Merge branch 'main' into radway_82_reslice_memory_estimator

Author: ferenc-gh

docs/source/howto/installation.rst

@@ -25,21 +25,22 @@ A virtual environment can be used if the following conditions are met:
    $ MPICC=$(type -p mpicc) pip install mpi4py==3.1.6
    $ pip install cython numpy pkgconfig setuptools # build dependencies of h5py
    $ CC=$(type -p mpicc) HDF5_MPI="ON" HDF5_DIR=/path/to/parallel-hdf5 pip install --no-build-isolation --no-binary=h5py h5py
-   $ pip install cupy-cuda12x # install cupy-cuda11x if CUDA library/CUDA toolkit version is 11.x
-   $ pip install aiofiles astra-toolbox ccpi-regularisation-cupy click graypy hdf5plugin loguru nvtx pillow pyyaml scikit-image scipy tomobar tqdm
+   $ pip install cupy-cuda13x # install cupy-cuda13x if CUDA library/CUDA toolkit version is 13.x
+   $ pip install aiofiles astra-toolbox click graypy hdf5plugin loguru nvtx pillow pyyaml scikit-image scipy tomobar tqdm
    $ pip install --no-deps httomo httomolib httomolibgpu httomo-backends
 
 Conda environment
 =================
 
-By default the :code:`cupy` installation will install the latest :code:`cuda-cudart`, which might be the higher CUDA version than on the system. One can specify the compatible to their system CUDA package, e.g., :code:`cuda-cudart==12.9.79`.
+By default the :code:`cupy` installation will install the latest :code:`cuda-cudart`. This can result in CUDA versions higher than the supported by the GPU device of the system. One can specify the compatible to their system CUDA package, e.g., :code:`cuda-cudart==12.9.79`.
 
 .. code-block:: console
 
    $ conda create --name httomo
    $ conda activate httomo
    $ conda install -c conda-forge cupy==14.0.1 openmpi==4.1.6 h5py[build=*openmpi*]
    $ conda install -c conda-forge tomopy==1.15.3 # optional
+   $ pip install aiofiles astra-toolbox click graypy hdf5plugin loguru nvtx pillow pyyaml scikit-image scipy tqdm
    $ pip install httomo httomolib httomolibgpu tomobar
 
 Setup HTTomo development environment: