On January 20th, Xiaopai Technology, a well-known VR headset manufacturer in the industry, held the "Virtual Reality • Xiaopai Technology 2022 Online Media Sharing Meeting" online, at which the company's CEO Ren Pan introduced his views and concepts on the metaverse and the VR industry, including the relationship between VR and the metaverse, the significance of immersion to the metaverse, the requirements of immersion for VR hardware and the industry standard of VR 3.0.
Sina VR helped everyone sort out the most dry part of the media sharing meeting, the following is Ren Pan's content sharing.
Pictured: Ren Pan, CEO of Xiaopai Technology
One. The form, purpose, and important influencing factors of the metaverse
1. The form of the metacosm
What exactly does the metaverse look like? Different industries and different companies may have different imaginations of the metaverse, and there is no definite statement at present. If I were to summarize the definition of the metacosm, the metacosm is a comprehensive revolution that incorporates information, the Internet, artificial intelligence, and VR, AR, and MR game engine technology. What the metacosm offers to humanity is the possibility of constructing a holographic digital world parallel to the traditional world.
Put simply, the current concept of a metacosm is like a hodgepodge of things that describe the prospects and possibilities of the Web 3.0 era. Web 1.0 refers to the desktop Internet era, relying on the hardware is the computer; Web 2.0 refers to the mobile Internet period, mainly relying on hardware is the mobile phone; the Web 3.0 era, is what we think of as the metaverse era. Its hardware interaction is done through VR devices and integrates 5G and blockchain technology as the underlying data technologies.
2. Explain the purpose of the metaverse from different dimensions: to satisfy desires, to break through limitations, and to enhance happiness
In the 5th century BC, the ancient Greek sage Protagoras had a famous philosophical proposition that man is the measure of all things.
The original meaning of the proposition is that we should measure the boundaries of our cognition with man, who is the measure of those things that already exist, and the measure of those things that do not exist. Although this proposition caused a lot of controversy at the time, it is very suitable for us to explain the purpose of the metacosm here, which teaches us to face our own needs from a human perspective. The metacosm satisfies man's desires and breaks man's limitations.
From the perspective of human development, competition is the nature on which humans can survive evolution and become higher species, but the development of civilization has suppressed human instincts. Individual human beings are constantly trying to break free from the many shackles of the real world, and we are constrained by various rules and regulations. The emergence of the metacosm offers the possibility for individual humans to break free from bondage. At the same time, life is only once, and it may not be satisfactory. The emergence of the metaverse also provides human beings with the possibility of experiencing more roles and lives in a limited life, thus increasing the richness of human life. People face multiple-choice questions all their lives, but in fact, we sometimes want to do multiple-choice questions? We "want", "want", "all want", we want to do this and that kind of thing, is it possible to do it all? The metacosm gives us this possibility.
We can also use Maslow's hierarchy of needs theory to explain the value of the metacosm. In 1943, Maslow pointed out that many of the needs of man are not reciprocal, but hierarchical. Human needs are often portrayed as pyramidal hierarchies. From the bottom of the pyramid from the bottom up, the needs are: physiology (food and clothing), safety (job security), love and belonging (friendship), respect, self-actualization. The first four levels are often referred to as defective requirements, while the highest levels are called growth requirements. Man's life is almost through the continuous climbing of this pyramid to ultimately achieve the purpose of self-realization.
The happiness gap that people have at different levels of the pyramid in real life is distributed in a spindle-shaped distribution: some people have just eaten enough, some people just have a sense of security, some people have received love, some people have been respected, and only a very small number of people have achieved self-realization. For example, Elon Musk and Zuckerberg, they have a "self-actualization" level of happiness that people in the lower layers cannot understand.
Through the various possibilities offered by metaversmic technology, it is very meaningful to improve the overall happiness of human beings and obtain the experience that Maslow needs three layers on the pyramid of levels.
