image. png

main points

Compared with AR (augmented reality) display, the supply chain of VR (virtual reality) display is more mature, because LCD or AMOLED display based on TFT drive line is widely used, and relevant panel manufacturers already exist, so production capacity and supply are not a problem.

The main components of AR display are micro display and optics. The two need to be combined to function. Ledos (LED on silicon, micro LED) has the advantage of high brightness and enough to resist the intensity of ambient light, but its colorization technology is not yet mature.

From the perspective of system design, wearable augmented reality (AR) and virtual reality (VR) devices include three main parts. The first is presentation or display, and the second is man-machine interface and machine sensing. Sensors that can sense the real world and be digitized are particularly important for high-end ar devices. The third part is computing and communication. To render computer generated objects for VR, AR or mixed reality (MR), computing performance is essential; The processor for computing can be the wearable device itself, or a mobile phone or personal computer using tethering. In the process of connection, it can be wired, or wireless connection such as Wi Fi 6 and 5g modem can be used to obtain a complete wireless experience.

Man machine interface and machine sensing can make use of existing sensor technology. However, 3D depth sensing, which digitizes the real world (or the surrounding world when using the device), is more challenging. Especially when users and devices are physically moving in the real world, slam (simultaneous localization and mapping) may also be designed and added. In the technology of three-dimensional depth sensing, considering the factors such as sensing principle, sensing distance, ambient light and the complexity of depth construction calculus, time of flight (TOF) may be more suitable than structured light technology. However, generally low-end VR or ar devices may not need to track or digitize the real world.

AR and VR have different display technologies in the display part, which depends on the environment and requirements of the application side. Wearable VR devices usually use closure, which is to improve the user's immersive experience. At the same time, because of this closed situation, the consideration of display is simpler than ar. When the wearable ar device is used, the main interference comes from the ambient light (possibly more than 800 NITS), or the complexity of the real world environment overlaps with the AR display, making it difficult for the user to identify. For example, black is not easy to appear in AR display, and dark objects may also be confused with dark objects in the environment. For the AR device design using OST (optical see through), the real world and virtual ar display are superimposed on the waveguide optics, and the above interference is more significant.

On the other hand, for devices that use video see through or passthrough (such as VR or some MR device designs), the real world is photographed and entered through the front lens of the device, and then overlapped with the virtual world through the image operation of the device. Therefore, the interference can be improved through image and processing. Wearable AR and VR devices have different applications and purposes, and there is not a relationship of mutual substitution. For example, when using "ingestion and viewing" devices, users' mobility and recognition ability in the real world will be limited, but "direct viewing" devices have less such restrictions. Therefore, the development of wearable AR and VR devices and the demand for display technology are not exactly the same. In addition, compared with the closed situation that VR needs to wear, the generation of AR does not necessarily depend on wearable devices. Smart phones and head up displays on vehicles can produce the effect and application of AR.

Displays for VR and AR applications

For VR displays, TFT LCD and AMOLED, which are already quite common in smart phones, are suitable for VR devices. However, some device manufacturers adopt "silicon-based OLED" (OLEDos, OLED on silicon or micro OLED). OLEDos has a driving circuit based on semiconductor CMOS silicon rather than TFT line. Therefore, OLEDos has the opportunity to have better specifications than AMOLED in resolution and size. VR display design can be single display or dual display. Both methods are adopted by manufacturers, mainly considering factors such as cost, field of vision and comfort during wearing. In addition, optical lenses, such as Fresnel lens or pancake lens, are used to better concentrate the display light and transmit it to the eyes. The main requirement for VR display is higher resolution to obtain higher "pixels per degree" and lower "screen door effect".

The "angle" in PPD is based on the included angle or field of view (FOV) between the display and the eye. The larger the FOV, the better the immersion. Low persistence of display is an important specification, because the content of most VR applications is computer images rendered by 3D operation. Usually, the VR display will emphasize the refresh rate of 90hz or 120Hz. However, in the actual design of VR device, we need to make some compromise considerations on the marketing positioning of the product; Higher display specifications mean higher display costs, higher power consumption and larger overall dimensions. For example, the first generation of oculus uses a 3.5-inch 1440x1600 AMOLED with dual screen design, while the more popular second generation is a 5.5-inch 3664x1920 LTPS TFT LCD with single screen design, so as to reduce the cost and price and improve the market penetration. The next generation of high-end products may return to the dual screen design.

Most ar displays are micro displays based on silicon circuits, such as micro display technology of micro electro mechanical systems (MEMS), including digital mirror device (DMD or DLP), laser beam scanning (LBS) - for Microsoft hololens, and self luminous silicon-based LED (ledos, led on silicon or micro LED). In the actual optical design, users do not directly stare at the micro display, but watch through additional optical elements, which is quite different from the common flat-panel displays (such as TFT LCD and AMOLED). Most of the early ar display optical designs could not give consideration to lightness and display area. For example, the patent of Google glass (us9013793) revealed the use of polarization beam splitter (PBS) to introduce light from the micro display into the prism with a certain thickness, and the optical element is located on the user's glasses. At present, the mainstream method is to use waveguide optics, which is satisfactory in thickness and shape.

