In the spacious workshop, rows of parts printed with the "mihoyo" logo are carefully transported into the assembly line and ready for use.
Soon, they will go through a series of delicate processes: material separation, wiring and assembly, be loaded into the game, and be lit up in a certain computer room, becoming part of the MiHoYo Cloud game world.
For many players, cloud gaming has an inexplicable attraction. This attraction does not come from the game, but more from the technology itself, which has a sense of the future that "although I cannot really aspire to it."
After all, streaming rendered games to local devices through high-speed networks in real time means that we no longer need to download and install, and we no longer need to consider the computing power configuration of local hardware. As long as we have a network, we can play freely. everything.
The distribution attributes of cloud games are almost perfectly in line with the expectations of all game manufacturers – game works can get rid of the threshold of equipment and channels to the greatest extent, and directly reach the maximum number of players.
But technology never arises out of thin air. On the occasion of the official launch of the cloud game version of "Honkai: Star Rail" – "Cloud Star Rail", miHoYo also released a documentary about the company's long journey in the field of cloud games. Explore and practice.
This is also a story about technology and dreams.
Cloud gaming that “carries forward with a heavy load”
I still remember that my first experience with cloud gaming was in 2017, when the "Final Fantasy 13" cloud gaming version was provided by the Japanese cloud gaming service provider "G-cluster".
However, the mosaic images and lags caused by severe network delays at that time really did not leave a good impression on me. Cloud gaming is more like a beautiful-looking concept that should be viewed from a distance but not played with.
Today, seven years later, you can find a thousand-yuan Android phone on the market, open the built-in browser, log in to "Cloud Star Railway", and you can enjoy the high image quality that is no less than that of the PC. With the modeling and light and shadow special effects, the characters’ every frown and smile are perfectly displayed, and the fight scenes with grand backgrounds are smooth and silky.
This directly breaks the shackles of the concept that "the quality of the gaming experience depends entirely on the quality of the hardware configuration" and opens the door to a new world.
Ideally, cloud gaming means that anyone can play games on all devices faster and more conveniently, which is completely different from our usual game playing process.
No downloads are required, no storage space is occupied, and hardware computing power is liberated, allowing players to experience high-quality games with low burden. This is the most direct attraction of cloud gaming.
For miHoYo, helping players "break through the computing power limitations of local hardware" is one of the important values of cloud gaming.
Behind every player who is playing, there is a server computer carrying a heavy load, and we are like opening a 24-hour Internet cafe.
MiHoYo's initial plan is to directly use public cloud servers on the market that are specifically provided to third-party institutions and enterprises to solve the huge computing power demand. This is also a solution that is generally favored in the industry.
However, as the project progressed, some pain points also emerged.
First of all, there were not many servers on the market for MiHoYo to choose from at that time, and high-performance public cloud servers that could meet the standards were more suitable for AI needs.
This also forces MiHoYo's cloud gaming team to optimize the underlying code of the game so that it can run on most public cloud servers.
But the second problem also arises – the high cost of server rental.
In MiHoYo's view, public cloud servers pursue large and comprehensive functions. Not only is it difficult to fully adapt to the game, but it will also cause a waste of computing power.
These redundant things make server costs very high and will also cause players to face high expenses. This is not what we want to see.
Basically, the operating and leasing costs of public cloud are also problems that all companies that purchase this type of business will inevitably face. If computing power cannot be reasonably scheduled and allocated, the costs originally borne for users will eventually become the source of their own downfall.
To sum up, for MiHoYo who wants to take their cloud gaming experience to the next level, the public cloud is only available, but not easy to use.
This is like buying a suit. Ready-made clothing is a "universal solution". As long as you don't have many requirements for size details, you can indeed buy a ready-made suit to take home.
But in comparison, tailor-made models are designed to serve buyers in terms of style, fabric, etc., and can better meet individual needs.
After careful consideration, it became MiHoYo's only choice to end up on his own and move towards self-research.
From zero to one, bet on self-research
The goal of the MiHoYo cloud gaming team is very clear, which is to create a cloud server hardware solution that meets computing power needs, is stable and reliable, can be deployed on a large scale, and is profitable.
Although we may not necessarily know the server best, there is no doubt that we know our games best.
For miHoYo, this is a big adventure out of the comfort zone.
In the past, Amazon, Facebook, Microsoft and other technology giants have tried to take advantage of their vast data processing centers to create cloud gaming systems.
However, many failure cases, including Google Stadia, reflect that cloud gaming platforms are by no means as simple as they seem. Some obstacles come from technology, while others are attributed to "lack of enough users." In the end, the business model collapses and becomes unsustainable.
There was a metaphor in the industry that cloud gaming is like a "meat grinder of resources and money." In addition to technical issues, the greater difficulty is how to find a balance between "high-quality gaming experience" and "hardware operating costs". This is the key to sustainable development.
MiHoYo is no exception. On the one hand, building your own server means you have more choices in hardware, but on the other hand, due to the lack of a universal solution, you are bound to face "manufacturer manufacturing delays, insufficient supply of unpopular parts," etc. Supply chain problems are common in real industries.
After studying the performance consumption of its own games and clarifying the product form and functions, the MiHoYo team found its own breakthrough point.
Similar to mobile phones and PCs, servers are also composed of multiple key components. Basically, except for the CPU, GPU, hard drive, and memory, any part that is "uneasy to use" will be included in their self-developed components one by one. Checklist.
