While we are on the subject, we thought it helpful to share a few facts concerning the AR/VR space.  In 2022 15 VR headset models were released from 10 brands, down slightly from 2021 when 17 models were released, again from 10 brands, with 2022 seeing 2 new brands releasing product and two that released in 2021, skipping 2022.  In the AR space, in 2022 17 models were released from 15 brands, up from 13 models in 2021 from 12 brands, with 8 new brands entering the AR space last year.  As smaller (and sometimes larger) companies tend to announce new models with release dates in the future, we enter 2023 with 22 announced VR models, 5 of which were announced in 2023, 11 last year, and 6 in 2021. 
While carrying an announced but unreleased model since 2021 might seem foolish, the Sony (SNE) Playstation VR2 headset was announced in October 2021 and will ship (hopefully) in late February, 730 days after its announcement, so while some announcements might fade into non-existence, we carry them until we know they are officially withdrawn.  The AR space enters 2023 with 14 announced but unreleased models from 13 brands, with two announced this year, 10 remaining from 2022, and 2 left over from 2021. 
One factor for AR/VR headsets that we monitor is whether the devices are stand-alone or tethered,, with standalone meaning they do not have to be attached (either by wire or wirelessly) to a separate device, such as a smartphone or PC.  There are both positives and negative to each mode, with the tethered devices requiring less local (on headset) computing capacity and power, usually equating to a lighter headset, at the cost of being less mobile.  Of the VR devices announced but unreleased this year 63.6% are standalone, while last year that spiked to 80.0%, with 2021 coming in at 64.7%,averaging 69.4% over the current and past two years.  Given the computing and power intensive imaging necessary for VR headsets, we would expect standalone VR headset share to remain between 65% and 75% this year.
In theory AR headsets should be less computing and power intensive and should therefore show a higher rate of standalone vs. tethered devices, however that is not the case as the current and past two year standalone average for AR headsets is 56.5%, lower than that of power hungry VR devices.  That said, the sequential change goes from 45.4% standalone in 2021 to 52.9% in 2022, and 71.4% this year, which establishes a significant trend toward standalone AR.  While we expect some of that trend can be accounted for as we noted above, it is important for consumer AR glasses to be ‘good-looking’, which implies thin, light, and unobtrusive, which we expect has pushed AR headset designers to trade off extended battery life in lieu of a more normal looking device that allows the user full freedom of movement, leading us to expect the trend for AR headsets to be standalone to continue.
Headset price is an obvious factor that has considerable influence on both AR and VR headset sales, but the price range for VR and AR headsets can be so varied that the inclusion of a single high-priced headset can skew the average drastically.  That said, with that understanding, the average price of VR headsets continues to decline, dropping from $3,083 in 2021 to $2,249 in 2022, and $1,374 for those expected to be released this year, although we note that there are a considerable number of potential price variables for some VR headsets, with prices ranging from $300 to $38,500.  AR headsets are generally less expensive that VR headsets, with less display and optical hardware and a more simplistic frame,, but again there are big variations in prices, with a range between $99 and $3,300, including those we are able to price that are to be released this year.  The average price for AR headsets has remained between $965 and $1,002 in 2021 and 2022, inclusive of a relatively small sample of those for release this year.
Even with a large number of variable for both AR and VR headsets there are few consistencies.  Every VR headset for which we have chipset data, produced since 2019 has used a Qualcomm (QCOM) chipset, and while the models have changed over time from the Snapdragon 835/845 to the XR1 and currently the XR2, Qualcomm has established itself as the basis for VR processing.  AR is a bit less narrow, with an occasional Intel (INTC) or AMD (AMD) chipset showing up, but again, Qualcomm is the de facto leader in the AR space.
Displays, especially in VR headsets continue to evolve, with all display types (LCD, OLED, Mini-LED, Micro-OLED, and QLED) all represented in this year’s potential devices, a change from the almost universal LCD displays used in VR in previous years.  This year’s AR headset displays are also quite varied as to display types, but given the lesser display requirements of AR, AR displays, all display types have been common in those headsets for a number of years, including LCoS, which is not used in VR.  All in, we track over 35 variables for AR and VR headsets in our database, and while it can be difficult to get data from some brands, there are many discernable trends that become visible.  More to come…

​Rumors and speculation continue to swirl around Apple’s (AAPL) plans for its entry into the world of AR/VR, with a multitude of hardware and options ranging from a relatively expensive (neighborhood of $3,000) and/or a low-priced alternative to compete with mass market VR headsets such as the Oculus (FB) Quest 2 or the Pico (pvt) Neo 3.  Apple’s entry into the AR/VR market will be a watershed event for the industry, but we see a trend that will blur the lines between AR and VR over the next year or two.  Currently there is a distinctive difference between headset designs for AR and VR, with AR headsets trending toward almost normal looking eyeglasses and VR headsets still more massive and obtrusive.  The technology behind AR was developed to combine the user’s view of the outside world, and an artificial image that can be superimposed on a normal visual image.  VR systems do not use external visual images, and in fact, require a darkened, enclosed space inside the headset to be fully effective.
This dichotomy is obvious in most cases but we have noticed that both VR and AR headset developers are more extensively looking over their shoulders and wondering if they each might be missing out on something.  In the vase of AR, we have seen a few systems that include technology that can darken the image an AR user sees, emphasizing the artificial overlay and reducing the impact of the pass-through image.  As the display overlay systems become more sophisticated in AR headsets, it seems AR developers are trying to gain at least some access to VR, albeit in a rudimentary way, as the AR market develops broader applications.
The same seems true for VR, where a number of systems either allow the user to remove the light blocking cover typically seen on VR headsets, which allows the system to function like an AR device, at least in a limited capacity and other systems use pass-through cameras to mix the outside world with VR.  While it might seem that giving a VR headset the ability to be used for AR is overkill, we expect the VR case is more likely used to allow a VR user the ability to intermittently ‘see’ someone or something in the user’s location, without removing the headset, but we expect the trend toward such mixed use devices will continue, making data collection a bit more difficult as AR and VR devices move toward each other’s home ground.
We believe both AR and VR headset developers will continue to improve this ‘mixed reality’ feature set and over the next two years, while there will still be dedicated AR and VR headsets, a greater number of new releases will gravitate toward becoming mixed reality devices, in order not to miss an unexpected growth path of application classification, and while Apple’s entry into the space seems to be inevitable, we expect Apple will be careful not to limit their market specifically to AR or VR.  Perhaps that might not be the case with the company’s first AR/VR iteration, but we believe Apple is smart enough not to pick a side and hope that consumers agree.

​​LG Electronics (066570.KS) released a number of updated laptops at CES 2023, as part of its Gram laptop line.  What makes these laptops unusual is the Gram 15.6” “Ultraslim” and the Gram 14” and 15” “Style” models now use OLED displays.  In itself, it is not surprising for LG Electronics to feature OLED displays as it is the leader in large panel OLED display production through its affiliate LG Display (LPL), who also produces small panel OLED displays for Apple and a number of other smartphone customers., but in the case of these new OLED laptops, parent LG Electronics is buying the OLED panels from rival Samsung Electronics. 
Samsung Display (pvt), affiliate of parent Samsung Electronics (005930.KS), is the leading producer of RGB OLED displays, primarily for use in smartphones and tablets, while more recently building out its RGB OLED display line to include larger panels suitable for the notebook market.  While smartphone OLED displays are typically produced on a flexible substrate, with the phone’s structure providing the rigidity, laptop displays tend to be produced on rigid substrates (glass), as the more structurally solid frame keeps the glass-based display from flexing.  Further, the production cost of glass OLED displays is lower than that of those produced on flexible substrates, as during the production process the flexible substrate material must be attached to a glass substrate to make sure it remains flat during deposition, and them must be removed using a LLO (laser lift-off) process to return the display to a flexible state.
LG Display does not produce rigid OLED displays, producing only those with flexible substrates for smartphones, watches, and automotive displays, so in order to keep down cost, but also provide an OLED alternative to its laptop line, LGE had no choice but to buy from SDC or go to a Chinese supplier.  Considering that it has had problems with the displays on earlier models in its laptop line, the choice of Samsung seems logical, albeit a bit embarrassing.  We believe Samsung Display has ~ an 85% share of the rigid OLED display market, with only Everdisplay (688538.CH) and Visionox (002387.CH) competing and has considerably more experience than the others in producing these large OLED displays.
What also stands out in this unusual relationship is that Samsung Electronics and LG Display were said to be in negotiations last year for a potential purchase by Samsung of a large (>1m) number of OLED TV panels, which never occurred.  Rumors that while LG Display was willing to lower the OLED panel price below the price at which it sells OLED panels to its parent, the two were still unable to reach an agreement and the deal was called off, but when it comes to CE products, margins rule. 