When these wearable ar devices are used, the image display is transmitted from the micro display; The light is then directed into the waveguide optics. The waveguide optical element is thin and nearly transparent, so users can see the superposition of objects in the real world and the virtual world in the waveguide optical element at the same time. The working principle of waveguide optical elements includes diffraction or holography. Unfortunately, the loss of light after passing through the waveguide is almost 99%, which will make the weak light look weak and unknown under the contrast of strong light in the real world. Therefore, high brightness is very important for AR display. In general, compared with AR display, the supply chain and color technology of VR display are quite mature. This is because LTPS TFT LCD or AMOLED displays that have been mass-produced are widely used, and many panel manufacturers already exist, such as JDI, sharp and Samsung display.

Figure 1: display design of wearable device

image. png

Source: Touch Panel Market Tracker, Omdia

Supply chain for VR and AR displays

There are some reasons why the AR display supply chain is not mature. First of all, the difficulty and accuracy of being applied to wearable devices and becoming micro displays are absolutely no less than that of general TFT displays. Moreover, even though there are some applicable micro display technologies, the industry has not yet reached the best technology maturity and consensus under the consideration of optical machine size, efficiency, color, cost and other factors. Of course, more importantly, the current shipments of wearable ar devices are too small. At present, most of the main applications are concentrated in vertical fields (such as industry, medical treatment, military, etc.), which can not persuade manufacturers to increase investment and production in the supply chain. At present, many manufacturers in this supply chain are small in scale, some employees are only about 100-200, and the mass production level of the equipment is completely different from that of TFT display industry. In order to promote adoption and ecosystem development, some manufacturers cooperate with each other and provide reference designs including display, optics and devices to encourage more brands to develop products.

The main components of AR display are micro display and optical devices, which are a combination. Ledos has the advantage of high brightness. Even after the weakening of waveguide optical elements (only about 1%), it has the opportunity to maintain it to 1000 nits. However, its RGB color technology is not yet mature, and the tried technical directions include color conversion, RGB stacking and RGB chips in a cube. In contrast, the color technology of OLEDos is more mature. It is common to mix white light and pass through the color filter, but this method has high light loss. In addition, there are also some methods of direct fine evaporation on OLEDos materials, stacks (e.g. kopin trio stack) or direct fine evaporation (e.g. emagin DPD) ?） Let's go. OLEDos is more suitable for closed device designs such as VR and MR, but its brightness value may be too low for AR devices (open design) passing through waveguide optical elements. Although LCOS, LBS and DMD are larger than ledos and OLEDos, their important advantage is that colorization and high brightness are feasible. All three can improve the brightness value through laser light source.

image. png

Source: Touch Panel Market Tracker, Omdia

Compared with ledos technology, which still needs further development, the relationship and series connection of OLEDos in the supply chain have taken shape, and the AMOLED panel factory of the existing TFT ecosystem has also shown interest. In the TFT ecosystem, DDIC (display driver chip) and AMOLED process (display pixel line and evaporation) are separated and perform their respective duties. DDIC is used to drive the display panel, and the panel factory is responsible for manufacturing the display pixel circuit based on TFT. However, this does not necessarily apply fully to OLEDos. Compared with AMOLED based on TFT, the supply chain of OLEDos is still in its infancy. There may be more business models between DDIC chip manufacturers and OLEDos panel manufacturers.

First, DDIC chip manufacturers design DDI and display pixels circuits on the wafer at the same time. Then, the OLEDos panel factory is only responsible for OLED evaporation on the wafer and dividing it into OLEDos chips. DDI circuit and display pixel circuit are located on one chip at the same time, which has become a single-chip design, and DDIC chip manufacturers play a key role.

Second, DDIC chip manufacturers only provide DDIC chips. This chip may be an ASIC commissioned by the panel factory or an existing product of a DDIC chip manufacturer. Either way, DDIC is separated from the display pixel circuit and has become a dual chip solution. DDIC can be attached to the display pixel chip, which is similar to COG (chip on glass) on TFT LCD.

Third, the panel factory is responsible for all design and manufacturing, from DDIC to display pixel circuit. This may be a single-chip or dual chip solution, and emphasizes the value of the panel factory in the supply chain. There is still debate about single-chip or dual chip OLEDos. When a higher resolution process, such as a 40nm semiconductor circuit, is usually required, especially when a higher resolution process is required. However, the display pixel circuit may be satisfied by using a lower level process, such as 90nm or um level. If the semiconductor process gap between the two is too large, single chip design is not necessarily more cost-effective than dual chip design.

At present, the shipment volume of VR devices is significantly higher than that of AR devices. In addition to the mature display technology and supply chain, the application end is more inclined to the consumer market is also an important factor. The potential of VR in games and virtual social networking is easy to see. However, after attracting early players, in the end, not most consumers are interested in wearing heavy devices for a long time and entering a virtual world. For the most famous brand, AR will take off