For example, in order to solve the remote management function, the MiHoYo team developed a universal BMC board called "MiLO". It is like the "big housekeeper" of the server. It not only allows operation and maintenance personnel to achieve a higher degree of freedom in operation and maintenance, but also expands MiHoYo's selection range of motherboards, eliminating the need to rely on expensive components from a single supplier. cut costs.
In terms of power management, the power supply used by MiHoYo not only had poor delivery time and was expensive, but also had a very single solution. For this reason, they also redesigned the power supply solution, developed a power conversion board, and modified the chassis power supply. The power supply of the entire machine has been adjusted, which has improved both cost and reliability.
After the hardware selection was determined, miHoYo's technical geeks even customized a chassis so that self-developed components could be put into standard server cabinets, effectively increasing the density of the server.
But designing a complete set of solutions is not enough. If cloud servers want to mature, they must also achieve production process verification, and eventually move toward mass production and scale.
Unlike assembling a server in a laboratory, mass production of thousands of machines in a factory is another level of difficulty.
In order to solve mass production, MiHoYo has specially built a cloud server production line. All parts will be shipped here and undergo several processes such as material distribution, wiring, assembly, and lighting testing.
After the games are installed, these machines will be put into a high-temperature aging laboratory for a 48-hour pressure test, and finally put into packaging boxes with the MiHoYo logo printed on them. Only then will the entire production process be completed and await shipment. .
At this point, the problem on the hardware side has been solved, but what about the experience side?
For players, whether cloud games can provide an experience close to that of local games is an important reason why they are willing to use cloud games – especially the lagging and screen tearing caused by network delays, which are also the objective "uncertainties" of the cloud gaming experience. sex".
Judging from today's technology, it is almost impossible to completely eliminate delays. In order to reduce this "uncertainty" to the extent acceptable to users, MiHoYo can only choose to increase the number of servers to improve players' connection experience.
This involves the large-scale laying of computer rooms, which naturally requires huge investments.
In China, miHoYo has built dedicated cloud server rooms in many cities in the north and south, and its service scope is sufficient to cover the entire country.
The servers follow the truck here and are split and put into cabinets one after another. After passing the software run test and verification, they will eventually be put to work and continue to provide stable cloud gaming services to surrounding areas in the next few years.
Currently, the MiHoYo team is still promoting next-generation hardware solutions, such as higher-density server solutions, multi-node redundant power supply solutions, and BMC boards with switch and network card functions…
They will continue to explore unknown areas, and a huge cloud gaming landscape is slowly unfolding.
Let one billion people become the "number one player"
This is not the first time miHoYo has made predictions about cutting-edge technology.
In the documentary "Chinese Game Chronicles", MiHoYo President Liu Wei, also known as "Big Viagra", once revealed that in January 2017 when the "Genshin Impact" project was established, even the iPhone 7, which had the best performance at the time, It’s hard to meet the standards of running an open-world mobile game.
To put it simply, MiHoYo at that time was already designing games for mobile terminals 2-3 years from now, and its confidence came from the deduction of the development of computing power:
I think it's a really big bet for us, but it's an option for our company.
Opportunities always come to those who are prepared, and MiHoYo did win the bet. "Genshin Impact" is regarded as a key node in miHoYo's technological progress, and the subsequent "Honkai Impact: Star Rail" also achieved remarkable results after its launch, once again becoming a concentrated expression of miHoYo's industrialization capabilities. .
Now deploying cloud games, miHoYo has once again stood at the forefront of technology prediction. Perhaps in 3-5 years, the goal of "cloud gaming becoming the main way to play games" will be difficult to achieve, but regardless of objective trends and subjective perceptions, we have to admit that cloud technology will have a place in the future.
In fact, among the world’s largest cloud computing service providers—Amazon, Microsoft, and Google—none of them started out as cloud services, but these three companies have captured 65% of this $500 billion market. . This also means that cloud gaming is a huge market that all giants are eyeing, and there are huge platform opportunities hidden there.
At a time when games as a service (GaaS) are becoming popular, cloud games are also the form that best fits its service spirit – using the most advanced technology to serve the broadest audience.
But is cloud gaming the ultimate goal of MiHoYo?
Looking back on the past two or three years, you will find that miHoYo's focus on "hard technology" has already gone beyond the scope of ordinary game companies – whether it is investing in research on brain-computer interfaces, fusion energy, or co-developing rocket companies , all show their pursuit of new technologies and cutting-edge science.
There is also Apple’s latest Vision Pro head display. While most companies are still waiting and watching in the space computing era, miHoYo has chosen to embrace the opportunities brought by new technologies – "Honkai Impact: Star Rail" has become the first The game titles that have been officially announced to be launched on visionOS have the implication of laying out the virtual world in advance.
Based on this consideration, I suddenly thought of a grand and charming vision that MiHoYo mentioned two years ago:
By 2030, create a virtual world that one billion people around the world are willing to live in.
This may be a gorgeous performance by a lone hero. In the eyes of outsiders, MiHoYo is more of a game company, but for MiHoYo itself, creating a true virtual digital world may be the only way to achieve this goal. The original intention of this company was to "save the world with technology nerds".
This may also be a spark that starts a prairie fire. A few years later, when we take out our portable devices, open the browser, and play a hearty game, we can’t help but look at the sky and remember this flower. The cloud called MiHoYo.
We still have to have dreams, what if they come true?
# Welcome to follow the official WeChat public account of aifaner: aifaner (WeChat ID: ifanr). More exciting content will be provided to you as soon as possible.