Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
Deviation card interpolation: Headings before the first card entry are not shifted · case 01
A card starting at 015 returns nothing for headings 000-014.
Deviation card interpolation: Headings before the first card entry are not shifted · case 02
A card starting at 015 returns nothing for headings 000-014.
Deviation card interpolation: Headings before the first card entry are not shifted · case 03
A card starting at 015 returns nothing for headings 000-014.
Deviation card interpolation: Headings before the first card entry are not shifted · case 04
A card starting at 015 returns nothing for headings 000-014.
Deviation card interpolation: Headings before the first card entry are not shifted · case 05
A card starting at 015 returns nothing for headings 000-014.
Deviation card interpolation: Unsorted cards are interpolated in entry order · case 01
Cards written clockwise from another heading give wrong deviations.
Deviation card interpolation: Unsorted cards are interpolated in entry order · case 02
Cards written clockwise from another heading give wrong deviations.
Deviation card interpolation: Unsorted cards are interpolated in entry order · case 03
Cards written clockwise from another heading give wrong deviations.
Deviation card interpolation: Unsorted cards are interpolated in entry order · case 04
Cards written clockwise from another heading give wrong deviations.
Deviation card interpolation: Unsorted cards are interpolated in entry order · case 05
Cards written clockwise from another heading give wrong deviations.
Deviation card interpolation: Interpolation assumes a 45 degree card spacing · case 01
Irregular cards give deviations outside the neighbouring values.
Deviation card interpolation: Interpolation assumes a 45 degree card spacing · case 02
Irregular cards give deviations outside the neighbouring values.
Deviation card interpolation: Interpolation assumes a 45 degree card spacing · case 03
Irregular cards give deviations outside the neighbouring values.
Deviation card interpolation: Interpolation assumes a 45 degree card spacing · case 04
Irregular cards give deviations outside the neighbouring values.
Deviation card interpolation: Interpolation assumes a 45 degree card spacing · case 05
Irregular cards give deviations outside the neighbouring values.
Anchor rode and swing radius: Rode length uses the lowest tide · case 01
At high water the anchor has too little scope and drags.
Anchor rode and swing radius: Rode length uses the lowest tide · case 02
At high water the anchor has too little scope and drags.
Anchor rode and swing radius: Rode length uses the lowest tide · case 03
At high water the anchor has too little scope and drags.
Anchor rode and swing radius: Rode length uses the lowest tide · case 04
At high water the anchor has too little scope and drags.
Anchor rode and swing radius: Rode length uses the lowest tide · case 05
At high water the anchor has too little scope and drags.
Anchor rode and swing radius: Bow roller height is left out of the scope · case 01
Scope ratios are computed on water depth only.
Anchor rode and swing radius: Bow roller height is left out of the scope · case 02
Scope ratios are computed on water depth only.
Anchor rode and swing radius: Bow roller height is left out of the scope · case 03
Scope ratios are computed on water depth only.
Anchor rode and swing radius: Bow roller height is left out of the scope · case 04
Scope ratios are computed on water depth only.
Anchor rode and swing radius: Bow roller height is left out of the scope · case 05
Scope ratios are computed on water depth only.
Anchor rode and swing radius: Rode rounds to the nearest chain mark · case 01
Rode can be shorter than the required scope.
Anchor rode and swing radius: Rode rounds to the nearest chain mark · case 02
Rode can be shorter than the required scope.
Anchor rode and swing radius: Rode rounds to the nearest chain mark · case 03
Rode can be shorter than the required scope.
Anchor rode and swing radius: Rode rounds to the nearest chain mark · case 04
Rode can be shorter than the required scope.
Anchor rode and swing radius: Rode rounds to the nearest chain mark · case 05
Rode can be shorter than the required scope.
Anchor rode and swing radius: Swing radius uses the high-water depth · case 01
The swing circle is underestimated and the boat can touch neighbours at low water.
Anchor rode and swing radius: Swing radius uses the high-water depth · case 02
The swing circle is underestimated and the boat can touch neighbours at low water.
Anchor rode and swing radius: Swing radius uses the high-water depth · case 03
The swing circle is underestimated and the boat can touch neighbours at low water.
Anchor rode and swing radius: Swing radius uses the high-water depth · case 04
The swing circle is underestimated and the boat can touch neighbours at low water.
Anchor rode and swing radius: Swing radius uses the high-water depth · case 05
The swing circle is underestimated and the boat can touch neighbours at low water.
Anchor rode and swing radius: Swing radius stops at the bow · case 01
The stern swings outside the planned circle.
Anchor rode and swing radius: Swing radius stops at the bow · case 02
The stern swings outside the planned circle.
