Adaptive methods

Embedded pairs and step-size control, measured off-archive.

An adaptive method uses an embedded pair to estimate each step's error, and a step-size controller retries any step that misses the tolerance. A record holds one set of weights and a fixed step count, so the run searches pairs in a lane of its own: cycles set aside for this class that write an unscored archive. The lane reports the modeled cycles a method needs to reach an accuracy target.

At a glance

family
An embedded pair on the dyadic lattice (order 3 with an order-2 estimate) and a step-size controller
search
Deterministic enumeration, plus the side-track jobs
question
How many modeled cycles does a method need to reach an accuracy target? (cycles to tolerance)
problems
The validation problems
arithmetic
float64; the Q15 adaptive solver runs one fixed pair
checked by
Nothing: not order-verified and not scored
cost basis
Float64 attempt counts, each attempt priced by the Q15 cost model
sources
adaptive_archive/elites.json, the side-track ledger, benchmark/results.json
points measured
56
one per parameter point in the plan; 24 of them were measured again after the executor changed
jobs
4
each closes one open design question
highest rejection share
0.15748
largest over the adaptive.controller_gains points. Arithmetic: float64 only; no Q15 effects are included
side-track code hash
305c692a1b5c
the executor and prototypes, at the latest measurement

Work against accuracy across the tolerance sweep

Achieved error against rhs evaluations as the tolerance tightens, one panel per problem on shared axes, both log. Pair: Bogacki-Shampine 3(2).
1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorbattery_2rcbattery_2rc: tolerance 0.001, error 0.00343958 after 28 rhs evaluations, 0 rejected stepsbattery_2rc: tolerance 0.0001, error 0.000673436 after 37 rhs evaluations, 0 rejected stepsbattery_2rc: tolerance 1e-05, error 0.000100945 after 55 rhs evaluations, 0 rejected stepsbattery_2rc: tolerance 1e-06, error 1.22824e-05 after 97 rhs evaluations, 0 rejected stepsbattery_2rc: tolerance 1e-07, error 1.44501e-06 after 199 rhs evaluations, 1 rejected stepsbattery_2rc: tolerance 1e-08, error 1.52719e-07 after 421 rhs evaluations, 2 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorbicycle_lateralbicycle_lateral: tolerance 0.001, error 0.000357318 after 43 rhs evaluations, 0 rejected stepsbicycle_lateral: tolerance 0.0001, error 3.03138e-05 after 67 rhs evaluations, 0 rejected stepsbicycle_lateral: tolerance 1e-05, error 4.76471e-06 after 139 rhs evaluations, 2 rejected stepsbicycle_lateral: tolerance 1e-06, error 5.85663e-07 after 274 rhs evaluations, 2 rejected stepsbicycle_lateral: tolerance 1e-07, error 1.08566e-07 after 571 rhs evaluations, 3 rejected stepsbicycle_lateral: tolerance 1e-08, error 1.50442e-08 after 1195 rhs evaluations, 3 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorbuck_converterbuck_converter: tolerance 0.001, error 0.00151497 after 91 rhs evaluations, 0 rejected stepsbuck_converter: tolerance 0.0001, error 0.000447659 after 181 rhs evaluations, 1 rejected stepsbuck_converter: tolerance 1e-05, error 7.38994e-05 after 376 rhs evaluations, 2 rejected stepsbuck_converter: tolerance 1e-06, error 8.24192e-06 after 787 rhs evaluations, 3 rejected stepsbuck_converter: tolerance 1e-07, error 8.87062e-07 after 1666 rhs evaluations, 3 rejected stepsbuck_converter: tolerance 1e-08, error 9.10891e-08 after 3571 rhs evaluations, 4 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorenzyme_qssaenzyme_qssa: tolerance 0.0001, error 1.7963e-05 after 1162 rhs evaluations, 2 rejected stepsenzyme_qssa: tolerance 1e-05, error 4.17984e-07 after 1165 rhs evaluations, 2 rejected stepsenzyme_qssa: tolerance 0.001, error 0.000638856 after 1165 rhs evaluations, 3 rejected stepsenzyme_qssa: tolerance 1e-06, error 8.16411e-07 after 1645 rhs evaluations, 109 rejected stepsenzyme_qssa: tolerance 1e-07, error 1.67823e-08 after 1957 rhs evaluations, 8 rejected stepsenzyme_qssa: tolerance 1e-08, error 2.38565e-09 after 3301 rhs evaluations, 5 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorglucose_minimalglucose_minimal: tolerance 0.001, error 0.000325491 after 49 rhs evaluations, 0 rejected stepsglucose_minimal: tolerance 0.0001, error 4.03546e-05 after 88 rhs evaluations, 0 rejected stepsglucose_minimal: tolerance 1e-05, error 5.1956e-06 after 181 rhs evaluations, 2 rejected stepsglucose_minimal: tolerance 1e-06, error 5.34465e-07 after 373 rhs evaluations, 2 rejected stepsglucose_minimal: tolerance 1e-07, error 5.58226e-08 after 787 rhs evaluations, 3 rejected stepsglucose_minimal: tolerance 1e-08, error 5.6319e-09 after 1675 rhs evaluations, 3 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorpll_lockpll_lock: tolerance 0.001, error 0.000746124 after 52 rhs evaluations, 0 rejected stepspll_lock: tolerance 0.0001, error 0.000145968 after 85 rhs evaluations, 0 rejected stepspll_lock: tolerance 1e-05, error 2.60402e-05 after 175 rhs evaluations, 2 rejected stepspll_lock: tolerance 1e-06, error 3.16661e-06 after 358 rhs evaluations, 2 rejected stepspll_lock: tolerance 1e-07, error 3.52712e-07 after 757 rhs evaluations, 3 rejected stepspll_lock: tolerance 1e-08, error 3.67843e-08 after 1603 rhs evaluations, 3 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorrobertson_scaledrobertson_scaled: tolerance 0.001, error 0.000597569 after 2353 rhs evaluations, 2 rejected stepsrobertson_scaled: tolerance 0.0001, error 3.26777e-05 after 2380 rhs evaluations, 6 rejected stepsrobertson_scaled: tolerance 1e-05, error 8.16161e-06 after 2425 rhs evaluations, 7 rejected stepsrobertson_scaled: tolerance 1e-06, error 5.66517e-07 after 2536 rhs evaluations, 7 rejected stepsrobertson_scaled: tolerance 1e-07, error 8.94056e-08 after 2857 rhs evaluations, 9 rejected stepsrobertson_scaled: tolerance 1e-08, error 3.81611e-09 after 3883 rhs evaluations, 55 rejected steps1e11e21e31e41e-91e-71e-51e-3rhs evaluationsachieved errorservo_load_stepservo_load_step: tolerance 0.001, error 0.00187753 after 3004 rhs evaluations, 2 rejected stepsservo_load_step: tolerance 0.0001, error 0.000138677 after 3007 rhs evaluations, 3 rejected stepsservo_load_step: tolerance 1e-05, error 3.17866e-05 after 3016 rhs evaluations, 4 rejected stepsservo_load_step: tolerance 1e-06, error 2.84762e-06 after 3028 rhs evaluations, 4 rejected stepsservo_load_step: tolerance 1e-07, error 4.45106e-08 after 3064 rhs evaluations, 5 rejected stepsservo_load_step: tolerance 1e-08, error 6.00325e-09 after 3145 rhs evaluations, 7 rejected steps

Source: side-track job adaptive.suite_sweep, 8 points. Arithmetic: float64 only; no Q15 effects are included.

