DESTROYING BOULDERS - NUMERICAL RESEARCH RELEASE 04
Research direction: Ashley Kalkowski. Computational engineering and drafting
used Codex assistance and established mechanics, identified in the references.

This release contains numerical source, tests and synthetic/reference results.
No physical specimen has been fractured or validated by these models. The beam,
disk and acoustic screens are separate reference problems, not one demonstrated
fracture machine. There are no instrument controls in this download.

Tested environment: Python 3.12.14, NumPy 2.3.5. Optional Matplotlib is needed for
some figure-generation commands; numerical checks do not require it.
From this extracted directory:
  python -m pip install -r requirements.txt
  python -m unittest discover -s efficiency -p "test_*.py" -v
  python -m unittest discover -s simulation -p "test_*.py" -v
  python -m unittest discover -s alternate-impact -p "test_*.py" -v
  python efficiency/adaptive_timing.py
  python efficiency/iteration4.py
  python efficiency/benchmark_suite.py --no-reuse
  python -m pip install -r requirements-figures.txt
  python generate_journey_figures.py

The supplied reports preserve the inputs used in the published release.
Re-running a script writes new local result files; retain the original extracted
copy to compare hashes. Current source-bound reports include source hashes; all files are identified in
the manifest. The three older quick-study snapshots lack embedded source hashes
and are historical result snapshots, not verified source-bound reruns.
Runtime varies by computer. Every computation is numerical; no output is a
physical actuator rating, fracture efficiency or minimum machine dimension.

Earlier source/report labels may say PRIVATE because they were developed before
the owner's project-specific public authorization. Those historical computation
flags do not mean that a script publishes anything; this reviewed release is
publicly available. Unrepaired source and private administration are not bundled.

The versioned journey explains both unfavorable results and software corrections.
Publication and source availability do not establish patentability, exclusive
ownership, a patent application or a blanket open-source license. Third-party
dependency licenses and attribution remain separate; see DEPENDENCIES.txt.
