1.1 Requisitos de hardware

Currently, the wgpu-py API supports the following platforms:

• Windows: DirectX12, Vulkan, or Angle.
• Linux & ndroid: Vulkan, GLES3, or Angle.
• MacOS & iOS: Metal.

Therefore, to run wgpu-py applications on your local machine, you need to install a modern graphics card with a modern graphics backend. The Nvidia GeForce or AMD Radeon graphics cards with the Vulkan, DirectX12, or Metal API are good candidates for testing wgpu-py applications.

On Windows 10+, wgpu-py applications should just work. As an example, I am using a Windows 10 desktop PC with two graphics cards. Fig.1-1 shows the Device Manager of my PC, where two graphics cards are displayed under the Display adapters: one is an Intel HD Graphics 4600 with a DirectX 12 backend and the other is an AMD Radeon R9 200 Series with a Vulkan API. The wgpu-py applications should be able to run on both graphics cards. Since the Intel graphics card, integrated with the motherboard, is the default graphics card, the wgpu-py* application will run on the Intel graphics card by default.

In this book, I will run our wgpu-py examples on the AMD Radeon graphics card because this AMD graphics card is more powerful than the default Intel graphics card. Unfortunately, Windows 10 does not allow you to easily switch the default graphics card. One way to do this is to disable the Intel graphics card from the Device Manager window, right click on Intel (R) HD Graphics 4600 and select Disable device, as shown in Fig.1-2. After this, my PC will use the AMD Radeon graphics card for its display.

On older Windows versions, you may need to install the Vulkan drivers.