Hi there,
I’m a real newbie concerning CUDA but I want to work with some Kernels to speed up my custom analysis. I’ve downloaded the CUDA Toolkit and I’m able to compile and run the sample projects. When I’m using the Projucer however I can’t set up Visual Studio 2017 in a way to use the template project provided by NVIDIA, therefore I got lots of Linker errors…
Has anyone done this before?
Thanks!
I’ve figured it out, however I’m going to use ArrayFire now, no need to reinvent the wheel and code my own kernels for now 
You said you figured it out. Can you share with us how you did it? I have the same problem.
I don‘t have that code anymore and I don‘t remember, sorry… Give it a try, it wasn‘t hard actually
I have no idea where to start because the Projucer regenerates the Visual Studio project each time.
Do you remember if you used CMake? Or did you edit some Projucer file?
I included the needed headers in Projucer, no CMake back then. Maybe search the forum but including dependencies should be trivial within the projucer 
It is the needed preprocessing of the .cu files which are the problem (for me). But maybe you do not remember those.
Sorry, I don’t, did you try ArrayFire? Maybe that’s easier
