PresentMon VS cr

Use PresentMon to figure out if it's a CPU or GPU issue. (GPUBusy)

EDIT - Intel's Presentmon has a "gpu busy" metric kinda designed to tell you if the gpu is waiting on the rest of the system and it's platform independent so should work with amd/nvidia stuff. https://game.intel.com/story/intel-presentmon/

You can also check the presentation latency with PresentMon

Borderless for DX12 and Vulkan games has no input delay. In fact those APIs don't have exclusive fullscreen support anymore. Why would the engineers who designed those get rid of "exclusive fullscreen" support if it wasn't useful? In fact most new games which offer a "fullscreen" option are faking it, you can use PresentMon to see which DXGI presentation mode is being used.

This behaviour can be monitored using PresentMon (best used on a second display) as well. This new DXGI layer would appear as Composed: Flip, Hardware: Independent Flip, or Hardware Composed: Independent Flip, while classic methods would appear as Hardware: Legacy Flip (if FSE was engaged) or Composed: Copy with GPU GDI.

I much prefer the approach taken by cr.h which monitors a "plugin" binary instead and acts as a simple wrapper around platform specific dll/dylib/so loading. You then have a multitude of strategies if you need persist state such as serialization, placing code or data on the host side and using virtual methods/fptrs, holding third party lib references to dlls on the host side or using the CR_STATE mechanism for simple things.

Depends on your use case, but I would second the option of just hot reloading c++. I have been experimenting with that (using cr.h) and its great - as long as you make sure your compile and link times are fast enough.