Installation failure: VS/CUDA/Matlab compatibility

[ckfaber]

I have been trying to get Kilosort installed for over a week and am at a total loss - I'm a total novice at this type of computing, and don't have any resources available at my institution to help, so I would really appreciate some insights!! I continue to have the following error when testing the configuration with mexgpuall, despite having reinstalled Visual Studio 2022, and selecting it for C and C++ comilation via mex -setup and mex -setup C++.

The error:

>> mexGPUall
Warning: The selected C++ compiler is not supported for CUDA compilation. Searching for a supported compiler. 
> In mexcuda (line 96)
In mexGPUall (line 7) 
Error using mex
Supported compiler not detected. For options, visit https://www.mathworks.com/support/compilers.

Error in mexcuda (line 168)
    [varargout{1:nargout}] = mex(mexArguments{:});

Error in mexGPUall (line 7)
    mexcuda -largeArrayDims spikedetector3.cu

My configuration:

gpuDevice returns the following:

>> gpuDevice

ans = 

  CUDADevice with properties:

                      Name: 'NVIDIA RTX A2000 8GB Laptop GPU'
                     Index: 1
         ComputeCapability: '8.6'
            SupportsDouble: 1
             DriverVersion: 12.2000
            ToolkitVersion: 11.2000

mex -setup returns:

>> mex -setup
MEX configured to use 'Microsoft Visual C++ 2022 (C)' for C language compilation.

mex -setup C++ returns:

>> mex -setup C++
MEX configured to use 'Microsoft Visual C++ 2022' for C++ language compilation.

• Matlab: R2022b (9.13.0.2166757 (R2022b) Update 4)
• CUDA : release 12.2, V12.2.91
• Visual Studio : Microsoft Visual Studio Community 2022 (64-bit) - Version 17.6.5

Please help! I have spikes to sort!!

[ckfaber]

Update: I seem to have resolved the issue by switching to Visual Studio 2019. At first, I thought the setup still had failed, as just hitting 'Run' on mexGPUall.m gave a mexcuda error. Stepping through mexGPUall.m line by line worked with no errors, however. Weird.