Every plotted point, in one table (48 rows)
problemtolerancerhs evaluationsachieved errorrejected steps
battery_2rc0.001280.003439580
battery_2rc0.0001370.0006734360
battery_2rc1e-05550.0001009450
battery_2rc1e-06971.22824e-050
battery_2rc1e-071991.44501e-061
battery_2rc1e-084211.52719e-072
bicycle_lateral0.001430.0003573180
bicycle_lateral0.0001673.03138e-050
bicycle_lateral1e-051394.76471e-062
bicycle_lateral1e-062745.85663e-072
bicycle_lateral1e-075711.08566e-073
bicycle_lateral1e-0811951.50442e-083
buck_converter0.001910.001514970
buck_converter0.00011810.0004476591
buck_converter1e-053767.38994e-052
buck_converter1e-067878.24192e-063
buck_converter1e-0716668.87062e-073
buck_converter1e-0835719.10891e-084
enzyme_qssa0.000111621.7963e-052
enzyme_qssa1e-0511654.17984e-072
enzyme_qssa0.00111650.0006388563
enzyme_qssa1e-0616458.16411e-07109
enzyme_qssa1e-0719571.67823e-088
enzyme_qssa1e-0833012.38565e-095
glucose_minimal0.001490.0003254910
glucose_minimal0.0001884.03546e-050
glucose_minimal1e-051815.1956e-062
glucose_minimal1e-063735.34465e-072
glucose_minimal1e-077875.58226e-083
glucose_minimal1e-0816755.6319e-093
pll_lock0.001520.0007461240
pll_lock0.0001850.0001459680
pll_lock1e-051752.60402e-052
pll_lock1e-063583.16661e-062
pll_lock1e-077573.52712e-073
pll_lock1e-0816033.67843e-083
robertson_scaled0.00123530.0005975692
robertson_scaled0.000123803.26777e-056
robertson_scaled1e-0524258.16161e-067
robertson_scaled1e-0625365.66517e-077
robertson_scaled1e-0728578.94056e-089
robertson_scaled1e-0838833.81611e-0955
servo_load_step0.00130040.001877532
servo_load_step0.000130070.0001386773
servo_load_step1e-0530163.17866e-054
servo_load_step1e-0630282.84762e-064
servo_load_step1e-0730644.45106e-085
servo_load_step1e-0831456.00325e-097

Controller gains and rejected steps

Share of attempted steps the controller rejected, by gains alpha (rows) and beta (columns). Every cell prints its share; on a log scale, the further a cell's color is from the page background, the larger the share. n/a means no run finished. Each point runs buck_converter, pll_lock, glucose_minimal at tolerance 1e-06.
betaalpha01/161/81/41/81/43/81/2alpha 1/8, beta 0: rejection share 0.02534; 13 rejected and 500 accepted steps, 1542 rhs evaluations, largest error 8.18309e-060.0253alpha 1/8, beta 1/16: rejection share 0.01793; 13 rejected and 712 accepted steps, 2178 rhs evaluations, largest error 2.7001e-060.0179alpha 1/8, beta 1/8: no run finished (status underflow)n/aalpha 1/8, beta 1/4: no run finished (status underflow)n/aalpha 1/4, beta 0: rejection share 0.01663; 7 rejected and 414 accepted steps, 1266 rhs evaluations, largest error 1.51356e-050.0166alpha 1/4, beta 1/16: rejection share 0.01562; 7 rejected and 441 accepted steps, 1347 rhs evaluations, largest error 1.23288e-050.0156alpha 1/4, beta 1/8: rejection share 0.01386; 7 rejected and 498 accepted steps, 1518 rhs evaluations, largest error 8.24192e-060.0139alpha 1/4, beta 1/4: no run finished (status underflow)n/aalpha 3/8, beta 0: rejection share 0.01018; 4 rejected and 389 accepted steps, 1182 rhs evaluations, largest error 1.87781e-050.0102alpha 3/8, beta 1/16: rejection share 0.00993; 4 rejected and 399 accepted steps, 1212 rhs evaluations, largest error 1.69917e-050.00993alpha 3/8, beta 1/8: rejection share 0.00957; 4 rejected and 414 accepted steps, 1257 rhs evaluations, largest error 1.51866e-050.00957alpha 3/8, beta 1/4: rejection share 0.00797; 4 rejected and 498 accepted steps, 1509 rhs evaluations, largest error 8.4292e-060.00797alpha 1/2, beta 0: rejection share 0.01309; 5 rejected and 377 accepted steps, 1149 rhs evaluations, largest error 2.1067e-050.0131alpha 1/2, beta 1/16: rejection share 0.01295; 5 rejected and 381 accepted steps, 1161 rhs evaluations, largest error 1.94951e-050.0129alpha 1/2, beta 1/8: rejection share 0.01269; 5 rejected and 389 accepted steps, 1185 rhs evaluations, largest error 1.82134e-050.0127alpha 1/2, beta 1/4: rejection share 0.15748; 80 rejected and 428 accepted steps, 1527 rhs evaluations, largest error 1.55843e-050.157

Source: side-track job adaptive.controller_gains, 16 points. Arithmetic: float64 only; no Q15 effects are included.

Every measured cell, in one table (16 rows)
alphabetarejected sharerejected stepsaccepted stepsrhs evaluations
1/800.02534135001542
1/81/160.01793137122178
1/81/8no run finishedn/an/an/a
1/81/4no run finishedn/an/an/a
1/400.0166374141266
1/41/160.0156274411347
1/41/80.0138674981518
1/41/4no run finishedn/an/an/a
3/800.0101843891182
3/81/160.0099343991212
3/81/80.0095744141257
3/81/40.0079744981509
1/200.0130953771149
1/21/160.0129553811161
1/21/80.0126953891185
1/21/40.15748804281527

What one Q15 attempt costs

Modeled Q15 cycles for one attempted step of bs32_q15, by part and problem size; the number at the end of each bar is the document's total. The pinned cost model prices no branch, so the error-magnitude bit scan is booked at its worst case, and each attempt can also run up to 47 conditional branches that no bar includes. Grade: upper_bound_unpriced_branches (upper bound within the model).
stage arithmeticerror estimatecontroller
01002003004002-state problems2-state problems: stage arithmetic, 78 cycles per attempt (30 rows)2-state problems: error estimate, 56 cycles per attempt (30 rows)2-state problems: controller, 100 cycles per attempt (30 rows)78561002343-state problems3-state problems: stage arithmetic, 117 cycles per attempt (10 rows)3-state problems: error estimate, 84 cycles per attempt (10 rows)3-state problems: controller, 100 cycles per attempt (10 rows)11784100301

Source: benchmark/results.json, adaptive_matched_tolerance: the 40 rows that carry a per-attempt price. Arithmetic: Q15.

Every plotted segment, in one table (6 rows)
problem sizepartcycles per attemptattempt totalrows
2-state problemsstage arithmetic7823430
2-state problemserror estimate5623430
2-state problemscontroller10023430
3-state problemsstage arithmetic11730110
3-state problemserror estimate8430110
3-state problemscontroller10030110
How this compares with SciPy RK23

SciPy RK23 runs the same Bogacki-Shampine 3(2) tableau our pair runs, so at rtol = atol = tol the two differ only in the step controller and the arithmetic: a median function-evaluation ratio of 1.27 (ours over RK23) over 39 cells and a median achieved-error ratio of 0.176 over the 33 cells where both sides recorded an error. A ratio above 1.0 means our side spent more or landed further from the reference. These ratios come from the matched-tolerance rows. Asking two solvers for the same tolerance does not put them at the same accuracy, so the matched-accuracy section below is the like-for-like comparison, and no wall-clock ratio is formed across the two implementations.

