typhoon.git/tests/ECM.md, branch master

Eval mobility/pseudo-mobility tweaks; drop unproductive move-ordering experiments.

2026-10-10T17:29:48Z

eval.c: add bishop/rook x-ray pseudo-mobility terms (squares seen past a directly-attacked higher-value piece), trim pawn-centrality edge penalty and bishop-pair bonus. chess.h/dynamic.c/generate.c/search.c/searchsup.c: revert three move-ordering experiments (recapture bonus, flee-to-safety GOOD_MOVE promotion, ply-2 move-pair bonus) that calibrated with real fail-high signal but never converted into a measurable node-count or solve-rate win on the curated suites -- see MOVE_ORDERING_IDEAS.md for the evidence and reasoning behind each drop. Add MOVE_ORDERING_EVIDENCE_COLLECTION as its own GNUmakefile flag, independent of PERF_COUNTERS, for any future such harness. command.c: remove the KING_DEFECT_DUMP-gated evalkingdump command, dead since the king-safety remine work moved to eval_tune/king_safety_data/. Co-Authored-By: Claude Sonnet 5

Dynamic move-ordering/LMR session: honest FH tracking, decoupled admission/magnitude, bugfixes, tooling

2026-10-04T15:17:52Z

Lands the 2026-09-28..10-03 move-ordering investigation (full history in lmr_testing/*.md). Summary of what ships: LMR/move-ordering redesign (searchsup.c, dynamic.c, data.c, chess.h): - GetLMRReduction's admission decision (do we reduce at all) is now decoupled from magnitude (how much): LMR_STATIC_ADMIT_THOUSANDTHS=500 gates admission via the 4D joint fail-high table (data.c) plus a countermove-match veto; magnitude is a flat Crafty/Ethereal-shaped log(depth)*log(moves) table (g_iLMRQuietReduction), which measured as good or better than the old per-tier joint-table payout while being structurally immune to the per-bucket miscalibration that sank the tiered approach. g_FailHighs honesty fix (real 0/0 init, not a fake prior) underlies the dynamic-danger signal. - Retired NumLeftoverMovesToSelect's depth-indexed selection budget -- evidence showed the leftover pool has real findable signal a bailout was discarding. - ShouldLatePrune (overt late-move pruning) implemented but disabled at its call site -- measured net-negative at the tested threshold; kept for reference and as a base for a future hang-bucket-9-targeted retry. - PRUNE_VERIFY, PRUNE_LIVE, DISABLE_LMR: opt-in diagnostic build flags (GNUmakefile), off in the standard profile. Bug fixes found and fixed while exercising this work: - JOINT_EXTRA_TIER_COUNT 4->6 (chess.h): fixed an out-of-bounds counter write reachable at remaining depth >= 14 (unreachable at sd10, routine in real search/self-play). - Node-type instrumentation (search.c, root.c, searchsup.c): fixed four compounding defects that produced impossible percentages (e.g. 106.2% of PV nodes) -- wrong denominator, qsearch misclassified as main search, untyped early-exit nodes, and fail-low tests using the in-place-raised alpha instead of the initial one. - FH-rate unit unification (search.c, searchsup.c, dynamic.c, chess.h): GetMoveFailHigh* returned per-mille while g_uJointFHThousandths is thousandths-of-a-percent, two different units sharing a "_THOUSANDTHS" suffix. Unified to one scale, renamed to "_PCT_THOUSANDTHS". RateBucket's band edges fixed to match. - QSearch option B (search.c): waives the opponent-stand-pat requirement at NODE_PV only (sound there since the exact score propagates; not sound at null-window nodes). iExtensionAmount unit fix (was 0/2 into a ONE_PLY-scaled field). pi->iEval now published when Eval() runs. - eval_tune/match_play.py local_ebf: was reporting 0.000 in every match since 2026-09-18 (regex only matched the trailing-stats parser, not the live move-scan loop; engines were also spawned with "nopost" so root.c never emitted the lines at all). True per-ply EBF is ~2.0, not the ~3.9 every prior match report implied. - eval_tune/test_vs_head.sh: positional-arg shift ordering bug caused every no-extra-args run to silently rebuild ./typhoon with a non-standard profile and, separately, could feed a flag's value in as the candidate binary path. - PERF_COUNTERS cleanup (eval.c, search.c): moved 3 leaked rdtsc sites to the existing EVAL_TIME guard, deleted two concluded one-off diagnostics (u64HeadStyle*, u64FHJointHist/RateBucket/AttemptsBucket) and their report plumbing, and re-gated a per-move hang-material estimate behind FH_TABLE_COLLECTION instead of running it unconditionally. Infrastructure: - match_play.py time-odds support (--odds/--st-head/--st-candidate), validated against the engine seat rather than color. - Wilson-upper-bound re-bake of the joint table (data.c): cells store a confidence-aware bound instead of a raw point estimate, so a consumer can threshold on the value alone; monotone, so every consumer became strictly more conservative. Verified behavior-neutral for LMR. Known-open items (not resolved here, see lmr_testing/GET_LMR_REDUCTION.md section 10 and 2026-10-03_UNCOMMITTED_TRIAGE.md): the self-play gate on this candidate is underpowered (+5 Elo, 95% CI roughly [-6, +17]); the curated-suite solve-count deltas used earlier in the investigation turned out to be noise-dominated cancellation when diffed per position, not the clean signal they first appeared to be. Co-Authored-By: Claude Sonnet 5

