From 1b658814ef57c458caaffd524bc0c88444572ea3 Mon Sep 17 00:00:00 2001 From: igneum-labs <337424239+igneum-labs@users.noreply.github.com> Date: Wed, 7 Oct 2026 22:09:23 +0000 Subject: [PATCH] adv-accept report: the five deepest seeds against the 0.995 floor (three false refusals, one true pass); log copied Co-Authored-By: Claude Fable 5.1 --- .../logs/adv-accept/gap-newlow5.log | 117 ++++++++++++++++++ .../cryptanalysis/report-acceptance-rule.md | 11 +- 2 files changed, 127 insertions(+), 1 deletion(-) create mode 100644 docs/analysis/cryptanalysis/logs/adv-accept/gap-newlow5.log diff --git a/docs/analysis/cryptanalysis/logs/adv-accept/gap-newlow5.log b/docs/analysis/cryptanalysis/logs/adv-accept/gap-newlow5.log new file mode 100644 index 000000000..068327869 --- /dev/null +++ b/docs/analysis/cryptanalysis/logs/adv-accept/gap-newlow5.log @@ -0,0 +1,117 @@ +lease: holding 8 pool cores (8,9,43,44,45,46,47,48, waited 0 s, class adv): adv-accept gap-newlow5 +gap: seed 122960 +gap: program id 4be7393ab6c84802 attempt 0 class mx8-era5800f15a+sh256x27 day 20730 (cache filled in 0.4 s) plant zero-dataset false +gap: mirror validated against hash_warp on 8 units (closed form and live) +gap: closed-form per-site distinct counts equal the library's distinct_indices_v4 on 4096 units +gap: (c) metrics on 64 units: closed distinct_mean 128.000 saturated 0 bias_max 52; live distinct_mean 128.000 saturated 0 bias_max 54 +gap: site 0 closed ratio 1.0000 (1040412) live ratio 1.0000 (1040386) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 1 closed ratio 1.0000 (1046542) live ratio 1.0000 (1046515) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 2 closed ratio 1.0000 (1046528) live ratio 1.0000 (1046577) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 3 closed ratio 1.0000 (1040378) live ratio 1.0002 (1040587) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 4 closed ratio 1.0000 (1044478) live ratio 1.0000 (1044490) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 5 closed ratio 1.0000 (1046515) live ratio 1.0000 (1046531) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 6 closed ratio 1.0000 (1046539) live ratio 1.0000 (1046516) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 7 closed ratio 0.9999 (1046443) live ratio 1.0000 (1046528) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 8 closed ratio 1.0001 (1046608) live ratio 0.9999 (1046470) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 9 closed ratio 1.0000 (1040417) live ratio 1.0001 (1040519) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 10 closed ratio 0.9999 (1044417) live ratio 1.0000 (1044430) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 11 closed ratio 1.0001 (1044609) live ratio 1.0000 (1044477) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 12 closed ratio 0.9824 (1028093) live ratio 0.9822 (1027920) diff +0.0002; reads on indices read >= 8 times: closed 1.3090% live 1.3251% +gap: site 13 closed ratio 1.0001 (1044617) live ratio 1.0001 (1044598) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 14 closed ratio 1.0000 (1040426) live ratio 1.0000 (1040402) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 15 closed ratio 1.0000 (1040387) live ratio 1.0001 (1040465) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: min ratio closed 0.9824 live 0.9822; largest site gap -0.0002 at site 3; verdict under the 0.98 floor: closed accept live accept; 11.8 s +gap DONE 4be7393ab6c84802 +gap: seed 130488 +gap: program id 24d5e6511a1b83dc attempt 6 class mx8-eraaa75d178+sh256x27 day 20730 (cache filled in 0.4 s) plant zero-dataset false +gap: mirror validated against hash_warp on 8 units (closed form and live) +gap: closed-form per-site distinct counts equal the library's distinct_indices_v4 on 4096 units +gap: (c) metrics on 64 units: closed distinct_mean 128.000 saturated 1 bias_max 70; live distinct_mean 128.000 saturated 0 bias_max 77 +gap: site 0 closed ratio 1.0000 (1046529) live ratio 1.0000 (1046535) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 1 closed ratio 1.0000 (1046565) live ratio 1.0000 (1046578) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 