Embedded pairs on the dyadic lattice

Lattice matrices that meet each condition an order-3 pair with an order-2 estimate needs, by stage count and lattice depth s. Counts are exact; the axis is log, so 0 has no bar. Neighboring conditions with the same count share a bar.
1101001,00010,000100,0003 stages, s ≤ 1matrices, order-2 consistent3 stages, s ≤ 1: 64 matrices, order-2 consistent64order-3 consistent, both solvable03 stages, s ≤ 2matrices, order-2 consistent3 stages, s ≤ 2: 512 matrices, order-2 consistent512order-3 consistent, both solvable3 stages, s ≤ 2: 4 order-3 consistent, both solvable43 stages, s ≤ 3matrices, order-2 consistent3 stages, s ≤ 3: 4,096 matrices, order-2 consistent4,096order-3 consistent, both solvable3 stages, s ≤ 3: 6 order-3 consistent, both solvable63 stages, s ≤ 4matrices, order-2 consistent3 stages, s ≤ 4: 32,768 matrices, order-2 consistent32,768order-3 consistent, both solvable3 stages, s ≤ 4: 24 order-3 consistent, both solvable244 stages, s ≤ 1matrices, order-2 consistent4 stages, s ≤ 1: 4,096 matrices, order-2 consistent4,096order-3 consistent, both solvable4 stages, s ≤ 1: 3,646 order-3 consistent, both solvable3,646

Not counted (space over the census cap of 100,000 matrices): 4 stages, s ≤ 2; 4 stages, s ≤ 3; 4 stages, s ≤ 4.

Source: side-track job adaptive.pair_census, 8 points. Arithmetic: exact over Fractions; no float and no Q15 arithmetic is involved.

Every plotted count, in one table (20 rows)
stages and lattice depthconditionmatrices
3 stages, s ≤ 1matrices64
3 stages, s ≤ 1order-2 consistent64
3 stages, s ≤ 1order-3 consistent0
3 stages, s ≤ 1both solvable0
3 stages, s ≤ 2matrices512
3 stages, s ≤ 2order-2 consistent512
3 stages, s ≤ 2order-3 consistent4
3 stages, s ≤ 2both solvable4
3 stages, s ≤ 3matrices4,096
3 stages, s ≤ 3order-2 consistent4,096
3 stages, s ≤ 3order-3 consistent6
3 stages, s ≤ 3both solvable6
3 stages, s ≤ 4matrices32,768
3 stages, s ≤ 4order-2 consistent32,768
3 stages, s ≤ 4order-3 consistent24
3 stages, s ≤ 4both solvable24
4 stages, s ≤ 1matrices4,096
4 stages, s ≤ 1order-2 consistent4,096
4 stages, s ≤ 1order-3 consistent3,646
4 stages, s ≤ 1both solvable3,646

Measured against real counterparts, at matched accuracy

Our solvers reached 47 of their 64 targets and the library solvers 64 of their 64; a target is one solver on one problem at one accuracy level. Where our runs carry a cycle count, a reached target costs a median of 3978 analytic cycles; the table gives each row's cost grade.

