By: Christoffer Holmgård, co-founder and CEO of modl.ai
Game development is a risky business. If the production goes over budget, ships late, ships with bugs, or is poorly received, it can cause huge financial losses to the developers and publishers. A studio is often only as successful as their last title, so it’s important that studios plan ahead and mitigate risk where possible.
Every decision during development, be it within game design, programming, graphics, sound design, marketing and more introduce risk. Testing is one of the constant elements woven throughout the entire process to ensure that the functionality is sound and that the quality of the game meets earlier aspirations. Sadly, testing can be extremely time-consuming and costly--plus it can shackle the people assigned to test with repetitive, monotonous tasks inherently prone to human error. As such, many game developers today are turning to automated testing tools that don't rely on manual efforts to improve accuracy and efficiency, and instead allow testers to use their intuition elsewhere.
Automated testing tools date back as far as the early 1990s, but as the field has matured, it has expanded in a variety of categories, each serving a specific purpose in the software development life cycle.
Unit Testing Tools: Are used to test the smallest parts of an application independently to make sure they operate as intended. As the name might imply, JUnit and NUnit are available for different programming languages.
Functional Testing Tools: These are used when we wish to check that the overall feature or component of a project operates as it was originally envisaged. Tools such as Selenium and TestComplete are very popular for this approach.
Performance Testing Tools: Evaluate how well a piece of software performs under different conditions, such as increased throughput, intermittent connectivity and busy servers. JMeter and LoadRunner are widely used for this purpose.
Integration Testing Tools: Are used when we have multiple components that arguably pass unit tests, but now we need to make sure they interact well. SoapUI and Postman are examples of such tools.
Security Testing Tools: As the name implies, these check for any security vulnerabilities and threats in the software that bad actors could exploit. OWASP ZAP and Nessus are key players in this space.
Code Review Tools: Code reviews are a critical part of agile development processes, whereby teams reinforce code quality by reviewing each other’s work. While this is available in many source control systems, external tools such as Crucible and CodeClimate can also be used.
Smoke Testing Tools: Perform a basic sanity check on the application to ensure that it is stable and meets the initial requirements. SeleniumIDE and Cucumber are commonly used for this purpose.
Compatibility Testing Tools: Validate if an application works across different devices, operating systems, browsers or network environments. BrowserStack and TestComplete Mobile are popular options in this category.
Regression Testing Tools: Are used to ensure that new changes or updates in an application do not affect the existing functionality. Ranorex and TestComplete are widely used for regression testing.
User Acceptance Testing Tools: These tools help in simulating real-world usage scenarios to check if the application meets user expectations. Applause and UserTesting are two well-known names.
Each of these tools performs a series of predefined tasks and checks the results without human intervention, making them valuable assets in the software development process. One the process is complete, then engineers can return to the code and examine what changes, if any, need to be made. Meaning their time was spent more effectively while the process was conducted. Automating testing not only saves time but it also improves accuracy and efficiency in finding and potentially fixing bugs that a manual review by a human might miss.
One of the main benefits of automated testing tools is their ability to perform repetitive tests quickly and accurately. This is a common requirement of testing both in video game development, and the wider software industry. Pressing the same button numerous times, or trying to collect an item in a game world to ensure the logic executes as desired. By moving towards automation, this allows developers and testers to focus on more important tasks, such as creating new features or addressing critical issues – both of which utilize the creativity and ingenuity of the developer more effectively. These tools also provide detailed reports and logs, making it easier to report back to the development team on where to identify and fix bugs.
Automated testing helps streamline the game development process and improve the final product. Automating repetitive tests not only accelerates the process but also ensures the consistency of testing conditions and removes human error (there’s only so many times a human tester will play the same part of a game before their focus is strained).
Meanwhile, larger-scale code analysis, such as regression tests, can be automated on build servers to save time and resources. This speaks to the second major benefit: automated testing tools can run on different devices and browsers simultaneously, be it a developer's local machine, or the company's server farms. This increases testing coverage and effectively reduces the testing time. Moreover, these tools can operate 24/7, performing tests outside of regular working hours and thus aiding in continuous development and integration processes.
The way they overlap with AI tools in gaming is also evolving. Automated testing tools are now incorporating machine learning and natural language processing capabilities to better understand and adapt to complex code. This allows for more comprehensive test coverage and improved accuracy in finding bugs. As the gaming industry continues to grow and evolve, the use of automated testing tools will only become more vital in ensuring high-quality game releases. Ultimately, these tools not only benefit developers but also enhance the overall gaming experience for players.
About the Author
Christoffer Holmgård is a co-founder and CEO of modl.ai, which empowers game developers and publishers with its state-of-the-art AI engine and machine learning models. In modl.ai, Christoffer bridges a deep understanding of game development processes and technology with the possibilities and limitations of modern artificial intelligence. In leading and building modl.ai, he leveraged his experience from Die Gute Fabrik and his background in organizational psychology. Christoffer holds a BA degree in Psychology from the University of Copenhagen and an MSc in Media Technology from the IT University of Copenhagen. In 2015, Christoffer earned a Ph.D. in Artificial Intelligence and Procedural Content Generation from the IT University of Copenhagen and a post-doctorate in Game Engineering from New York University.
No comments:
Post a Comment