On Thu Jul 31 16:30:39 2025 +0000, Tomasz Pakuła wrote:
I think the main issue is that just watching the USB traffic from windows is not enough to understand what's being done by the Dinput and what's just done by the driver AND when a new device is plugged in on Windows, it can do whatever it wants to do, and what the driver need to do to prepare it. Win Wine, the device is already prepared by the system and drivers, so some things should simply not be replicated. The autocentering is one of the better examples.
The tests use a virtual HID device and capture/compare what DInput requests to the device driver directly on the HID level.
It has then also been manually compared against USB traffic by other people, and with HID PID-capable non-Microsoft devices like Simucube.