prospects for gate array ASICs
for the next five years. It also updates the annual forecast of the entire
ASIC marketplace, with quarterly projections through 2003 and yearly data
through 2006. This update also includes an in-depth look at Lambda systems
for optical communications -- one of the hardest-hit ASIC market sectors
during the 2001 downturn.
--------------------------------------
Copyright 2010 The WAVE Report
To subscribe to the WAVE Report go to
To unsubscribe also use the Wave Report Home page or send the preformatted UNSUBSCRIBE message:
Previous issues of WAVE, as well as other info can be found at
Comments on or questions about the WAVE may be sent to:
John N. Latta - Editor-In-Chief
The WAVE Report may be redistributed in full for individual readership and posted to newsgroups, Web, and FTP sites. This publication may not be reprinted or redistributed for profit. Short quotes are permitted but must be attributed to the WAVE Report.