FSR 2.0: How AMD Increases Game Resolution by 50%

Contents:

• Upscaling and Anti-Aliasing: Relationship and Importance
• FSR 1.0: Fundamentals of Upscaling Technology
• Conclusion: The Future of FSR and Its Evolution
• FSR 2.0: The Next Level of Upscaling

Upscaling and Anti-Aliasing: The Relationship and

Modern graphics technologies in the gaming industry are impossible to imagine without analyzing two fundamental aspects: upscaling and anti-aliasing. These methods are crucial for producing high-quality images in games. Upscaling improves image resolution, making it sharper and more detailed, while anti-aliasing helps eliminate visual artifacts such as jagged edges. For a deeper understanding of their interaction and impact on graphics quality, we recommend reading our previous review of DLSS technology.

Anti-aliasing is an important process designed to eliminate aliasing, also known as "jaggies." This effect appears at the boundaries of objects and makes the image less appealing. To eliminate it, pixels of intermediate colors are used, which help soften sharp contrast transitions. Anti-aliasing improves the visual perception of images by creating smoother lines and edges. This is especially relevant in graphic design and computer animation, where image quality plays a key role. Proper use of anti-aliasing allows you to achieve a more professional and aesthetically pleasing result.

Anti-aliasing, despite its simplicity, is challenging to achieve high-quality results. Directly blending contrasting areas can result in a loss of image sharpness. However, this method does not require significant computational resources, making it a popular choice in a number of video games. To improve graphics quality and enhance visual perception in games, it is important to consider various anti-aliasing techniques that minimize artifacts and preserve image details.

To achieve higher-quality antialiasing of visual effects, the engine requires more information about the frame. This can be achieved by rendering the image at a high resolution and then downscaling it to the required size. This process allows the system to obtain more data, which helps it more accurately identify areas that require antialiasing. Using this approach significantly improves graphics quality and makes visual elements smoother and more realistic.

The anti-aliasing landscape changed significantly with the introduction of temporal anti-aliasing (TAA) in the early 2010s. This method uses information from previous frames to improve image quality. TAA applies subtle shifts that are imperceptible to the eye but effectively address rendering issues. This approach results in smoother and more detailed images, which is especially important in modern video games and graphics-intensive applications. As a result, TAA has become a popular tool for developers seeking to improve the visual quality of their projects.

Despite its advantages, TAA is not without its drawbacks. Many gamers note that this method can cause image blur and artifacts, especially in fast-moving scenes. Fast-moving objects can leave trails, which negatively affects the overall quality of the graphics. Such problems can reduce the immersion in the game and make the gameplay less enjoyable. Discussing these drawbacks is important for gamers looking to optimize graphics and improve the visual experience in games.

Developers use heuristic algorithms to minimize image quality issues. These algorithms can improve visual perception even when the image has imperfections. The use of such techniques helps to increase clarity and detail, which makes images more appealing to users.

For a more complete understanding of anti-aliasing and techniques like TAA, it is recommended to study additional resources that explain these technologies in detail. These materials will help you better understand how anti-aliasing works and its impact on visual quality, and introduce you to the latest advancements in computer graphics.

Learn about anti-aliasing, TAA, and other technologies that are shaping the future of video game graphics. These techniques improve image quality by reducing noise and increasing clarity. Anti-aliasing helps make the edges of objects smoother, while TAA (Temporal Anti-Aliasing) uses information from previous frames to achieve a more stable and high-quality image. Understanding these algorithms will help you better appreciate modern games and their visual achievements.

FSR 1.0: Upscaling Basics

FSR, or FidelityFX Super Resolution, is an upscaling technology developed by AMD. Unlike modern solutions such as NVIDIA's DLSS, FSR 1.0 uses traditional interpolation methods. This results in less efficient production of high-quality images. However, FSR offers gamers the opportunity to improve performance across multiple platforms. This technology helps improve the visual experience in games by increasing frame rates and delivering better image quality without the need for expensive hardware.

From a technological perspective, upscaling and anti-aliasing share similar goals: both procedures aim to supplement scene information to maintain high quality while minimizing resource expenditure. However, FSR 1.0 uses a method that is more in line with post-process anti-aliasing than modern algorithms that use temporal data. This difference highlights the importance of choosing the right approach depending on performance and image quality requirements.

FSR 1.0 initially used interpolation to fill the gaps between pixels, which limited its ability to create detailed images. Unlike more advanced algorithms such as DLSS, which relies on data from previous frames, FSR is unable to provide similar quality in displaying fine details and text. This limitation makes FSR less effective for users who prefer high definition and detail in graphics.

