In December last year, UWA DAY 2021 just ended. Only after 5 months has UWA DAY 2022 entered the preparatory stage again. After “Driving Game Development into the Era of Industrialization”, this year’s UWA DAY will be “Practicing the Industrialization of game R&D”. With the theme of “industrialization”, it seems that UWA is determined to carry out the industrialization of game R&D to the end!
However, GameLook also noticed that there are still many misunderstandings in the industry that industrialization is equated with technological empowerment. In fact, the most mature optimization tools and production processes still have to serve developers and specific project development. In other words, the industrialization of game research and development requires the joint participation and practice of the entire industry in order to ultimately achieve a comprehensive improvement in the level of research and development.
For this reason, UWA Technology, which is rooted in game research and development, has adopted “Practicing” as the keyword for 2022. This company, which has provided game performance solutions and technical support for nearly 9,000 games in the past six years, has become more and more aware of the significance and value of practicing game R&D industrialization with a large number of specific practices.
In order to understand the pace of UWA’s implementation of industrialization, we interviewed Zhang Xin, founder, and CEO of UWA Technology again, to talk about the standardization, normalization, specialization, automation, and scale of game development.
Q: The theme of this year’s UWA DAY is “Practicing the Industrialization of Game R&D”. What topics will be presented this year to highlight the theme of the conference?
A: UWA DAY is already in its sixth year this year. Every year, we will focus on a theme and invite industry experts to share new technologies, new experiences, and new trends in the industry over the past year with practitioners in the game industry. This year, many experts from companies including ByteDance, NetEase, Tencent, Lilith, etc. have been confirmed as lecturers. They will share their knowledge and experience in how to implement industrialized deployment in their respective teams, including central control, performance optimization, TA, and test development.
UWA lecturers will also focus on topics including how game development standardization, normalization, specialization, and automation are implemented in the process of project development and management; benchmarks based on the establishment of industry standards. They will also introduce our experience in project optimization by illustrating the classic cases of UWA deep optimization adopted in Unity and Unreal projects. The lecturers will demonstrate the importance of the industrialization of game R&D through multi-dimensional speeches, which will be a very exciting experience for everyone.
Q: What kind of deployment did UWA make to drive industrialization within the past year?
A: We launched the UWA Pipeline product in the second half of last year, aiming to improve the R&D process of a large number of small and medium-sized game teams. Because we found that many performance or technical problems are actually caused by the imperfection of the R&D process. Once the process is reasonably controlled and scientifically developed, many subsequent problems can be avoided. When Pipeline was 1.0, many teams still used it as a tool for daily packaging of the design and test team. This year, UWA Pipeline has reached the 2.0 stage, which is not only a Game CI that meets the needs of the project itself but also adds a new Private cloud real machine remote debugging function. In the company’s intranet environment, with Pipeline as the platform, a set of “test equipment management” systems is built, which can access mobile devices that may be scattered throughout the company, display the use status and parameter information of each model in real-time, and can call the required test device at any time.
For example, in our cooperation with “Figure Story”, we can restore the operations of a QA member on the mobile phone side on the PC side, and use automated test scripts to test the function and performance of batch devices. A test case can obtain reports of functional testing and UWA performance evaluation data. If you think that the writing of automated scripts is difficult to master, UWA also has training and scripting services for you.
At the same time, another major function of the UWA Pipeline 2.0 stage is that it can be seamlessly linked with various SaaS services of UWA, and completed automatically through tasks. For example, GOT Online and real device testing services for static resources, code detection, performance evaluation, etc can be automatically run, and problems can be found and solved in a timely manner. All of these UWA SaaS services can be tried on our website, and interested teams are welcome to visit our website and use a trial.
These products and services are the first steps towards standardization and automation in our efforts to industrialize game development. UWA Pipeline can connect a lot of work, automate the fixed work as much as possible, and release more manpower.
In terms of standardization, we launched the “macro” standards for project quality and performance last year and gave professional ratings through the evaluation of the project’s recent performance and the use of package resources. For projects with lower ratings, it is not recommended to largely invest in user acquisition, because the worse the performance and quality ratings are, the higher the cost of user acquisition and the lower the price/performance ratio will be. Through in-depth cooperation over the past year, many projects have reached the “Double-A Standard” (Quality Rating A and Performance Rating A, as shown in the figure below) on our UWA platform.
Project Quality Rating Chart
Report Performance Rating Figure
Additionally, in order to further help the R&D team improve optimization efficiency, we have continuously provided users with more detailed “micro” standards through massive performance data analysis over the past year. These standards can be refined not only to each game module such as rendering, physics, UI, etc. but also to important performance variables in each module (such as Draw Call, Triangle, number of particle systems, loading initialization, texture memory usage, etc.). These standards can be divided not only according to the computing power of different models but also according to the different goals set by users. For example, if a user tests the game on Xiaomi Mi 10 and wants the frame rate to reach 60 frames, then we can tell the user within 1 second which module’s performance index exceeds the standard, how much exceeds the standard, and how should it be optimized. All of these are based on our massive performance data and hardware data in the background as well as our understanding of performance after optimizing thousands of games, and our own data analysis capabilities. We already released this feature to all clients last month, and we believe that this will greatly facilitate the promotion of R&D standardization in the entire game industry, especially benefiting many small and medium-sized teams that have little experience in this area.
In terms of specialization, we have developed a more in-depth GPU performance analysis tool, which not only supports in the Unity engine, but also in an active expansion to include game projects of the Unreal engine. At present, we have cooperated with many game companies such as Hero Games, Zulong Games, and Lilith Games to carry out in-depth cooperation for their Unreal projects.