2 closed ratio 1.0001 (1040505) live ratio 1.0001 (1040460) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 3 closed ratio 1.0001 (1040487) live ratio 1.0000 (1040383) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 4 closed ratio 1.0000 (1044450) live ratio 1.0000 (1044437) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 5 closed ratio 0.9898 (1035816) live ratio 0.9899 (1035924) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 6 closed ratio 1.0000 (1044483) live ratio 1.0001 (1044574) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 7 closed ratio 1.0000 (1044456) live ratio 1.0001 (1044594) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 8 closed ratio 1.0000 (1046579) live ratio 1.0001 (1046609) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 9 closed ratio 1.0001 (1044544) live ratio 0.9999 (1044416) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 10 closed ratio 1.0001 (1040508) live ratio 1.0000 (1040412) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 11 closed ratio 1.0000 (1040378) live ratio 1.0000 (1040423) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 12 closed ratio 1.0001 (1044573) live ratio 1.0000 (1044445) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 13 closed ratio 0.9999 (1044363) live ratio 0.9999 (1044407) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 14 closed ratio 1.0000 (1044461) live ratio 1.0001 (1044543) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 15 closed ratio 0.9999 (1046441) live ratio 1.0000 (1046495) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: min ratio closed 0.9898 live 0.9899; largest site gap -0.0001 at site 7; verdict under the 0.98 floor: closed accept live accept; 10.9 s +gap DONE 24d5e6511a1b83dc +gap: seed 19921 +gap: program id fe35af21bc43a4cc attempt 7 class mx8-era551f20b4+sh256x27 day 20730 (cache filled in 0.6 s) plant zero-dataset false +gap: mirror validated against hash_warp on 8 units (closed form and live) +gap: closed-form per-site distinct counts equal the library's distinct_indices_v4 on 4096 units +gap: (c) metrics on 64 units: closed distinct_mean 128.000 saturated 0 bias_max 45; live distinct_mean 128.000 saturated 0 bias_max 58 +gap: site 0 closed ratio 1.0000 (1044469) live ratio 1.0000 (1044476) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 1 closed ratio 1.0000 (1046508) live ratio 1.0000 (1046521) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 2 closed ratio 1.0000 (1046506) live ratio 1.0000 (1046512) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 3 closed ratio 1.0000 (1046485) live ratio 1.0000 (1046553) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 4 closed ratio 0.9912 (1037277) live ratio 0.9911 (1037205) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 5 closed ratio 1.0000 (1046520) live ratio 0.9999 (1046458) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 6 closed ratio 1.0000 (1046491) live ratio 1.0000 (1046558) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 7 closed ratio 1.0001 (1044578) live ratio 0.9999 (1044375) diff +0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 8 closed ratio 0.9999 (1046376) live ratio 1.0000 (1046571) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 9 closed ratio 1.0000 (1044452) live ratio 1.0001 (1044609) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 10 closed ratio 1.0001 (1044584) live ratio 1.0000 (1044498) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 11 closed ratio 1.0001 (1044594) live ratio 0.9999 (1044409) diff +0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 12 closed ratio 1.0001 (1040490) live ratio 1.0001 (1040446) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 13 closed ratio 1.0001 (1044628) live ratio 1.0000 (1044521) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 14 closed ratio 1.0000 (1040406) live ratio 1.0001 (1040509) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 15 closed ratio 0.9998 (1040162) live ratio 1.0002 (1040550) diff -0.0004; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: min ratio closed 0.9912 live 0.9911; largest site gap -0.0004 