Anchor rode and swing radius: Swing radius stops at the bow · case 03
The stern swings outside the planned circle.
Anchor rode and swing radius: Swing radius stops at the bow · case 04
The stern swings outside the planned circle.
Anchor rode and swing radius: Swing radius stops at the bow · case 05
The stern swings outside the planned circle.
Elliptic Kepler solver: Newton correction is added instead of subtracted · case 01
The iteration walks away from the root and returns a huge anomaly.
Elliptic Kepler solver: Newton correction is added instead of subtracted · case 02
The iteration walks away from the root and returns a huge anomaly.
Elliptic Kepler solver: Newton correction is added instead of subtracted · case 03
The iteration walks away from the root and returns a huge anomaly.
Elliptic Kepler solver: Newton correction is added instead of subtracted · case 04
The iteration walks away from the root and returns a huge anomaly.
Elliptic Kepler solver: Newton correction is added instead of subtracted · case 05
The iteration walks away from the root and returns a huge anomaly.
Elliptic Kepler solver: Mean anomaly is not reduced to one revolution · case 01
Multi-revolution or negative mean anomalies return eccentric anomalies outside [0,2pi).
Elliptic Kepler solver: Mean anomaly is not reduced to one revolution · case 02
Multi-revolution or negative mean anomalies return eccentric anomalies outside [0,2pi).
Elliptic Kepler solver: Mean anomaly is not reduced to one revolution · case 03
Multi-revolution or negative mean anomalies return eccentric anomalies outside [0,2pi).
Elliptic Kepler solver: Mean anomaly is not reduced to one revolution · case 04
Multi-revolution or negative mean anomalies return eccentric anomalies outside [0,2pi).
Elliptic Kepler solver: Mean anomaly is not reduced to one revolution · case 05
Multi-revolution or negative mean anomalies return eccentric anomalies outside [0,2pi).
Elliptic Kepler solver: Newton loop stops at a millimetre-scale tolerance · case 01
Eccentric anomalies are only good to about three decimals.
Elliptic Kepler solver: Newton loop stops at a millimetre-scale tolerance · case 02
Eccentric anomalies are only good to about three decimals.
Elliptic Kepler solver: Newton loop stops at a millimetre-scale tolerance · case 03
Eccentric anomalies are only good to about three decimals.
Elliptic Kepler solver: Newton loop stops at a millimetre-scale tolerance · case 04
Eccentric anomalies are only good to about three decimals.
Elliptic Kepler solver: Newton loop stops at a millimetre-scale tolerance · case 05
Eccentric anomalies are only good to about three decimals.
Elliptic Kepler solver: Iteration cap is too small for near-parabolic orbits · case 01
Highly eccentric orbits return unconverged anomalies.
Elliptic Kepler solver: Iteration cap is too small for near-parabolic orbits · case 02
Highly eccentric orbits return unconverged anomalies.
Elliptic Kepler solver: Iteration cap is too small for near-parabolic orbits · case 03
Highly eccentric orbits return unconverged anomalies.
Elliptic Kepler solver: Iteration cap is too small for near-parabolic orbits · case 04
Highly eccentric orbits return unconverged anomalies.
Elliptic Kepler solver: Iteration cap is too small for near-parabolic orbits · case 05
Highly eccentric orbits return unconverged anomalies.
Elliptic Kepler solver: Parabolic eccentricity is accepted by the elliptic solver · case 01
e=1 returns a meaningless eccentric anomaly.
Elliptic Kepler solver: Parabolic eccentricity is accepted by the elliptic solver · case 02
e=1 returns a meaningless eccentric anomaly.
Elliptic Kepler solver: Parabolic eccentricity is accepted by the elliptic solver · case 03
e=1 returns a meaningless eccentric anomaly.
Elliptic Kepler solver: Parabolic eccentricity is accepted by the elliptic solver · case 04
e=1 returns a meaningless eccentric anomaly.
Elliptic Kepler solver: Parabolic eccentricity is accepted by the elliptic solver · case 05
e=1 returns a meaningless eccentric anomaly.
True anomaly from eccentric anomaly: Cosine formula loses the descending half of the orbit · case 01
After apoapsis the true anomaly folds back instead of continuing past pi.
True anomaly from eccentric anomaly: Cosine formula loses the descending half of the orbit · case 02
After apoapsis the true anomaly folds back instead of continuing past pi.
True anomaly from eccentric anomaly: Cosine formula loses the descending half of the orbit · case 03
After apoapsis the true anomaly folds back instead of continuing past pi.
