By Ross Young
The times they are always changing in the display market.
A couple of days after this blog was posted, it was published and later denied by a company official that Samsung will be moving its a-Si TFT LCD operations out of Samsung Display and into Samsung Electronics. This aligns perfectly with its intention indicated below to close its a-Si fabs as in a couple of years its a-Si TFT LCD operations may be limited to providing panels to its TV business.
Samsung Display, one of the leading a-Si TFT LCD manufacturers for the past 20 years, is shutting down its 1st 7th gen line, Line 7-1, from as early as the end of July and there are rumors of a few more fab closures to come. Its a-Si TFT LCD footprint is expected to shrink dramatically over the next few years with potentially just one a-Si fab in operation in 2018 to serve its market leading TV business. Panasonic LCD is also rumored to be shutting down most or all of its 8.5G line. Based on the near term decline in a-Si TFT LCD capacity, Credit Suisse upgraded AUO and Innolux on this news and their shares surged.
The a-Si Fabs They Are A Closing
Samsung’s Line 7-1, which began operation in October 2004 and production in May 2005, reached capacity of over 150K 1870 x 2200mm substrates per month and is rumored to be shutting down from July 30 through the end of the year. It is a unique glass size which was the largest in the world at the time of its introduction and is optimized for both 40” and 46” LCD TV panel production. It was also the site of the Samsung-Sony joint venture, S-LCD, established in April 2004. This line enabled Samsung and Sony to rapidly take share from smaller 6G fabs optimized for 32” and 37”. During that time, the two companies competed aggressively in retail while establishing 40” and 46” as mainstream sizes. The JV with Sony ended in 2011. S-LCD also built an 8th gen fab that began production in Q3’07 and Samsung also built a second 7th gen line, L7-2 on its own, with even more capacity at over 170K substrates per month which started production in Q3’08.
Line 7-1 , which reportedly accounts for around 3% of global production and 15% of Samsung’s production, was primarily producing 40” LCD TV panels which have been losing ground to 42” and 43” panels produced on larger, more cost effective fabs. As a result, this fab was most likely one of its worst performers and it made sense to close it down as more capacity comes online from China targeting the TV market over the next couple of years.
Displays are already a weak spot in Samsung’s financial performance. It caused the device solutions division to be the only division to suffer a loss among all major divisions within Samsung in Q1’16 and its revenues of 6.04 trillion won ($5.2B) were the lowest in 2 years. The display division alone posted an operating loss of 270B won ($233M) in Q1’16 resulting in an operating margin of -4.4%. While it has a number of technology and capacity advantages in OLEDs for mobile displays, it’s a-Si TFT LCD business has matured and is increasingly becoming commoditized as new entrants from China with government support are taking share. In addition, most of the a-Si TFT LCD market segments have become stagnant or are in decline which points to more losses given the growing Chinese competition. On the other hand, OLEDs and flexible OLEDs are poised for rapid growth as they take share. As a result, it makes good business sense for Samsung to exit much of the a-Si TFT LCD market.
There are rumors that Samsung intends to sell the equipment from Line 7-1 to Twinstar in India which plans to spend $10B with government support to establish an LCD manufacturing center in Maharashtra, India. Samsung is rumored to be converting this facility to a 6G OLED line to support demand from Apple.
Line 7-2, newer and larger than Line 7-1, is also rumored to be for sale and is expected to be shut down in 2017 but it likely depends on profitability. Line 7-2 has been producing 3-4M 40” panels, 300K 75” panels as well as monitor panels. Twinstar is also a potential customer for the manufacturing equipment and this line may also be utilized for OLED production in the future.
Samsung is expected to begin shipping for the first OLED iPhone in volume from Q3’17 with input capacity from the modified Line 7 ramping up to between 90K and 120K substrates per month. Apple is expected to purchase 70M OLED panels in 2017 according to sources.
That is not all however. Line 6, its last 5th gen line with 1100 x 1300mm substrates and around 190K substrate per month capacity, is also expected to be shut down according to industry insiders. That line, which features a-Si and oxide TFT LCD production, has been producing tablet, notebook PC and LCD monitor panels. Those markets have also been hit by stagnant demand, growing competition from China, commoditization and declining margins. This line is expected to be shut down between the end of 2016 and 2017. Line 6 is expected to be converted to a semiconductor facility, making application processors.
