More small fixes

job_install_hpc.sh

@@ -25,7 +25,7 @@ wget https://github.com/conda-forge/miniforge/releases/latest/download/Mambaforg
 bash Mambaforge-Linux-x86_64.sh -bfp ${MSBS_ROOT}/mambaforge
 
 # Activate Mamba-forge.
-eval "$(${ROOT}/mambaforge/bin/conda shell.bash hook)"
+eval "$(${MSBS_ROOT}/mambaforge/bin/conda shell.bash hook)"
 
 # Make sure your base environment is up-to-date.
 mamba update --all -y
@@ -34,7 +34,7 @@ mamba update --all -y
 mamba create -n openmm python cudatoolkit git jupyterlab numpy pandas scipy matplotlib ipympl rdkit openbabel openmm mdtraj nglview pymbar pdbfixer parmed
 
 # Activate the OpenMM environment
-conda activate openmm
+mamba activate openmm
 
 # Test the installation
 python -m openmm.testInstallation

setup_hpc.md

@@ -45,17 +45,28 @@ We assume the following:
    The last part should be something along the following lines:
 
    ```
-   OpenMM Version: 7.7
-   Git Revision: 130124a3f9277b054ec40927360a6ad20c8f5fa6
+   OpenMM Version: 8.0
+   Git Revision: a7800059645f4471f4b91c21e742fe5aa4513cda
 
-   There are 2 Platforms available:
+   There are 4 Platforms available:
 
    1 Reference - Successfully computed forces
    2 CPU - Successfully computed forces
+   3 CUDA - Successfully computed forces
+   1 warning generated.
+   1 warning generated.
+   1 warning generated.
+   1 warning generated.
+   4 OpenCL - Successfully computed forces
 
    Median difference in forces between platforms:
 
-   Reference vs. CPU: 6.30535e-06
+   Reference vs. CPU: 6.2894e-06
+   Reference vs. CUDA: 6.74325e-06
+   CPU vs. CUDA: 7.39638e-07
+   Reference vs. OpenCL: 6.74399e-06
+   CPU vs. OpenCL: 7.76254e-07
+   CUDA vs. OpenCL: 2.11141e-07
 
    All differences are within tolerance.
    ```
@@ -80,7 +91,7 @@ Submit a simple job script on the queue, in which you perform the same test.
    # Activate the OpenMM software.
    MSBS_ROOT=${VSC_DATA}
    eval "$(${MSBS_ROOT}/mambaforge/bin/conda shell.bash hook)"
-   conda activate openmm
+   mamba activate openmm
    # Set the number of threads
    export OPENMM_CPU_THREADS=${SLURM_CPUS_ON_NODE}
    # Run the test.
@@ -103,23 +114,7 @@ Submit a simple job script on the queue, in which you perform the same test.
    After the job is submitted, you can safely log out (type `exit`), even switch off your laptop, and connect later to check the status of the job.
    When the job status is `C` (completed), or it is no longer present in the output of `squeue`, the job was executed and corresponding output files will be created.
 
-1. The output of the test job should be similar to:
-
-   ```
-   OpenMM Version: 7.7
-   Git Revision: 130124a3f9277b054ec40927360a6ad20c8f5fa6
-
-   There are 2 Platforms available:
-
-   1 Reference - Successfully computed forces
-   2 CPU - Successfully computed forces
-
-   Median difference in forces between platforms:
-
-   Reference vs. CPU: 6.30535e-06
-
-   All differences are within tolerance.
-   ```
+1. The output of the test job should be the same as that of the previous test.
 
 1. Add the following to your `.bashrc` file, if you want to facilitate the activation of Mamba-forge, without having it always active:
 

setup_laptop.md

@@ -1,7 +1,7 @@
 # Setup of OpenMM and Jupyterlab on your laptop
 
 To run OpenMM simulations on your laptop, we will work with a standardized Python environment, Anaconda, which is supported on all major operating systems.
-For reasons of efficiency and robustness, we will in fact use a derivative of Anaconda, called Mamba-forge.
+For reasons of efficiency and robustness, we will in fact use a derivative of Anaconda, called Mambaforge.
 
 :warning: **As of May 2022, the installation of OpenMM on Windows is broken.**
 Technical details of this issue can be found below.
@@ -15,7 +15,7 @@ Most of the simulation software is developed for Unix platforms (macOS and Linux
 
 Take the following steps:
 
-1. Download the [Mamba-forge installer](https://conda-forge.org/miniforge/)
+1. Download the [Mambaforge installer](https://conda-forge.org/miniforge/)
    that matches the operating system and CPU architecture of your laptop.
 
 1. Run the Mambaforge installer.
@@ -28,7 +28,7 @@ Take the following steps:
      ```bash
      bash Mambaforge*.sh -b -p ${HOME}/mambaforge
      ```
-     Add the following line to your `~/.bashrc` file, which makes it convenient to activate the Mamba-forge installation:
+     Add the following line to your `~/.bashrc` file, which makes it convenient to activate the Mambaforge installation:
      ```bash
      alias m='eval "$(${HOME}/mambaforge/bin/conda shell.bash hook)"'
      ```
@@ -45,7 +45,7 @@ Take the following steps:
    - **Linux and WSL.**
      Open your preferred virtual terminal and enter the command alias `m`.
 
-1. Configure conda and install OpenMM (and other useful tools).
+1. Configure Mambaforge and install OpenMM (and other useful tools).
 
    In this step, you all need to enter commands in a virtual terminal, the software equivalent of terminal computers from the 70s.
    Every command you enter will be executed after you press `Enter`.
@@ -75,7 +75,7 @@ Take the following steps:
    ```
 
    You may have to close your terminal and re-open a new one,
-   run the commands `m` (Linux and macOS only) and `conda activate openmm` again before the following steps work.
+   run the commands `m` (Linux and macOS only) and `mamba activate openmm` again before the following steps work.
 
 1. Test your OpenMM installation by entering the following command on the command prompt: