The Ryzen AI 300 processors were launched on the market some time ago. AMD also recently introduced the new Fluid Motion Frames 2 (AFMF 2) technology, which is intended to offer a number of improvements and take the user experience to a new level. This technology will now also be usable with the Ryzen AI 300 processors.

AMD plans to test the cooperation between Variable Graphics Memory (VGM) and AFMF 2. But what is VGM? This is a new technology, especially for the Ryzen AI 300 processors, which should enable users to convert up to 75% of the system RAM into dedicated graphics memory, i.e. vRAM. However, AMD also emphasizes that this is not to be confused with shared graphics memory, which is already available on the newer processors. The company also recommends that at least 16 GB of RAM should be kept free for the CPU.
What does this mean in concrete terms? On a system with 32 GB RAM, which VGM has set to “medium”, 8 GB RAM should be converted into 8 GB vRAM for the integrated AMD Radeon 890M GPUs, for example, after a restart. This represents an increase of about 7.5 GB compared to the default vRAM allocation of 512 MB.

To illustrate this technology once again with graphics, benchmark tests were carried out. A resolution of 2880 x 1800 and medium graphics settings were used, which led to a noticeable increase in performance. However, it should also be mentioned that AMD’s FSR technology was also used. This was set to “Balanced” for this test. The high preset was selected for 1080p benchmarking and this also offered a significant increase. However, as previously mentioned, the FPS was increased slightly with FSR in “Quality” mode. In the game Cyberpunk 2077, AMD recorded a performance increase of 78% with a Ryzen 9 HX 370 processor with a Low preset and Balanced FSR.
So that you can see for yourself, AFMF 2 is now available for these processors as a technical preview in AMD’s Adrenalin Edition software.
Source: AMD
4 Antworten
Kommentar
Lade neue Kommentare
Urgestein
Mitglied
Mitglied
Mitglied
Alle Kommentare lesen unter igor´sLAB Community →