Makes recogn.c use g_iDrawScore. Some refactoring in the pawn

2026-09-20T15:13:57Z

hash / _EvalPawn including some scoring changes. Self-play positive.

ECM.md: classify f698c02's new ECM solve losses (5 legit, 4 flaky)

2026-09-19T03:11:14Z

Investigated per the established HEAD-confidence methodology after the f698c02 checkpoint's aggregate sd10/sn5M drop (98/191->94/191, 99/191->98/191) looked concerning despite winning self-play against old HEAD, typhoon_conservative, and typhoon_liberal. Reran archived head_reference/archive/typhoon-4a7ec9a against all three curated suites to get real before/after logs (the in-place head_reference/logs/ had already been overwritten by update_head_reference.sh). 5 of the new losses are real (ECM.746/280/597, plus already-known 736/251); 4 are suite noise (ECM.357/717 abandon a confirmed move for an unconfirmed one at the last iteration, ECM.601 has no confirmation signature at all, ECM.193 wavers to the node-budget cutoff) -- roughly flat net real signal, not the large regression the raw aggregate suggested. Co-Authored-By: Claude Sonnet 5

Add king-safety mate-mining regression pipeline; retune file-storm/

2026-09-14T06:18:07Z

queen-proximity king-safety weights and add a back-flight bonus from it. Mines real human-game checkmates and matched sound-draw controls (eval_tune/mine_king_mates.py, shard_mine.py, extract_king_features.py, match_controls.py, train_king_safety_model.py), then fits a logistic regression against _EvalKing's raw per-defect breakdown (engine-side support: chess.h's KING_DEFECT_DUMP_INFO, eval.c's KING_DEFECT_DUMP instrumentation, command.c's evalkingdump command, all gated behind the KING_DEFECT_DUMP=1 build flag) to see which defects actually predict getting mated. See eval_tune/king_safety_data/HANDOFF.md for the full methodology and findings. Two rounds of retuning landed from this data so far, both validated via sd10 curated-suite comparison plus per-position Crafty corroboration (tests/ECM.md) and gated self-play: - KingFileDefects/KingStormingPawnDefects (_ComputeFileStormDefects) bumped ~1.75-2x relative to the rest of the ring-scan defects -- the regression's most dominant, most stable tier-1 predictors, previously weighted the same as every other hand-picked defect. - KING_QUEEN_PROXIMITY_DANGER rescaled from -8/count to -32/count, and a new KING_BACK_FLIGHT_BONUS (+1/square) term added for empty, unattacked ring squares on the side of the king away from the attacker -- a real, non-collinear protective signal the current EmptyUnattackedSquare table didn't credit at all. Both calibrated by unstandardizing the regression's coefficients and anchoring against KING_MISSING_ONE_CASTLE_OPTION's already-live -23 value (same direct-EVAL_TERM unit family). Co-Authored-By: Claude Sonnet 5 Claude-Session: https://claude.ai/code/session_01P9EDZVcqu6E2bNRm4zDrTf