You're grabbing the frame from the framebuffer for processing on an ultra-parallel device with direct access to the framebuffer. It's always going to be faster than FRAPS, provided you notice the performance impact of either.
OS: Windows 10 64 bit Professional
CPU: AMD Ryzen 5900X
RAM: 16GB
GPU: Radeon Vega 56
CPU: AMD Ryzen 5900X
RAM: 16GB
GPU: Radeon Vega 56
