Microsoft has released a new DirectX 12 Agility development kit that includes new features that should be key in the hands of programmers. improve game performance.

You already know that DX 12 is the latest version of Microsoft’s Application Programming Interface (API). They are libraries used for creating multimedia and video applications, especially video games, and although there are other development standards (eg OpenGL and its promising version Vulkan), DX is heavily used by the industry and is one of the keys for Windows to maintain its huge desktop computer market share. Every piece of news it provides is therefore of general interest.

DirectX 12 Agility 1.710.0

The new version of this development kit for libraries launches two functions called Heaps of GPU uploads and Non-normalized sampling. They are intended for programmers, but should improve game performance at the end-client level by making better use of PC hardware resources.

The first function is particularly important, so CPU and GPU access VRAM (graphics memory) simultaneously.. It should be explained that historically, the processor did not have access to the graphics memory of the dedicated processor, which forced programs to copy large amounts of data to the GPU via the PCI bus.

Most modern GPUs have implemented a VRAM Variable Size Base Address Register (BAR) that allows Windows to manage GPU VRAM with WDDM 2.0 or later drivers. The latest version of the DirectX 12 Agility SDK will allow the CPU to access graphics memory at the same time as the graphics card itself. there is no need to copy data from the CPU to the GPU.

Microsoft says that this optimization “can offer many performance benefits in PC games”in a context where video games are huge memory hogs as their sophistication and visual complexity increases and VRAM is key to their performance.

Shared memory between CPU and GPU eliminates the need to duplicate data and it will reduce the flow of data that must be transferred via the PCI-Express interface and it should also free up the RAM used by the CPU in this type of task. Memory speed shouldn’t be an issue since the dedicated memory that comes with modern graphics cards is very fast GDDR6. In fact, latency and access times are likely to improve.

Microsoft has already delivered a new version of the SDK to developers, and it should only be a matter of time before games take advantage of these types of optimizations. It will also be supported by three major graphics card manufacturers.. NVIDIA has already included it in its Game Ready and Studio drivers (version 531.41 or later), as has Intel for its A and Xe series (with driver 31.0.101.4255 or later). AMD Radeons will soon support it and the end customer just needs to update the drivers.

It’s hard to quantify the performance improvement that can be achieved with this feature, and it will depend a lot on each game and the ability of developers to take advantage of it. Promising on paper anyway.