MLS Co. (002745.CH), China’s largest LED packager has agreed to form a joint venture with Rhode Island based Lighting Science Group (LSCG), with MLS providing lower cost LEDs to Lighting Science’s US LED bulb product customer base. MLS will invest $4.9m for a 49% stake in the JV, with LSCG owning 51%.  Lighting Science has been a private label supplier of LED bulbs to Home Depot (HD), although in 2Q ’16, HD elected to purchase bulbs directly from Asian suppliers, reducing LSCG’s revenue stream, but allowing them to pursue other big box retailers, which was prohibited under the old agreement.  The agreement with MLS could give LSCG a bit more cost control over its products, making them more viable to new potential customers.

According to the Chinese trade press, LG Electronics (066570.KS) will sell an ink-jet printing encapsulation tool to Beijing based Tianma Microelectronics (000050.CH) for use in the production of OLED displays.  The tool, which is expected to ship in 2Q, would be part of Tianma’s build-out of its Gen 6 OLED fab, which is expected to be in operation around September of this year.  Tianma currently has a small Gen 4.5 OLED pilot line.  We believe the tool mentioned is the same one LG Display (LPL) is testing in their E5 OLED fab.
What makes this information unusual is that it would seem strange for LG Electronics to own the tool, as LG Electronics is an assembler and brand marketing company, while LG Display is the display producer.  Perhaps the writer confused the two, but what makes it more confusing is why LG Display (should the confusion be correct) would be selling the tool if it is testing the tool on a production basis, and in many cases tool manufacturers will ‘loan’ a tool to panel producers to test, keeping ownership with the vendor if the tool is rejected.
Also questionable is the idea that LG, a South Korean firm with very significant OLED manufacturing expertise, would want to sell equipment to a potential Chinese competitor.  It wouldn’t be the 1st time, as Samsung Display (pvt) has sold equipment to Chinese producers, but it is certainly frowned upon by the South Korean government, as the fear that Chinese producers will gain some knowledge that will close the experience gap between the two countries.  South Korea has seen this happen years ago when Chinese entities bought South Korean panel producer Hydis, and closed the company after stripping out the IP and physical assets.  While blocking the equipment sale by LG would likely be extreme, the government of South Korea has made it plain that they disapprove of anything that will help Chinese display industry participants gain traction.

As part of the Chinese Ministry of Science & Technology R&D program, BOE (000725.CH) is developing a small panel (5”) quantum dot display.  The device, which is said to be produced using an ink-jet printing process, has no backlight, indicating that the device is generating light directly from the quantum dots.  While there is little information about the project, as it is only one of 863 programs sponsored by the ministry, and the most recent ministry newsletter is from 2012, we expect the device is a preliminary prototype, and far from near commercial production. 

As we have noted in the past, the current use of quantum dots is in films that enhance the color accuracy of LED backlight devices, with the next timeline inflection point being the use of the materials directly as LED phosphors, with the final goal, a self-emitting device, such as the one mentioned above.  The development of such a device, and its eventual commercialization remains a few years away according to our contacts in the industry, but we do expect to see similar articles, demos, and IP toward that goal, especially from Chinese display participants, as the playing field in QDs is far more level than in OLED, where South Korean producers dominate the space.  While we expect the BOE project is a very early stage device, where there’s a will, there’s a way…

A group of Chinese hackers known as the Hong Kong Confederation, are predicting a full-scale cyber attack on unspecified targets in South Korea to protest the deployment of the THAAD (Terminal High Altitude Area Defense) anti-ballistic missile system.  The system, designed by the US Army is used to shoot down short and intermediate range missiles, in this case potentially coming from North Korea, although the system has been installed in a number of US locations and in a number of Middle Eastern countries.

