According to Trendforce, memory prices remained flat in June, which is typically the case in months 2 and 3 of each quarter.  As price negotiations for Q3 are currently underway, expectations are that July will see DDR4 8Gb average monthly pricing increase between 3% and 8%, which would then set the price for the quarter.  NAND Flash 128Gb prices were also flat in June, but no estimate was given for July.

We recently noted statistics from Semi.org concerning the 19 new fab projects that are underway this year and an anticipated 10 new projects to begin next year, generating ~$16b (+31% y/y)in fab equipment spending this year and a record ~$23.8b (+45%) next year.  Semi provides their initial forecast in March and has just updated that data, so we went back to compare the new (June) with the old (March) to see if anything changed over the last 90 days.
Much was the same, however after a 16% increase in fab equipment spending last year the March prediction for equipment spend for this year was 15.5% y/y growth and that has changed quite considerably to 30% this year, almost double the previous growth estimate, and underlying the March estimate for fab equipment spending growth in 2022 of 12% was a year-end 2022 spend of ~$80b.  While the fab equipment growth estimate for 2022 decreased by 1% to 11% growth, likely due to 2021 acceleration, the ~$80b target for 2022 is now expected to be reached and exceeded ($84b) this year, now estimated to reach ~$93b in 2022.
The granularity of the spending remains similar in that 300mm semi capacity was expected to grow 8% this year and 8% in 2022, which now is estimated to be 8% this year and 7% in 2021, but 200mm capacity, which was expected to grow 4% this year and 5% next year, is now expected to grow 5% this year and next, and Semi’s fab database, which includes almost 1,400 lines and 111 that are in planning stage, added 12 new lines between March and June, and removed 1.  We show the Rate of Change in fab equipment spending and the change in the Semi forecast in Fig. 1.

​While we seem to be obsessive about Mini-LED TVs, we justify that preoccupation with the fact that while Chinese brand TCL (000100.CH) was the first to offer consumer Mini-LED TVs and is in their 3rd iteration of the technology, TV market leaders Samsung Electronics (005930.KS) and LG Electronics (066570.KS) are in early release mode for such products.  As we have noted, Samsung released, at least announced, its consumer Mini-LED/QD TV line earlier this year but did not reveal set pricing until the 3rd week in May.  Since then there have been a number of price reductions across both the 8K and 4K Mini-LED/QD line which we have noted a number of times, the latest on Monday. 
While LG Electronics has also announced their Mini-LED/QD TV line, those sets are just beginning to appear and pricing is beginning to surface.  LG seems to have learned from Samsung’s obvious mis-pricing and has priced both its 8K and 4K Mini-LED/QD lines at or below Samsung’s original entry prices, with the 8K line being significantly discounted from Samsung’s original and current prices, and the 4K line is priced roughly the same as Samsung’s current pricing, sort of a tacit acceptance of what both companies believe to be acceptable prices.
As we have noted before, these sets are not exact duplicates, with some features available to a particular brand or model, and Samsung has considerably more diversity in terms of the number of offerings, 15 to LG’s 6, but the basic features such as resolution and size are comparable, and the table below makes those comparisons.  LG offers the only 120Hz. refresh rate models, but based on the original Samsung price, the LG models average 23.1% lower in price, while against the current price they are only 10.3% lower on average, with all of the discount coming from the 8K line, which is priced 22.4% below Samsung’s current price on average.  TCL’s Mini-LED/QD offerings are also available but only at 4K resolution and range in price from $3,000 (75” High-end) to $950 (55” Low-end).  LG will begin making the sets available in the US in July.  A more detailed price/feature comparison is available on request.

