These are "instructions for the impatient": shortcuts for getting Amber
installed on popular operating systems. For more details, see Chapter 2
of the Amber 2018 Reference Manual, or visit
Jason Swails' wiki.
Installing Amber on MacOS
Install XCode: Xcode contains Apple's development tools. You can
download this from the Mac App store (just search for "xcode"). After
installing XCode, you need to agree to the license. So run the following
command in a terminal:
Then, read through the license (or type q to skip through it), and type "agree"
when prompted. After this, you need to install the command-line tools.
Running the following command in the terminal installs the command-line tools:
sudo xcodebuild -license
Install XQuartz: Starting with Mac OS X 10.8, Apple stopped including
X11 with their OS. Instead, you have to download the XQuartz program. You can
download it here.
Install gfortran: You will need to install a Fortran compiler, since
Apple doesn't supply one. Get one for your MacOS version
Install Amber: Now you are ready to proceed with the installation
instructions in the Amber manual. The configure command should be this:
./configure -macAccelerate clang
Notes about Mojave (OSX 10.14)
A few tweaks are needed for Mojave; see this
mailing list post.
We are in the process (Nov. 2018) of trying to assemble an update that will
work for all versions of MacOS, but for now, you will need to follow the notes
on the mailing list for machines running Mojave.
Installing Amber on Ubuntu Linux and Windows 10
All of Amber will compile and run on
Ubuntu Linux platforms (and Debian
lubuntu, Kubuntu and xubuntu, and the Windows 10 "subsystem for Linux"). The
following command should get you the needed packages:
sudo apt-get install csh flex patch gfortran g++ make xorg-dev \
libbz2-dev zlib1g-dev \
libboost-dev libboost-thread-dev libboost-system-dev
Note: Starting with AmberTools16, you don't need to install the
python libraries if you don't wish to: the configure script will offer to
download a compatible python for you.
Caution: depending on the graphics hardware (specifically, onboard
intel chipsets) and associated drivers for your particular setup, installing
the xorg-dev package on Ubuntu has been reported to sometimes also require a
reinstall of the xserver (xserver-xorg) and associated machine-dependent
3D acceleration libraries. If you aren't ready to do this, then just
configure AMBER with the -noX11 option.
If you want to install Amber in parallel, you can probably use the apt-get
versions of MPI:
sudo apt-get install openmpi-bin libopenmpi-dev
Some users might prefer mpich to openmpi. In that case:
sudo apt-get install mpich libmpich-dev
If you have trouble with these, go to
$AMBERHOME/AmberTools/src, and execute either
Installing Amber on Fedora core Linux
Instructions are generally the same as above, except that yum is used
as a package manager. The following command should get you the needed
sudo yum install gcc gcc-gfortran gcc-c++ flex tcsh zlib-devel \
bzip2-devel libXt-devel libXext-devel libXdmcp-devel \
tkinter openmpi openmpi-devel perl perl-ExtUtils-MakeMaker \
patch bison boost-devel
The above should also work on CentOS and Red Hat Enterprise Linux systems as
Installing Amber on openSUSE Linux
Instructions are generally the same as above, except that
zypper is used as a package manager. The following command should
get you the needed libraries:
sudo zypper install gcc gcc-c++ gcc-fortran libtool \
xorg-x11-devel xorg-x11-libXt-devel libbz2-devel \
patch make flex glibc-devel-static boost-devel