True anomaly from eccentric anomaly: Cosine formula loses the descending half of the orbit · case 04
After apoapsis the true anomaly folds back instead of continuing past pi.
True anomaly from eccentric anomaly: Cosine formula loses the descending half of the orbit · case 05
After apoapsis the true anomaly folds back instead of continuing past pi.
True anomaly from eccentric anomaly: Half-angle factors are swapped · case 01
The spacecraft appears to linger near periapsis instead of apoapsis.
True anomaly from eccentric anomaly: Half-angle factors are swapped · case 02
The spacecraft appears to linger near periapsis instead of apoapsis.
True anomaly from eccentric anomaly: Half-angle factors are swapped · case 03
The spacecraft appears to linger near periapsis instead of apoapsis.
True anomaly from eccentric anomaly: Half-angle factors are swapped · case 04
The spacecraft appears to linger near periapsis instead of apoapsis.
True anomaly from eccentric anomaly: Half-angle factors are swapped · case 05
The spacecraft appears to linger near periapsis instead of apoapsis.
True anomaly from eccentric anomaly: Full angles are used in the half-angle formula · case 01
True anomaly runs twice as fast as the orbit.
True anomaly from eccentric anomaly: Full angles are used in the half-angle formula · case 02
True anomaly runs twice as fast as the orbit.
True anomaly from eccentric anomaly: Full angles are used in the half-angle formula · case 03
True anomaly runs twice as fast as the orbit.
True anomaly from eccentric anomaly: Full angles are used in the half-angle formula · case 04
True anomaly runs twice as fast as the orbit.
True anomaly from eccentric anomaly: Full angles are used in the half-angle formula · case 05
True anomaly runs twice as fast as the orbit.
True anomaly from eccentric anomaly: True anomaly is left in (-pi, pi] · case 01
Descending-branch anomalies are reported as negative angles.
True anomaly from eccentric anomaly: True anomaly is left in (-pi, pi] · case 02
Descending-branch anomalies are reported as negative angles.
True anomaly from eccentric anomaly: True anomaly is left in (-pi, pi] · case 03
Descending-branch anomalies are reported as negative angles.
True anomaly from eccentric anomaly: True anomaly is left in (-pi, pi] · case 04
Descending-branch anomalies are reported as negative angles.
True anomaly from eccentric anomaly: True anomaly is left in (-pi, pi] · case 05
Descending-branch anomalies are reported as negative angles.
True anomaly from eccentric anomaly: Radius ratio uses the wrong sign on the eccentric term · case 01
The spacecraft is placed at apoapsis distance at periapsis.
True anomaly from eccentric anomaly: Radius ratio uses the wrong sign on the eccentric term · case 02
The spacecraft is placed at apoapsis distance at periapsis.
True anomaly from eccentric anomaly: Radius ratio uses the wrong sign on the eccentric term · case 03
The spacecraft is placed at apoapsis distance at periapsis.
True anomaly from eccentric anomaly: Radius ratio uses the wrong sign on the eccentric term · case 04
The spacecraft is placed at apoapsis distance at periapsis.
True anomaly from eccentric anomaly: Radius ratio uses the wrong sign on the eccentric term · case 05
The spacecraft is placed at apoapsis distance at periapsis.
Mean anomaly propagation between epochs: Epoch difference ignores leap days · case 01
Propagation across 29 February drifts by a whole day of motion.
Mean anomaly propagation between epochs: Epoch difference ignores leap days · case 02
Propagation across 29 February drifts by a whole day of motion.
Mean anomaly propagation between epochs: Epoch difference ignores leap days · case 03
Propagation across 29 February drifts by a whole day of motion.
Mean anomaly propagation between epochs: Epoch difference ignores leap days · case 04
Propagation across 29 February drifts by a whole day of motion.
Mean anomaly propagation between epochs: Epoch difference ignores leap days · case 05
Propagation across 29 February drifts by a whole day of motion.
Mean anomaly propagation between epochs: Revolutions are scaled by 2pi into a degree angle · case 01
Mean anomaly advances by radians while labelled degrees.
Mean anomaly propagation between epochs: Revolutions are scaled by 2pi into a degree angle · case 02
Mean anomaly advances by radians while labelled degrees.
Mean anomaly propagation between epochs: Revolutions are scaled by 2pi into a degree angle · case 03
Mean anomaly advances by radians while labelled degrees.
Mean anomaly propagation between epochs: Revolutions are scaled by 2pi into a degree angle · case 04
Mean anomaly advances by radians while labelled degrees.
Mean anomaly propagation between epochs: Revolutions are scaled by 2pi into a degree angle · case 05
Mean anomaly advances by radians while labelled degrees.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