​Apple (AAPL) is slow in making transitions.  Whether a function of a cautious attitude toward new technology or a desire to put its own stamp on what might already be available, they take their time, where others tend to jump as quickly as possible.  Of course it helps to have 1.66b active Apple users, creating a very ‘sticky’ base that is typically willing to wait for Apple to move forward, so Apple has, at the least, the luxury of exploring new technologies for a while before pushing them out to their customer base.  What does make a big difference however is when Apple does adopt a new technology, whether it is early or late, it means significant unit volumes and a blessing for the Apple supply chain.
That said, Apple was a slow starter with OLED, adopting the technology in late 2017 with the iPhone X, while leaving the other iPhone models with LCD displays.  In 2018 two iPhone models had OLED displays, and by last year all iPhones sported OLED displays, and while Apple sells lots of smartphones, they also have other products that could easily adopt OLED displays going forward (The Apple Watch also uses OLED).  For much of this year rumors have been circulating that Apple will be offering its first iPad with an OLED display next year and would continue to spread the technology across the iPad line in 2023, likely to all but the iPad Mini, which is a smaller and lower priced offering.
Those rumors have surfaced again, with current expectations that Apple will offer a 10.86” iPad version next year with an OLED screen, albeit based on relatively mundane OLED technology.  The OLED display is expected to be based on a rigid substrate (glass) but instead of being encapsulated with glass (typical for rigid OLED), it would use the same thin-film encapsulation used on most flexible OLED displays.  The TFT structure is expected to be typical LTPS, the most common form of OLED backplane, with the displays being made exclusively for Apple by Samsung Display (pvt). 
The reason this rumor is gaining strength comes from Samsung’s push to commercialize its IT OLED panel business, which has been exclusive to parent Samsung Electronics, and given Samsung’s goal of shipping 1m OLED notebook displays this year, we expect they have plans to increment that number again in 2022 and 2023.  While we expect SDC is actively seeking a broad customer base, a deal with Apple for the iPad, even for one of four Apple iPad models, would go a long way toward both filling existing OLED capacity and justifying the expansion of same.  Apple has financed such capacity expansions in the past, particularly for LG Display (LPL) and Japan Display (6740.JP), and while we expect Samsung is not in a position to need or accept financing for small panel OLED capacity, Apple would likely contract dedicated capacity, which they have done with SDC in the past, and such contracts, being as close to ‘take-or-pay’ as possible in  the CE space, are a guarantee that SDC could maintain profitability for that new capacity, without significant market risk. 
We note also that the surface area of a 10.86” iPad display (4:3 aspect ratio) would be equal to the surface area of four iPhone 12s (9:19.5 aspect ratio), and the relatively simplistic structure of the supposed OLED iPad would likely allow Samsung to maintain high yields and profitability, so there is some logic to the expectations.  In 2023 however the rumor suggests that Apple will expand its iPad offerings to 11” and 12.9” (iPad Pro current sizes) offerings using flexible OLED and an LTPO backplane structure, which should be able to maintain profitable yields at that time, and in order to offset any criticism of the OLED iPads being less bright than their former LCD brethren, the OLED emitting structure itself is said to be a double RGB stack, a process that is being adopted by both SDC and LG Display for ‘certain’ customers. 
Such a dual stack process can be used in near-eye displays that must be both high resolution and high brightness, but research has been focused on photolithography techniques for such a process that would require a significant process change and cost for high volume production.  Both OLED producers have been working toward being able to run the substrate through the deposition process a 2nd time to create the double stack, without compromising the first layer rather than using a more complex approach, which is why that process is scheduled for 2023 rather than next year, giving SDC and LGD time to perfect the process.
All in, the current rumor does carry some weight in terms of meshing with expansion plans for SDC and LGD, and on a more general basis, Apple’s desire to continually improve their displays, but while this track toward an OLED iPad line is being developed, so is the work toward Mini-LEDs, which are based on LCD architecture.  This gives Apple some leverage with OLED display producers as adapting mini-LEDs to notebook sized LCD panels is a relatively easy task that can be accomplished using existing technology.  OLED producers must push hard to further develop OLED technology knowing they have competition from the ranks of LCD producers who are eager to keep their investment active and competitive against OLED and can do so with relatively little investment and new infrastructure.  That said, as always, time lines are developed to be changed, so rumors need verification that is hard to come by when dealing with companies that are obsessive about project security, but at least the current Apple/iPad/OLED rumors have some basis on which to be made, albeit one that could change daily..

​Back in January (see our notes 1/14/21 and 1/25/21) we noted that Apple (AAPL) had added a ‘caution’ to its devices that instituted its MagSafe magnetic charging system, including the iPhone 12 family.  At issue were the increased number and strength of the ring of magnets that were responsible for holding the device in the correct location during charging, with the warning that anyone using an implanted devices, such as a defibrillator or pacemaker should keep any device or accessory using MagSafe more than 6 inches from the implanted item.
This means not keeping an iPhone 12 in a breast pocket or putting the phone next to you when you are sleeping.  The FDA agreed with Apple’s warning, and at the time of the study was unaware of any adverse reactions between the iPhone 12 and any implantable devices.  A recent study by the American Heart Association however, disagrees with that conclusion.  The AHA study found that the iPhone Pro Max resulted in ‘identifiable magnetic interference in 3 out of 3 tests where the phone was place directly over an implanted device and 8 out of 11 (72.7%) of ex-vivo tests, meaning those where the phone was placed next to an implantable device that was not implanted.  The conclusion was, “As a result of these tests, the researchers concluded that Apple’s iPhone 12 Pro Max MagSafe technology can cause magnet interference on CIEDs and has the potential to inhibit lifesaving therapy.”
As noted, the tests were done by placing the phone directly over the device on the patient’s skin, although it was found that with some models the effect was seen at 1.5cm (0.6”), so in more practical terms, the risk would be more from the possibility of an accidental contact, however the AHA did go on to compare the results for the iPhone 12, which Apple insists does not carry any greater risk than older generation iPhones (non-MagSafe), against a study done that found no cases of magnetic issues in the 148 patients that were tested. 
All in, even if the implants are affected by the iPhone 12, the interference will not cause a heart attack or any other symptoms that might be dangerous to the patient.  During the interference, the implant would however be in backup mode and would not be able to provide the patient with the necessary assistance during tachycardia or other arrhythmias, which could have serious consequences.  The bottom line is if you have an implanted device, keep your iPhone (or other similarly charging device) in your pants pocket or handbag.

There have been reports over the last year that high-tech money is no longer flowing to what might be considered ‘older’ technology and is only earmarked for semiconductors and AI.  While we are sure that there are projects that have seen lower funding from local governments, but when it comes to China’s big boys, the funding seems to be available and even ahead of where we would have expected funding to be at this mid-point of the year.
When speaking of LEDs in China, the country’s largest LED chip manufacturer is Sanan (600703.CH), who falls within the top 10 globally, having generated $8.35b last year, growing 13.5% y/y, and while that certainly gives the company status, Sanan is spread across much of China and has to garner favor with each local or provincial government in the regions where it is located or expanding to.   While the ‘elite’ status with the central government gives Sana cache, each subsidy deal has to be worked through with the particular government support group one by one.
That said, since the last day of 2020 Sanan has received the following subsidies, with each allocated to the particular Sanan subsidiary in that region, and while we understand that these subsidies are at the behest of each local or provincial government, we would expect that the rate of endowment established thus far in 2021, would continue through the year, although the largest listed below is for a specific project currently under construction, so the amount of future subsidies could change as the project progresses, in either direction.  Again, Sanan is a company well recognized by all governments as one that is representative of China’s ability to become a global force in the global LED/semiconductor market, and is therefore more likely to get funding than start-ups with a leading edge process, but the idea that subsidies are only going to semiconductor foundry, advanced packaging, or Ai companies is a false one.

On 6/08/21 we noted that expectations for the Chinese 6/18 shopping holiday were diminished as TV panel and set prices continue to rise globally, making usual holiday discounts less available as TV brands struggled to keep margins from declining.  We also noted that what we believe was a false sense of optimism among TV set producers in China had built inventory levels into the 6/18 holiday, increasing the risk of a slowdown if that inventory was not sold through.  According to early data comparing the extended period leading up to and including the holiday, on-line unit volumes declined as noted below, while on-line sales value increased by 25.7% y/y.  Off-line units also declined by 26.9% y/y and off-line sales value declined by 5.2%.

While we would typically view an increase in sales value as a positive, especially when viewed with the ASP increases seen above, however when unit volume data is added, the conclusion becomes less positive.   While the value of holiday sales grew 12.2% y/y and ASP’s grew 43.2%, unit volumes fell by 21.5%.  Given the optimistic view that Chinese TV brands held going into the holiday, this would imply that not all inventory was sold and the typical inventory refresh following large shopping holidays could be less than expected or eliminated.
While much of global demand will still rely on the US TV market, given its strength over the last few quarters, we have seen weakness in the US TV market recently (see our note 6/22/21) as COVID-19 restrictions are lifted and US government subsidies are ended.  While the effects of weaker TV demand globally will take some time to filter through the supply chain, we expect panel producers will continue to shift production away from TV and toward monitors and notebooks.  That said, we would expect to see TV panel price increases diminish as we noted in our note last Friday (6/25/21), with only driver or other component shortages pushing brands to hold excess inventory.
While it is still early to predict the environment during the 2021 holiday season, given the number of variables facing the CE space currently, just the data above seems to be pointing toward price elasticity being the biggest determinant for the holidays., where a 43.2% increase in ASP led to a 21.5% drop in unit volume, and while that ASP rise led to a 12.2% increase in sales value, keeping TV brands at least somewhat satisfied, the unit volume weakness will have a more profound effect on the TV supply chain, and could help to alleviate some of the component shortages that have helped to push prices up. 
The TV space does not operate in a vacuum, so as noted above, panel producers will continue to shift production to those products with the highest return, which tends to be smaller display products, meaning monitors and notebooks.  The gaming monitor space is growing and there are still country-wide programs that require notebooks or Chromebooks, but any disturbance in demand for those products would begin to show as lower panel production utilization, which is typically not the case during the holiday build period.  Perhaps the industry can scrape by this season but our Magic 8 Ball is moving

From This:

To this:

If you were one of those that collects digital music, photos, and any other file type and has amassed a large collection of same, you have probably considered purchasing an external hard drive to bring your data out of the cloud, where it is not only vulnerable to potential server issues but potentially hackable.  As the cost of storage has been declining since hard drives were developed, the cost of a 2Tb drive, a size reserved for server farms only a few years ago, is now available for $230 and a 3Tb external drive for ~$375.  There are lots of hard drive manufacturers but Seagate (STX) and Western Digital (WDC) are the share leaders, so many branded consumer drives will be based on their devices, but a particular Western Digital product seems to be having a difficult time on its own.
Western Digital’s My Book and My Book Duo drives  seem to have a vulnerability that allows access through commands generated by ‘outside sources’ that cause the entire contents of the drive to be erased by executing a factory reset, unbeknownst to the user  Western Digital has issued a notice to all users , especially those that have their drives directly connected to the internet, such as with a direct connection to a router, to disconnect their drives from the internet and follow steps to connect the drive directly to your computer via an Ethernet cable, rather than accessing the drive via a network or the internet.
According to the WD forum, multiple users have found their drives wiped clean while they were away from the system, with many unable to recover their data.  Those most vulnerable seem to be those that enable port sharing, a way for multiple users (processes) to share a port and IP address.  Using a technique called port scanning, hackers can find those devices that are using this protocol and access the drive with a relatively simple exploit.  What makes this a bit unusual is that WD stopped supporting My Book Live in 2015 with one last firmware update and no breech of either cloud services or the devices has been reported.  Of course those who have not backed up their drives are paying the price, but where do you backup a 2Tb drive to, another 2Tb drive or the cloud, which was the point of moving the data to the drive in the first place.