China has been quite vocal in its opposition to the system’s installation in South Korea, but not because they are worried about Chinese missiles being shot down by the system.  In fact, THAAD shoots missiles down in their terminal phase, meaning as they are approaching their target and is not a long-range defense system, however the tracking system that is part of THAAD is what bothers the Chinese government.  They are concerned that the data produced by the system on Chinese missile launches will give the US and its allies a ‘detection’ advantage, however similar systems in Guam and Japan have been in operation, and the addition of the systems in South Korea would add only marginally different information.

That said, THAAD has become a rallying cry in China, and unfortunately a diplomatic sticking point in negotiations between the US, China, South Korea, and North Korea, as can be seen by the entry of Chinese hackers into the fray under the guise of nationalistic ‘defense’.  The hacker group put out a request on their website for additional hackers to join them saying “Lets join the attack on Korean sites to defend the country”, and set the attack to begin ‘earlier than the 24th and before 7:30PM on the 31st’.  The hacker group, also known as the Red Hackers, applies a DDoS[1] attack on Japanese websites each year on Manchuria Incident Day (September 18) to protest the 1931 Japanese takeover of the Chinese Manchurian province, but rarely publicly recruits participants or sets dates for attacks.

[1] DDoS – Distributed Denial of Service

Back on 2/27/17 we noted that Samsung Electronics (005930.KS) had verified that it would release a new Galaxy Tab S3 at Mobile World Congress, returning its tablet business to OLED displays.  Samsung had used an OLED display in its Tab s (7/2014) and Tab S2 (9/2015), but returned to LCD technology for the Tab A (3/2016) and Tab E (1/2016).  We believe Samsung Display (pvt) will be sticking with OLED displays for new tablet models and is considering using OLED for some laptop production this year.

We believe Samsung Display will be converting at least a portion of its L6 LCD production line from a-Si[1] to Oxide[2] backplane production and OLED display lines.  The L6 fab had a raw capacity of 268,125 m2 as of November 2016, which represented 7.5% of Samsung Display’s overall capacity, and 1.3% of worldwide capacity, and we believe had begun to reduce production on at least one of the three L6 lines at the end of last year.  We expect that line to be closed by May and the conversion to Oxide and OLED to begin.  We expect phase 1 of the oxide conversion to be fully ramped by April/May 2018, with lower output levels beginning this summer, and the ramp down of L6 phase 2 starting in July of this year, completed by year-end, and the conversion of L6 phase 2 to be completed by July/August 2018.  The L6 phase 3 ramp down will begin at the end of this year, with the conversion completed by 1Q 2019.

Samsung Display has yet to signal whether the L6 conversion will change the substrate size from Gen 5, so we build the changes into our model using a Gen 5 format.  As the upgraded fab will be supplying both IGZO backplanes and OLED displays, we expect the actual OLED capacity to be considerably lower than what the fab had been producing for LCDs.  We believe that Samsung will shift its LCD tablet and notebook production to its L8 Gen 5.5 lines, albeit at a reduced rate, as it continues to supply Apple (AAPL) with panels for the iPad, along with Sharp (6753.JP) and LG Display (LPL).

While expectations are that Samsung will release the previously mentioned Galaxy Tab S3 this month, there have been rumors that they will also release a Win 10 tablet line this year, although that remains unconfirmed, and we believe has supplied OLED displays to HP (HPE) and Lenovo (992.HK) for their Spectre  and Yoga Thinkpad laptops.  But the laptop market is far different than the smartphone market, and the necessity for OLED displays is less urgent than for smartphones, where the device is on for a large portion of the day.  That said, there was little OLED capacity for laptop display sizes, and yields were lower than those for smartphones, so Samsung Display had less of an incentive to expand that segment.  So what makes Samsung Display willing to step up its ‘non-smartphone’ OLED production capacity?  Most likely it is interest from Apple for its iPad line, and while an immediate move to OLED would be somewhat premature, Samsung Display could be responding to both its parent company’s demand and the potential for Apple to move to at least an OLED iPad variant sometime in the future.  Apple will be announcing new iPads at a press event in April.

