Display Forum 2002
by John Latta
Wave Issue 0304 2/14/03

December 11 - 12, 2002
Nice, France

As the readers of the WAVE Report know we are a fan of the conferences put on by DisplaySearch. The quality of the market assessments are excellent and the industry coverage comprehensive. Display Forum was a joint conference organized by Meko and DisplaySearch. Although the focus was on European display market there was much to be gained from fresh market dynamic insights.

Having only come from OLED 2002 a few weeks before we were wondering - what is new to learn about in displays? Wrong. There was much to be gained from Display Forum.

Eastman Kodak

Les Polgar, President of Eastman Kodak Displays, gave the keynote on day two. It was an excellent summary of Kodak's view of the technology and market. Key points made include:

Kodak estimates that above $100m/year is being spent on OLED research. This is significantly higher that the estimate from OLED 2002.

There are 9 companies that have or are shipping OLED displays as of December 2002. Of these, 8 of the 9 use Kodak's small molecule technology. The companies are: Pioneer, TDK, SNMD, RitDisplay, Nippon Seiki, eMagin, Sanyo and Kodak.

Although there is a perception that small molecule and polymer technologies differ in the following ways:

Rigid vs. flexible substrates
Glass vs. plastic
Vacuum deposition vs. solution processing

Les cautioned not to assume that small molecule cannot be done on flexible substrates. Thus, to bin the attributes of polymer as exclusive of small molecule would be a mistake. Note the announcement of the previous day by UDC with DuPont for solution based processing. (see below)

Kodak does not quote lifetimes openly because there are so many factors present. He cited some of these factors as: initial luminance, color or white, test cell or actual product conditions. One must understand all of these parameters to determine if OLED materials are suitable for an application.

He cautioned that the current business model for much of the industry, which separates suppliers from users, has fundamental flaws. Suppliers include: IP and those that sell materials to licensees. The users are those that make and sell display products. The issue centers around the objectives of each segment. While suppliers, such as CDT, are focused on a "liquidity event," the users seek to make profits from production. Les stated we have seen both sides. First, when Kodak only licensed its IP and then when the joint agreement was put in place with Sanyo where it became a user. This fundamentally changed how Kodak viewed the market because, in part, it now had to make the technology work in products.

One issue lost on the IP companies is that they must focus on total systems cost. This is what their customers see--not only, for example, just the cost of the materials.

Les held up the Nintendo Game Boy into which had they had inserted an OLED. This was the same unit widely shown in the SID 2002 booth. He cited this as an example of where OLEDs do not have an application. "This is a bad example for OLED because the price situation does not align. In order to do this a supplier company would have to sell at a tremendous loss." However, these comments were inconsistent with his remarks that the industry needs to get real about applications and recognize where OLEDs can be applied and where they do not fit.

Kodak developed a roadmap for the application of OLEDs in March 2001 for its management. Some 21 months later this in unchanged. Les cited that track record is quite consistent - it takes from 30 - 39 months from the time a demonstration of technology is made to the time it takes to go into production. According to his road map the following is expected:

2002 - Cell phones which use 2.2" OLED displays
2003 - 2.16" and 4" displays for cameras and camcorders
2004 - 5" to 7" displays for handheld TVs and DVD players
2005 - 9.5" to 15" full color displays for notebooks and PC Monitors
2006 - >15" full color television

The cost of the Fab 1 and Fab 2 OLED facilities was modest by most standards - $350m.

The comments they received over and over from CEATEC and the OLED television they showed was consistent - "This is the best television we have ever seen."

Now it is the time for the industry to buckle down and do the hard work making products with efficient processes.

The display division at Kodak has now moved to a group president and this is reflective of the increasing importance that displays play at Kodak.

Ross Smith - FPD Forecast

DisplaySearch and the presentations set the stage for the market dynamics which will create a major shift in display technology and market by 2005 - 2006. TFT LCD capacity will go from 10m sq m in 2002 to 40m sq m in 2006. This is a 41% growth rate. Yet, as this capacity is brought on-line the companies are investing in LTPS so that either OLED backplanes can be created or TFT LCDs. LG/Philips is investing in 3 5th generation plants for AMOLED. The driving force behind these huge investments is television set market. In 2005 it is expected that 180m televisions will be sold. Ross Smith, President of DisplaySearch, showed an economic model for LCD TFT production. He forecast a 32" FPD television for $999 in 2006. Ross considers this the sweet spot of the market where demand will be dramatic and it is from this point the FPD televisions will become a mass market reality.

Using that as the foundation of his talk, Ross laid out the current and projected spending for FPD fabrication facilities to 2006. This provided a capacity forecast based on the attributes of the facilities. From this he then estimated the unit and revenue for FPD. Then came the supply and demand assessment. Part of this was a DisplaySearch model for forecasting the cost of FPD modules - 32" and 40". The key points reached were:

6th and 7th generation fabs are required to make a profit from these large panels. 6th generation begins with glass substrates of: 1500mm X 1800mm while 7th generation is: 1800mm X 2100mm.

Taiwan and Korea is expected to account for 80% of the production for large area panels by 2005.

However, in the small to medium panel market Japan will retain 76% share to 2006.

The past crystal cycles, where over capacity created low prices and losses and then shortages followed by higher prices and another fab build cycle, is expected to dampen significantly from 2002 to 2006.

By 2006 15" XGA LCD monitor panel pricing is expected to decline to $25 by 2006, while 18" panels will be approximately $220.

The major market target is the entry of FPD into the television market. For both the 32" and 40" panels profits can only be made with 6th and 7th generation fabs.

OLED remains a dark horse. If its many issues are resolved, especially lifetime, OLEDs offer some significant advantages. These include: compelling image characteristics and the potential for lower cost. It was stated that LCD TFT costs for a 42" display are on a par with PDP of the same cost, the reason being the backlighting cost. This is very significant in that unless the component costs can be significantly reduced TFT LCD panels will be competing with other existing technologies. Yet, OLEDs, because they do not need backlighting, offer a significant advantage.

It is expected that there will be considerable push towards LED backlighting, also to reduce the cost in LCD TFT displays.

By 2006 the FPD market will reach $63b.

Screen Technology

A start up in Cambridge, England, Screen Technology, presented a tiling technology which is based on using conventional LCD TFT panels. This uses low cost plastic light guides to effectively expand the area of the panel to allow multiple panels to be combined.

Tony Lowe, a non-executive director gave an interesting talk on a different approach to tiling displays. The essence of the technique is to use conventional LCD panels. In front of the panel is placed an ITrans module. This is essentially a light pipe assembly which fans out from each pixel on the TFT LCD panel. The new panel, defined by the front screen of the ITrans module, exactly fills the dimensions of the tile which, at least, extends beyond the bezel of the TFT LCD. Thus, multiple modules can be combined adjacent to each other with precise alignment.

The difficulty with tiling is that the edge problem is quite difficult to solve. The requirements include:

Invisible join between tiles
No detectable changes in luminance across the display
The angular distribution of light must match for each adjacent pixel which are in alignment between the tiles.

These are very difficult to accomplish due to the sensitivity of the eye to discontinuities in brightness.

Other companies have attempted such displays but failed in part due to the edge requirements. Other issues include color fringing at the edges.

So far they have assembled a 2 X 1 prototype and are working on a 4 X 3 52" viability demonstrator both of which use 12.1" displays. They do not expect to show their technology until SID 2004.

They claim, due to the construction of the ITrans design that the depth of the ITrans and the $/m are independent of the display diagonal.

The company was initially funded with 500,000 £ and seeking funding in Mid 2003.

Although they would not discuss the cost of the module I learned from other sources it the adds about 15% to the cost of the display. It is claimed the ITrans module can be easily fabricated from injection molded plastic.

UDC (Universal Display Corporation)

During their presentation UDC announced a joint development agreement between it and DuPont to create soluble small molecule phosphorescent OLED materials. It is intended that phosphorescent, printable OLEDs could result. Note that to-date when small molecule materials are put into a solution the light emitting characteristics stop. This agreement is an indication that these companies may have a solution, pun intended, to this problem.

Sony on Wireless Strategy

Much to our surprise Sony's Marketing Director for Information Technology, Andreas Ditter, gave a talk on Sony's Wireless Challenge. Yes, it was out of place at a display conference but we found it interesting none the less. The theme of Andreas's talk was on Sony's strategy for wireless. Sony placed a strong emphasis on 802.11a/b in the home and the role it plays in creating the broadband home. He went so far as to state that Sony would have dual mode a/b for the home in Europe in 2003. Sony also made a strong case for Bluetooth. Many of the points that he made were an advance indicator of what Sony showed at CES.

Nokia

Antti Laaperi, Nokia, VP of Mobil Phones R&D gave an interesting keynote. One of the more provocative charts showed the growth in SMS usage on phones as a function of penetration by country. What was so striking is that the usage of SMS by the Nordic countries far out distanced those in Greece, UK, France, Italy and Spain. I described this as the Mediterranean effect and there was no good explanation of this offered.

Antii had as a premise in his talk, that phones would be moving to multimedia capability - specifically the support of imaging and transmission over the phone network. Yet, the SMS chart was cause for pause. That is, are we likely to see similar cultural issues on high end phones, with imaging for example, which are just not accepted by other cultures? In response, Antti agreed. Later in off-line discussions, he cited for example that Saudi Arabia could well ban such phones. This only emphasized a point that local cultures have a significant role in determining technology adoption.

In the PDA section of the conference much was made about the rush to converged devices. The obvious example is the PDA and the phone. In this case the display plays a central role in enabling for expanded functionality of such devices. I asked the question - why are we sure this is the case? Just the question provoked an extended discussion. In the end Antti had the best observation - Nokia is now on the third generation of such converged phones and they have found they are no more than niche products.

Selected Highlights

DisplaySearch predicts that the ASP for PDA displays will remain nearly flat from 2002 to 2006 at $40. I doubted this. The reason is that as Gen 5 & 6 LCD fabs come on line this will further free up Gen 3 and 4 fabs and this excess capacity will drive the prices down. Yet, Barry Young countered that converting a Gen 4 fab, the first to make large size panels >10", to one that makes 3" panels not so easy. He cited one example, where a Gen 4 fab is producing small displays but ships its sheet production to another company for cutting and final display fabrication. It is easier to do this than convert a large display fab to a small display fab, in spite of the generation of the fab.

An emerging trend, with a focus on China, are desknote computers. This is a computer in a portable form factor that just looks like a notebook. It is not a notebook but a portable computer that has the capabilities of a desktop. That is, there is no battery, the processor is a desktop processor, as is the graphics chip.

DisplaySearch cautions on the Tablet PC that the form factor and price are just not in line with what the market needs. The form factor is too bulky for the Tablet PC to feel like a pad of paper and the size should also be larger like a tablet of paper. The price is out of line, by being more expensive than a notebook - the pricing should go down not up from the notebook.

What is Missing?

In spite of all the high points, we were disappointed with the CD-ROM conference proceedings that was mailed subsequent to the conference. Contrary to the tradition at DisplaySearch, the presentations were in Adobe Acrobat format, rather than PowerPoint. As a result, one of the best slides at the conference, Ross Young's dramatic illustration of the size of the glass panels for Gen 4 to 7, was masked into one slide. The utility of the proceedings was significantly limited by the omission of the native presentation materials.