We can see that the world has now said that there has been a lot of polarization. In Western capitalist countries, the gap between the rich and the poor is severely divided, resulting in the rich getting richer and the poor getting poorer. In the past, when we were studying, the books in the middle school in the 90s said that the contradiction of social development was the contradiction between the people's growing material and cultural needs and the current low level of material culture in the country. So what is the main contradiction now? It is the contradiction between people's pursuit of the meaning of life and the original meaninglessness of life, so what kind of world do we need to construct to solve this problem?
We think the metaverse is the answer. The metaverse will eventually enhance the overall happiness of human beings, so that everyone has more opportunities to rise to the level of self-worth realization and respect, and make their lives more valuable. So we think that the value of the metacosm is to improve the overall state of human existence.
3. Immersion is the cornerstone of the metacosm
What are the cornerstones of the metacosm? What exactly is driving the metacosm concept to heat up? This starts with the eight elements of the metaverse. Roblox once defined the eight elements of the metaverse as: identity, friends, immersion, civilization, economic system, low latency, diversity, and anytime, anywhere. We believe that these eight elements are not parallel, but have a primary and secondary distinction. In the desktop Internet era in which the PC is located and the mobile Internet era in which the mobile phone is located, the elements of identity, friends, civilization, economic system, low latency, diversification, and anytime, anywhere have long existed. We have been able to do identity authentication on WeChat and high-quality interaction in the circle of friends.
However, desktop and mobile Internet have no "immersion" element. We believe that the fundamental reason why PCs and mobile phones do not open the metaverse may be that users "observe" the three-dimensional world through two-dimensional devices, lacking enough "immersion". Without immersion, we cannot substitute ourselves into the virtual world, and we cannot believe that the metaverse is real.
Here you can share a cold knowledge, Virtual Reality is usually literal translation is virtual reality, there may be a translation misunderstanding here. In fact, the English Word Virtual means "in fact close, almost actual", while virtual means "fictitious, false". Virtual Reality should be called "near reality" if direct translation, and the real purpose of the VR industry is to be close to reality, close to reality.
Many philosophical and psychological theories can be used to explain the importance of "immersion." The three greatest Jews in the world were Einstein, the great physicist Albert Einstein, the great philosopher Marx, and the famous psychologist Freud. Freud once proposed in psychodynamicism that "the three major parts of the spirit consist of the id, the ego, and the superego." This concept is used to explain the formation and interaction of the conscious and subconscious minds. The "id" (the complete subconscious) represents desire and is contained by consciousness; the "self" (mostly conscious) is responsible for dealing with things in the real world; and the "superego" (partially conscious) is a conscience or inner moral judgment. Freud believed that in addition to the "self" that can actively carry out cognition, there is also a subconscious existence in our hearts. Among them, the "original self" is the "bottom me", which determines our desires and needs, like icebergs, 90% below sea level, not easy to detect, but the source of all our emotions.
When we encounter the opposite sex we want, we will begin to blush before we have a rational judgment on the appearance of the opposite sex, because the neurons and facial blood circulation system controlled by the self allow us to start accelerating blood circulation. And the "self" is to feel this, we will have the consciousness of "seeing the TA blushing heartbeat", and finally under the action of the self and the self, the idea of the upper "superego" is "I seem to be in love with the ta".
In the immersive personality model, it is shown that both the "self" and the "superego" are produced under the "original self", and these three parts exist independently and influence each other.
VR technology simulates the physical environment - light, sound, smell, taste, movement, and uses the simulated physical environment to affect our eyes, ears, nose, tongue, somatosensory, tactile, resulting in chemical hormones and electrical signals to affect the self, and then affect the self and the superego. This is the psychological basis for VR to make people have a "sense of immersion".
The reason why mobile phones and PCs cannot produce "immersive" is because mobile phones and PCs only simulate light sources and sounds, and cannot simulate the entire physical environment like VR. Of course, the simulation of the physical environment by VR technology is not the same as the brain-computer interface, and the brain-computer interface will eventually directly cross our senses, generating chemical signals and electrical signals to stimulate the brain, so VR is considered to be the last generation of physical hardware devices before the brain-computer interface.