Bilinear interpolation is one of the most popular interpolation methods, also used in FSR 1.0 technology. When scaling a 2x2 pixel image to 3x3 pixels, the original pixels are placed at the corners, and the intermediate values ​​are calculated by averaging. This approach is often used in photo editors because it allows for fast image processing. However, it is worth noting that bilinear interpolation does not always ensure high image quality, especially when significantly increasing the size.

There are many interpolation algorithms, each differing in the mathematical methods used to calculate unknown values ​​but providing similar results. For example, the nearest-neighbor method copies data from the nearest pixel. This approach can result in angular and sharp images, which significantly impacts visualization quality. These characteristics make the method less suitable for tasks requiring high image clarity and smoothness.

FSR 1.0, despite its simplicity, uses more complex filters to improve sharpness and eliminate artifacts that arise when using the Lanczos method, which is designed to smooth rounded shapes. This results in higher-quality images, but the end result is often noticeably inferior to native high-resolution images. FSR 1.0 provides users with the ability to improve graphics performance, but to achieve maximum image quality, it is worth considering the benefits of native resolution.

The first version of FSR significantly lagged behind its competitors, including DLSS 2.0, which already delivered high-quality images as close to native as possible. Users noted that FSR 1.0 resembled outdated interpolation technologies, such as bilinear, which negatively impacted the overall visual experience. Improving image quality and increasing performance became necessary steps to make FSR technology more competitive in the market.

FSR had certain limitations in use, as it required a pre-antialiased image to operate effectively. In contrast, DLSS could simultaneously improve image quality and antialias frames, providing a higher-quality visual experience. This created additional difficulties, especially for players using TAA, as artifacts on the screen became more noticeable. Thus, the choice between FSR and DLSS depends on user preferences and image quality requirements in games.

Conclusion: The Future of FSR and Its Evolution

Despite some limitations of FSR 1.0, this technology became the basis for future improvements. Subsequent versions, such as FSR 2.0 and 3.0, significantly increased functionality and quality, allowing them to meet modern upscaling standards. AMD is expected to continue to develop its technologies to effectively compete with solutions from NVIDIA and other manufacturers. Improvements to FSR can also contribute to wider adoption of the technology in the gaming industry, which in turn will positively impact the user experience.

FSR 2.0: The Next Level of Upscaling

AMD's FSR 2.0 technology is a significant advancement in image upscaling. Unlike its predecessor, the first version of which failed to fully meet user expectations, FSR 2.0 uses modern temporal techniques, which deliver higher image quality with minimal performance loss. This allows gamers to enjoy sharper and more detailed graphical effects in games without the need for powerful hardware. The implementation of FSR 2.0 opens up new possibilities for developers and improves the overall visual experience in computer games.

Temporal upscaling is a method that combines multiple low-resolution images into a single high-quality image. This process involves shifting each frame relative to the previous one by a subpixel, allowing the technology to more accurately capture scene details. The resulting image is comparable in quality to real high-resolution frames. Temporal upscaling is widely used in various fields, including video games and the film industry, to improve visual effects and provide a smoother viewing experience.

When overlaying images without pre-processing, as the TAA algorithm does, the quality of the resulting image can significantly decrease. The TAAU algorithm has been in use in the industry for several years, but despite being available for Unreal Engine, it has not seen widespread use due to its insufficient output image quality. Developing more efficient image processing algorithms is becoming a pressing issue, as they can improve visual perception and enhance the overall performance of graphics applications.

NVIDIA efficiently handles graphics processing by using neural networks that analyze each frame individually. In contrast, FSR 2.0 technology uses motion vectors and depth data provided by the game engine, allowing for more accurate assessment of changes between frames. This approach simplifies the process of image compositing and improves the quality of visualization in games.

DLSS uses motion data and has a neural network architecture, which allows for the automation of many processes. In contrast, FSR 2.0 requires manual calculations. The FSR 2.0 algorithm must evaluate the significance of new information and carefully clean the image of artifacts using tone and depth maps. This makes DLSS a more effective tool for enhancing graphics quality in games, providing users with a smoother and more detailed experience.

FSR 2.0 actively addresses the shortcomings of TAAU, using new algorithms and data from the game engine. Image quality improves with each new version, but the manual approach doesn't always produce perfect results. Even modern versions of FSR 2 can encounter issues such as artifacts and flickering objects, especially in scenes with complex geometry. To achieve the best results, it is important to continue optimizing and adapting technologies, which will minimize visual flaws and improve overall graphics stability.

FSR continues to be one of the most accessible and popular upscaling tools. While it remains inferior to DLSS, the differences between the two technologies are narrowing every year. Currently, FSR significantly improves image quality, and its availability on modern consoles allows developers to implement new features without fear of losing their audience. This makes FSR an essential tool for improving the performance and visual experience of games, which contributes to wider adoption of the technology among developers and players.