As you may remember, AMD introduced its Ryzen AI 300 integrated processors in early Junejust weeks before the arrival of summer. And just weeks before the end of said season, it confirmed its arrival on the market in multiple portable devices, which are added to the Copilot+ PC platform and which we could see, for the most part, at IFA 2024, the Berlin technology event that marks the return to activity after the inevitable summer break which, however, is a little less so each year than the previous one.
As is inevitable in these times, the focus of this new generation of AMD APUs has put itself in its capabilities for AIwhich in this case are very directly related to the platform defined by Microsoft and which emphasizes the need for the computing capacity it considers necessary to use the Windows 11 functions with ease to depend exclusively on the NPU of said integrated circuits. However, not everything is Copilot nor should AI-based functions depend exclusively on the NPU. And, even more, artificial intelligence is not everything, although it is beginning to seem otherwise.
The most recent example can be found in a recent announcement by AMD, which has just confirmed the arrival of Fluid Motion Frames 2 (AFMF 2) and Variable Graphics Memory (VGM) to the new laptops equipped with Ryzen AI 300 APUs, two technologies that promise to substantially improve the user experience of these devices (as well as those based on other integrated processors compatible with them), especially with regard to the intensive use of their graphics section, that is, mainly for gaming.
AMD Fluid Motion Frames 2
AMD Fluid Motion Frames 2 is, as you may already know, the second generation of AMD technology that marks this technology’s response to NVIDIA DLSS 3 Frame Generation, that is, is capable of generating “intermediate” frames, exclusively with the GPU and making use of AI, which is interspersed between those generated in the traditional way, that is, also using the CPU. These frame generation technologies, which are able to take advantage of the specialized functions of graphic adapters, contribute to significantly reducing the well-known CPU bottleneck.
At the end of July AMD published a technical preview of AFMF 2 and, from what we could see at that time, It seems that those from Santa Clara had stepped on the accelerator after a somewhat improvable debut of this technology in its first generation. As we told you then, the announcement of DLSS 3 caught AMD off guard, which forced this technology company to make an announcement more than a year before the arrival of AFMF to users. And, even so, with its debut it was evident that there was still a lot of work to be done.
Now with AFMF 2, which is now available in preview for users of systems built around the Ryzen AI 300 APUs, Everything indicates that AMD has caught upand is therefore now in a position to offer an alternative to NVIDIA’s proposal. With the added benefit that, as was the case with its first generation, Fluid Motion Frames 2 has a much broader scope and, according to AMD, its implementation is much simpler.
Let’s remember that these were the main new features that, according to AMD, would arrive with AFMF 2:
- AI-optimized enhancements – AFMF 2 provides significant improvements to frame rate generation using AI optimization when developing the updated algorithm, resulting in smoother frame rate. Additionally, to allow for greater control over how AFMF works with your games, AFMF 2 introduces two new modes where optimal settings are automatically enabled, but advanced users can fine-tune them to their preferences.
- Performance improvements – AFMF 2 features a new “Performance” mode that reduces the overhead of AFMF 2 to help make high frame rate gaming experiences more achievable on a wider range of devices. This is especially beneficial when using AFMF 2 with integrated graphics cards and is now the default “Auto” setting on AMD Ryzen processors that support Radeon Graphics.
- Lower Latency Frame Generation – Experience significant improvements in reducing latency added by frame generation with the latest version of AFMF. These improvements apply across the board when using AFMF 2, regardless of the configuration mode, resolution, or hardware used.
- Other updates and improvements – AFMF 2 now supports borderless fullscreen mode when using AMD Radeon RX 7000 and Radeon 700M series graphics cards. Additionally, AFMF 2 now supports games that utilize Vulkan and OpenGL, further expanding the already vast number of games that can be used with AFMF. Additionally, interoperability with AMD Radeon Chill has been enabled.
Variable Graphics Memory
Less known, but no less important, Variable Graphics Memory represents a significant evolution over the shared memory technology already present in AMD integrated circuits.As you can imagine, both have a common root, which consists of “stealing” part of the system’s RAM to add it to the system’s VRAM, thus increasing the system’s capabilities when carrying out tasks (and in tasks, obviously, I include gaming) that are especially demanding on the graphics section and/or on the system’s VRAM.
Thus, in this case we find a proposal that allows you to “convert” up to 75% of RAM into graphics memoryand as its name indicates, it allows us to make this type of adjustment dynamically, depending on the needs of the game we are going to use (or application, which I can also think of some cases). And this is especially useful if we are talking about laptops with a graphics adapter integrated into the APU, as is the case with systems with the Ryzen AI 300, in which the VRAM tends to be low and, therefore, not cover the minimum requirements of many games.
Another use in which the VGM can be especially practical is, indeed, the one referred to above. to applications that employ local AI modelsand as you may already know, these types of applications (or to be more exact, the models they use) tend to be very demanding with the amount of graphic memory in the system. Thus, with this technology that is already available in the Ryzen AI 300, its users will be able to bypass this limitation.
These two functions, as I indicated at the beginning, are available in one Preview version included in a special edition of Adrenalinwhich you can download from here. Keep in mind, however, that although we can expect them to be quite polished, we are not yet looking at the final versions of these technologies, so it is possible that they will not function 100% as expected.
Source: www.muycomputer.com