Achieved error against rhs evaluations for each problem, both axes log (a work-precision view). A mark is the cheapest run that met a target; lower left is better. Blue is ours, orange the library, and the shape names the solver.
bs32_float (ours, float64)bs32_q15 (ours, q15)RK23 (library, compiled_float64)RK45 (library, compiled_float64)
1e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorbattery_2rcbattery_2rc, bs32_float (ours): error 0.00408717 after 25 rhs evaluations; meets target 0.015625battery_2rc, bs32_float (ours): error 0.00325413 after 28 rhs evaluations; meets target 0.00390625battery_2rc, bs32_float (ours): error 0.00047517 after 40 rhs evaluations; meets target 0.000976562battery_2rc, bs32_float (ours): error 0.000157722 after 49 rhs evaluations; meets target 0.000244141battery_2rc, bs32_q15 (ours): error 0.00140763 after 31 rhs evaluations; meets target 0.015625, 0.00390625battery_2rc, RK23 (library): error 0.0137779 after 29 rhs evaluations; meets target 0.015625battery_2rc, RK23 (library): error 0.00241547 after 32 rhs evaluations; meets target 0.00390625battery_2rc, RK23 (library): error 0.000244299 after 50 rhs evaluations; meets target 0.000976562battery_2rc, RK23 (library): error 3.23714e-05 after 83 rhs evaluations; meets target 0.000244141battery_2rc, RK45 (library): error 0.00137537 after 44 rhs evaluations; meets target 0.015625, 0.00390625battery_2rc, RK45 (library): error 0.000102248 after 50 rhs evaluations; meets target 0.000976562, 0.0002441411e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorbicycle_lateralbicycle_lateral, bs32_float (ours): error 0.0150249 after 31 rhs evaluations; meets target 0.015625bicycle_lateral, bs32_float (ours): error 0.00344055 after 37 rhs evaluations; meets target 0.00390625bicycle_lateral, bs32_float (ours): error 0.000358791 after 43 rhs evaluations; meets target 0.000976562bicycle_lateral, bs32_float (ours): error 0.000141989 after 49 rhs evaluations; meets target 0.000244141bicycle_lateral, bs32_q15 (ours): error 0.000138421 after 52 rhs evaluations; meets target 0.000244141bicycle_lateral, bs32_q15 (ours): error 0.000696358 after 52 rhs evaluations; meets target 0.015625, 0.00390625, 0.000976562bicycle_lateral, RK23 (library): error 0.00208275 after 56 rhs evaluations; meets target 0.015625, 0.00390625bicycle_lateral, RK23 (library): error 9.35999e-05 after 71 rhs evaluations; meets target 0.000976562, 0.000244141bicycle_lateral, RK45 (library): error 0.00174484 after 62 rhs evaluations; meets target 0.015625, 0.00390625bicycle_lateral, RK45 (library): error 0.000468714 after 74 rhs evaluations; meets target 0.000976562bicycle_lateral, RK45 (library): error 3.68415e-05 after 92 rhs evaluations; meets target 0.0002441411e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorbuck_converterbuck_converter, bs32_float (ours): error 0.00404258 after 55 rhs evaluations; meets target 0.015625buck_converter, bs32_float (ours): error 0.000804876 after 64 rhs evaluations; meets target 0.00390625, 0.000976562buck_converter, bs32_q15 (ours): error 0.00185816 after 82 rhs evaluations; meets target 0.015625, 0.00390625buck_converter, bs32_q15 (ours): error 0.000827936 after 118 rhs evaluations; meets target 0.000976562buck_converter, RK23 (library): error 0.0018877 after 65 rhs evaluations; meets target 0.015625, 0.00390625buck_converter, RK23 (library): error 0.000676007 after 173 rhs evaluations; meets target 0.000976562buck_converter, RK23 (library): error 0.00012305 after 320 rhs evaluations; meets target 0.000244141buck_converter, RK45 (library): error 0.00110774 after 122 rhs evaluations; meets target 0.015625, 0.00390625buck_converter, RK45 (library): error 0.000128402 after 158 rhs evaluations; meets target 0.000976562, 0.0002441411e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorenzyme_qssa (stiff)enzyme_qssa, bs32_float (ours): error 0.00327363 after 1156 rhs evaluations; meets target 0.00390625enzyme_qssa, bs32_float (ours): error 0.0137765 after 1156 rhs evaluations; meets target 0.015625enzyme_qssa, bs32_float (ours): error 0.000457048 after 1162 rhs evaluations; meets target 0.000976562enzyme_qssa, bs32_float (ours): error 0.000151936 after 1165 rhs evaluations; meets target 0.000244141enzyme_qssa, bs32_q15 (ours): error 0.0126544 after 1249 rhs evaluations; meets target 0.015625enzyme_qssa, RK23 (library): error 0.00150861 after 1163 rhs evaluations; meets target 0.015625, 0.00390625enzyme_qssa, RK23 (library): error 4.22153e-06 after 1178 rhs evaluations; meets target 0.000976562, 0.000244141enzyme_qssa, RK45 (library): error 0.000128097 after 1928 rhs evaluations; meets target 0.000976562, 0.000244141enzyme_qssa, RK45 (library): error 0.00329778 after 1958 rhs evaluations; meets target 0.015625, 0.003906251e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorglucose_minimalglucose_minimal, bs32_float (ours): error 0.0100669 after 25 rhs evaluations; meets target 0.015625glucose_minimal, bs32_float (ours): error 0.00294387 after 28 rhs evaluations; meets target 0.00390625glucose_minimal, bs32_float (ours): error 0.000790258 after 40 rhs evaluations; meets target 0.000976562glucose_minimal, bs32_float (ours): error 0.000168746 after 58 rhs evaluations; meets target 0.000244141glucose_minimal, bs32_q15 (ours): error 0.000870334 after 46 rhs evaluations; meets target 0.015625, 0.00390625, 0.000976562glucose_minimal, RK23 (library): error 0.00249821 after 29 rhs evaluations; meets target 0.015625, 0.00390625glucose_minimal, RK23 (library): error 0.000695673 after 47 rhs evaluations; meets target 0.000976562glucose_minimal, RK23 (library): error 9.94644e-05 after 80 rhs evaluations; meets target 0.000244141glucose_minimal, RK45 (library): error 0.000515792 after 32 rhs evaluations; meets target 0.015625, 0.00390625, 0.000976562glucose_minimal, RK45 (library): error 5.90612e-05 after 44 rhs evaluations; meets target 0.0002441411e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorpll_lockpll_lock, bs32_float (ours): error 0.00587411 after 34 rhs evaluations; meets target 0.015625pll_lock, bs32_float (ours): error 0.00324962 after 37 rhs evaluations; meets target 0.00390625pll_lock, bs32_float (ours): error 0.000704462 after 52 rhs evaluations; meets target 0.000976562pll_lock, bs32_float (ours): error 0.000172239 after 82 rhs evaluations; meets target 0.000244141pll_lock, bs32_q15 (ours): error 0.00146873 after 40 rhs evaluations; meets target 0.015625, 0.00390625pll_lock, RK23 (library): error 0.0108605 after 41 rhs evaluations; meets target 0.015625pll_lock, RK23 (library): error 0.00104873 after 56 rhs evaluations; meets target 0.00390625pll_lock, RK23 (library): error 0.00026392 after 86 rhs evaluations; meets target 0.000976562pll_lock, RK23 (library): error 5.51968e-05 after 152 rhs evaluations; meets target 0.000244141pll_lock, RK45 (library): error 0.00103685 after 62 rhs evaluations; meets target 0.015625, 0.00390625pll_lock, RK45 (library): error 0.000699328 after 74 rhs evaluations; meets target 0.000976562pll_lock, RK45 (library): error 6.36869e-05 after 86 rhs evaluations; meets target 0.0002441411e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorrobertson_scaled (stiff)robertson_scaled, bs32_float (ours): error 0.000620332 after 2353 rhs evaluations; meets target 0.000976562robertson_scaled, bs32_float (ours): error 0.00265531 after 2356 rhs evaluations; meets target 0.00390625robertson_scaled, bs32_float (ours): error 0.00906587 after 2362 rhs evaluations; meets target 0.015625robertson_scaled, bs32_float (ours): error 6.24164e-05 after 2374 rhs evaluations; meets target 0.000244141robertson_scaled, RK23 (library): error 0.000785984 after 2372 rhs evaluations; meets target 0.015625, 0.00390625, 0.000976562robertson_scaled, RK23 (library): error 0.000137615 after 2402 rhs evaluations; meets target 0.000244141robertson_scaled, RK45 (library): error 0.000154804 after 4826 rhs evaluations; meets target 0.015625, 0.00390625, 0.000976562, 0.0002441411e11e21e31e41e-61e-51e-41e-31e-21e-1rhs evaluationsachieved errorservo_load_step (stiff)servo_load_step, bs32_float (ours): error 0.00045421 after 3004 rhs evaluations; meets target 0.000976562servo_load_step, bs32_float (ours): error 0.00348294 after 3004 rhs evaluations; meets target 0.00390625servo_load_step, bs32_float (ours): error 0.0070112 after 3004 rhs evaluations; meets target 0.015625servo_load_step, bs32_float (ours): error 9.32705e-05 after 3007 rhs evaluations; meets target 0.000244141servo_load_step, bs32_q15 (ours): error 0.0154087 after 3130 rhs evaluations; meets target 0.015625servo_load_step, RK23 (library): error 0.00105884 after 3023 rhs evaluations; meets target 0.015625, 0.00390625servo_load_step, RK23 (library): error 0.000126277 after 3035 rhs evaluations; meets target 0.000976562, 0.000244141servo_load_step, RK45 (library): error 0.00607231 after 5204 rhs evaluations; meets target 0.015625servo_load_step, RK45 (library): error 0.000200382 after 5234 rhs evaluations; meets target 0.00390625, 0.000976562, 0.000244141

Source: benchmark/results.json, matched_accuracy (adaptive rows); arithmetic per solver in the legend.

Every plotted mark, in one table (78 rows)
problemsolversiderhs evaluationsachieved errortargets met
battery_2rcbs32_floatours250.004087170.015625
battery_2rcbs32_floatours280.003254130.00390625
battery_2rcbs32_floatours400.000475170.000976562
battery_2rcbs32_floatours490.0001577220.000244141
battery_2rcbs32_q15ours310.001407630.015625, 0.00390625
battery_2rcRK23library290.01377790.015625
battery_2rcRK23library320.002415470.00390625
battery_2rcRK23library500.0002442990.000976562
battery_2rcRK23library833.23714e-050.000244141
battery_2rcRK45library440.001375370.015625, 0.00390625
battery_2rcRK45library500.0001022480.000976562, 0.000244141
bicycle_lateralbs32_floatours310.01502490.015625
bicycle_lateralbs32_floatours370.003440550.00390625
bicycle_lateralbs32_floatours430.0003587910.000976562
bicycle_lateralbs32_floatours490.0001419890.000244141
bicycle_lateralbs32_q15ours520.0001384210.000244141
bicycle_lateralbs32_q15ours520.0006963580.015625, 0.00390625, 0.000976562
bicycle_lateralRK23library560.002082750.015625, 0.00390625
bicycle_lateralRK23library719.35999e-050.000976562, 0.000244141
bicycle_lateralRK45library620.001744840.015625, 0.00390625
bicycle_lateralRK45library740.0004687140.000976562
bicycle_lateralRK45library923.68415e-050.000244141
buck_converterbs32_floatours550.004042580.015625
buck_converterbs32_floatours640.0008048760.00390625, 0.000976562
buck_converterbs32_q15ours820.001858160.015625, 0.00390625
buck_converterbs32_q15ours1180.0008279360.000976562
buck_converterRK23library650.00188770.015625, 0.00390625
buck_converterRK23library1730.0006760070.000976562
buck_converterRK23library3200.000123050.000244141
buck_converterRK45library1220.001107740.015625, 0.00390625
buck_converterRK45library1580.0001284020.000976562, 0.000244141
enzyme_qssabs32_floatours11560.003273630.00390625
enzyme_qssabs32_floatours11560.01377650.015625
enzyme_qssabs32_floatours11620.0004570480.000976562
enzyme_qssabs32_floatours11650.0001519360.000244141
enzyme_qssabs32_q15ours12490.01265440.015625
enzyme_qssaRK23library11630.001508610.015625, 0.00390625
enzyme_qssaRK23library11784.22153e-060.000976562, 0.000244141
enzyme_qssaRK45library19280.0001280970.000976562, 0.000244141
enzyme_qssaRK45library19580.003297780.015625, 0.00390625
glucose_minimalbs32_floatours250.01006690.015625
glucose_minimalbs32_floatours280.002943870.00390625
glucose_minimalbs32_floatours400.0007902580.000976562
glucose_minimalbs32_floatours580.0001687460.000244141
glucose_minimalbs32_q15ours460.0008703340.015625, 0.00390625, 0.000976562
glucose_minimalRK23library290.002498210.015625, 0.00390625
glucose_minimalRK23library470.0006956730.000976562
glucose_minimalRK23library809.94644e-050.000244141
glucose_minimalRK45library320.0005157920.015625, 0.00390625, 0.000976562
glucose_minimalRK45library445.90612e-050.000244141
pll_lockbs32_floatours340.005874110.015625
pll_lockbs32_floatours370.003249620.00390625
pll_lockbs32_floatours520.0007044620.000976562
pll_lockbs32_floatours820.0001722390.000244141
pll_lockbs32_q15ours400.001468730.015625, 0.00390625
pll_lockRK23library410.01086050.015625
pll_lockRK23library560.001048730.00390625
pll_lockRK23library860.000263920.000976562
pll_lockRK23library1525.51968e-050.000244141
pll_lockRK45library620.001036850.015625, 0.00390625
pll_lockRK45library740.0006993280.000976562
pll_lockRK45library866.36869e-050.000244141
robertson_scaledbs32_floatours23530.0006203320.000976562
robertson_scaledbs32_floatours23560.002655310.00390625
robertson_scaledbs32_floatours23620.009065870.015625
robertson_scaledbs32_floatours23746.24164e-050.000244141
robertson_scaledRK23library23720.0007859840.015625, 0.00390625, 0.000976562
robertson_scaledRK23library24020.0001376150.000244141
robertson_scaledRK45library48260.0001548040.015625, 0.00390625, 0.000976562, 0.000244141
servo_load_stepbs32_floatours30040.000454210.000976562
servo_load_stepbs32_floatours30040.003482940.00390625
servo_load_stepbs32_floatours30040.00701120.015625
servo_load_stepbs32_floatours30079.32705e-050.000244141
servo_load_stepbs32_q15ours31300.01540870.015625
servo_load_stepRK23library30230.001058840.015625, 0.00390625
servo_load_stepRK23library30350.0001262770.000976562, 0.000244141
servo_load_stepRK45library52040.006072310.015625
servo_load_stepRK45library52340.0002003820.00390625, 0.000976562, 0.000244141

Ours on these rows means run by this harness through its own solver path, which includes the classical rk4 control, so it does not mean discovered.

Library integrators choose their own step counts, so their work is context and never a same-work comparison with any fixed-step run on this site.

Every row, with its cost grade (128 rows)

Each row fixes a target for the achieved final-state error and reports what the solver spent to reach it. A row that did not reach its target stays in the table, and its status column says so.

