Accelerating Fortran DO CONCURRENT with GPUs and the NVIDIA HPC SDK

I am beginning my learning in programming NVIDIA GPUs, and just read this very instructive and useful article. I would like to use DO CONCURRENT and the HPC SDK to accelerate several of my Fortran programs, which need to run on Windows, using GPUs. I saw that the HPC SDK is not available yet for Windows. I would like to ask the authors @gozen or @grahamlopez if there is any estimate of when this Windows version will become available.

I also saw on another post (WSL and PGI compiler works great!) that one can enable the Windows Subsystem for Linux (WSL) on a Windows 10 machine, install the HPC SDK in WSL, compile Fortran code with the NVFORTRAN compiler, and then (apparently) invoke the resulting executable from Windows. This may provide a solution to accelerate my Fortran code under Windows until a HPC SDK version for Windows is released, but only for Windows 10. Is there any way to run the resulting Linux executable under previous Windows versions, in which WSL is not available?