From the above, we conclude that VR simulates the physical environment, makes people feel immersed, existing, and makes the self produce a sense of reality, such simulation is based on the high immersion provided by VR technology. Immersion provides the self with the sensory information it needs, creating a sense of reality for the self and the superego. Therefore, immersion is the cornerstone of the metaverse, and without immersion, there is no metacosm. At this stage, any metaverse concept that does not talk about immersion technology is most likely a pseudo-metacosm.
Take the famous game "Resident Evil 7" as an example: this game has two versions, 3D and VR, but the game experience of the two versions is very different, which is a perfect example of A/B Testing. When I played the 3D version, the experience was very smooth, and like the Resident Evil I played before, it was to pick up a gun and shoot zombies. But in the VR version of Resident Evil, it was difficult for me to smoothly carry out the game.
The reason is that in the VR version, I feel like zombies are right around me, and they feel like they're right behind the next door. Although I know that as long as you push open the door in the game and pick up a gun to shoot the zombies out, you can pass the level. But I can't do it because VR is immersive. Different from ordinary flat games, VR provides enough visual presence and surround sound to create a real feeling, which is the sense of immersion brought about by the advancement of VR technology. This sense of immersion deceives the original self, makes people's adrenaline hormones greatly increase, and makes it difficult to distinguish between true and false. Even if the ego can rely on reason to know that this is just a game, the self is still afraid. From here, we can think of one of Oculus's VR design principles: "Never scare users in VR content, because that will really scare people to death." The concept of the metaverse has recently become popular because the real immersion provided by VR hardware allows people to truly "enter" the virtual world that has existed for a long time. "Man is the measure of all things". With the occupancy of people, the "virtual world" becomes the "metacosm".
The change that VR immersion brings to the experience is not just like Tesla, allowing people to switch from oil cars to new energy vehicles, but more like a subversive revolution - immersion brings us a sense of immigration, like the vision provided by Space X, human beings will eventually go to Mars. Through VR technology, human beings will eventually migrate to the metaverse, which is the mission of the entire VR technology. At the same time, unlike the immigration mars can only buy a one-way ticket, the immigrant meta-universe, you can return to the real universe at any time.
To sum up: the maturity of VR hardware provides enough immersion for the metaverse, so the concept of the metacosm is on fire. We believe that among the eight elements of the metaverse, immersion is the baseplate and cornerstone, and with the real immersion provided by VR hardware, people can enter the metaverse.
Two. With the help of the true immersion of VR hardware, you can enter the metaverse
1. High immersion needs to meet the technical standards of VR 3.0: natural, free, self
Immersion is a core element that underpins the metaverse concept and requires high-performance VR hardware to achieve. This requires the current VR vendors on the market to achieve revolutionary technological upgrades. Xiaopai summarizes the goals of this series of technical improvements into the technical standards of VR 3.0, that is, in the three aspects of visual immersion, physical immersion, and cognitive immersion (corresponding to the nature, freedom, and self in the definition of Xiaopai VR 3.0, respectively).
Visual immersion corresponds to the "natural" dimension in the VR 3.0 definition. UVI (Unit of Visual Immersion), or visual immersion equivalent, is an important measure of visual immersion. We usually think that PPD (Pixels per Degree) is the angular resolution, FOV (Field of View) is the field of view angle, refresh rate these three indicators have an absolute ceiling level: PPD = 60 represents the retina level; FOV = 220 degrees represents the natural field of view of the human eye; the refresh rate is greater than 180Hz The human eye is basically indistinguishable. The three indicators of any VR headset are divided by the ceiling level of the indicator (that is, the 100% level) to get a percentage, which represents the performance of the headset on the indicator. The weighted harmonized average of these three percentages is UVI. We think this model is a good measure of visual immersion levels. 100% UVI means that the human eye is completely unable to distinguish whether to see the image in the headset or the real world. This standard sounds sci-fi, but it's actually no longer far away. Xiaopai has released a UUVI of up to 73% for the Reality series of 12K QLED headsets, which is very close to the "excellent" standard.
