Minor changes to documentation following recent tests, including NumPy v2, and OpenMPI update. This closes #22

Author: Michael-P-Allen

GUIDE.md

@@ -392,7 +392,7 @@ for all this output to be collated (in an undefined order) and written to a sing
 was carried out using Open MPI, which allows the program to be run with a command line which includes
 an option for each process to write to separate files, similar to the following:
 ```
-mpirun -np 4 -output-filename out ./mc_nvt_lj_re < mc.inp
+mpirun -np 4 --output dir=out ./mc_nvt_lj_re < mc.inp
 ```
 whereby the output files are placed in subdirectories,
 identified by process rank,

README.md

@@ -59,7 +59,7 @@ A few examples consist of individual routines or modules,
 rather than working programs,
 so there is no need to build them.
 
-The build process for the Fortran examples has been tested using SCons v4.8.0
+The build process for the Fortran examples has been tested using SCons v4.8.1
 (and some earlier versions back to v2.5.1 with minor changes to the `SConstruct` file).
 
 If you don't like using SCons, or can't get it to work,
@@ -78,7 +78,7 @@ it is advisable to __compile each example in its own build directory__
 (which is what the `SConstruct` file is configured to do)
 or to delete all intermediate files before each individual compilation.
 
-We have used gfortran v13.2.0 (and earlier versions back to v6.3) for testing,
+We have used gfortran v14.2.0 (and earlier versions back to v6.3) for testing,
 but have attempted to stick to code which conforms to the Fortran 2018 standard.
 Note that, by default, we do not select any optimization options in compilation.
 If you are using a different compiler,
@@ -89,8 +89,8 @@ Unfortunately, due to the enormous variety of computing platforms and compilers,
 __we cannot offer more specific advice on the build process.__
 
 The Python versions do not require building, they are simply run through the Python interpreter.
-They have been tested with Python 3.12.2 and NumPy 1.26.4
-(and previously, Python versions back to 3.6.0).
+They have been tested with Python 3.13.3 and NumPy 2.2.5
+(and previously, Python versions back to 3.6.0, and NumPy v1 up to 1.26.4).
 
 ## Random number generators
 

SConstruct

@@ -4,8 +4,8 @@
 
 import os, sys
 
-# This has been tested using SCons v4.6.0, gfortran v13.2.0,
-# using MacOS (Ventura 13.6.3, Xcode 15.1) with compilers and libraries installed through MacPorts.
+# This has been tested using SCons v4.8.1, gfortran v14.2.0,
+# using MacOS (Sequoia 15.5.3, Xcode 16.3) with compilers and libraries installed through MacPorts.
 # It may not work on your system. It is possible that you can get it to work by
 # changing the flags and library/include paths defined in the following few statements.
 # The most likely trouble spots are the programs that use the non-standard

python_examples/GUIDE.md

@@ -28,6 +28,9 @@ especially for combining arrays.
 On top of this, NumPy and the [SciPy](https://www.scipy.org/ "SciPy home page") package
 provide a huge variety of scientific libraries,
 and our Python examples make extensive use of these.
+Our codes were originally written using NumPy v1,
+but the transition to NumPy v2 has involved very few, minor, changes to them.
+
 Nonetheless,
 any code which cannot be handled in vectorized form by NumPy will still run slowly.
 Several strategies are available to address these issues
@@ -398,7 +401,7 @@ for all this output to be collated (in an undefined order) and written to a sing
 was carried out using Open MPI, which allows the program to be run with a command line which includes
 an option for each process to write to separate files, similar to the following:
 ```
-mpiexec -n 4 -output-filename out ./mc_nvt_lj_re.py < mc.inp
+mpiexec -n 4 --output dir=out ./mc_nvt_lj_re.py < mc.inp
 ```
 whereby the output files are placed in subdirectories,
 identified by process rank,

python_examples/README.md

@@ -31,6 +31,6 @@ The [User Guide](./GUIDE.md) contains some comments on the Python language,
 some notes to assist in running the programs,
 and some typical results.
 The Python versions do not require building, they are simply run through the Python interpreter.
-They have been tested with Python 3.12.2 and NumPy 1.26.4
-(and previously, Python versions back to 3.6.0).
+They have been tested with Python 3.13.3 and NumPy 2.2.5
+(and previously, Python versions back to 3.6.0, and NumPy v1 up to 1.26.4).