Due to the content limitation, please feel free to read the full content with graphics and datas at UWA official webpage at
https://blog.en.uwa4d.com/2022/05/1...does-uwa-practice-the-game-industrialization/
However, GameLook also noticed that there are still many misunderstandings in the industry that industrialization is equated with technological empowerment. In fact, the most mature optimization tools and production processes still have to serve developers and specific project development. In other words, the industrialization of game research and development requires the joint participation and practice of the entire industry in order to ultimately achieve a comprehensive improvement in the level of research and development.
For this reason, UWA Technology, which is rooted in game research and development, has adopted “Practicing” as the keyword for 2022. This company, which has provided game performance solutions and technical support for nearly 9,000 games in the past six years, has become more and more aware of the significance and value of practicing game R&D industrialization with a large number of specific practices.
In order to understand the pace of UWA’s implementation of industrialization, we interviewed Zhang Xin, founder, and CEO of UWA Technology again, to talk about the standardization, normalization, specialization, automation, and scale of game development.
Q: The theme of this year’s UWA DAY is “Practicing the Industrialization of Game R&D”. What topics will be presented this year to highlight the theme of the conference?
A: UWA DAY is already in its sixth year this year. Every year, we will focus on a theme and invite industry experts to share new technologies, new experiences, and new trends in the industry over the past year with practitioners in the game industry. This year, many experts from companies including ByteDance, NetEase, Tencent, Lilith, etc. have been confirmed as lecturers. They will share their knowledge and experience in how to implement industrialized deployment in their respective teams, including central control, performance optimization, TA, and test development.
UWA lecturers will also focus on topics including how game development standardization, normalization, specialization, and automation are implemented in the process of project development and management; benchmarks based on the establishment of industry standards. They will also introduce our experience in project optimization by illustrating the classic cases of UWA deep optimization adopted in Unity and Unreal projects. The lecturers will demonstrate the importance of the industrialization of game R&D through multi-dimensional speeches, which will be a very exciting experience for everyone.
Q: What kind of deployment did UWA make to drive industrialization within the past year?
A: We launched the UWA Pipeline product in the second half of last year, aiming to improve the R&D process of a large number of small and medium-sized game teams. Because we found that many performance or technical problems are actually caused by the imperfection of the R&D process. Once the process is reasonably controlled and scientifically developed, many subsequent problems can be avoided. When Pipeline was 1.0, many teams still used it as a tool for daily packaging of the design and test team. This year, UWA Pipeline has reached the 2.0 stage, which is not only a Game CI that meets the needs of the project itself but also adds a new Private cloud real machine remote debugging function. In the company’s intranet environment, with Pipeline as the platform, a set of “test equipment management” systems is built, which can access mobile devices that may be scattered throughout the company, display the use status and parameter information of each model in real-time, and can call the required test device at any time.
For example, in our cooperation with “Figure Story”, we can restore the operations of a QA member on the mobile phone side on the PC side, and use automated test scripts to test the function and performance of batch devices. A test case can obtain reports of functional testing and UWA performance evaluation data. If you think that the writing of automated scripts is difficult to master, UWA also has training and scripting services for you.
At the same time, another major function of the UWA Pipeline 2.0 stage is that it can be seamlessly linked with various SaaS services of UWA, and completed automatically through tasks. For example, GOT Online and real device testing services for static resources, code detection, performance evaluation, etc can be automatically run, and problems can be found and solved in a timely manner. All of these UWA SaaS services can be tried on our website, and interested teams are welcome to visit our website and use a trial.
These products and services are the first steps towards standardization and automation in our efforts to industrialize game development. UWA Pipeline can connect a lot of work, automate the fixed work as much as possible, and release more manpower.
In terms of standardization, we launched the “macro” standards for project quality and performance last year and gave professional ratings through the evaluation of the project’s recent performance and the use of package resources. For projects with lower ratings, it is not recommended to largely invest in user acquisition, because the worse the performance and quality ratings are, the higher the cost of user acquisition and the lower the price/performance ratio will be. Through in-depth cooperation over the past year, many projects have reached the “Double-A Standard” (Quality Rating A and Performance Rating A, as shown in the figure below) on our UWA platform.
Project Quality Rating Chart
Report Performance Rating Figure
Additionally, in order to further help the R&D team improve optimization efficiency, we have continuously provided users with more detailed “micro” standards through massive performance data analysis over the past year. These standards can be refined not only to each game module such as rendering, physics, UI, etc. but also to important performance variables in each module (such as Draw Call, Triangle, number of particle systems, loading initialization, texture memory usage, etc.). These standards can be divided not only according to the computing power of different models but also according to the different goals set by users. For example, if a user tests the game on Xiaomi Mi 10 and wants the frame rate to reach 60 frames, then we can tell the user within 1 second which module’s performance index exceeds the standard, how much exceeds the standard, and how should it be optimized. All of these are based on our massive performance data and hardware data in the background as well as our understanding of performance after optimizing thousands of games, and our own data analysis capabilities. We already released this feature to all clients last month, and we believe that this will greatly facilitate the promotion of R&D standardization in the entire game industry, especially benefiting many small and medium-sized teams that have little experience in this area.
In terms of specialization, we have developed a more in-depth GPU performance analysis tool, which not only supports in the Unity engine, but also in an active expansion to include game projects of the Unreal engine. At present, we have cooperated with many game companies such as Hero Games, Zulong Games, and Lilith Games to carry out in-depth cooperation for their Unreal projects.
Due to the content limitation, please feel free to read the full content with graphics and datas at UWA official webpage at
https://blog.en.uwa4d.com/2022/05/1...does-uwa-practice-the-game-industrialization/
