Why 0.9 only support vs2015?

i can not select vs 2013/2012 other than vs2015,but 2012 and 2013 has more users than vs2015,can i change it??? or other methods can i create project for vs2012?

All the libraries that OF uses have to be built for one version of Visual Studio.
If we build libs for VS 2012 they won’t link in 2013 and also 2015.

So we usually always support the latest IDE at the time of release.
In this case it is VS 2015

I apologize that we can’t support multiple VS IDEs - we would love to!
Its the fact that the libs can’t be used across different VS editions that stops us from doing that.

All the best!

If you desperately need VS 2012 support, for instance if you need to use a library that is super hard to compile for VS 2015 (like me!) then you can always develop against a previous point in the commit history.

I have a project going using something close to 0.9 in VS 2012 using this branch:

Obviously comes with it’s own drawbacks, but can sort you out with a working VS2012 project file if needed.


Is there a reason you need OF0.9 to work in VS2012?

OF0.8.4 was built for VS2012 so if you need to use that IDE you can just use the older release. We would need a better reason than user base to compile the libraries for older VS IDEs especially when you can have multiple installed and even more so since VS2015 is free. You don’t even need the other ones you just need the compiler so you can use 2015 and compile with the older VSMC versions by changing the settings in the project properties.

We also have code in the main library that isn’t supported by VS2012 namely c++11/14 functionality that has been since added into 2013 and 2015.

Is openFrameworks possible to support vs2013? I need to use CUDA 7.5 library with OF0.9.2, but CUDA 7.5 only support Visual Studio up to 2013(http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/index.html#abstract).

How can I build OF0.9.2 library for vs2013 by myself?

Check out the Apothecary scripts for help compiling libs that OF requires. Its how we built them for 2015 but you will need to find the proper libs in order to support 2013.

An alternative would to be to use a OF version before 0.9.0 since those supported 2012 and thus have CUDA support. That is probably the easier method.