​Quantum Dots are almost passe in CE circles, with hundreds of millions of displays using these particles to enhance both LCD and OLED technology.  Originally, quantum dots were used in films that took the light from an LED source (usually blue LEDs) and created a broader color source that passed through liquid crystal and a phosphor-based color filter.  The QDs converted the blue light into other colors (red & green primarily), which added to the color that was generated when the light passed through the red and green dots of the color filter.  Color filters however are subtractive in that a red dot in a color filter allows red to pass through while blocking blue and green, which reduces the overall brightness and forces the set to rely on phosphor formulation for its color purity.  QDs convert colors rather than filter them which means  less total color degradation when QDs are part of the LCD or OLED stack.
Samsung Electronics was and continues to be a strong promoter of QD technology, first because it differentiated Samsung’s display products, and second because they did not have a non-LED TV technology that could compete with LG Display’s (LPL) WOLED TV technology, which has been promoted as having more vibrant colors than typical LCD TVs.    Last year Samsung took QDs a step further and combined them with OLED technology to create QD/OLED displays, that use a combination of blue and green OLED emitter material to generate light, and quantum dots to convert that light into colors, producing an OLED based display that does not use a color filter.  When compared to typical WOLED displays, which use OLED materials to create a white light source but use a phosphor-based color filter to create colors, the QD/OLED display is, in theory, brighter, with the QDs (additive) replacing the phosphors (subtractive).  Samsung Display (pvt) the producer of QD/OLED displays, has a single production fab for this new technology and is in the process of deciding if and when to expand production, but a bit further down the road is another path for quantum dots, and one that has even more implications for the display space.
While QDs are typically stimulated into operation by light, they also have the characteristic of being able to be stimulated by an electrical charge, similar to the way OLED materials generate light.  In this case the QDs are self-emissive or electro-luminescent, and present another alternative to existing and potential EL display technologies, such as micro-LEDs and OLEDs.  That said, as QDs are relatively new to display applications, much work needs to be done to design QD materials and driving circuitry to bring EL QDs to market as a competitive display technology, but given the relatively simple production process for quantum dots and what seems to be a simple stack structure, they are particularly attractive as a potentially disruptive display technology.
Nanosys (pvt), the leader in QD technology, and a licensor to Samsung, privately showed a 6” QD EL display at CES, which, while still in development, is a step toward the commercialization of this technology, with the promise that more information about the technology would be shared in a few months by a yet unnamed manufacturing partner.  Nanosys was careful to point out that true EL-QD commercialization is still a few years out (2025 -2026 at the earliest), but that the demo was a pure QD EL, using no other materials or filters.  Considering that QDs are already being printed in the Samsung QD/OLED process, the production costs for EL QDs, when they are ready for mass production, should be low and applicable to almost any surface, so some of the drawbacks facing other existing and potential display technologies are less of a roadblock to EL-QD commercialization.
All in, while the display and CE industries are not going to change overnight, there are a number of new display technologies vying for the display technology throne and while EL-QDs are still toward the back of the line, recent progress establishes a timeline.  In the display space timelines are more often missed than met, but at least there is a finite goal and a technology that does not have to start from scratch  today.  

