Tutorials¶

Six step-by-step tutorials. The first four build on each other and use the files in docs/tutorials/files/; tutorials 5 and 6 use the files of examples 22 and 30. Every file is ready to run.

	what you learn	files
1. Your first simulation	write a system file (.ufs) and a run spec (run.json), run them with the CLI, read the results, plot them	plate.ufs, run.json
2. Diagnostics	put instruments where you would put them on the table; probes; spatial models; near/far field and apertures (sampling)	run-diagnostics.json
3. Larger systems	stretcher, amplifier, compressor and SHG; probes and overrides; the same system from Python, with a parameter scan	cpa.ufs, run-cpa.json, cpa.py
4. Your own measured data	load CSV / NPZ / JSON recordings, compare them with the model, calibrate unknown parameters with error bars, predict what you did not measure	make_data.py, data/, run-measured.json, run-fit.json
5. Contrast of an amplifier chain	ghosts, pedestals and a third-order cross-correlator; how a post-pulse becomes a pre-pulse; scanning a parameter with a study	../22-contrast-chain/
6. Diffraction by an iris	Gaussian, flat-top, uniform and square beams through an iris, imaged at several distances; choosing a grid for a hard aperture; checking against the Fresnel integral; the same from Python	../30-iris-diffraction/

Before you start: build the engine once (see Getting started):

dotnet build UltrafastEngine.sln -c Release
pip install -e "python[hosted,plot]"          # for ultrafast-plot and the Python parts
python python/tools/build_engine.py

In these tutorials, ultrafast means the command-line tool. From a checkout without installing anything:

alias ultrafast="dotnet $PWD/src/UltrafastEngine.Cli/bin/Release/net10.0/ultrafast.dll"    # macOS / Linux

On Windows, use dotnet run --project src\UltrafastEngine.Cli -c Release -- or publish the CLI (see Getting started). Run the commands from docs/tutorials/files/.