Xiao Pi believes that the horizontal field of view of the headset in line with VR 3.0 should reach 200 degrees, the field of view angle of the coincident area is close to 120 degrees, and the upper and lower field of view angles reach 135 degrees. The refresh rate is greater than 180, and the human eye resolution is not perceived. Let's do a DEMO on the picture, 144 Hz compared to 90 Hz, the latency is about 11 milliseconds lower. The experience is much faster than 110 Hz, with a latency of about seven or eight milliseconds, and the gap is noticeable when playing games.
The bottleneck of UVI improvement is mainly the overall limited GPU computing power. GPU hash rate determines the number of frames (i.e., refresh rate) generated per unit of time multiplied by the total number of pixels in a single frame.
In the case of a certain total number of pixels, under the premise that the aspect ratio is fixed, PPD and FOV are inversely proportional, and PPD needs to be reduced to increase FOV, and vice versa. That is to say, there must be direct trade-offs and compromises between PPD and FOV.
For example, the double eyes 4K monocular 2K, basically in 20PPD and 100FOV. Double eyes 8K single eye 4K, you can do 24PPD, 170FOV.
We calibrate and compare the headsets on the market with the angle of view (FOV, Field-of-View) and angular resolution (PPD, Pixel-per-Degree), which play a decisive role in the visual immersion of VR headsets, and use FOV and PPD as the vertical and horizontal axes of the chart, any resolution is equal to FOV multiplied by PPD. When the total number of horizontal pixels is fixed (such as 2K/4K/8K...) ), the angle of view and the angular resolution are inversely proportional.
We can see that the products that reach the resolution of 4K have entered the qualified area, and under the VR 2.0 standard, the resolution should reach more than 8K, and under the VR 3.0 standard, the resolution should reach more than 12K. In order to increase both the field of view angle and angular resolution, the total number of pixels must be increased (for example, from 4K to 8K, or even 12K). Only when the total number of pixels increases, there is more room to trade off between FOV and PPD to define different types of VR headset products that are visually immersive enough to meet the needs of different user groups.
To break through the bottleneck of UVI's increased computing power, it is mainly necessary to optimize algorithms to improve the utilization efficiency of limited computing power, some of which rely on additional hardware support.
For example, the effective clear area of the human eye in the entire FOV is actually very small, and the clear area of text reading that is almost effective is only about 15 degrees. The clear display area of the picture is only about 60 degrees, so it leaves us with a lot of optimization space for the algorithm. Through the fixation rendering technology, we can greatly reduce the computing power burden on the GPU side. There are many similar optimization algorithms developed and adopted by Xiaopai, such as ASW, ATW, DSC. Through these optimization algorithms, we can reduce the hash power burden on the GPU by more than half, and conversely, on the same computing power platform, we can support twice the PERFORMANCE indicators of VR headsets.
Physical immersion corresponds to the dimension of "freedom" in VR3.0. VR hardware needs to achieve 6DOF positioning (six degrees-of-freedom tracking), that is, the positioning tracking of the head and hands, while the device needs to be sufficiently optimized in terms of wireless, spatial influence, somatosensory peripherals, ergonomic design, weight and volume control and distribution. In the VR 2.0 period, the positioning scheme is mainly outside-in, the advantage of its scheme is accurate positioning, the disadvantage is that the price is relatively high, it needs to be equipped with two lighthouse base stations, and the movement is very inflexible; the inside-out positioning scheme is the mainstream development direction, its accuracy is high enough, the price is cheap, the use is convenient, and the accompanying see-through function has formed the prototype of the full-field of view AR function, laying a good foundation for vr's future transition to MR.
Cognitive immersion corresponds to the "self" dimension of VR3.0. To judge cognitive immersion, you can use the Mirror Test. The Mirror Test is often used to test whether an animal is smart enough to know that the image in the mirror is itself and not some other individual. We borrow this concept to refer to whether you can see your own expressions, eyes, and body movements in the VR mirror to form a sufficient sense of self-existence. In addition to the tracking positioning of the head hand 6DOF mentioned above, it also requires the support of many functions such as face tracking (face tracking), eye tracking (eye tracking) and body tracking (full body motion capture) and even emotional tracking. Only with the support of full cognitive immersion can it be possible to form a truly effective social scene in VR.