​A few days before CES this year Chinese CE brand TCL (000100.CH) indicated that it would be shipping it first QD/OLED TV by the end of this year at a press briefing and in materials available to the press before the show began.  While the technology behind QD/OLED is broad enough that patent protection could likely be relatively easily side-stepped, TCL, or one of its subsidiaries in the display space, most likely Chinastar (pvt) or China Display Optronics (334.HK) would have to build at least a pilot line to produce such a device, with enough capacity to at least place product in conspicuous locations.  Certainly we are not saying that such a pilot line has not been built, but Chinese brands are want to publicize achievements and little has been said about such a development at TCL, although they have made considerable progress in developing Mini-LED technology.
As it turns out, the TCL statements turned out to be a mistake, and the company indicated that such plans were incorrect and its intention is to continue to develop and promote its mini-LED display technology this year, leaving Samsung Display as the sole producer of the technology and Samsung Electronics and Sony (SNE) as the only brands currently offering QD/OLED TVs.  While TCL is best known for their lower priced but higher quality TVs, and being particularly fierce competitor in the mini-LED TV space, it looks like TCL’s marketing guys got a bit carried away pre-CES, and we will have to wait a bit to see how far along TCL is concerning QD/OLED display technology, at least until 2024.
 
 

We are big on CE products that are practical, perform functions efficiently, and are not exorbitantly expensive, all relatively simple goals, although there are many CE products that do not fill even these basic parameters.  While we have yet to put our hands on the specific Samsung (005930.KS) device mentioned below, that was announced at CES 2023, we see the device, if it lives up to its promotion, as one that might fill the bill.  The device is called the Samsung Smartthings Station and will be released in Korea (and likely the US) next month.  This is a relatively simple device that represents the ‘new world’ in smart home devices promised by the new ‘Matter’ connectivity standard, that allows Matter enabled products to interoperate with other devices and controllers, as opposed to forcing the users to use proprietary controlling devices.
In the pre-Matter world, you might have a lighting dimmer system that you controlled with wall dimmers, a handheld controller, or a smartphone app, and at the same time you might have automated curtains that operated with another handheld controller or app.  Then there was your smart TV, with its own remote, unless you are using an external Roku (ROKU) device, for which there was another controller, and the list goes on with doorbells, leak sensors, smart refrigerators, and a variety of other devices, all of which needed some sort of controller or app, mostly proprietary.  There are many ‘hubs’, which act as central control points for a number of devices, but again most are able to work only with devices made by a specific brand, so smart homes tend to be a mass of proprietary systems that did not talk to each other.
Matter is to change that, with a standard that can be adopted by any device, brand, or smart product, giving it almost universal communication ability with other devices, and allowing the user to decide which controller he or she might want to use once they become ‘Matter’ enabled.  So, in theory, users should eventually be able to control all of the smart devices in their homes with a single controller (of their choice), although adoption will take some time and some older legacy existing devices might never become Matter enabled.
Samsung’s device is a simple one, and supports Matter, Thread[1], and Zigbee, and at ~$60 is the least expensive border router currently available that supports Thread and Matter, allowing home devices to be connected in a mesh configuration.  This allows for better communication between devices on the network and in most cases a larger reach for the network.  Samsung’s iOS is also the only OS that currently supports Matter, other than Apple’s (AAPL) iOS, with Google (GOOG) Home and Amazon (AMZN) Alexa, expected to add Matter support to their OS’s later this year.  Samsung is spreading out the Matter coverage across its own product line, with the Samsung Family Fridge line, , Samsung TVs, and Samsung monitors getting the upgrade in March and others to follow later this year.
The device itself is a wireless charger and supports three ‘routines’ with a simple press of the device.  The routines are set by the user and can operate a single function, such as turning on the lights, or a sequence of functions, such as turning on lights while opening curtains and turning on a smart speaker.  But here’s where it gets better…  Scenarios can be programmed into the device that are actuated by placing your phone on the charger, such as one might do before going to bed.  This can be set to trigger a routine that could close curtains, turn off lights, and turn down the heat, all of which would be reversed automatically when the phone is picked up from the charger the next morning, and we note that as long as the devices are Matter enabled, it does not matter which brand they are produced by, they would all operate under the Samsung device.
Not to be outdone by Apple, Samsung also includes the ability to find Matter devices within the home network range, so misplaced smartphones can be triggered to ring wherever they might be, and items equipped with smart tags, will contact the system and sends an e-mail or text to the user whenever they leave or enter the network area, so if a dog with a tagged collar gets out, the user will be immediately notified.  Of course this means that anyone trying to sneak out of the house or sneaking back in will generate the same notification if they have their smartphone with them, so errant spouses and sneaky teenagers beware…
All in, at least on the surface, Samsung has taken a set of simple but useful functions and put them in what is a relatively low-priced package that has at least some intrinsic value on its own (wireless charger).  Our only concern would be how difficult the device is to pair (attach to other devices) and program, which have been stumbling blocks for earlier smart devices, although Matter itself is said to make those functions a bit more intuitive.  More philosophically, Samsung, who abandon the smart home hardware business roughly three years ago, seems to have thought this re-entrance through, even stating that the company would rather embed its smart home technology in its products rather than creating branded hardware, a bit of a change for a company known for its CE hardware, although the SmartThings Station is just that, a branded product.  That said, with Matter leveling the playing field, consumers will have infinitely more choices about sensors, controllers, and applications, and the first Matter offering from Samsung seems to show that they are working toward making their smart home products practical rather than forcing consumers toward expensive proprietary hardware.