​Conversations with Chinese display producers can be challenging both in that they are careful to choose which questions they answer, and those answers they do give can be quite difficult to decipher as wording is carefully chosen.  That said, even with carefully chosen wording, Chinese panel producers are usually proud to discuss progress that is being made on fab capacity or new projects that project the company in a positive light and through these conversations and Q&A sessions, we can grab a few tidbits about the working of the display space on the Mainland.
BOE (200725.CH), as the largest of the Chinese panel producers has a number of projects at various stages across its swath of fabs, R&D centers, and other production lines, which are in a constant state of change, and those changes are of great importance to BOE’s supply chain.  We recently noted that Corning (GLW) had officially opened its 6th glass plant in China, this one being in Wuhan on the site of BOE’s Gen 10.5 LCD fab, and is the primary (and likely sole) supplier of substrate glass to that fab, as local glass producers in China tend to be reliable for glass sizes up to Gen 6.  The fab itself is in the process of its phase 2 expansion, which BOE expects to be completed this month, giving the fab a stated capacity of 120,000 sheets/month, which should be reflected in Corning’s volumes going forward.  We note that a 30,000 sheet/month increase in capacity at this plant (Gen 10.5) consumes 80.1% more glass surface area than a 30,000 sheet/month expansion at a Gen 8.5 LCD fab, so the impact of such an increase is significant to any supplier that works with BOE on an area basis (glass, LCD materials, backplane, et.).
BOE is also in what it calls ‘ramp-up’ mode at its B7 Gen 6 OLED fab in Chengdu and its B11 Gen 6 OLED fab in Mianyang, both of which are in production but are expanding capacity, and the company is developing another Gen 6 OLED fab in Chongqing, which is currently under construction.  While the two OLED expansions are proceeding ‘smoothly’ according to the company, the Chongqing fab, when the construction is completed, will begin ramp-up mode and  be ’carried out according to customer needs’, which are ‘climbing’.  We expect this to be a reference to Apple (AAPL) and its quest to further diversify its OLED display supplier list, which now consists of Samsung Display (pvt) and LG Display (LPL), with BOE providing relatively small volumes, primarily for screen replacement.  As we have noted, BOE is trying to become a full-scale supplier to Apple’s OLED smartphone production, but has faced difficulties in maintaining yields comparable to others.  We expect the Chongqing fab is intended to prove to Apple that BOE is capable of becoming a primary supplier. 
BOE expects to increase its shipments of flexible OLED displays by more than 100% this year, but much will depend on the Gen 6 fab ramps and demand from customers other than Huawei (pvt), who has been facing extreme component and semiconductor shortages due to US sanctions.  Last year Huawei released 32 smartphone models with OLED displays.  This year they have released 7 to date, which makes it incumbent on BOE to broaden its OLED customer base in order to meet its goal.
BOE is also developing a 12” silicon substrate line for the production of ultra-high resolution AR/VR displays to complement its existing 8” line from which it says it has shipped both ‘fast LCD’ and micro-OLED displays, however while the company has shown demos of silicon-based OLED AR displays at shows over the years at high resolutions, we have yet to see any actual product above 5,644 ppi, although the company says they have a number of sizes under development.  In terms of micro-LED BOE says they are mass producing a 75” COG (Chip-on-Glass) micro-LED panel, although we believe it is a modular component that can be assembled into a larger system.
As noted, BOE is very aggressive in terms of growing its capacity, not only in display production but in a number of associated areas, albeit most associated with BOE’s LCD or OLED capacity in some way, yet as always in the CE space, much is done with smoke and mirrors.  Non-specific measurements are the stock and trade of panel producers and are helped by rags whose headlines often read, “…expanding capacity…” only to find out that the company mentioned is ‘considering expanding capacity if customer demand warrants the expansion and capital can be secured…’   That’s why analysts have grey hair at an early age.

We have mentioned Samsung’s FE (Fan Edition) smartphones as recently as 6/14 in reference to the initial FE, a refurbished version of the infamous and recalled Galaxy Note 7, and the more recent Galaxy S21 FE released last October, but it seems that while Samsung is expecting to release the next Galaxy S21 FE this coming October, it might not be available to all that would like it.  We noted on the 14th that Samsung had halted production of the S21 FE and was considering allocating the phones application processor, the Qualcomm (QCOM) Snapdragon 888 to other products.  Now it seems that Samsung is considering releasing the S21 FE but only in the US and Europe and excluding South Korea, Japan, and Australia as a possible alternative, with the lack of foundry capacity at the root of the problem, although there is still hope for you ‘secondary countries’ in that Samsung will decide some time before December whether it will release the S21 FE in countries other than the US and Europe, likely based on available capacity at Qualcomm and Samsung’s AP inventory levels after the initial release.