2. R&D threshold for VR hardware
VR hardware research and development involves many technical fields, and the threshold is extremely high. Recently, many people are asking us: What is the threshold for doing VR hardware? Is the threshold for VR hardware not high? Is it that mobile phone manufacturers can also do it, and can also do VR hardware?
Simple answer: Obviously, this is not the case. The technical threshold of VR is very high, and the relevance with the mobile phone industry is not high.
We can look at the technologies involved in VR, including software, optics, sensing, perceptual science, computer vision, computer graphics, hardware, display, UI, OS, wear, and audio.
First of all, I want to express that VR puts forward higher requirements for technology in all aspects of these aspects, and promotes the improvement of these fields, which are not so high in the field of mobile phones.
Take a simple example, the field of optics. The optical part of VR accounts for a large proportion of the product. At the same time, the distortion and chromatic aberration can be controlled at the large field of view angle, and the volume and weight can be controlled, which puts forward very high requirements for optics. This does not exist in the mobile phone industry. Since the demand does not exist, the upstream manufacturers do not have similar specifications of the product, which requires our whole machine manufacturers to develop their own.
The display is even more so, we have to do a large field of view, high resolution, the road is still very long. In the future, it will take 16K to meet our requirements. But we also know that the existing screen is only about 4K, just getting started. For mobile phones, the 2K screen is basically the ceiling, and it makes no sense to be high.
Looking at positioning, the VR industry generally uses SLAM to do inside-out tracking. Some people think it's not a big deal, but there are a lot of technical improvements behind it. SLAM As we all know, the original use scenario is robot navigation and positioning. Sweeping robots are positioned with SLAM. For it, low latency is not just needed. But for VR, using ordinary SLAM technology directly can cause severe vertigo, because its MTP is simply less than 20 milliseconds. Without in-depth optimization, there is no way to directly apply it to the inside-out tracking of VR devices. SLAM positioning in VR must achieve sub-millimeter-level high precision, millisecond-level low latency, ultra-high robustness, that is, no loss, no drift, no card. This all means a very high technical threshold.
Looking at the operating system, the traditional display scheduling mechanism in the operating system can not meet the requirements of VR's MTP latency.
VR aims to wear for a long time in terms of wear and ergonomics, and female users have also raised the need not to mess with the discovery and makeup. There are also no such problems with mobile phones.
In summary, we believe that the VR industry and the mobile phone industry are two very different industries. If anyone thinks it is, then I think he may have misunderstood his judgment on this matter a bit.
The VR industry, like the mobile phone industry, will be divided into the first half and the second half. Now, we think VR is still in the first half of the process of solving pain points. The deep integration of the industrial chain and the innovation of business models should be left to the second half. It's more about providing some hi-points, generating large-scale economic benefits, and then diluting the costs.
At this stage, the most important thing in the VR industry is that the whole machine manufacturer and the upstream manufacturer work together to improve the ability of the entire industry chain through joint research and development and rapid iteration, and those manufacturers that have grown up in the mature industry chain, such as smart phone manufacturers, may be very uncomfortable at this time.
For example, Xiaomi entered the production of smartphones around 2011. At that time, Qualcomm and Google decided to develop Android together, and there was already a precedent for the success of Apple's business model. But imagine what would happen if Xiaomi came in in 2003? Will the mobile Internet occur at this time? No, because the advent of 4G, new smartphones, multi-core chips, and the smart operating system Linux will be needed to enable the birth of the mobile Internet.
This stage of VR is the same, the first half of the need for hardware companies, if someone thinks that mobile phone manufacturers can easily enter the VR industry, I think it is also very good, at least it proves that everyone can come in and look at the industry. But I think it may be a tortuous road for mobile phone manufacturers to switch to VR hardware.