If Samsung closes down Line 6 and Lines 7-1 and 7-2, it’s a-Si production will be down to just two 8.5 gen fabs, one in Suzhou, China and one in Tangjong, Korea. The Suzhou fab, ramping to over 100K substrates/month, is also rumored to be a candidate for sale with Chinese and Indian TFT LCD suppliers potential customers. If Samsung licenses it’s latest a-Si TFT LCD or oxide TFT LCD technology, that would certainly make the sale more attractive. This means Samsung would have just one a-Si TFT LCD fab, Line 8.
I believe Line 8 will be sufficient for Samsung to support its internal TV brand with leading edge panels for the TV market as it has 400K, 2200 x 2500mm (8.5G) substrate per month capacity. Samsung would likely continue to outsource its low end panels for its TVs to other LCD manufacturers and focus this fab on higher margin products such as 8K, HDR, QD WCG, etc. I cannot see Samsung abandoning LCD TV panel production until it believes OLEDs are cost effective enough to takeover the mid-range and high end of the TV market. This likely means until OLED materials can be inkjet printed which dramatically reduces OLED material costs, Samsung will still likely focus all its TV panel production on LCDs.
If this occurs as described, Samsung Display's output 42" and larger would be LCD and would be OLED below 42", although it may not have much production between 15" and 40". This would be an amazing, but expected transformation given the supply/demand and profitability outlook for the a-Si market.
Panasonic is also following Samsung’s lead. It is expected to reduce the glass input at its 8.5G line from around 50K/month to just 10K/month in September. Given the limited scale and costs of this Japan factory and its loss of share of its TV brand in many regional TV markets, the company is better off purchasing panels from LG and lower cost Chinese suppliers than trying to run a high cost fab in Japan in a market where commoditization is becoming increasingly common and a large wave of new, lower cost capacity from China is on its way. It may keep 10K of capacity for some period of time for Panasonic branded products, but it is hard to imagine operating a fab at just 10K/month for long as their lack of volume will increase their costs relative to other higher volume players.
Interestingly, with both Samsung and Panasonic reducing their a-Si capacity in the near term, Credit Suisse has upgraded their outlook for AUO and Innolux to outperform and both companies have seen significant share price increases as a result. Credit Suisse indicated that the closure of L7-1 will reduce supply by 12M 40” panels or 1/3 of the 40” market which will tighten supply in the 40”-43” market and produce a better pricing environment in this segment. AUO’s EPS was raised from NT$10.2 to NT$13 while Innolux’s EPS was raised from NT$10.10 to NT$14. Both companies stock prices have taken off with AUO’s stock price rising by 22% in the US in 3 days.
As Japanese and Korean players take more a-Si capacity offline, it will create a better environment for the Taiwan and Chinese suppliers, enabling the market to better digest all the new capacity being brought online in China. However, there is a tremendous amount of capacity coming and much of this older capacity from Samsung and Panasonic may get sold and reintroduced. As pointed out at SID, by 2019 China is expected to have:
2016 and 2017 are expected to be back-to-back years of over $12B in LCD/OLED equipment spending. Unless these fabs struggle with their ramp and their yields, the a-Si market will continue to be oversupplied although conditions should improve in the 40"-43" market in the second half of 2016.
Rather than battle it out with government subsidized Chinese players, Samsung appears to be accelerating its exit of the increasingly commoditized a-Si TFT LCD market. This should please its shareholders, especially if it can sell all its old equipment and license it’s a-Si technology for a royalty, but will be harmful to some of its suppliers. It is also likely going to protect its TV business by holding onto its most competitive fab, but will increasingly become an OLED company. Interesting questions as a result of these moves include:
By Bob O'Brien
In the continuous effort to improve LCD TV, to stave off the threat of OLED TV as well as to try to keep TV price declines to a minimum, manufacturers are pushing innovation in many ways. High Dynamic Range and Quantum Dot technologies are minimizing, if not eliminating, the OLED advantages in contrast and color. Another approach is to match the ultra-slim form factor of OLED TV with another technology highlighted at the SID Display Week last month: glass light-guide plates (LGP).