problemsidesolverarithmetictarget errorachievedstepsrhs evaluationsanalytic cyclescost gradestatus
battery_2rclibraryRK23compiled_float640.0002441413.23714e-052783n/anonereached
battery_2rclibraryRK23compiled_float640.0009765620.0002442991650n/anonereached
battery_2rclibraryRK23compiled_float640.003906250.002415471032n/anonereached
battery_2rclibraryRK23compiled_float640.0156250.0137779829n/anonereached
battery_2rclibraryRK45compiled_float640.0002441410.000102248850n/anonereached
battery_2rclibraryRK45compiled_float640.0009765620.000102248850n/anonereached
battery_2rclibraryRK45compiled_float640.003906250.00137537744n/anonereached
battery_2rclibraryRK45compiled_float640.0156250.00137537744n/anonereached
battery_2rcoursbs32_floatfloat640.0002441410.0001577221649n/anonereached
battery_2rcoursbs32_floatfloat640.0009765620.000475171340n/anonereached
battery_2rcoursbs32_floatfloat640.003906250.00325413928n/anonereached
battery_2rcoursbs32_floatfloat640.0156250.00408717825n/anonereached
battery_2rcoursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
battery_2rcoursbs32_q15q150.000976562n/an/an/an/aupper_bound_unpriced_branchesnever_reached
battery_2rcoursbs32_q15q150.003906250.0014076310313010upper_bound_unpriced_branchesreached
battery_2rcoursbs32_q15q150.0156250.0014076310313010upper_bound_unpriced_branchesreached
bicycle_laterallibraryRK23compiled_float640.0002441419.35999e-052371n/anonereached
bicycle_laterallibraryRK23compiled_float640.0009765629.35999e-052371n/anonereached
bicycle_laterallibraryRK23compiled_float640.003906250.002082751656n/anonereached
bicycle_laterallibraryRK23compiled_float640.0156250.002082751656n/anonereached
bicycle_laterallibraryRK45compiled_float640.0002441413.68415e-051592n/anonereached
bicycle_laterallibraryRK45compiled_float640.0009765620.0004687141274n/anonereached
bicycle_laterallibraryRK45compiled_float640.003906250.001744841062n/anonereached
bicycle_laterallibraryRK45compiled_float640.0156250.001744841062n/anonereached
bicycle_lateraloursbs32_floatfloat640.0002441410.0001419891649n/anonereached
bicycle_lateraloursbs32_floatfloat640.0009765620.0003587911443n/anonereached
bicycle_lateraloursbs32_floatfloat640.003906250.003440551237n/anonereached
bicycle_lateraloursbs32_floatfloat640.0156250.01502491031n/anonereached
bicycle_lateraloursbs32_q15q150.0002441410.00013842117523978upper_bound_unpriced_branchesreached
bicycle_lateraloursbs32_q15q150.0009765620.00069635817523978upper_bound_unpriced_branchesreached
bicycle_lateraloursbs32_q15q150.003906250.00069635817523978upper_bound_unpriced_branchesreached
bicycle_lateraloursbs32_q15q150.0156250.00069635817523978upper_bound_unpriced_branchesreached
buck_converterlibraryRK23compiled_float640.0002441410.00012305106320n/anonereached
buck_converterlibraryRK23compiled_float640.0009765620.00067600757173n/anonereached
buck_converterlibraryRK23compiled_float640.003906250.00188772065n/anonereached
buck_converterlibraryRK23compiled_float640.0156250.00188772065n/anonereached
buck_converterlibraryRK45compiled_float640.0002441410.00012840226158n/anonereached
buck_converterlibraryRK45compiled_float640.0009765620.00012840226158n/anonereached
buck_converterlibraryRK45compiled_float640.003906250.0011077420122n/anonereached
buck_converterlibraryRK45compiled_float640.0156250.0011077420122n/anonereached
buck_converteroursbs32_floatfloat640.000244141n/an/an/an/anonenever_reached
buck_converteroursbs32_floatfloat640.0009765620.0008048762164n/anonereached
buck_converteroursbs32_floatfloat640.003906250.0008048762164n/anonereached
buck_converteroursbs32_floatfloat640.0156250.004042581855n/anonereached
buck_converteroursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
buck_converteroursbs32_q15q150.0009765620.000827936391189126upper_bound_unpriced_branchesreached
buck_converteroursbs32_q15q150.003906250.0018581627826318upper_bound_unpriced_branchesreached
buck_converteroursbs32_q15q150.0156250.0018581627826318upper_bound_unpriced_branchesreached
enzyme_qssalibraryRK23compiled_float640.0002441414.22153e-063841178n/anonereached
enzyme_qssalibraryRK23compiled_float640.0009765624.22153e-063841178n/anonereached
enzyme_qssalibraryRK23compiled_float640.003906250.001508613841163n/anonereached
enzyme_qssalibraryRK23compiled_float640.0156250.001508613841163n/anonereached
enzyme_qssalibraryRK45compiled_float640.0002441410.0001280972921928n/anonereached
enzyme_qssalibraryRK45compiled_float640.0009765620.0001280972921928n/anonereached
enzyme_qssalibraryRK45compiled_float640.003906250.003297782921958n/anonereached
enzyme_qssalibraryRK45compiled_float640.0156250.003297782921958n/anonereached
enzyme_qssaoursbs32_floatfloat640.0002441410.0001519363851165n/anonereached
enzyme_qssaoursbs32_floatfloat640.0009765620.0004570483841162n/anonereached
enzyme_qssaoursbs32_floatfloat640.003906250.003273633831156n/anonereached
enzyme_qssaoursbs32_floatfloat640.0156250.01377653821156n/anonereached
enzyme_qssaoursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
enzyme_qssaoursbs32_q15q150.000976562n/an/an/an/aupper_bound_unpriced_branchesnever_reached
enzyme_qssaoursbs32_q15q150.00390625n/an/an/an/aupper_bound_unpriced_branchesnever_reached
enzyme_qssaoursbs32_q15q150.0156250.01265443591249102258upper_bound_unpriced_branchesreached
glucose_minimallibraryRK23compiled_float640.0002441419.94644e-052680n/anonereached
glucose_minimallibraryRK23compiled_float640.0009765620.0006956731547n/anonereached
glucose_minimallibraryRK23compiled_float640.003906250.00249821929n/anonereached
glucose_minimallibraryRK23compiled_float640.0156250.00249821929n/anonereached
glucose_minimallibraryRK45compiled_float640.0002441415.90612e-05744n/anonereached
glucose_minimallibraryRK45compiled_float640.0009765620.000515792532n/anonereached
glucose_minimallibraryRK45compiled_float640.003906250.000515792532n/anonereached
glucose_minimallibraryRK45compiled_float640.0156250.000515792532n/anonereached
glucose_minimaloursbs32_floatfloat640.0002441410.0001687461958n/anonereached
glucose_minimaloursbs32_floatfloat640.0009765620.0007902581340n/anonereached
glucose_minimaloursbs32_floatfloat640.003906250.00294387928n/anonereached
glucose_minimaloursbs32_floatfloat640.0156250.0100669825n/anonereached
glucose_minimaloursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
glucose_minimaloursbs32_q15q150.0009765620.00087033415463510upper_bound_unpriced_branchesreached
glucose_minimaloursbs32_q15q150.003906250.00087033415463510upper_bound_unpriced_branchesreached
glucose_minimaloursbs32_q15q150.0156250.00087033415463510upper_bound_unpriced_branchesreached
pll_locklibraryRK23compiled_float640.0002441415.51968e-0550152n/anonereached
pll_locklibraryRK23compiled_float640.0009765620.000263922886n/anonereached
pll_locklibraryRK23compiled_float640.003906250.001048731756n/anonereached
pll_locklibraryRK23compiled_float640.0156250.01086051241n/anonereached
pll_locklibraryRK45compiled_float640.0002441416.36869e-051486n/anonereached
pll_locklibraryRK45compiled_float640.0009765620.0006993281274n/anonereached
pll_locklibraryRK45compiled_float640.003906250.001036851062n/anonereached
pll_locklibraryRK45compiled_float640.0156250.001036851062n/anonereached
pll_lockoursbs32_floatfloat640.0002441410.0001722392782n/anonereached
pll_lockoursbs32_floatfloat640.0009765620.0007044621752n/anonereached
pll_lockoursbs32_floatfloat640.003906250.003249621237n/anonereached
pll_lockoursbs32_floatfloat640.0156250.005874111134n/anonereached
pll_lockoursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
pll_lockoursbs32_q15q150.000976562n/an/an/an/aupper_bound_unpriced_branchesnever_reached
pll_lockoursbs32_q15q150.003906250.0014687313403042upper_bound_unpriced_branchesreached
pll_lockoursbs32_q15q150.0156250.0014687313403042upper_bound_unpriced_branchesreached
robertson_scaledlibraryRK23compiled_float640.0002441410.0001376157872402n/anonereached
robertson_scaledlibraryRK23compiled_float640.0009765620.0007859847852372n/anonereached
robertson_scaledlibraryRK23compiled_float640.003906250.0007859847852372n/anonereached
robertson_scaledlibraryRK23compiled_float640.0156250.0007859847852372n/anonereached
robertson_scaledlibraryRK45compiled_float640.0002441410.0001548046034826n/anonereached
robertson_scaledlibraryRK45compiled_float640.0009765620.0001548046034826n/anonereached
robertson_scaledlibraryRK45compiled_float640.003906250.0001548046034826n/anonereached
robertson_scaledlibraryRK45compiled_float640.0156250.0001548046034826n/anonereached
robertson_scaledoursbs32_floatfloat640.0002441416.24164e-057862374n/anonereached
robertson_scaledoursbs32_floatfloat640.0009765620.0006203327822353n/anonereached
robertson_scaledoursbs32_floatfloat640.003906250.002655317832356n/anonereached
robertson_scaledoursbs32_floatfloat640.0156250.009065877852362n/anonereached
robertson_scaledoursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
robertson_scaledoursbs32_q15q150.000976562n/an/an/an/aupper_bound_unpriced_branchesnever_reached
robertson_scaledoursbs32_q15q150.00390625n/an/an/an/aupper_bound_unpriced_branchesnever_reached
robertson_scaledoursbs32_q15q150.015625n/an/an/an/aupper_bound_unpriced_branchesnever_reached
servo_load_steplibraryRK23compiled_float640.0002441410.00012627710013035n/anonereached
servo_load_steplibraryRK23compiled_float640.0009765620.00012627710013035n/anonereached
servo_load_steplibraryRK23compiled_float640.003906250.0010588410013023n/anonereached
servo_load_steplibraryRK23compiled_float640.0156250.0010588410013023n/anonereached
servo_load_steplibraryRK45compiled_float640.0002441410.0002003827625234n/anonereached
servo_load_steplibraryRK45compiled_float640.0009765620.0002003827625234n/anonereached
servo_load_steplibraryRK45compiled_float640.003906250.0002003827625234n/anonereached
servo_load_steplibraryRK45compiled_float640.0156250.006072317615204n/anonereached
servo_load_stepoursbs32_floatfloat640.0002441419.32705e-059993007n/anonereached
servo_load_stepoursbs32_floatfloat640.0009765620.000454219993004n/anonereached
servo_load_stepoursbs32_floatfloat640.003906250.003482949983004n/anonereached
servo_load_stepoursbs32_floatfloat640.0156250.00701129983004n/anonereached
servo_load_stepoursbs32_q15q150.000244141n/an/an/an/aupper_bound_unpriced_branchesnever_reached
servo_load_stepoursbs32_q15q150.000976562n/an/an/an/aupper_bound_unpriced_branchesnever_reached
servo_load_stepoursbs32_q15q150.00390625n/an/an/an/aupper_bound_unpriced_branchesnever_reached
servo_load_stepoursbs32_q15q150.0156250.01540879793130244062upper_bound_unpriced_branchesreached

Three grades of cycle number sit in this document and each row says which it is: the pinned model cross-checked against the reference listing for Q15 explicit rows, a design estimate resting on a software-reciprocal guess for the implicit rows, and an upper bound with unpriced branches for the Q15 adaptive rows. Library rows carry no cycle number at all, because SciPy has no cost model and inventing one for it would be inventing the comparison.

