Release notes here.
Please download MCX and MCXLAB v0.9.7-2 at our registration/download page. We are greatly appreciated if you can tell us a little bit about you and your related research by registering your copy of MCX. Binary executable (for 32bit and 64bit machines) and source code packages are both provided. For MCXLAB, a single package containing the pre-compiled Matlab/Octave MEX files for 32/64bit Linux and 32/64bit Windows is provided.
First, make sure you have installed svn client on your system. If you are using a Debian-based GNU/Linux system (such as Ubuntu, Knoppix etc), you can install svn by
sudo apt-get install subversion
if you are using a Redhat-based GNU/Linux system (such as Fedora, CentOS etc), you can do this by
su -c 'yum install subversion'
If your operating system is Windows, we recommend you installing TortoiseSVN.
The latest code snapshot can be checked out from project's SVN repository. This can be done anonymously by the following command.
svn co https://svn.code.sf.net/p/mcx/svn/mcextreme_cuda/trunk mcx
The password is the same as the username. After you checking out the code, you can compile and run the software.
If you are one of the developers who have SVN write-permission, you can checkout the latest code with the same command above.
Once checkout, you can simply make changes or debug your changes in your local copy. If your local changes have passed review by other co-developers, you can then commit your changes back to the svn repository by
cd /path/to/mcx/local/copy svn commit -m "a short summary of your changes"
You can also find more algorithm and implementation details about MCX in this paper.