[1] Thread is a Wi-Fi mesh network protocol that extends and enhances IoT devices that can extend wireless networks of varying types including Wi-Fi, Ethernet, LTE, 5G an others, using low power and low latency connections between all devices, essentially a self-healing network. 

​More on Matter:
“Will it Matter?” – 07/21/22
“Apple Makes it Easy to Fix an iPhone, Mostly” – 09/20/22
“Will It Matter?” – 10/06/22
“More On Matter” – 10/13/22
“It’s a Complicated Matter” – 11/14/22
“What’s the Matter Now?” – 11/30/22
 

Starting the new year, large panel LCD prices changed modestly, down 0.1% m/m, a far cry from the volatility seen last year (Jan ’22 was down 4.3% m/m), with only tablets seeing prices decline out of our forecasted range.  While not in our large panel data, mobile phone panel prices also dropped more than our expectations, as weakness in the smartphone sector saw the weakness continue into the new year.   While panel shipment and revenue data is typically a month behind panel price data, December panel producer sales were up 2.0% m/m (down 28.7% y/y), while unit shipments were up 3.4% m/m (down 20.6% for the year), pushing large panel overall ASP down by 1.3% for the month.  Tablet and notebook shipments were up 8.6% and 11.7% m/m in December, which would portend a relatively weaker shipment month in January, especially given the early Chinese New Year. 
We believe inventory levels for most IT panels are close to normal, as are TV panels, but demand is still an issue, which leads to a basically flat January in terms of panel prices, with some seasonal weakness in shipments through the Chinese holiday.  All in, a quiet start to the new year, and with few surprising announcements at CES earlier this month, little new product to attract consumers.  Samsung Electronics (005930.KS) is expected to release it next iteration of its Galaxy S flagship smartphone line in early February, but without something unexpected, such as a new Galaxy Fold model (The ‘Folds’ tend to be released around August), price will most likely be the deciding factor for most smartphone buyers, rather than new features.  