Lane elites, ranked by cycles to tolerance

The archive fixes a cycle budget and measures the error a method reaches. The lane fixes an accuracy target on the axis-T ladder and measures the modeled cycles a candidate needs to reach it, so neither number converts into the other. Every run is float64. The pinned order-condition checker never saw these coefficients, so none is order-verified.

candidates measured
3,519
under the lane search code hash below
candidates ranked
147
the subset this document put in order; the leading 32 are listed below, the document's cap
source
rk-work/adaptive_archive/elites.json
elite target
0.0009765625 final-state error
axis-T ladder
targets of 1/64, 1/256, 1/1024 and 1/4096 final-state error (benchmark/results.json)
lane search code hash
0c67fece6592f9ad
document written
cycle 3928, 2026-09-10 09:51 CT
The 32 entries plotted here, each at its median modeled cycles at the elite target against its best achieved error (log axis); left is cheaper and lower is more accurate. Every plotted entry uses controller gains alpha 1/4, beta 1/8 and safety 7/8.
360036503700375038001e-61e-51e-4median cycles at the elite targetbest achieved errorrank 1: median 3623.5 cycles at the elite target, best achieved error 2.04078e-05rank 5: median 3658 cycles at the elite target, best achieved error 1.02219e-05rank 2: median 3658 cycles at the elite target, best achieved error 1.39166e-05rank 6, 10: median 3658 cycles at the elite target, best achieved error 1.67009e-05 (2 entries)rank 12: median 3658 cycles at the elite target, best achieved error 2.04078e-05rank 9: median 3658 cycles at the elite target, best achieved error 2.05342e-05rank 11: median 3658 cycles at the elite target, best achieved error 2.61467e-05rank 4, 7: median 3658 cycles at the elite target, best achieved error 3.71639e-06 (2 entries)rank 3: median 3658 cycles at the elite target, best achieved error 5.25029e-05rank 8: median 3658 cycles at the elite target, best achieved error 5.73765e-06rank 23: median 3692.5 cycles at the elite target, best achieved error 1.02219e-05rank 20: median 3692.5 cycles at the elite target, best achieved error 2.04078e-05rank 13, 15, 16, 22: median 3692.5 cycles at the elite target, best achieved error 2.05342e-05 (4 entries)rank 14, 18, 27: median 3692.5 cycles at the elite target, best achieved error 2.29076e-05 (3 entries)rank 21: median 3692.5 cycles at the elite target, best achieved error 2.30341e-05rank 17, 19: median 3692.5 cycles at the elite target, best achieved error 3.71639e-06 (2 entries)rank 24: median 3692.5 cycles at the elite target, best achieved error 4.36623e-05rank 25: median 3692.5 cycles at the elite target, best achieved error 5.19462e-05rank 26: median 3692.5 cycles at the elite target, best achieved error 5.32916e-06rank 29: median 3727 cycles at the elite target, best achieved error 1.14176e-05rank 28: median 3727 cycles at the elite target, best achieved error 2.05342e-05rank 30, 32: median 3761.5 cycles at the elite target, best achieved error 2.05342e-05 (2 entries)rank 31: median 3761.5 cycles at the elite target, best achieved error 2.61467e-05

Source: rk-work/adaptive_archive/elites.json. float64 runs, cycles priced on float_trajectory_q15_attempt_cost; not order-verified.

Every plotted mark, in one table (23 rows)
rankmedian cycles at the elite targetbest achieved errorentries
13623.52.04078e-051
536581.02219e-051
236581.39166e-051
6, 1036581.67009e-052
1236582.04078e-051
936582.05342e-051
1136582.61467e-051
4, 736583.71639e-062
336585.25029e-051
836585.73765e-061
233692.51.02219e-051
203692.52.04078e-051
13, 15, 16, 223692.52.05342e-054
14, 18, 273692.52.29076e-053
213692.52.30341e-051
17, 193692.53.71639e-062
243692.54.36623e-051
253692.55.19462e-051
263692.55.32916e-061
2937271.14176e-051
2837272.05342e-051
30, 323761.52.05342e-052
313761.52.61467e-051
The 32 listed entries, the ranking rule and how the lane prices a cycle

Ranking rule, quoted from the document: lowest median cycles at the elite target across the problems where the status is reached; a candidate that reached the elite target nowhere ranks below every candidate that reached it somewhere; ties broken by the count of targets reached, highest leading, then by record_hash. No clock is read and no value is sampled, so the order is total and two runs over the same records agree.

Rank is the position the document gives under the rule above. Each row shows the candidate's shape, the median modeled cycles it needed at the elite target, how many problems it reached that target on, and the cost basis its cycles were priced on.

rankcandidateorderembedded orderstagessafetyalphabetamedian cycles at targetproblems at targettargets reachedcost basis
1b11a9878e9b49d4a3247/81/41/83623.58 of 832 of 32float_trajectory_q15_attempt_cost
21adcafbcd48261773247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
3232db335f2d2ccce3247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
47be462fcf9627d5f3247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
59ea7e8b6d3aaafbc3247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
67b40405e658627343247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
76a4036c7af53fb773247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
8f93332ea5461fde03247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
9b1aa4d85f8abc57b3247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
10b7082fb5196873503247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
115d99a0a2d0ba50953247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
125b76d6663b758c933247/81/41/836588 of 832 of 32float_trajectory_q15_attempt_cost
1337d56db1bb12d79f3247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
14f31a49507df5537d3247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
153655b7e0f20b05593247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
1661d2b841e34b57603247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
1766f7ac19aab94da23247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
186db7bc20e505f2833247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
19464e660ea2e569a23247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
202b6ba4affd5a69493247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
21f4fa8dfd89794ecf3247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
227180e5b05a7824653247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
23e0e6443513fbf8703247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
24a8247015672fda063247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
25cf6c8c02d003e7b73247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
263ae7129c2e6290833247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
2733e21b90a917c9513247/81/41/83692.58 of 832 of 32float_trajectory_q15_attempt_cost
287924019c170895b33247/81/41/837278 of 832 of 32float_trajectory_q15_attempt_cost
29b1ced1c87e1a2d083247/81/41/837278 of 832 of 32float_trajectory_q15_attempt_cost
3006556ac45308781c3247/81/41/83761.58 of 832 of 32float_trajectory_q15_attempt_cost
313bf6d70fcd0b554a3247/81/41/83761.58 of 832 of 32float_trajectory_q15_attempt_cost
322a630167053c66ad3247/81/41/83761.58 of 832 of 32float_trajectory_q15_attempt_cost

How this lane prices a cycle

float_trajectory_q15_attempt_cost

direction
upper bound within the model, over a trajectory the model did not produce
excludes
conditional branches, which costmodel prices for no class, so the bit scan is booked at its worst case; derivative evaluations, so FSAL is worth zero cycles here; and the Q31 time register terms
grade
an upper bound in the Q15 cost model over an attempt count measured in float64, which is a mixed number and is labelled as one
includes
the stage work of the propagating formula through the pinned cycle_count, the error estimate at one priced block per nonzero d term scaled by the state count, and one controller update, charged on every attempt whether accepted or rejected
source
costmodel.cycle_count plus costmodel.count_sequence over adaptive_q15.reference_sections, with adaptive.solve_adaptive supplying the attempt count
why_mixed
adaptive_q15.solve_adaptive_q15 is hard-wired to the BS32 pair through A_REPS, B_REPS and D_REPS, so a candidate pair has no Q15 run available. The attempt count comes from the float64 controller and the price of an attempt comes from the Q15 model. A Q15 run of the same pair would reject more attempts than the float run does, so the attempt count is the optimistic half of the figure and the per-attempt price is the pessimistic half