at site 15; verdict under the 0.98 floor: closed accept live accept; 17.4 s +gap DONE fe35af21bc43a4cc +gap: seed 600738 +gap: program id ef6d49f3871af472 attempt 4 class mx8-era190e00ce+sh256x27 day 20730 (cache filled in 0.4 s) plant zero-dataset false +gap: mirror validated against hash_warp on 8 units (closed form and live) +gap: closed-form per-site distinct counts equal the library's distinct_indices_v4 on 4096 units +gap: (c) metrics on 64 units: closed distinct_mean 127.999 saturated 23 bias_max 50; live distinct_mean 128.000 saturated 16 bias_max 71 +gap: site 0 closed ratio 1.0001 (1044552) live ratio 1.0000 (1044519) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 1 closed ratio 1.0000 (1040387) live ratio 1.0001 (1040456) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 2 closed ratio 1.0001 (1040452) live ratio 1.0000 (1040333) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 3 closed ratio 1.0000 (1040400) live ratio 1.0002 (1040621) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 4 closed ratio 1.0000 (1040390) live ratio 1.0002 (1040543) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 5 closed ratio 1.0000 (1046535) live ratio 1.0000 (1046503) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 6 closed ratio 1.0001 (1040495) live ratio 0.9998 (1040221) diff +0.0003; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 7 closed ratio 1.0002 (1040586) live ratio 1.0000 (1040399) diff +0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 8 closed ratio 0.9934 (1039606) live ratio 0.9933 (1039499) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 9 closed ratio 0.9999 (1044362) live ratio 1.0000 (1044485) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 10 closed ratio 1.0000 (1040377) live ratio 1.0002 (1040566) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 11 closed ratio 1.0001 (1046594) live ratio 1.0000 (1046500) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 12 closed ratio 1.0000 (1044514) live ratio 1.0001 (1044594) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 13 closed ratio 1.0000 (1044440) live ratio 1.0001 (1044552) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 14 closed ratio 1.0001 (1046593) live ratio 1.0000 (1046569) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 15 closed ratio 1.0000 (1046509) live ratio 1.0000 (1046522) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: min ratio closed 0.9934 live 0.9933; largest site gap +0.0003 at site 6; verdict under the 0.98 floor: closed accept live accept; 10.8 s +gap DONE ef6d49f3871af472 +gap: seed 602822 +gap: program id d8b5571c7abc80df attempt 6 class mx8-era7d3d7ad9+sh256x27 day 20730 (cache filled in 0.4 s) plant zero-dataset false +gap: mirror validated against hash_warp on 8 units (closed form and live) +gap: closed-form per-site distinct counts equal the library's distinct_indices_v4 on 4096 units +gap: (c) metrics on 64 units: closed distinct_mean 128.000 saturated 0 bias_max 59; live distinct_mean 128.000 saturated 0 bias_max 58 +gap: site 0 closed ratio 1.0001 (1040442) live ratio 1.0002 (1040572) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 1 closed ratio 1.0000 (1046498) live ratio 1.0000 (1046564) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 2 closed ratio 1.0000 (1046576) live ratio 1.0000 (1046476) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 3 closed ratio 0.9951 (1039315) live ratio 0.9951 (1039414) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 4 closed ratio 1.0001 (1040449) live ratio 1.0000 (1040401) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 5 closed ratio 1.0000 (1044504) live ratio 1.0000 (1044510) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 6 closed ratio 0.9999 (1040285) live ratio 1.0001 (1040495) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 7 closed ratio 1.0000 (1044485) live ratio 1.0000 (1044526) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 8 closed ratio 1.0000 (1040341) live ratio 1.0001 (1040518) diff -0.0002; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 9 closed ratio 0.9999 (1046458) live ratio 1.0000 (1046547) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 10 closed ratio 1.0000 (1046546) live ratio 1.0000 (1046553) diff -0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 11 closed ratio 1.0001 (1044550) live ratio 1.0000 (1044461) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 12 closed ratio 1.0001 (1046591) live ratio 1.0000 (1046513) diff +0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 13 closed ratio 1.0001 (1040449) live ratio 1.0000 (1040416) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 14 closed ratio 1.0000 (1044458) live ratio 1.0000 (1044446) diff +0.0000; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: site 15 closed ratio 1.0000 (1044480) live ratio 1.0001 (1044596) diff -0.0001; reads on indices read >= 8 times: closed 0.0000% live 0.0000% +gap: min ratio closed 0.9951 live 0.9951; largest site gap -0.0002 at site 6; verdict under the 0.98 floor: closed accept live accept; 10.6 s +gap DONE d8b5571c7abc80df +lease: released 8 pool cores after 77 s, exit 0 diff --git a/docs/analysis/cryptanalysis/report-acceptance-rule.md b/docs/analysis/cryptanalysis/report-acceptance-rule.md index 22cbb4de2..d68ed0e97 100644 --- a/docs/analysis/cryptanalysis/report-acceptance-rule.md +++ b/docs/analysis/cryptanalysis/report-acceptance-rule.md @@ -277,12 +277,21 @@ distinct-index ratio at the acceptance's own 2^20 sample, against the class v5 f | seed 1738 (gap-hot3.log, build-1, 21:37 BST) | 33ece54e2df5ea44, 4 | 13 (instr 54) | 0.9868 / 0.9870 | refused | hot set, +0.102%, 1.68x | | seed 170 (gap-hot4.log, build-1, 21:48 BST) | 53a02a7c9c9b86f3, 4 | 2 (instr 11) | 0.9880 / 0.9877 | refused | hot set, +0.143%, 2.02x | | seed 122960 (gap-122960.log, build-2, 22:41 BST; the deepest tail seed) | 4be7393ab6c84802, 0 | 12 (instr 45, src r6, full window) | 0.9824 / 0.9822; 1.31% of the site's reads on indices read 8 or more times | refused | hot set, +0.111%, 1.54x; hottest item 0.022% of all reads | +| seed 130488 (gap-newlow5.log, box 2, 23:08 BST) | 24d5e6511a1b83dc | 5 | 0.9898 | refused | CLEAN live: 1.0002x, X_0.1% +0.000% (a false refusal) | +| seed 19921 | fe35af21bc43a4cc | 4 | 0.9912 | refused | CLEAN live: 1.0005x (a false refusal) | +| seed 600738 | ef6d49f3871af472 | 8 | 0.9934 | refused | CLEAN live: 1.0002x (a false refusal) | +| seed 602822 | d8b5571c7abc80df | 3 | 0.9951 | at or above 0.995: passes | CLEAN live: 1.0003x (a true pass) | | Devnet 3 epoch 0, class v4 | fce15bf61030be57, 0 (day 20733) | 0 (instr 3, src r6) | 0.9992 / 0.9993 | AT OR ABOVE: a residual the floor misses | adv-cache-2: site 0 bit law P(bit 0..2 = 1) = 0.250 / 0.375 / 0.4375, chi2/dof 3.70 at 2^26 | | adv-cache-2 era-fixed-20 | 11f9f955b21d56c9, 1 | 11 (instr 46) | 0.9997 / 0.9997 | AT OR ABOVE: a residual | adv-cache-2: top 0.1% 1.26x its window control, one item read 512 times | | adv-cache-2 era-drawn-2 | 7ceb797d31eedb3e, 0 | 13 (instr 55) | 0.9992 / 0.9993 | AT OR ABOVE: a residual | adv-cache-2: chi2/dof 2.04, top 0.1% 1.27x | Reading. The floor at 0.995 refuses every program whose live hot set passes the f8 test X_f >= f (all -eight of this lane's, minimum sites 0.9821 to 0.9919), and does not reach the three milder concentrations (top 0.1 +eight of this lane's, minimum sites 0.9821 to 0.9919). It also refuses programs with NO live concentration: +of the five deepest 256-unit seeds, three (130488, 19921, 600738) read 0.9898 to 0.9934 at their minimum +site and are clean on the live set (1.0002x to 1.0005x), so a site can be measurably less than uniform at +2^20 on the stand-in without concentrating item reads live; these are the floor's false refusals, the +population it pays as its 2.4 percent clean rejection rate, and they cost the chain nothing but a redraw. +The floor does not reach the three milder concentrations (top 0.1 percent share 1.26x to 1.27x, chi2/dof 2 to 3.7) whose minimum site reads 0.9992 to 0.9997, inside the clean population's spread at 2^20 (clean median 0.9999, clean first percentile about 0.9993 over 29,032 programs). Devnet 3's own first program is one of those three. What those residuals are worth to a chip: a 1.26x share