Among the 2016 Display Industry Awards presented, both Asahi and Corning received awards for Display Components of the Year for Glass LGPs, Asahi for its XCV Glass LGP and Corning for its Iris Glass LGP. Both products are offering similar benefits: a glass LGP allows for a thin profile with a narrow bezel, in comparison with product designs using polyamide LGPs.
According to Corning, Iris Glass is 36 times stiffer than plastic, with 90% lower thermal expansion. Similarly, Asahi says that “Thanks to the rigidity of the glass, we can realize the thinner TV without additional support structure to secure enough rigidity”, which is also explained in a promotional video from Corning:
Products using glass LGPs were first shown at the CES in 2015, and introduced later last year. Corning’s Iris glass has appeared in TVs primarily sold in China, including Sharp 60” and 70” UHD models and a 70” product from LeTV, while Asahi’s XCV glass appeared in Sony’s flagship X900C series distributed worldwide, with an edge thickness of only 4.9mm. For folks like me who started in the display industry in CRTs, the thin design of these products represents quite an achievement: a 65” TV with the profile of an iPhone.
So what’s not to like? Well, one of the first problems is that the initial products using a glass LGP did not include High Dynamic Range. The edge-lit LED design did not allow for local dimming. This may be addressed going forward; more on that later.
Another problem area experienced in the first Sony product iteration was in the implementation. Many consumers complained about “Light Bleed”. A quick check on amazon.com on the Sony X900C product reveals that the average review was only 2.7 stars out of 5, and reviews titled “Light Bleed is TOTALLY UNACCEPTABLE” are typical (somewhat better reviews on bestbuy.com, at 4 stars out of 5, but still the lowest rated 65” on offer and critical users cited light bleed). I’m sure that’s NOT what Sony wanted or expected with such a flagship product.
In Sony’s refresh of the product for 2016, the X930D, they’ve introduced a slim design with a dual-layer LGP that allows for local dimming. Unfortunately for Asahi, this product does not use their glass (Corning also confirmed that Sony’s product does not use Iris glass). My source at Sony could not disclose the material used for the LGP, but Asahi confirmed that it’s not XCV. Whatever the material is, Sony appears to have solved its light bleed issues, based on the amazon reviewers, with a healthy 4.4 stars and no light bleed comments.
Although not in the Sony 930, Asahi is still optimistic about the glass LGP category, and promised that “several manufacturers have applied XCV for new models”, with details coming out later this year. Corning, if anything, may be more bullish on the category. Corning has been receiving interest from several regions and anticipates solid growth in the category, with additional models to be introduced in 2H’16.
According to Corning, one of the key elements of the product roadmaps involving glass LGP involves a continuing push to thinner glass. This allows the end device to get even thinner, allows for an easier implementation in curved displays, and drives the cost down. Corning is even discussing product configurations in the IT space: an All-in-One product with a glass LGP would allow a cost-effective implementation in a relatively high ASP product category with a premium on design. Corning is building its organization for LGP to take advantage of the opportunity posed by Iris glass LGP.
Is there a solution for glass LGP implementation that includes local dimming? In my opinion, this will be a
critical success factor for this category going forward. High Dynamic Range is becoming (or soon will become) a must-have feature for a high-end TV set. Although none of my contacts could discuss the material that Sony is using in the X930D, it apparently is not glass. That doesn’t mean that such a solution using glass is impossible; the reasons for choosing another material are unclear. Corning has discussed products using glass LGP which apply 1D local dimming; this may be a path to enable a product which meets a High Dynamic Range criterion. Corning is also examining opportunities to add additional functionality to the LGP; a surface treatment or other optical mechanism may open up the possibility to remove existing layers of film, for example.
