Blendyn has been ported to Blender >=2.80. Blendyn is currently still compatible with Blender versions from 2.50 to 2.79, through the code in the branch blendyn27. Be careful though, that only occasional bugfixes will be applied in that branch, as the main development will be carried out in the devel and master branches.

The basic installation of the add-on is very simple. Download the repository as a .zip file using the Download ZIP button at the top right of the repository main page, under the Code menu.

NOTE: please check that the current active branch is "master": that is the one you want to download, unless you are an MBDyn developer or, in general, have access to the bleeding edge MBDyn's source code.

In Blender, go to Edit -> Preferences -> Add-ons -> Install and locate the .zip file that you have downloaded. The add-on will be shown under the Physics category. You will have to activate it in the usual way, by clicking on the checkbox on the right.

If, upon the activation of the add-on, complaints that the numpy package is not installed, please follow the instructions below to install pip, if it is not already present in your system, and then install numpy with

# pip install numpy

or

# pip3 install numpy

Additionally, for some elements (like beams, for example) to be visualized correctly, you need to activate the Auto Run Python Scripts option in the Save & Load tab of the Blender User Preferences, as shown in the picture below.

A few optional Python packages are needed to enable some additional features in Blendyn:

• the netcdf4 package is needed to import MBDyn results in NetCDF format
• the pygal and cairosvg packages are needed to enable the plotting of MBDyn output from the Blender UI
• the psutil package is needed to enable running MBDyn from the Blender UI

Blendyn depends also on numpy, which should be bundled in recent Blender versions, but may be missing on some platforms.

In the current version, Blendyn is equipped with an automatic system to check for the presence of the packages required by the additional features, and install them. The panel is accessed through the Blendyn addon section in the Edit -> Preferences -> Add-ons Blender menu.

Below the Blendyn info, a Preferences section, dedicated to optional features, is displayed:

if the packages required by an additional feature are missing, an Install dependencies button is displayed at the end of the related block. Clicking on it will result in Blendyn checking for and, if needed, installing pip and using it to install the required packages. The latter will be installed in the modules folder aside of Blender user addons folder (e.g., for GNU/Linux systems, /home/user/.config/blender/v.vv/scripts/addons/modules, where v.vv is the installed Blender version).

If the packages are found, the Install dependencies button is hidden and a checkmark appears next to the Python packages names.

WARNING: after installing and additional dependency, you should restart Blender to enable Blendyn to recognise and load it.

NOTE 1: Using MBDyn NetCDF output with Blendyn is highly recommended.

The following guide² will provide help in the installation of both MBDyn and Blendyn in Windows 10/11 operating system, using the Windows Subsystem for Linux feature. It assumes that WSL is installed, so please refer to the linked page on the Microsoft documentation to get started from scratch. This guide also assumes that the distribution installed in the WSL is Ubuntu, but keep in mind that other are available, including some (albeit not officially supported by Microsoft) based on Arch (for example this). If you choose one of the latter, you can directly head over to the Linux installation guide.

The part related to the MBDyn installation can be used also as a reference for a Blendyn-friendly MBDyn installation in Ubuntu (and most of Debian and Ubuntu based distributions).

Open WSL and install the required (first apt install line) and additional (second apt install line) packages for MBDyn:

$ sudo apt update
$ sudo apt install autoconf libtool make gcc g++ gfortran
$ sudo apt install liblapack-dev libnetcdf-dev [ libnetcdf-cxx-legacy-dev | libnetcdf-c++4-dev ]

meaning that you can either install libnetcdf-cxx-legacy-dev or libnetcdf-c++4-dev, as MBDyn supports both. The latter is actually to be preferred.

Optional but recommended, to be able to perform eigenanalyses, load runtime module and enable some additional linear solvers (refer to the Linux installation guide for details):

$ sudo apt install libltdl-dev libsuitesparse-dev liblapack-dev libarpack2-dev

Download MBDyn, ore clone it from the Gitlab repository (install git if needed before cloning)

$ git clone https://public.gitlab.polimi.it/DAER/mbdyn.git
$ cd mbdyn/

Generate the configure scripts:

$ sh bootstrap.sh

You are now ready to configure, build, and install the MBDyn package:

$ LDFLAGS="-rdynamic -lsuitesparseconfig" \
  CPPFLAGS="-I/usr/local/include/suitesparse" \
  ./configure \
 --prefix=/usr/local \
 --enable-netcdf=yes \
 --with-umfpack=yes \
 --with-lapack=yes \
 --with-arpack=yes \
 --with-module="hid hfelem" \
 --enable-runtime-loading
$ make -j4
$ sudo make install

the choice of changing the prefix to /usr/local is arbitrary: one can choose to keep the default prefix /usr/local/mbdyn and add the /usr/local/mbdyn/bin directory to the systems' PATH environment variable (look here, for example, for more information).

Also the choice of building the hid and hfelem modules it is just an example: you can choose which ones are of interest for you, or just remove the --with-module="" switch entirely if you do not wish to build any of them.

Head over to the Blender website and download the most recent release.

Download the .zip from this repo and install the addon following the instructions at the top of this page.

1: Thanks to Stefano Francesco Pitton for the tip.

2: The original version of this guide was kindly sponsored by Jacopo Surra.