All in, Samsung Display and parent Samsung Electronics, continue to push OLED displays further into their mobile lines, and reduce their exposure to LCD in small panel devices.  If we are correct in our assumptions about the L6 conversion, Samsung Display should be able to produce the equivalent of 2.6m 13.5” OLED laptop screens/quarter at the completely converted facility[3].  Apple has been averaging about 11.5m iPad units/quarter for the last two years, so we have to assume that only a portion of the iPad line could be converted to OLED over the next few years, as Samsung Display will also use the new capacity for its captive customer, but even the possibility that Apple might use OLED for its iPad line keeps a fire burning under the OLED space, and will continue to drive Samsung Display and others to add capacity.



[1] A-Si – amorphous silicon – commonly used as an LCD backplane
[2] Indium Gallium Zinc Oxide – aka ‘Oxide’ – another type of backplane
[3] 100% yield – Actual rates will be lower

March display panel pricing saw mixed results.  Monitors (-0.2%), tablets (-0.8%), and smartphone (-0.9%) declined, while notebooks remained flat, while TV panel prices rose 0.7%.  While most categories are seeing seasonal declines, the TV panel space is still a seller’s market.  Chinese buyers have cut back TV panel orders in order to work down inventory levels, but are still willing to pay higher prices, fearing shortages might continue into 2H, with brands now facing prospects of very low or no profitability for the year.  Price increases at the set level continue, as TV brands try to offset the continuing component cost increases, but the early results, seen during the Chinese New Year holiday reflect disappointing sales as consumers see little incentive to pay up for current TVs.

The risk of a continuing inventory build in the TV space continues to increase, and if demand proves weaker than expected as the year develops, panel producers will lose their ability to raise prices or even maintain prices, as brands lower expectations.  There will remain some TV panel sizes that are in short supply due to Samsung LCD plant conversions to small panel OLED and other factors, but the massive TV panel price increases that were seen late last year have subsided, and the effect on TV brands is becoming more obvious, along with the retail malaise. 

There are two possible scenarios for TV panel pricing (for panel producers) this year.  One where overall panel prices remain flat or move up slightly.  In that scenario brands will continue to suffer and consumers will react poorly to any set level price increases, which has the potential to negatively impact holiday 2017 sales.  Alternatively, If brands react earlier to slower demand, there is the chance that TV panel prices will decline, allowing brands to at least make some profit or lower set prices to stimulate sales.  While panel producers will lose a bit of margin, this scenario would be beneficial to the industry, allowing both panel producers and brands to profit somewhat and for consumers to step up purchases.  Of course, there is some middle ground or variation in these two scenarios, but if the general trend continues, it looks like panel producers will have a good year and brands and consumers will not.  This tends to set the stage for the following year (2018), which will likely see a more severe reduction in panel demand from brands, and a more aggressive decline in TV panel prices.  Hopefully the second scenario, or something close to it, will play out this year, and the reaction in 2018 will be more positive, but it is still a bit early to tell.  Right now we would weight the chances about 60% for the first scenario and 40% for the second.

The smartphone world is abuzz as leaked photos of the upcoming Samsung (005930.KS) Galaxy S8 reveal a new button on the side of the device.  What secrets does it hold?  Will it cure cancer?  No, but it will give you direct access to “Bixby”, Samsung’s version of Apple’s (AAPL) 2011 AI platform based Siri.  Of course you can access Bixby vocally, similar to the ‘hello google’ voice command, but Bixby is so important that it gets its own button! In the rarified world of AI based digital assistants, Bixby is king. That said, the app will likely be released in a slightly reduced form, as the development of the tool is still continuing.  Updates will add to Bixby’s usability.

What is really at stake here is the entire display industry, as vocal interaction with search functions supplants a visual search, and while the necessity for eventually looking at whatever it is that you searched for is still a part of most searches, the use of voice activated search assistants lowers the value of displays.  While the use of voice to command interaction with computers has been around for many years, and the use of AI in search algorithms has also, the functionality, or ability to use natural language has improved steadily over the last few years. 