Finally, one additional factor to consider when thinking about this category is the potential impact to display glass (or cover glass) supply. Although TVs with glass LGP are only a small portion of the market, they will have an outsize impact on glass supply because of their thickness and size. The typical glass LGP is 2.0mm, compared to the standard display glass substrate thickness of 0.5mm (combined 1.0mm, including both TFT and CF substrates). Furthermore, glass LGP is used on the largest screen sizes, typically 55” and up, so the combined impact on glass supply is likely 4-5x as big as an average (~40”) TV. Although all the major display glass makers are managing with some idle capacity, this could be consumed quickly if glass LGP TVs could capture even 2-3% of the TV market (~10% of the 55”+ market). Both Corning and Asahi maintain that although glass LGP is made on the same manufacturing platform as display glass (fusion melting for Corning, float process for Asahi), they can effectively balance supply/demand across their manufacturing platforms and don’t anticipate any supply problems.
In a display industry that is struggling to find areas for growth, and with the threat that OLED TV could reduce the demand for display glass (if using metal encapsulation, as LG does today), glass Light Guide Plates represent a promising growth area. These SID award winners look like they will make an enduring mark on the display industry.
(Full disclosure: the author is a former Corning employee and holds some Corning stock. The author also has a non-disclosure agreement regarding sensitive information learned while he was a Corning employee, but no confidential information is included in this report.)
For more insight, visit displaysupplychain.com.
Be sure to sign up or send us an email to receive free information on the display supply chain.
I’m Back And Excited to Assess the Impact of New Form Factors on Existing and New Markets with Bob O’Brien
By Ross Young
Eight years after leaving DisplaySearch and four years after leaving IMS Research, both of which have become part of IHS Corp., I have re-entered the display market research and consulting business. It has been a good eight years. I travelled, got into photography, exercised (11 IRONMANs in 3 years), lost weight, gained weight, got healthy and got injured (11 IRONMANs in 3 years), etc.
What brought me back? Well, my triathlon career seems to have ended after sustaining a torn hip labrum, torn adductors (sports hernia) and a torn disc from playing basketball with my 16 year old in combination with completing and training for all those triathlons. I can still bike and swim quite fast, but I can’t run without pain now and the best method to prevent further hip damage without at least one surgery seems to be to stop running.
But I am also a little bored and legitimately excited about what is occurring in the display industry. While display revenues and end market revenues have become stagnant, innovation has continued and we are on the cusp of flexible/foldable/unfoldable/rollable displays which can potentially breathe new life into many of these stagnant markets. Samsung is rumored to be introducing the first foldable AMOLED products next year which will lead to important changes in the display supply chain. First, it could certainly create the next must have gadget whether from Samsung or another company and it could cause a rebound in the handset/phablet/tablet markets.
In addition, it will cause some disruption upstream. What will happen to ITO touch screens? What about cover glass? What products will replace them? These are markets that could get seriously impacted as a growing number of handhelds and other products adopt this form factor over the next 5 years.
While I was seriously burned out after running DisplaySearch for 12 years, I am refreshed and energized and looking forward to bringing the passion I had when creating DisplaySearch, most of its product categories and many of its most successful products. Some of you may remember that I was authoring up to 14 reports in a quarter before I left. I was also happy to hear from many friends and potential clients at SID 2016 that an alternative to the leading display research firm was needed due to their significant price increases. I am happy to deliver valuable syndicated and custom content at more reasonable prices, not Veritas et Visus prices, but still very reasonable...:)
I am also excited to be working with Bob O’Brien. Many of you know Bob. He has decades of experience turning market and business analysis into strategic insights at Corning, LG Display and Philips Display Components. He essentially ran a smaller version of DisplaySearch with analysts in the US and Asia performing channel checks throughout the display supply chain. He also developed an analytical model for predicting glass demand based on the supply/demand dynamics of the display supply chain. He also carried out consumer survey work on the TV replacement cycle and helped explore and develop the Gorilla glass business in emerging markets.
In addition, I will soon be introducing the Display Supply Chain (DSC) Monitor, 20 years after The DisplaySearch Monitor was introduced. This product will focus on key issues and opportunities facing the display supply chain along with analysis and insights from Bob and I and other experts. This product will be targeted at marketing managers, sales professionals, procurement managers, financial analysts and others looking to understand what is happening throughout the display supply chain. If you would like to know when the first issue becomes available, please contact us. In addition, if there are certain specific things you would like to see in this and other products or you would just like to make contact, feel free to email me at firstname.lastname@example.org or via phone at 512-577-3672. Great to be back in this exciting space and hope to hear from you.