It takes a lot of miscalculation to be out by over a year in consumer electronics, and right now executives from Toshiba and Canon must be starting to understand what it must have been like to be at Sony these last few, unsuccessful years.
Toshiba and Canon openly admitted that all the hoopla and predictions and conference gloating and investment that the two have jointly put into surface-conduction electron-emitter display (SED) TVs, is now so much hot air.
They will delay the launch of their SED until the end of 2007, and at Faultline we suspect that by then the technology will be dead, with combined losses across the two companies of more than $2 billion.
SED panels will go into manufacture in the middle of 2007, some 16 months from now. At a paper given early in 2005 Toshiba told the world that the price per panel shipped of SEDs would be around 67,000 Yen by 2010, something like $600, which, considering this is supposed to be the top end technology that is going to offer a quality ceiling to the TV market, is frighteningly low.
SED should work fine up to 50 inch devices straight away. They would be under 1 inch thick, and by 2010 that means they would retail for $1,200. Now it appears that Toshiba and Canon can’t hit those sorts of numbers. The plan was always to introduce SEDs at a much higher price at the top end, and then gradually find manufacturing efficiencies and drive the unit price of the entire TV market down.
What has happened, and we go back to our own predictions in the matter, is that prices have fallen faster in LCD and Plasma than Toshiba and Canon could have predicted. This surprises us not at all. Back in February 2004, we asked the question ‘So what do you think of these Flat TVs? Too expensive?,’ with an article that suggested that prices would fall faster than leading researcher DisplaySearch, and the equipment manufacturers believed, simply because everyone already in the market was doubling capacity, and about 5 new entrants were joining the fray. It was clear that the numbers just didn’t add up.
The common consent was that 2005 would be a 15 million unit year for LCD and Plasma combined rising to just 25 million units in 2008 with unit prices around $880 to the store and retail sales price of around $1,760.
We suggested that was too low, and that what might happen in reality was closer to 45 million units being by 2008. High availability of panels would crash pricing to around $500 shipped to the store, and under $1,000 retail. If that happens, then is it comparable pricing to the failed SED plan.
During 2005 the market beat the 15 million with 21.2 million units mark for LCD TV alone not counting Plasma, vindicating the Faultline opinion but even our estimates now appear low.
Currently we are on track for a 42 million unit year for LCDs alone during 2006, with Plasma still riding high at around 6 million units last year, sitting at the top end of the market, and with both markets set to grow 100% plus next year. That means that 2006 is likely to be a 54 million unit market for just these two technologies. LCDs already take 15% of the total TV market and plasma is holding up at 4%.
In the coming two years, LCD manufacturers will spend another $10 billion on next generation fabrication plants, which will bring down the cost for LCD TVs further and further, and take greater market share. It is easier for operations that are already earning $billions in TV sales to ramp their operations with this kind of spend, compared to companies like Canon that are new to the business and are gambling costs with no current income. If they could spend as much on SEDs as others can spend on LCDs, then there would have been no problem, but they can’t. Not even Toshiba can fund both an LCD program and an SED program to replace it under current market conditions.
All this is bad news not only for SEDs but also for any other replacement TV technology. The only way into the current TV market is via the top end, and that means technologies that are stable at 50 inches and above, and which can give enough of a quality and reliability advantage over Plasma that they can charge top rates.
SED TVs produce very bright and clear pictures because the technology is a way of emulating cathode ray tube technology, they also use up to a third less power than plasma panels but most consumers have cheap enough electricity not to worry about that aspect.
In order to recoup their investment in their plant and R&D in two to three years, Toshiba and Canon will need to ship something like 1 million devices. They would need to ship from the factory at something like $3,000 and sell in stores at $6,000 from day one, with perhaps 50% of factory revenue going towards repaying factory investment, the rest for material and labor, and general Operating Expenditure.
It was critical that the Toshiba Canon factory was not delayed and that the ramp from 15,000 devices a month (which it said originally it would begin with) needs to be fast up to the maximum it had planned of 70,000. Now, beginning the process a year later, it will take at least 3 years to repay their start up costs, and if LCD pricing continues to plummet, they may never do enough business to repay R&D. So we can’t see them risking any further funds.
Toshiba Canon insists that these devices will fetch a premium over similar sized Plasma and LCD devices, but these are falling in price towards $2,000 for the sub 40 inch devices, and LCD, as we have said, is targeting larger and larger form factors. If SED technology cannot get a substantial premium against Plasma and LCD, it will never be launched.
And if even if SEDs find a small home at the top end of the market, it will have to be fast. Just 18 months ago LCD was targeting the 30 inch device market. Now it is already eating up the 40 inch market and rising. LCD panel manufacture is going through the efficiencies that chip manufacture went through in the 1970s and 1980s, where better and better engineering principles can be applied to raise yields, stability and predictability of output.
New technologies always hit the market less ready, and less understood. While the SEDs that have been made, look good against Plasma, especially in the fact that they don’t suffer from burn in on the screen or from a drop in emissions density, that doesn’t excuse them of hitting a reasonable price point.
