# Quickstart This is the shortest path from a fresh clone to a finished refinement, on both **Linux** and **Windows**. It takes a few minutes, most of which is the one-time environment install. ## Prerequisites - [pixi](https://pixi.sh) — the only tool you need to install yourself. It manages the whole environment, **including GSAS-II** (installed automatically; you do not install GSAS-II separately). - `git`. The `pixi run ...` commands below are identical on Linux and Windows — run them in your terminal (bash/zsh on Linux, PowerShell or Command Prompt on Windows) from the repository root. ## 1. Clone and install ```bash git clone https://github.com/NSLS2/PowderLine.git cd PowderLine pixi install ``` `pixi install` creates the environment and installs GSAS-II. The first run downloads packages and takes a few minutes; later runs are instant. ## 2. Run an example refinement PowderLine ships real refinement cases under `examples/`. Run the LaB6 calibrant: ```bash pixi run kicker examples/example_LaB6/input.json ``` You'll see progress and a final weighted-profile R-factor (Rwp). That's a complete Rietveld refinement driven entirely by the JSON recipe. > **One recipe, multiple engines.** GSAS-II is the default engine, but the same > recipe also runs through Bruker TOPAS v7 or the open-source easydiffraction. > See [Getting Started](getting-started.rst) for `engine="topas"` / > `engine="easydiffraction"`. ## 3. Read the output Results are written to `examples/example_LaB6/output/`: | File | What it is | |---|---| | `refined_parameters.csv` | Every refined parameter with its ESD (9-column table) | | `LaB6_unit_cell_report.csv` | Unit-cell parameters + ESDs for the phase | | `LaB6_peak_list_report.csv` | Reflection list (hkl, d-spacing, 2θ, intensity) | | `fit_profile.txt` | Observed / calculated / background / difference intensities | | `dummy.lst` | Human-readable GSAS-II log — open this first for Rwp and parameter tables | | `dummy.gpx` | GSAS-II project file — reopen it in the GSAS-II GUI to inspect visually | Open `dummy.lst` in any text editor to see the refinement summary, or load `dummy.gpx` in GSAS-II to explore the fit graphically. ## 4. Validate a recipe (no refinement) To check that a recipe is well-formed without running it: ```bash pixi run kicker --validate-only examples/example_LaB6/input.json ``` This is the fastest way to confirm your own recipe is valid before a full run. ## 5. Make it your own - Copy an example directory (e.g. `examples/example_LaB6/`) and edit its `input.json` — replace the `xrd_data`, structure, and refinement flags with your own. `examples/example_template/` is a blank skeleton to start from. - Send results elsewhere with `--output DIR`. - See more detail in each example's `DESCRIPTION.md`. ## Next steps - **[Getting Started](getting-started.rst)** — full recipe structure, every output file, and the `powderline.run()` Python API. - **[Troubleshooting](TROUBLESHOOTING.md)** — if a run fails or a recipe is rejected. - **[Schema History](SCHEMA_HISTORY.md)** — the recipe schema (current: 0.26.0) and how to migrate older recipes.