Will it obviate the need for visual displays?  Not likely, but it could have an effect on the overall habits of mobile and stationary device users over a generation.  Will your kids be satisfied that their smartphone AI told them that a restaurant got 5  ‘poor’ rating on Yelp, or will they have to read the reviews themselves?  Natural language interfaces have been tried before yielding less than spectacular results, but algos get better over time.  Does this mean we not only lose the use of our legs, see our hands evolve into a single pointy finger, and now see our ears enlarge and lose the ability to see?  Probably not, but speaking is easier than typing and listening is easier than looking, so we would have to expect some long-term changes to the display space should Bixby, Cortina, Alexa, and others become our children’s companions…

Taiwan based Digitimes indicated in an article that Samsung Display (pvt) and LG Display (LPL) are decreasing their reliance on outside sources for organic materials used to produce flexible OLED displays.  While this is ultimately true, that South Korean firms would ideally like to have ultimate control over the materials used in the production of flexible (and other) OLEDs, the article did little to specify any changes that have not been mentioned previously, generally citing mergers and investments by both producers that would help them gain the necessary IP to bring production of flexible OLED materials to Samsung SDI (006400.KS) and LG Chem (051910.KS) respectively.

The article specified that both producers obtained polyamide varnish from Japanese supplier Ube Industries (4208.JP) and a number of other TFT backplane production materials from South Korean suppliers ENF Technology (102710.KS), Dongjin Semichem (005290.KS) and Dongwoo Fine-Chem (pvt), a division of Sumitomo (8053.JP), none of whom struck us as different than would have been expected.  Further, deposition materials, a far more important component of the OLED production process, were specified coming from Idemitsu Kosan (5029.JP), Universal Display (OLED), Dow Chemical (DOW), and Merck (MRK), but hinted that the Samsung purchase of Novaled and an investment in Sun Fine Chem, a division of Hodogaya Chemical (4112.JP) were leading to the acquisition of IP that would allow these organics to be made internally, and finally that flexible substrates and cover glass were sourced from 3M (MMM) and Corning (GLW) respectively.

While there are a great number of materials and suppliers used in the OLED production process, this article did little to shed any light on current or potential changes that have been implemented by South Korean producers.  Samsung purchased Germany based Novaled for $347m back in August 2013, a company known for its ‘PIN OLED’ material that is used to enhance the electron mobility of certain OLED stack layers.  It doesn’t replace other stack components, but works with other stack materials to give better overall stack results, and an investment by Samsung in SFC (Sun Fine-Chem) a division of Hodogaya Chemical, goes back to 2011.

We believe that the only thing relating to OLED materials that has changed recently, other than the usual vying for a position in the OLED stack of these South Korean producers, is the change in backplane made by Samsung Display earlier this year.  This change, to the M8 backplane did make some changes in OLED material suppliers, with red host material moving from Dow to Duksan Neolux (213420.KS) and green host moving from Samsung SDI to Nippon Steel (5401.JP), while other emitter and host materials suppliers remained the same.  While on a long-term basis, OLED producers are always looking to reduce costs, using their leverage to pit one supplier against the other, that doesn’t always lead to using home country or in-house suppliers, as is seen in the change above where green host moved from an in-house South Korean supplier to a Japanese supplier, and both producers evaluate many materials from suppliers to find the right combination of performance and cost.  If a local or in-house supplier can meet those needs, everyone is satisfied, if not, there is no hesitation on the part of Samsung Display and LG Display to use outside suppliers, and that has not changed over the last few years.

Consumer Electronics companies can be likened to professional surfers.  They are always looking for the next big wave and since growth in some of the more traditional CE product groups has been slowing, CE companies have been jumping on the ‘connected home’ bandwagon, a subset of iOt.  Connected products are being released daily, and while only 10% of US, UK, and Australian consumers indicated that they were using ‘connected solutions’ in a Gartner survey last year, the hype from the CE industry would indicate that nothing is more important than owning a Zencrate Anti-Anxiety Smart Dog Crate (pvt – kickstarter) that will calm your pup with music, fans, and a camera (so you can see him/her shaking in the box) during thunderstorms.

While we are very definitely fans of things that can make life a bit easier (we started with BSR remote controlled outlets in 1975), much of what is touted today is pure marketing, hopefully leading to margin improvement by producers.  The kitchen seems to be a big focus for such products, so we look at the rational, or lack of, for this location first, and will follow up with other ‘rooms’ in subsequent notes.  Here are the primary categories for connected kitchen items and our comments:

Coffee Maker – “knows when you wake up and starts brewing you favorite brand”[i] – Sounds good and is a bit of an improvement over those ‘old style’ coffee makers that work off a clock setting, but you still have to load it up the night before.  No longer will you have to adhere to your coffee maker’s schedule, but little else has changed, except your coffee maker will pester you via smartphone until you fill it for the next day.

Convection Oven – “Starts heating your toast when you step out of the shower each morning” – Wonderful, but the toast would have to be loaded the night before and would be sitting in the convection oven for your 8 hour sleep time.  Don’t forget that the convection oven will also remind you to load it up the night before.

Crock Pot – “Starts cooking dinner while you are at work so its ready when you get home” – This has possibilities, although most crock pots run for ~9 hours anyway, but being able to start the crockpot via your smartphone might help if you are going to be early or late.  That said, whatever your cooking is going to have to be put in the crock pot by you, either the night before, or before you leave for work, so don’t be too late in starting the cooking or your food will have spoiled before its even been cooked.

Lights – “Automatically turn on when you enter and automatically turn off when you leave”. Finally, and if you have kids you understand immediately, a practical idea, but one that has been around for many years and is easily implemented in a variety of ways.

TV -  “Automatically turns on the traffic report when you enter in the morning” – OK, some possibilities here, although not having to press an ‘on’ button doesn’t seem like much of a time or effort saver.

Stove – “Receive an alert when your stove or oven is left unattended” – A good idea, but one that would have to be adjusted to whatever the timeframe is for the items being cooked.  If you are cooking a roast for an hour, you would need to be notified if the oven is unattended for more than an hour, but if you are reheating a meal, the oven would have to understand that it should contact you much sooner.

Cabinets & Drawers – “Cabinets and drawers would notify you if small children are accessing dangerous or off-limit areas like cleaning supplies” – An excellent idea, although a bit of overkill considering the many ways in which cabinets can be baby-proofed currently, but has promise if it can understand who is allowed to touch what cabinet.

Many of these items, and this is just the kitchen, are interesting but largely impractical, and will likely make only a slight difference to your daily routine. So the real question will be whether consumers will pay extra for such connected devices.  The most practical, auto-lighting can be accomplished by a small sensor in the light switch rather than a ‘connected’ system, unless it is important to you to turn on the lights just before you enter the room with your smartphone…



[i] Quotes are from Twice@Home 3/13/17

The Korean press has noted that Apple (AAPL) has received criticism for ‘disregarding Korea’ by announcing an additional $508m investment in two R&D centers in China, following the promise to build two other R&D centers in Beijing and Shenzhen.  Apple has yet to open an Apple store in South Korea, although one is being developed in Seoul although no date for opening has been set.  However they have 40 stores in China, with the R&D centers seen as a way to build what has been a declining share on the Chinese mainland.

Despite the country rhetoric, there are real reasons why Apple has not focused on South Korea, with the most obvious being it is the home of rivals Samsung Electronics and LG Electronics (066570.KS), who have a distinct home court advantage.  But the country points to the fact that South Korea has the highest smartphone penetration rate (88%) of any country worldwide, while China is in the middle of the pack at 58.5%.  That misses the point however that South Korea has a population of ~50.5m, while China has a population of 1.38b, so the simple math says South Korea has 44.4m individual smartphone users, while China has 808.5m smartphone users, or 18.2x as many, making the decision relatively simple from an economic standpoint.