Samsung had some trouble pricing its inityial release of its quantum dot/OLED TV last year, originally listing the 55” and 65” models at $$2,400 and $3,500 respectively, prior to their April release, however the final pricing at release was lower for both, at $2,200 and $3,000, indicating the companywas still trying to best realize the most attractive price points, both from a profit perspective and from the consumer’s perspective.  As the QD/OLED concept was new for consumers it was up to Samsung to establish a point where these new sets could be placed in their TV line-up, but at the time Samsung Display (pvt), Samsung’s display affiliate, was having problems producing the  displays, with yields below 50%, and with Samsung Electronics targeting between 45m and 50m sets to be shipped for the year, such low yields meant that Samsung would have little reason to market sets that had such a small impact on the overall TV line.
Samsung Display, knowing without the support of its parent, having only Sony (SNE) as a 2nd customer for the TV panels, pushed its manufacturing lines to solve the yield problems, which began to see improvement around mid-year.  With yields improving, Samsung could justify the QD/OLED product line, and while it was still a small part of the entire Samsung TV line, the volumes were increasing at SDC and the potential for the QD/OLED line was finally acknowledged.
The problem however was that the combination of Samsung’s lack of marketing, weak overall TV set sales, and worsening TV panel pricing, left the QD/OLED sets in a netherworld, unrecognized by consumers, despite very favorable technical reviews.  Samsung seemed to recognize that they had a choice between a massive advertising campaign and setting a more attractive price point, and given the fact that even with the improving yields SDC had relatively limited display production capacity compared to other TV set modalities, the latter was chosen and the price of the sets was reduced.
While the price on Samsung’s website for both sets did change, the price on Amazon (AMZN), particularly those from 3rd party sellers, was much more telling and makes obvious the price point at which the sets became attractiuve to consumers, as shown in Figure 5.  While some of that volume can be attributed to improving yield at SDC, it seems that by October the price points (~$1,500 for the 55” sets & ~$1,890 for the 65” sets) had been reached and sales improved significantly,  We note also that during the earlier part of the year, Sony’s share was increasing and in September both Samsung and Sony had equal shipment share, but while Samsung continued to search for a more compatable price point, Sony did little to change the price of their QD/OLED sets, and share declined rapidly, hitting a low of 11.3% in November. 
We expect Sony’s purchase targets from SDC were set early in the year and the improvememnt in SDC’s yield did little to change that, with Samsung Electronics getting the bulk of the increasing shipments, but the chart makes it convincing that such a steep increase in QD/OLED shipments was directly influenced by the price, especially as it leveled off when the increases occurred.  3rd party pricing on Amazon does not always reflect list or discounted pricing om brand sites, but it seems a better indicator than Samsung’s own ‘brand’ pricing, which did decline from the initial $2,200 & $3,000 to $1,500 and $2,000 during the holidays last year (up slightly since), so even with the vaguries of 3rd party pricing on Amazon, it seems obvious that Samsung found the price point it needed for the QD/OLED sets to compete with other TV set brands.
The question remains however, wher=ther the relatively small number of QD/OLED displays that can be produced by SDC will be enough for parent Samsung to truly support the technology as a viable part of its TV product line.  Samsung has indicated that it will also be producing a 77” QD/OLED model this year, which is a size format that can be shared with 49”/50” displays cut from a single sheet, so we would expect to se additional QD/OLED panel sizes later this year, but most important would be the OK from both SDC and Samsung Electronics concerning the expansion of caapcity for the QD/OLED process, and that is a hard decision to be making during the weakest part of the yearly CE cycle.  Hopefully that decision has already been made and suppliers contacted if such capacity could be put on this year, but until SDC makes such a commitment, QD/OLED will remain a small part of Samsung’s TV set line.

It is Chinese New Year, which runs all this week, so to celebrate the Year of the Rabbit, we asked two AI websites to create original artwork using the phrase “Year of the Rabbit”.  Here are two examples of what was generated by these sites, both of which are free, using the same text.  With considerable new coverage appearing toward AI assisted art and writing applications that the general public is just becoming aware of, we thought it appropriate to generate our Chinese New Year art through such systems, rather than through the usual Google (GOOG) image search.  Are they better, more appropriate, or more to the point than original (human) art?  “Art is in the eye of the beholder, and everyone will have their own interpretation”[1]


[1] E.A. Bucchianeri – “Brushstrokes of a Gadfly”

