Tweak github release text

ericpre · Mar 17, 2024 · 7e0cf4b · 7e0cf4b
1 parent fc6c03a
commit 7e0cf4b
Showing 1 changed file with 11 additions and 7 deletions.
diff --git a/RELEASE_TEXT.md b/RELEASE_TEXT.md
@@ -1,20 +1,24 @@
-The HyperSpy bundle is similar to the [Miniforge](https://github.com/conda-forge/miniforge)/[Anaconda](https://docs.anaconda.com/anaconda/) distribution and uses the `conda` or `mamba` package manager.
+Download one of the installers in the `"Assets"` section.
 
-**[Installation instructions](https://hyperspy.org/hyperspy-bundle/install.html)**
+See below the information to select the most suitable installer for a given platform (`Windows`, `Linux` or `Mac`) and architecture / type (`AMD`, `Intel`, `Silicon`, `Portable`).
 
 ## Recommended: Anaconda-type distribution (Linux, MacOS and Windows)
 
-For Windows and Linux, there is the choice between two variants, which are optimised for Intel or AMD CPUs:
-- `Intel`: with the MKL library,
-- `AMD`: with the openblas library.
+The HyperSpy bundle is similar to the [Miniforge](https://github.com/conda-forge/miniforge)/[Anaconda](https://docs.anaconda.com/anaconda/) distribution and uses the `conda` or `mamba` package manager.
+
+**[Installation instructions](https://hyperspy.org/hyperspy-bundle/install.html)**
+
+For Windows and Linux, there is the choice between two variants:
+- `Intel`: optimised for Intel architecture (uses the MKL library),
+- `AMD`: optimised for AMD architecture (uses the openblas library).
 
 For Mac, there are also two variants available:
-- `Intel` for Mac computers with Intel CPU, with MKL library
+- `Intel` for Mac computers with Intel CPU, with the MKL library
 - `Silicon` for Mac computers with Apple Silicon (M1, M2, etc.) CPU, with the Accelerate library
 
 For included packages, [see documentation](https://hyperspy.org/hyperspy-bundle/index.html#included-software-and-libraries).
 
-This distribution is built using [constructor 3.6.0](https://github.com/conda/constructor).
+This distribution is built using [constructor 3.6.0](https://conda.github.io/constructor).
 
 ## Portable (Windows only)
 The portable distribution is based on the [WinPython](https://winpython.github.io) distribution and is a self-extracting archive. Running the installer will install the distribution in the current directory. Since the distribution is portable, it can be moved to any directory or run from an external drive.