Toshiba engineers previously thought they had this under control. The trick to making SED’s is to make the electron emitters, which fire electrons at a phosphor covered screen, cheaply enough. To do this they need to make a tiny 4 to 6 nanometer sized gap to release electrons through.
In SED manufacture this is done using a combination of two processes called conductance forming and conductance activation. The first process evaporates a tiny 30 nm to 50 nm thick layer of Carbon, while the other process deposits Carbon to fill up the holes. By keeping both processes going at once, at the right level, the holes formed settle at just the right width all on their own. No nano engineering needed.
In order to control the two processes, Toshiba uses a secret organic gas and the concentration of this gas, and the voltage input to the film, between them, define the exact size for the emitter hole.
Since the initial layer is put down using inkjet technology and the emitter holes are not made mechanically or using photolithography, the entire process is automatic and therefore exceptionally cheap. Toshiba say it is just like ‘blowing a fuse.’
After 60,000 hours of testing SEDs, lose about 10% of their emissions making then effectively as reliable as a CRT. But if SEDs can only compete at the top end of the market, then they can’t compete in the 280 million general TV device market, but in just the 5 million, rising to 10 million, 50 inch plus device segment.
Sony was caught by a similar problem back in 2003 when it underestimated the top end of the TV market and its shift to plasma and LCD, and it completely lost control of the TV market. The resulting financial mess was referred to as Sony Shock, as it went into losses and its share price crashed, as the world realized just how badly it had under-estimated the flat TV screen market. It swiftly cut a deal with Samsung to jointly fund a 7th generation LCD panel plant in Korea, and the result has been that Sony has risen back to the top of the global TV rankings, completing the transformation just this quarter, on top of the pile in global TV sales, and taking Samsung with it.
Both Philips and Thomson are just about through difficult transitions where they rid themselves of their CRT screen operations in return for Korean and Chinese partnerships in flat TV screens, gifting their CRT operations to partners who will continue them for third world markets and pockets of the Far East.
Sharp on its own is spending as much on refitting plant for LCDs as Toshiba and Canon are spending on introducing the new SED technology, while Panasonic, yet again, is spending that much again on upgrading its factories to double its output of Plasma devices.
Back in September Sony confirmed that it was going to try to get into the market for flat screen technologies, using CRT class images. This was announced by new CEO Howard Stringer as he told the world what his turnaround plan was for Sony.
The future display technology that Sony is targeting is called Organic Light Emitting Diodes (OLED) which use a self-luminous flat panel display. Sony has formed a special Display Device Development Group that directly reports to the CEO of its Electronics group to accelerate development.
But even this may never see the light of day if LCDs continue into an 8th and 9th generation shedding costs on the way.
OLED came originally out of Kodak about 25 years ago but has been really tricky to bring to a bulk manufacturing process. Like the SED technology that Canon and Toshiba have just about failed with, this is another flat version of the Cathode Ray Tube which effectively creates thousands of tiny CRTs behind a flat piece of glass, which is coated with an emissive layer. This class of technologies have the clarity of CRT, but are usually under an inch thick, and can be made far wider, more easily, than CRTs and also LCDs.
In Sony’s mind this technology was on the blocks to arrive in the 2007/2008 timeframe, but it looks like Stringer is accelerating its arrival. So far, Sony was been loathe to even talk about it publicly. We believe that Sony is fully aware of how tricky these technologies are, especially on pricing points, and it doesn’t want to promise a savior technology before it is ready. And that’s what Toshiba and Canon have gone and done.
There are also other technologies sitting waiting for takers, which may promise better pictures and eventually easier manufacturing, but will anyone ever dare to invest in them if SEDs falter.
US company Nano-Proprietary which claims to own many of the patents used in making carbon nanotubes, one way of building flat TV screens. Last year Nano-Proprietary took legal action against Canon in Texas over patent infringement. It has a patent sharing agreement with Toshiba, but not Canon.
But if it doesn’t win its suit or get a royalty agreement, then there is little chance that another company is going to take similar technology and run with it, at least not in flat TV sets.
Similarly Motorola has been trying to license a breakthrough process for building 3.3 millimeter thick, flat TV screens using a technology called Nano Emissive Displays (NEDs), where it grows thousands of tiny little cathode ray tubes straight out of the glass substrate to hold the required vacuum between the phosphor and the electron emitter.
Motorola Labs showed just a 5 inch segment of such a color display working in May, and hopes to begin signing licensees for the process.
Motorola even went to the trouble of asking researcher DisplaySearch to apply its own economic modeling to the process, a model that we now know to be flawed, given that it failed to predict the price falls and higher volumes on LCDs. Motorola claimed that the cost of manufacture for 40 inch screens could go as low as $400 from the factory, which could lead to retail pricing of $800 and below.
But so far Motorola has announced no new takers and as LCDs get better and better it looks more and more likely that’s its the technology that will drive TVs for the foreseeable future.