While rumors that Apple (AAPL) has postponed its AR device project in favor of an XR device to be released this year or next, there has been considerable activity concerning AR devices before and during CES 2023.  While we still favor AR over VR in terms of practical applications, the extended reality universe has been gradually attracting larger CE companies, and with those companies comes the need for high volume products.  Gaming is the most obvious application for VR, particularly as enthusiasm for the Metaverse wanes a bit, and as a ~$200b market, the gaming software market makes the hardware that feeds such a large space quite attractive to CE manufacturers looking to become a new player or expand their share of CE hardware.
That said, while VR is the focus for most hardware manufacturers, the more recent trend has been toward creating a mixed environment that allows VR headsets to also be used for AR, and while this is really a convenience for VR users, it adds a bit more credibility to the AR space, which saw 17 models from 15 brands released in 2022.  In that group were 8 brands that were new to the AR space, although some have VR offerings, and of those eight, three are likely recognizable to our readers, Oppo (pvt), Xiaomi (1818.HK), and Huawei (pvt), all well-known Chinese CE companies with representation in a number of CE segments.  Further, announcements of upcoming AR devices were made by 13 brands, including TCL (000100.CH) and Lenovo (992.HK), also major Chinese CE companies, indicating the shift from small, private brands to large Chinese high-volume manufacturers, although we do note that the earliest supporters of AR hardware were Google (GOOG), Microsoft (MSFT), and Meta (FB) years ago.
We believe it is important for investors to understand a bit about both the AR and VR space, both from the perspective of being able to understand trends in the industry and to better understand how new products will affect the space and the private and public companies involved.  SCMR LLC has developed a comprehensive database of both AR and VR products (headsets) which breaks down the hardware into relatively simple categories, but before we detail some of that data, we believe it would be helpful for investors to understand the basics concerning AR headsets.
The idea of Augmented Reality, in a limited form has been around for many years but much of the early work on developing a system that could provide such visual information to a user was developed by and for the military, either as a ‘simulator’ or a HUD (Heads-up Display), with much of the research being done at various universities.  Early pioneers such as Evans & Southerland (pvt) and Boeing (BA) were limited by the need for computing power and graphic capabilities, but in 2013 Google opened the concept of AR to the consumer world with the release of Google Glass to a limited group of ‘Explorers’ (You could ‘explore’ for $1,500 & ~8,000 did), and roughly a year later to the general public, and followed with ‘Enterprise’ and ‘Edition 2’ models in 2017 and 2019, although consumer oriented production was ended in 2015.  The device was able to project data to the user’s eye using either a touch system on the side of the device or by voice commands, and could take pictures and short videos and was able to project requested directions, messaging, Google search results, or connect to your smartphone.
Unfortunately, the cost was high and while the headset was remarkably sophisticated for its time, the concept of walking down a street with such glasses while viewing data was a bit ahead of its time, and privacy concerns about not knowing when the Google Glass user was recording, left many concerned over privacy.  No matter what the problems were with Google Glass at the time, you have to give Google considerable credit for trying to bring AR to the consumer market, and despite Google’s problems with ‘glass’, Microsoft released the Hololens in 2016, a Windows AR device that was oriented toward business applications (~$3,000), and Oculus (now Meta) released the ‘2’, a less expensive (~$1,500) but bulkier AR headset in the same year.  From that point forward AR device releases came primarily from small, VC or private investor based firms, and while both Microsoft and Google released updated models in 2019, the number of devices released each year was minimal.  That changed in 2020 with over 10 models released, 2021 with 13, and last year with 17 new models.

What makes AR different from VR is that AR is not an immersive experience and allows the user to see and interact with the real world while seeing projected data or images overlaid, essentially a ‘pass-through’ device.  Simplistic uses, such as directions, similar to automobile HUDs, give the user the ability to follow directional indicators that move with the user without holding a smartphone, and more sophisticated uses, such as language translation are basic to AR (Watch the video here to see the translation function in action) and have real world uses that over time, will become ubiquitous with over 7,000 languages spoken across the globe and a global population of over 430m considered deaf (ASL can be translated). 
Marketing is an even more obvious use, allowing a shopper to look at an object, such as through a store window, and see information about that item instantly, sort of having a personal assistant available 24/7, with sophisticated retailers opening the door to consumers ordering items as they wander through a store, without the need for ‘advocates’ to help them make a physical sale.  More commonly seen uses for AR are in the business world, where manuals, data, and other information about complex systems can be projected directly on such items, giving workers the ability to see how repairs are made, or procedures for maintenance, and during meetings, users can see data and images while also seeing facial expressions and objects around them, rather than avatars in an unreal VR environment.

Of course, AR devices vary considerably, both in feature sets and price, and with 14 AR headset models already announced but unreleased and the potential for Apple to enter the market at some point, it is important to understand the potential differences and a bit about how AR devices work.  Among the most important items to understand is whether the device is a standalone one, or one that needs to be tethered to another device, in many cases a smartphone.  This is important as non-tethered headsets would need the ability to communicate with the outside world directly, while tethered devices could use the communications hardware of a smartphone or similar device.  The hardware needed for direct communication would add weight, bulk, cost, and power usage to the device, but would free the user from some mobility constraints.
There are a number of display technologies used in AR headsets.  Some display the AR image on a transparent display, some use non-transparent displays combined with camera images, and some project the AR image directly on the user’s eye, but the quality of the display, the cameras, and or the optics are important factors in determining the quality of the device and what its uses might be.   The display itself is important and a variety of display technologies are used in current AR headsets (OLED, Micro-LED, Micro-OLED, and LCD), but combining camera images, and AR images requires optical systems that take up room and add weight, which makes the development of consumer AR not only based on the quality of the display, but on the optical technology. 
Along with the display and optical hardware, the systems needs to be able to understand the user’s position in space, so it can move the overlayed image as the user’s gaze moves.  This is done with tracking cameras, accelerometers, and other sensors, and in some cases eye tracking.  All of this and built-in speakers, a microphone, memory & storage capacity, at least some CPU capacity, and a battery, must fit in a lightweight and thin frame (averages ~70 grams or 2.5 oz. as opposed to VR which averages ~17.5 oz.) that looks (hopefully) almost like normal glasses.
As with VR, the resolution of the displays themselves is important, especially as they are so close to the user’s eye, with typical AR devices sporting full HD (1920 x 1080) displays, but as AR displays are used in normal or even bright light situations, display brightness is also a factor that must be considered when evaluating AR headsets. While we do not trust advertised ‘peak brightness’ metrics used by many brands, brightness levels (peak) have been increasing over time, with Micro-OLED (OLED on Silicon) and Micro-LED displays offering the brightest current headsets. 
There is one headset characteristic that is important to both AR and VR, and that is FOV or field of view.  In VR headsets an FOV of 90° or more is essential, as the brain is more easily fatigued when it receives a more narrow visual field, and while AR devices are essentially visual pass-through devices, the larger the FOV in AR devices the more natural the overlay seems and a full view requires less head movement but at a higher cost.  Therefore, the cost of AR headsets can vary considerably based on hardware, with 2022 released devices ranging from $349 to $3,300 (~$500 average), and as larger CE companies move into the space, the ability to lower overall product costs is improved.
While we believe VR is still a technology looking for an application, we see so many use cases for AR that we wonder why AR headset brands seem less inclined to promote the technology to the general public than VR brands.  We do note that some AR applications require software developed specifically for that use case, such as AR used in a surgical suite or use in vehicle repair and maintenance, but when looking at such a basic use as language translation, much of the software has been written and is in the public domain or easily licensed.  A reasonable looking AR headset that sells for $150, would be able to solve communication issues for millions of individuals, families, and organizations, and while we are not quite there yet, we believe such a target is not that far off based on the progress made over the last few years.

​Key:
Top Left – ThirdEye Razor
Top Right – Vuzix Blade 2
Bottom Left – Magic Leap 2
Bottom Right – Huawei Vision Glass