What these numbers are not

Every measured point (56 points)

One table per job and one row per point, from its newest ledger line. A point counts as measured only under the code hash that measured it, so an older line from earlier code is left out; that rule and the retry policy are in the measurement ledger. Plan, job catalogue and invariants: docs/SIDETRACK-AUTOMATION.md in the harness repository. Designs these feed: docs/EPOCH2-DESIGN.md and docs/EPOCH3-DESIGN.md.

adaptive.controller_gains

16 points, from 32 ledger lines

Closes: EPOCH2-DESIGN section 4: the dyadic PI gains rest on one measurement, and the classical alpha for a 3(2) pair is 1/3 rather than 1/4

Arithmetic: float64 only; no Q15 effects are included

pointcyclepointsfinishedmax_errorrejection_ratetotal_acceptedtotal_fevalstotal_rejected
a1o2_b0o12520332.1067e-050.0130937711495
a1o2_b1o162520331.94951e-050.0129538111615
a1o2_b1o42520331.55843e-050.15748428152780
a1o2_b1o82520331.82134e-050.0126938911855
a1o4_b0o12520331.51356e-050.0166341412667
a1o4_b1o162520331.23288e-050.0156244113477
a1o4_b1o4252030n/an/an/an/an/a
a1o4_b1o82520338.24192e-060.0138649815187
a1o8_b0o12520338.18309e-060.02534500154213
a1o8_b1o162520332.7001e-060.01793712217813
a1o8_b1o4252030n/an/an/an/an/a
a1o8_b1o8252030n/an/an/an/an/a
a3o8_b0o12520331.87781e-050.0101838911824
a3o8_b1o162520331.69917e-050.0099339912124
a3o8_b1o42520338.4292e-060.0079749815094
a3o8_b1o82520331.51866e-050.0095741412574

adaptive.pair_census

8 points

Closes: EPOCH2-DESIGN section 2: the search-space note assumes a dyadic A admits both an order-p b and an order-(p-1) b_hat with a free parameter left, and that has not been counted

Arithmetic: exact over Fractions; no float and no Q15 arithmetic is involved

pointcycleb_hat_exactb_hat_freeboth_solvabled_forced_zerofsalmatricesspace_sizestatus
st3_s1252064640006464ok
st3_s22520384512400512512ok
st3_s325202048409660040964096ok
st3_s42520102403276824003276832768ok
st4_s125204096409636460040964096ok
st4_s22520n/an/an/an/an/a0262144capped
st4_s32520n/an/an/an/an/a016777216capped
st4_s42520n/an/an/an/an/a01073741824capped

adaptive.q15_estimate_floor

24 points

Closes: EPOCH2-DESIGN section 3: the roughly 2 LSB floor-bias figure for the Q15 error estimate is an estimate, and section 5 needs the scored tolerance ladder to sit above whatever that floor turns out to be

Arithmetic: Q15 fixed point throughout, on the pinned primitives in rk_harness.fixedpoint, with floor multiply; the reference float mode used to reproduce the frozen curve is float64

the bit scan is booked at its worst case and its branches cannot be priced by the pinned cost model, so the sequence figure is an upper bound rather than an exact count. The factor table is held in Q14 because a Q15 table cannot represent a factor of 1.0 at all; the encoding is an open item in EPOCH2-DESIGN section 4.

pointcyclepointsfinishedstatusesflattening_tol_lsbmean_bias_lsb_pooledmean_bias_lsb_totaloverflow_rejectedtables_agree
battery_2rc_s12520108attempt_cap, ok512-0.00486615-2.035640false
battery_2rc_s22520108attempt_cap, ok512-0.00466504-1.895320false
battery_2rc_s42520108attempt_cap, ok512-0.0126933-1.9430false
bicycle_lateral_s12520108attempt_cap, ok256-0.0168511-1.594510false
bicycle_lateral_s22520108attempt_cap, ok512-0.00646408-1.289540false
bicycle_lateral_s42520108attempt_cap, ok512-0.00734325-1.228970false
buck_converter_s12520108attempt_cap, ok512-0.0504386-1.81190false
buck_converter_s22520108attempt_cap, ok512-0.0101078-1.840620false
buck_converter_s42520108attempt_cap, ok512-0.016269-1.813960false
enzyme_qssa_s12540107attempt_cap, ok, step_underflow256-0.200438-1.8909135false
enzyme_qssa_s22540107attempt_cap, ok, step_underflow512-0.12472-1.84144136false
enzyme_qssa_s42540107attempt_cap, ok, step_underflow512-0.0850521-1.81784140false
glucose_minimal_s12520107attempt_cap, ok, step_underflow512-0.0394501-1.917370false
glucose_minimal_s22520107attempt_cap, ok, step_underflow512-0.0237129-1.938910false
glucose_minimal_s42540107attempt_cap, ok, step_underflow512-0.0134941-1.904170false
pll_lock_s12520107attempt_cap, ok, overflow512-0.104358-1.947281false
pll_lock_s22520107attempt_cap, ok512-0.00373301-1.895910false
pll_lock_s42520108attempt_cap, ok512-0.00638674-1.820130false
robertson_scaled_s12540107attempt_cap, ok, step_underflow128-0.166833-1.8887110false
robertson_scaled_s22540106attempt_cap, ok, overflow, step_underflow64-0.112156-1.8856520false
robertson_scaled_s42540105attempt_cap, ok, overflow, step_underflow32-0.0732931-1.757532false
servo_load_step_s12540107ok, overflow, step_underflow256-1.58357-1.727873false
servo_load_step_s22540107ok, overflow, step_underflown/a-1.52041-1.591913false
servo_load_step_s42540107ok, overflow, step_underflown/a-1.45251-1.481714false

adaptive.suite_sweep

8 points, from 16 ledger lines

Closes: EPOCH2-DESIGN section 5: the scored tolerance ladder is unchosen, and the published curve covers 3 of the 8 validation problems

Arithmetic: float64 only; no Q15 effects are included

pointcyclepointsfinishedstatusesfevals_maxfevals_minrejection_rate_max
battery_2rc252066ok421280.0151515
bicycle_lateral252066ok1195430.0434783
buck_converter252066ok3571910.0166667
enzyme_qssa252066ok330111620.198905
glucose_minimal252066ok1675490.0333333
pll_lock252066ok1603520.0344828
robertson_scaled252066ok388323530.0425039
servo_load_step252066ok314530040.00667939
Further evidence, not published here
documentstatewhat it holds
rk-work/validation/axes.jsonnot written yetthe cycles-to-tolerance rows for this class, and the measured floor of the Q15 error estimate that bounds how tight a tolerance can mean anything
rk-work/adaptive_archive/YYYY-MM-DD.jsonlwrittenthe lane's per-day search log: one line per candidate the lane enumerated, ranked against nothing
rk-work/schedule/shares.jsonwrittenthe per-cycle lane log summarised per method class

The traceability rule lists key_findings.json, validation/results.json, benchmark/results.json, the side-track ledger with its artifacts, and the two lane elites documents, adaptive_archive/elites.json and implicit_archive/elites.json. Each document named above sits outside that list and says so in its own schema, so this page names it and reports whether it exists without publishing a number from it. Only the owner can add one to the list.