RERUN_LOG.txt -- re-runs of 9 October 2026 (Python 3.13.5, numpy 2.5.2, scipy 1.18.1, mpmath 1.3.0, macOS,
one laptop, one process at a time)

Output of `sh run_quick.sh`, run from an extracted copy of the revised archive in its state before this log was
written.  After that run only main.tex, verification_report.md, README.md and independent_run_2/README.md were
edited (wording), and this log was replaced; all programs and all recorded outputs of the final archive are
byte-identical to those of the archive used.  Total time 24 minutes.
"identical": the new output is byte-identical to the recorded file.  "identical up to times": identical after
removing the fields that record running times.  "passed": the program reported that all its tests passed.

Earlier re-runs of the same day, each with 0 comparisons not identical: two when the note was written (from a
copy of this folder and from an extracted copy of the first archive), and one by the third verification run from
an extracted copy of the archive as it was before the revision (without the folder independent_run_2/ and with
the first version of writing_stage/check_prop_E.py).

Not repeated by run_quick.sh: the 26 optimisation runs of verification run A (about 55 minutes with 6
processes).  Their recorded raw results (verification_run_A/outputs/run_data) were post-processed again by
post_vA.py and summ_table_vA.py; the outputs are identical to the recorded ones (see below).  Three of the 26 runs
(single_AP, lacdiag_AP, flat8_P) were repeated separately by the third verification run, with the arguments of
jobs_vA.txt: their logs are identical to the recorded logs up to the fields "sec=", and all arrays of their result
files are identical to the recorded ones.

== writing_stage/ (programs written with the note)
identical             writing_stage/constants_table.out
identical             writing_stage/figure_tikz.tex
identical             writing_stage/check_prop_E.out
passed                check_prop_E.py
== original/ (programs with which the results were first obtained)
identical             original check_lemmas.out
passed                check_lemmas.py
identical             original constants.out
identical             original lower_bound_illustration.out
identical             original check_lemma6_adversarial.out
passed                check_lemma6_adversarial.py
identical             original growth_dense1_s1.0.out
identical             original growth_dense2_s1.0.out
identical             original growth_dense4_s1.0.out
identical             original growth_dense8_s1.0.out
identical             original growth_lac4_s1.0.out
identical             original growth_lac6_s1.0.out
== verification_run_A/
identical             A consts_vA.out
identical             A b1_moment_check_vA.out
identical             A lemma_checks_vA.out
passed                lemma_checks_vA.py (ALL PASS)
identical             A validate_optimiser_vA.out
identical             A modsaw_vA.out
identical             A growth_vA.out (post-processing of the recorded runs)
identical             A growth_summary_vA.md
== verification_run_B/
identical up to times B propC_check_B.out
identical up to times B remarkE_check_B.out
identical up to times B second_route_check_B.out
== independent_run_2/ (the verification run made on the final text)
identical             independent_run_2 run2_constants.out
identical             independent_run_2 run2_lemmas.out
identical             independent_run_2 run2_prop91.out
identical             independent_run_2 run2_lower_bound.out
passed                independent_run_2 (four programs)
== done: 0 comparison(s) not identical
